01_neural_network_regression_with_tensorflow.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "01_neural_network_regression_with_tensorflow.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyPz98rxwXhaiQJchd9eFIrY",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/ngcheeyuan/CDC-Voucher-Merchant-Web-App/blob/main/01_neural_network_regression_with_tensorflow.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#Introduction to regression with neural network in tensorflow"
      ],
      "metadata": {
        "id": "pEHr5t7IqKmP"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "oMQtoivzmZP8",
        "outputId": "15b3fcdd-ee9b-4637-fe91-130775bbf779"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "2.7.0\n"
          ]
        }
      ],
      "source": [
        "import tensorflow as tf\n",
        "print(tf.__version__)"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Creating data to fit"
      ],
      "metadata": {
        "id": "2M6eOvlDqZ27"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "import matplotlib.pyplot as plt"
      ],
      "metadata": {
        "id": "WZQ1dNk6qTpL"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X = np.arange(-100,100,3,dtype = float)\n",
        "y = X+10"
      ],
      "metadata": {
        "id": "vvorxCwTqZvG"
      },
      "execution_count": 42,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X = tf.constant(X, dtype = tf.float16)\n",
        "y = tf.constant(y, dtype = tf.float16)"
      ],
      "metadata": {
        "id": "LTk2RzApv4k8"
      },
      "execution_count": 45,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plt.scatter(X,y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 286
        },
        "id": "WJKtC6-UqX0K",
        "outputId": "eb6a550d-785f-417a-e07b-149be5e17275"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.collections.PathCollection at 0x7fef5fd18b90>"
            ]
          },
          "metadata": {},
          "execution_count": 8
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "y == X + 10"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "WZZ31sj9qz0j",
        "outputId": "b4a36feb-9416-47ed-fb34-e1998484afe6"
      },
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tf.Tensor: shape=(67,), dtype=bool, numpy=\n",
              "array([ True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
              "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
              "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
              "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
              "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
              "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
              "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
              "        True,  True,  True,  True])>"
            ]
          },
          "metadata": {},
          "execution_count": 46
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Input and output shapes"
      ],
      "metadata": {
        "id": "8Xwld6hCq7d2"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#Create a demo tensor for our housing price prediction problems\n",
        "house_info = tf.constant(['bedroom','bathroom','garage'])\n",
        "house_price = tf.constant([9512312])\n",
        "house_info,house_price"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_khuPZmSrAcN",
        "outputId": "fc43231d-30f3-45e6-ee44-2196bf6e31b1"
      },
      "execution_count": 47,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(<tf.Tensor: shape=(3,), dtype=string, numpy=array([b'bedroom', b'bathroom', b'garage'], dtype=object)>,\n",
              " <tf.Tensor: shape=(1,), dtype=int32, numpy=array([9512312], dtype=int32)>)"
            ]
          },
          "metadata": {},
          "execution_count": 47
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "input_shape = X.shape\n",
        "output_shape = y.shape\n",
        "input_shape , output_shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "N0I1sNpmrOnw",
        "outputId": "37042e66-60bf-4dce-b665-2a1010d6c9c2"
      },
      "execution_count": 48,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(TensorShape([67]), TensorShape([67]))"
            ]
          },
          "metadata": {},
          "execution_count": 48
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X[0].ndim"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PqvDXxV2rVwo",
        "outputId": "375ccfca-dec8-4891-ae51-75847d57d915"
      },
      "execution_count": 49,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0"
            ]
          },
          "metadata": {},
          "execution_count": 49
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X[0],y[0]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "U2xtyOb3ri06",
        "outputId": "96f8ebc6-0b3d-4c91-d030-c2a091e54801"
      },
      "execution_count": 50,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(<tf.Tensor: shape=(), dtype=float16, numpy=-100.0>,\n",
              " <tf.Tensor: shape=(), dtype=float16, numpy=-90.0>)"
            ]
          },
          "metadata": {},
          "execution_count": 50
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X = tf.constant(X)\n",
        "y = tf.constant(y)\n",
        "X,y"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KfndgNhbsCct",
        "outputId": "3f81b842-7938-48a1-c89c-c243fee0a7af"
      },
      "execution_count": 51,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(<tf.Tensor: shape=(67,), dtype=float16, numpy=\n",
              " array([-100.,  -97.,  -94.,  -91.,  -88.,  -85.,  -82.,  -79.,  -76.,\n",
              "         -73.,  -70.,  -67.,  -64.,  -61.,  -58.,  -55.,  -52.,  -49.,\n",
              "         -46.,  -43.,  -40.,  -37.,  -34.,  -31.,  -28.,  -25.,  -22.,\n",
              "         -19.,  -16.,  -13.,  -10.,   -7.,   -4.,   -1.,    2.,    5.,\n",
              "           8.,   11.,   14.,   17.,   20.,   23.,   26.,   29.,   32.,\n",
              "          35.,   38.,   41.,   44.,   47.,   50.,   53.,   56.,   59.,\n",
              "          62.,   65.,   68.,   71.,   74.,   77.,   80.,   83.,   86.,\n",
              "          89.,   92.,   95.,   98.], dtype=float16)>,\n",
              " <tf.Tensor: shape=(67,), dtype=float16, numpy=\n",
              " array([-90., -87., -84., -81., -78., -75., -72., -69., -66., -63., -60.,\n",
              "        -57., -54., -51., -48., -45., -42., -39., -36., -33., -30., -27.,\n",
              "        -24., -21., -18., -15., -12.,  -9.,  -6.,  -3.,   0.,   3.,   6.,\n",
              "          9.,  12.,  15.,  18.,  21.,  24.,  27.,  30.,  33.,  36.,  39.,\n",
              "         42.,  45.,  48.,  51.,  54.,  57.,  60.,  63.,  66.,  69.,  72.,\n",
              "         75.,  78.,  81.,  84.,  87.,  90.,  93.,  96.,  99., 102., 105.,\n",
              "        108.], dtype=float16)>)"
            ]
          },
          "metadata": {},
          "execution_count": 51
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "input_shape = X[0].shape\n",
        "output_shape = y[0].shape\n",
        "input_shape , output_shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MgRCZkOJsIwe",
        "outputId": "676499c2-d4b2-40dc-84f9-4bff850f467f"
      },
      "execution_count": 52,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(TensorShape([]), TensorShape([]))"
            ]
          },
          "metadata": {},
          "execution_count": 52
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# ## OLD\n",
        "# # Fit the model\n",
        "# model.fit(X, y, epochs=5) # this will break with TensorFlow 2.7.0+\n",
        " \n",
        "# ## New\n",
        "# # Fit the model\n",
        "# model.fit(tf.expand_dims(X, axis=-1), y, epochs=5) # <- updated line\n",
        "# The code adds an extra dimension to X on the last axis, turning it ndim=1 to ndim=2 (what the model requires).\n",
        "\n",
        "# To see a full explanation of this refer to the following:"
      ],
      "metadata": {
        "id": "H1LNpa6Mrj2a"
      },
      "execution_count": 53,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Steps in modelling with tensorflow.\n",
        "1. Create a model - define the input and output layer , as well as the hidden layers\n",
        "2. Compiing a model - define the loss function and the optimizer and evaluation metrics\n",
        "3. Fitting a model - letting the model try to find pattersn between X & y"
      ],
      "metadata": {
        "id": "y2V1LWcBsLj7"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# set random seed\n",
        "tf.random.set_seed(42)\n",
        "#1. Create a model using the Sequential API\n",
        "model = tf.keras.Sequential()\n",
        "model.add(tf.keras.layers.Dense(1))\n",
        "model.compile(loss = tf.keras.losses.mae,\n",
        "              optimizer = tf.keras.optimizers.SGD(lr = 0.0001),\n",
        "              metrics = ['mae'])\n",
        "\n",
        "model.fit(tf.expand_dims(X, axis=-1), y, epochs=100)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "aPbmpKd3sb1m",
        "outputId": "0fbfc33e-5b69-4168-b7ed-591dbcdce8de"
      },
      "execution_count": 61,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/keras/optimizer_v2/gradient_descent.py:102: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead.\n",
            "  super(SGD, self).__init__(name, **kwargs)\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 22.5374 - mae: 22.5374\n",
            "Epoch 2/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 21.8308 - mae: 21.8308\n",
            "Epoch 3/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 21.2423 - mae: 21.2423\n",
            "Epoch 4/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 20.4671 - mae: 20.4671\n",
            "Epoch 5/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 19.8703 - mae: 19.8703\n",
            "Epoch 6/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 19.2563 - mae: 19.2563\n",
            "Epoch 7/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 18.4696 - mae: 18.4696\n",
            "Epoch 8/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 17.7081 - mae: 17.7081\n",
            "Epoch 9/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 17.0610 - mae: 17.0610\n",
            "Epoch 10/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 16.4374 - mae: 16.4374\n",
            "Epoch 11/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 15.8376 - mae: 15.8376\n",
            "Epoch 12/100\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 15.2171 - mae: 15.2171\n",
            "Epoch 13/100\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 14.5658 - mae: 14.5658\n",
            "Epoch 14/100\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 14.0465 - mae: 14.0465\n",
            "Epoch 15/100\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 13.6059 - mae: 13.6059\n",
            "Epoch 16/100\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 13.0795 - mae: 13.0795\n",
            "Epoch 17/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 12.6653 - mae: 12.6653\n",
            "Epoch 18/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 12.0812 - mae: 12.0812\n",
            "Epoch 19/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 11.6296 - mae: 11.6296\n",
            "Epoch 20/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 11.2105 - mae: 11.2105\n",
            "Epoch 21/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 10.9886 - mae: 10.9886\n",
            "Epoch 22/100\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 10.6548 - mae: 10.6548\n",
            "Epoch 23/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 10.6218 - mae: 10.6218\n",
            "Epoch 24/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 10.5483 - mae: 10.5483\n",
            "Epoch 25/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 10.3659 - mae: 10.3659\n",
            "Epoch 26/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 10.2132 - mae: 10.2132\n",
            "Epoch 27/100\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 10.1024 - mae: 10.1024\n",
            "Epoch 28/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 10.0418 - mae: 10.0418\n",
            "Epoch 29/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 9.9379 - mae: 9.9379\n",
            "Epoch 30/100\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 9.9417 - mae: 9.9417\n",
            "Epoch 31/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.9335 - mae: 9.9335\n",
            "Epoch 32/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.9252 - mae: 9.9252\n",
            "Epoch 33/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 9.8990 - mae: 9.8990\n",
            "Epoch 34/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.9055 - mae: 9.9055\n",
            "Epoch 35/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.8976 - mae: 9.8976\n",
            "Epoch 36/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.9142 - mae: 9.9142\n",
            "Epoch 37/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.8953 - mae: 9.8953\n",
            "Epoch 38/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 9.8973 - mae: 9.8973\n",
            "Epoch 39/100\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 9.8998 - mae: 9.8998\n",
            "Epoch 40/100\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 9.9030 - mae: 9.9030\n",
            "Epoch 41/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 9.8941 - mae: 9.8941\n",
            "Epoch 42/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 9.8931 - mae: 9.8931\n",
            "Epoch 43/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 9.8975 - mae: 9.8975\n",
            "Epoch 44/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 9.8979 - mae: 9.8979\n",
            "Epoch 45/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.8927 - mae: 9.8927\n",
            "Epoch 46/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.8930 - mae: 9.8930\n",
            "Epoch 47/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.8964 - mae: 9.8964\n",
            "Epoch 48/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 9.8937 - mae: 9.8937\n",
            "Epoch 49/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 9.8910 - mae: 9.8910\n",
            "Epoch 50/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.9037 - mae: 9.9037\n",
            "Epoch 51/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.8914 - mae: 9.8914\n",
            "Epoch 52/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.8945 - mae: 9.8945\n",
            "Epoch 53/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.8925 - mae: 9.8925\n",
            "Epoch 54/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.8970 - mae: 9.8970\n",
            "Epoch 55/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 9.8949 - mae: 9.8949\n",
            "Epoch 56/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.8985 - mae: 9.8985\n",
            "Epoch 57/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 9.8983 - mae: 9.8983\n",
            "Epoch 58/100\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 9.8951 - mae: 9.8951\n",
            "Epoch 59/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 9.9010 - mae: 9.9010\n",
            "Epoch 60/100\n",
            "3/3 [==============================] - 0s 3ms/step - loss: 9.9006 - mae: 9.9006\n",
            "Epoch 61/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.9169 - mae: 9.9169\n",
            "Epoch 62/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.9073 - mae: 9.9073\n",
            "Epoch 63/100\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 9.9125 - mae: 9.9125\n",
            "Epoch 64/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 9.8996 - mae: 9.8996\n",
            "Epoch 65/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.9014 - mae: 9.9014\n",
            "Epoch 66/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.9024 - mae: 9.9024\n",
            "Epoch 67/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.9079 - mae: 9.9079\n",
            "Epoch 68/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.9056 - mae: 9.9056\n",
            "Epoch 69/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.9065 - mae: 9.9065\n",
            "Epoch 70/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.9065 - mae: 9.9065\n",
            "Epoch 71/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.9106 - mae: 9.9106\n",
            "Epoch 72/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 9.9148 - mae: 9.9148\n",
            "Epoch 73/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 9.9059 - mae: 9.9059\n",
            "Epoch 74/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 9.9040 - mae: 9.9040\n",
            "Epoch 75/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 9.8997 - mae: 9.8997\n",
            "Epoch 76/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 9.9021 - mae: 9.9021\n",
            "Epoch 77/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.9018 - mae: 9.9018\n",
            "Epoch 78/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.9006 - mae: 9.9006\n",
            "Epoch 79/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 9.9094 - mae: 9.9094\n",
            "Epoch 80/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.9057 - mae: 9.9057\n",
            "Epoch 81/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.9058 - mae: 9.9058\n",
            "Epoch 82/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 9.9055 - mae: 9.9055\n",
            "Epoch 83/100\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 9.9196 - mae: 9.9196\n",
            "Epoch 84/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.9062 - mae: 9.9062\n",
            "Epoch 85/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.9001 - mae: 9.9001\n",
            "Epoch 86/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.9003 - mae: 9.9003\n",
            "Epoch 87/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 9.9084 - mae: 9.9084\n",
            "Epoch 88/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.8968 - mae: 9.8968\n",
            "Epoch 89/100\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 9.8935 - mae: 9.8935\n",
            "Epoch 90/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.8943 - mae: 9.8943\n",
            "Epoch 91/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.8906 - mae: 9.8906\n",
            "Epoch 92/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.8955 - mae: 9.8955\n",
            "Epoch 93/100\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 9.8917 - mae: 9.8917\n",
            "Epoch 94/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.8931 - mae: 9.8931\n",
            "Epoch 95/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.8838 - mae: 9.8838\n",
            "Epoch 96/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 9.8827 - mae: 9.8827\n",
            "Epoch 97/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 9.8873 - mae: 9.8873\n",
            "Epoch 98/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 9.8934 - mae: 9.8934\n",
            "Epoch 99/100\n",
            "3/3 [==============================] - 0s 3ms/step - loss: 9.8867 - mae: 9.8867\n",
            "Epoch 100/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 9.8849 - mae: 9.8849\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7fef5a448350>"
            ]
          },
          "metadata": {},
          "execution_count": 61
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X,y"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gKD4rBw7tIBN",
        "outputId": "fc1fb555-4e53-4b44-b917-e77582bde23e"
      },
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(<tf.Tensor: shape=(8,), dtype=float64, numpy=array([-7., -4., -1.,  2.,  5.,  8., 11., 14.])>,\n",
              " <tf.Tensor: shape=(8,), dtype=float64, numpy=array([ 3.,  6.,  9., 12., 15., 18., 21., 24.])>)"
            ]
          },
          "metadata": {},
          "execution_count": 34
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "y_pred = model.predict([17])\n",
        "y_pred"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "l_E8x3y-ubo9",
        "outputId": "1121e82d-120e-4eb8-f52b-21fdee777216"
      },
      "execution_count": 55,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[16.543104]], dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 55
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Improving our model\n",
        "\n",
        "We can improve our model, by altering the steps we took to create a model.\n",
        "1. Creating a model -We might add more layers, increase the number of hidden units within each of the hidden layers, change the activation function of each layer.\n",
        "\n",
        "2. Compiling a model - we might change the optimization function of the learning rate of the optimization function\n",
        "\n",
        "3. Fitting a model - here we might fit a model for more epochs, leaving it training for longer) or on more data (giving the model more examples to learn from)"
      ],
      "metadata": {
        "id": "tafj-jD7ueoY"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model = tf.keras.Sequential([tf.keras.layers.Dense(1)])\n",
        "model.compile(loss =tf.keras.losses.mae,\n",
        "              optimizer = tf.keras.optimizers.SGD(lr = 0.01),\n",
        "              metrics = ['mae']\n",
        "              )\n",
        "\n",
        "model.fit(tf.expand_dims(X, axis=-1), y, epochs=100)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KNq88OXwutID",
        "outputId": "19a705f2-354f-4d6c-bf19-dbf1ff701ac5"
      },
      "execution_count": 78,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/keras/optimizer_v2/gradient_descent.py:102: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead.\n",
            "  super(SGD, self).__init__(name, **kwargs)\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 25.5596 - mae: 25.5596\n",
            "Epoch 2/100\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 12.1583 - mae: 12.1583\n",
            "Epoch 3/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 10.8061 - mae: 10.8061\n",
            "Epoch 4/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 11.5632 - mae: 11.5632\n",
            "Epoch 5/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.9675 - mae: 9.9675\n",
            "Epoch 6/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 9.9103 - mae: 9.9103\n",
            "Epoch 7/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 10.2118 - mae: 10.2118\n",
            "Epoch 8/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 20.2324 - mae: 20.2324\n",
            "Epoch 9/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 18.2016 - mae: 18.2016\n",
            "Epoch 10/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 12.2472 - mae: 12.2472\n",
            "Epoch 11/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 14.5281 - mae: 14.5281\n",
            "Epoch 12/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 15.9454 - mae: 15.9454\n",
            "Epoch 13/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 20.2656 - mae: 20.2656\n",
            "Epoch 14/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 16.9827 - mae: 16.9827\n",
            "Epoch 15/100\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 10.0472 - mae: 10.0472\n",
            "Epoch 16/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 10.3309 - mae: 10.3309\n",
            "Epoch 17/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 11.7562 - mae: 11.7562\n",
            "Epoch 18/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 20.1025 - mae: 20.1025\n",
            "Epoch 19/100\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 13.0249 - mae: 13.0249\n",
            "Epoch 20/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 19.2698 - mae: 19.2698\n",
            "Epoch 21/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 10.9298 - mae: 10.9298\n",
            "Epoch 22/100\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 9.7434 - mae: 9.7434\n",
            "Epoch 23/100\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 16.3032 - mae: 16.3032\n",
            "Epoch 24/100\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 15.6600 - mae: 15.6600\n",
            "Epoch 25/100\n",
            "3/3 [==============================] - 0s 9ms/step - loss: 14.8658 - mae: 14.8658\n",
            "Epoch 26/100\n",
            "3/3 [==============================] - 0s 10ms/step - loss: 11.7532 - mae: 11.7532\n",
            "Epoch 27/100\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 11.9614 - mae: 11.9614\n",
            "Epoch 28/100\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 11.9356 - mae: 11.9356\n",
            "Epoch 29/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 13.3822 - mae: 13.3822\n",
            "Epoch 30/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.4742 - mae: 9.4742\n",
            "Epoch 31/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 21.9680 - mae: 21.9680\n",
            "Epoch 32/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 9.6731 - mae: 9.6731\n",
            "Epoch 33/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 17.7882 - mae: 17.7882\n",
            "Epoch 34/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 11.5323 - mae: 11.5323\n",
            "Epoch 35/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 16.1924 - mae: 16.1924\n",
            "Epoch 36/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 16.1501 - mae: 16.1501\n",
            "Epoch 37/100\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 14.1256 - mae: 14.1256\n",
            "Epoch 38/100\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 9.9843 - mae: 9.9843\n",
            "Epoch 39/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 12.3995 - mae: 12.3995\n",
            "Epoch 40/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 11.9929 - mae: 11.9929\n",
            "Epoch 41/100\n",
            "3/3 [==============================] - 0s 3ms/step - loss: 15.3357 - mae: 15.3357\n",
            "Epoch 42/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 9.1575 - mae: 9.1575\n",
            "Epoch 43/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 14.9027 - mae: 14.9027\n",
            "Epoch 44/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 13.8845 - mae: 13.8845\n",
            "Epoch 45/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 9.0766 - mae: 9.0766\n",
            "Epoch 46/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.1680 - mae: 9.1680\n",
            "Epoch 47/100\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 20.9585 - mae: 20.9585\n",
            "Epoch 48/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 21.1653 - mae: 21.1653\n",
            "Epoch 49/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 11.0145 - mae: 11.0145\n",
            "Epoch 50/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 15.7910 - mae: 15.7910\n",
            "Epoch 51/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 8.9229 - mae: 8.9229\n",
            "Epoch 52/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 12.7047 - mae: 12.7047\n",
            "Epoch 53/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.2248 - mae: 9.2248\n",
            "Epoch 54/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 14.1020 - mae: 14.1020\n",
            "Epoch 55/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 12.0564 - mae: 12.0564\n",
            "Epoch 56/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 12.7270 - mae: 12.7270\n",
            "Epoch 57/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 19.5684 - mae: 19.5684\n",
            "Epoch 58/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 11.7114 - mae: 11.7114\n",
            "Epoch 59/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 8.1577 - mae: 8.1577\n",
            "Epoch 60/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 15.0795 - mae: 15.0795\n",
            "Epoch 61/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 17.6183 - mae: 17.6183\n",
            "Epoch 62/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.2998 - mae: 9.2998\n",
            "Epoch 63/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 9.2675 - mae: 9.2675\n",
            "Epoch 64/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 11.6491 - mae: 11.6491\n",
            "Epoch 65/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 19.8620 - mae: 19.8620\n",
            "Epoch 66/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 13.2366 - mae: 13.2366\n",
            "Epoch 67/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 12.1571 - mae: 12.1571\n",
            "Epoch 68/100\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 11.5674 - mae: 11.5674\n",
            "Epoch 69/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 8.9372 - mae: 8.9372\n",
            "Epoch 70/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 9.2547 - mae: 9.2547\n",
            "Epoch 71/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 12.6185 - mae: 12.6185\n",
            "Epoch 72/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 15.4166 - mae: 15.4166\n",
            "Epoch 73/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 14.3826 - mae: 14.3826\n",
            "Epoch 74/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 8.6627 - mae: 8.6627\n",
            "Epoch 75/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 14.2764 - mae: 14.2764\n",
            "Epoch 76/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 18.1280 - mae: 18.1280\n",
            "Epoch 77/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 8.4220 - mae: 8.4220\n",
            "Epoch 78/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 8.3828 - mae: 8.3828\n",
            "Epoch 79/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.3452 - mae: 9.3452\n",
            "Epoch 80/100\n",
            "3/3 [==============================] - 0s 9ms/step - loss: 19.1755 - mae: 19.1755\n",
            "Epoch 81/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 13.2935 - mae: 13.2935\n",
            "Epoch 82/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 8.3768 - mae: 8.3768\n",
            "Epoch 83/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 11.5314 - mae: 11.5314\n",
            "Epoch 84/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 14.6761 - mae: 14.6761\n",
            "Epoch 85/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 8.4408 - mae: 8.4408\n",
            "Epoch 86/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.0371 - mae: 9.0371\n",
            "Epoch 87/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 14.9529 - mae: 14.9529\n",
            "Epoch 88/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 8.7323 - mae: 8.7323\n",
            "Epoch 89/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 14.6523 - mae: 14.6523\n",
            "Epoch 90/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 23.4177 - mae: 23.4177\n",
            "Epoch 91/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 8.3662 - mae: 8.3662\n",
            "Epoch 92/100\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 17.9168 - mae: 17.9168\n",
            "Epoch 93/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 14.0646 - mae: 14.0646\n",
            "Epoch 94/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 16.0807 - mae: 16.0807\n",
            "Epoch 95/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 14.7488 - mae: 14.7488\n",
            "Epoch 96/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 9.2547 - mae: 9.2547\n",
            "Epoch 97/100\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 10.2886 - mae: 10.2886\n",
            "Epoch 98/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 8.8090 - mae: 8.8090\n",
            "Epoch 99/100\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 22.3183 - mae: 22.3183\n",
            "Epoch 100/100\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 11.0838 - mae: 11.0838\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7fef5568ac50>"
            ]
          },
          "metadata": {},
          "execution_count": 78
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model.predict([17])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GO8mm4GCw6Kj",
        "outputId": "ca0f1de5-fcae-4d87-e327-87217c8d5467"
      },
      "execution_count": 86,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[37.196354]], dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 86
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model = tf.keras.Sequential([tf.keras.layers.Dense(30,activation =None),\n",
        "                             tf.keras.layers.Dense(1)])\n",
        "\n",
        "model.compile(loss =tf.keras.losses.mae,\n",
        "              optimizer = tf.keras.optimizers.Adam(lr = 0.001),\n",
        "              metrics = ['mae']\n",
        "              )\n",
        "\n",
        "model.fit(tf.expand_dims(X, axis=-1), y, epochs=500)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hWKDYOSzxkQF",
        "outputId": "34ed1d6b-294e-439d-fe5b-2061100697df"
      },
      "execution_count": 93,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/keras/optimizer_v2/adam.py:105: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead.\n",
            "  super(Adam, self).__init__(name, **kwargs)\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 70.8430 - mae: 70.8430\n",
            "Epoch 2/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 68.8618 - mae: 68.8618\n",
            "Epoch 3/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 66.8397 - mae: 66.8397\n",
            "Epoch 4/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 64.8952 - mae: 64.8952\n",
            "Epoch 5/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 62.9196 - mae: 62.9196\n",
            "Epoch 6/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 60.9778 - mae: 60.9778\n",
            "Epoch 7/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 58.9935 - mae: 58.9935\n",
            "Epoch 8/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 56.9949 - mae: 56.9949\n",
            "Epoch 9/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 54.9971 - mae: 54.9971\n",
            "Epoch 10/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 52.9829 - mae: 52.9829\n",
            "Epoch 11/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 51.0438 - mae: 51.0438\n",
            "Epoch 12/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 49.1359 - mae: 49.1359\n",
            "Epoch 13/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 47.1519 - mae: 47.1519\n",
            "Epoch 14/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 45.2304 - mae: 45.2304\n",
            "Epoch 15/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 43.3270 - mae: 43.3270\n",
            "Epoch 16/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 41.4953 - mae: 41.4953\n",
            "Epoch 17/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 39.5765 - mae: 39.5765\n",
            "Epoch 18/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 37.7116 - mae: 37.7116\n",
            "Epoch 19/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 35.7574 - mae: 35.7574\n",
            "Epoch 20/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 33.8858 - mae: 33.8858\n",
            "Epoch 21/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 31.9712 - mae: 31.9712\n",
            "Epoch 22/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 30.1300 - mae: 30.1300\n",
            "Epoch 23/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 28.3056 - mae: 28.3056\n",
            "Epoch 24/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 26.3901 - mae: 26.3901\n",
            "Epoch 25/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 24.6643 - mae: 24.6643\n",
            "Epoch 26/500\n",
            "3/3 [==============================] - 0s 10ms/step - loss: 22.6952 - mae: 22.6952\n",
            "Epoch 27/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 20.8770 - mae: 20.8770\n",
            "Epoch 28/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 19.0252 - mae: 19.0252\n",
            "Epoch 29/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 17.1198 - mae: 17.1198\n",
            "Epoch 30/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 15.2905 - mae: 15.2905\n",
            "Epoch 31/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 13.5765 - mae: 13.5765\n",
            "Epoch 32/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 12.0942 - mae: 12.0942\n",
            "Epoch 33/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 10.8232 - mae: 10.8232\n",
            "Epoch 34/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 9.9013 - mae: 9.9013\n",
            "Epoch 35/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.6305 - mae: 9.6305\n",
            "Epoch 36/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.6169 - mae: 9.6169\n",
            "Epoch 37/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.5808 - mae: 9.5808\n",
            "Epoch 38/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 9.5570 - mae: 9.5570\n",
            "Epoch 39/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 9.5422 - mae: 9.5422\n",
            "Epoch 40/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 9.5214 - mae: 9.5214\n",
            "Epoch 41/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 9.4853 - mae: 9.4853\n",
            "Epoch 42/500\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 9.4482 - mae: 9.4482\n",
            "Epoch 43/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.4098 - mae: 9.4098\n",
            "Epoch 44/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.3695 - mae: 9.3695\n",
            "Epoch 45/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 9.3354 - mae: 9.3354\n",
            "Epoch 46/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 9.2994 - mae: 9.2994\n",
            "Epoch 47/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 9.2658 - mae: 9.2658\n",
            "Epoch 48/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 9.2305 - mae: 9.2305\n",
            "Epoch 49/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.1944 - mae: 9.1944\n",
            "Epoch 50/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 9.1631 - mae: 9.1631\n",
            "Epoch 51/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.1193 - mae: 9.1193\n",
            "Epoch 52/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 9.0910 - mae: 9.0910\n",
            "Epoch 53/500\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 9.0468 - mae: 9.0468\n",
            "Epoch 54/500\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 9.0104 - mae: 9.0104\n",
            "Epoch 55/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 8.9858 - mae: 8.9858\n",
            "Epoch 56/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 8.9451 - mae: 8.9451\n",
            "Epoch 57/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 8.9129 - mae: 8.9129\n",
            "Epoch 58/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 8.8875 - mae: 8.8875\n",
            "Epoch 59/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 8.8636 - mae: 8.8636\n",
            "Epoch 60/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 8.8287 - mae: 8.8287\n",
            "Epoch 61/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 8.8302 - mae: 8.8302\n",
            "Epoch 62/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 8.7955 - mae: 8.7955\n",
            "Epoch 63/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 8.7582 - mae: 8.7582\n",
            "Epoch 64/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 8.7312 - mae: 8.7312\n",
            "Epoch 65/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 8.6999 - mae: 8.6999\n",
            "Epoch 66/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 8.6695 - mae: 8.6695\n",
            "Epoch 67/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 8.6376 - mae: 8.6376\n",
            "Epoch 68/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 8.6067 - mae: 8.6067\n",
            "Epoch 69/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 8.5818 - mae: 8.5818\n",
            "Epoch 70/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 8.5545 - mae: 8.5545\n",
            "Epoch 71/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 8.5243 - mae: 8.5243\n",
            "Epoch 72/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 8.5258 - mae: 8.5258\n",
            "Epoch 73/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 8.4799 - mae: 8.4799\n",
            "Epoch 74/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 8.4495 - mae: 8.4495\n",
            "Epoch 75/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 8.4218 - mae: 8.4218\n",
            "Epoch 76/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 8.3846 - mae: 8.3846\n",
            "Epoch 77/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 8.3544 - mae: 8.3544\n",
            "Epoch 78/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 8.3242 - mae: 8.3242\n",
            "Epoch 79/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 8.2991 - mae: 8.2991\n",
            "Epoch 80/500\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 8.2623 - mae: 8.2623\n",
            "Epoch 81/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 8.2373 - mae: 8.2373\n",
            "Epoch 82/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 8.2134 - mae: 8.2134\n",
            "Epoch 83/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 8.2060 - mae: 8.2060\n",
            "Epoch 84/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 8.1619 - mae: 8.1619\n",
            "Epoch 85/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 8.1293 - mae: 8.1293\n",
            "Epoch 86/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 8.0893 - mae: 8.0893\n",
            "Epoch 87/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 8.0688 - mae: 8.0688\n",
            "Epoch 88/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 8.0142 - mae: 8.0142\n",
            "Epoch 89/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 7.9843 - mae: 7.9843\n",
            "Epoch 90/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 7.9510 - mae: 7.9510\n",
            "Epoch 91/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 7.9342 - mae: 7.9342\n",
            "Epoch 92/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 7.8913 - mae: 7.8912\n",
            "Epoch 93/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 7.8601 - mae: 7.8601\n",
            "Epoch 94/500\n",
            "3/3 [==============================] - 0s 3ms/step - loss: 7.8319 - mae: 7.8319\n",
            "Epoch 95/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 7.8030 - mae: 7.8030\n",
            "Epoch 96/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 7.7685 - mae: 7.7685\n",
            "Epoch 97/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 7.7272 - mae: 7.7272\n",
            "Epoch 98/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 7.6992 - mae: 7.6992\n",
            "Epoch 99/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 7.6711 - mae: 7.6711\n",
            "Epoch 100/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 7.6375 - mae: 7.6375\n",
            "Epoch 101/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 7.6119 - mae: 7.6119\n",
            "Epoch 102/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 7.5622 - mae: 7.5622\n",
            "Epoch 103/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 7.5459 - mae: 7.5459\n",
            "Epoch 104/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 7.4899 - mae: 7.4899\n",
            "Epoch 105/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 7.4691 - mae: 7.4691\n",
            "Epoch 106/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 7.4226 - mae: 7.4226\n",
            "Epoch 107/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 7.4032 - mae: 7.4032\n",
            "Epoch 108/500\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 7.3613 - mae: 7.3613\n",
            "Epoch 109/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 7.3342 - mae: 7.3342\n",
            "Epoch 110/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 7.3049 - mae: 7.3049\n",
            "Epoch 111/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 7.2860 - mae: 7.2860\n",
            "Epoch 112/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 7.2488 - mae: 7.2488\n",
            "Epoch 113/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 7.2172 - mae: 7.2172\n",
            "Epoch 114/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 7.1792 - mae: 7.1792\n",
            "Epoch 115/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 7.1700 - mae: 7.1700\n",
            "Epoch 116/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 7.1376 - mae: 7.1376\n",
            "Epoch 117/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 7.1214 - mae: 7.1214\n",
            "Epoch 118/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 7.0897 - mae: 7.0897\n",
            "Epoch 119/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 7.0577 - mae: 7.0577\n",
            "Epoch 120/500\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 7.0359 - mae: 7.0359\n",
            "Epoch 121/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 6.9991 - mae: 6.9991\n",
            "Epoch 122/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 6.9687 - mae: 6.9687\n",
            "Epoch 123/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 6.9432 - mae: 6.9432\n",
            "Epoch 124/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 6.9109 - mae: 6.9109\n",
            "Epoch 125/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 6.8809 - mae: 6.8809\n",
            "Epoch 126/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 6.8225 - mae: 6.8225\n",
            "Epoch 127/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 6.8343 - mae: 6.8343\n",
            "Epoch 128/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 6.7480 - mae: 6.7480\n",
            "Epoch 129/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 6.7354 - mae: 6.7354\n",
            "Epoch 130/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 6.7289 - mae: 6.7289\n",
            "Epoch 131/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 6.6515 - mae: 6.6515\n",
            "Epoch 132/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 6.6179 - mae: 6.6179\n",
            "Epoch 133/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 6.5892 - mae: 6.5892\n",
            "Epoch 134/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 6.5462 - mae: 6.5462\n",
            "Epoch 135/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 6.5158 - mae: 6.5158\n",
            "Epoch 136/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 6.4817 - mae: 6.4817\n",
            "Epoch 137/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 6.4672 - mae: 6.4672\n",
            "Epoch 138/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 6.4088 - mae: 6.4088\n",
            "Epoch 139/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 6.3885 - mae: 6.3885\n",
            "Epoch 140/500\n",
            "3/3 [==============================] - 0s 10ms/step - loss: 6.3719 - mae: 6.3719\n",
            "Epoch 141/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 6.3302 - mae: 6.3302\n",
            "Epoch 142/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 6.3020 - mae: 6.3020\n",
            "Epoch 143/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 6.2631 - mae: 6.2631\n",
            "Epoch 144/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 6.2462 - mae: 6.2462\n",
            "Epoch 145/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 6.2239 - mae: 6.2239\n",
            "Epoch 146/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 6.1816 - mae: 6.1816\n",
            "Epoch 147/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 6.1619 - mae: 6.1619\n",
            "Epoch 148/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 6.1252 - mae: 6.1252\n",
            "Epoch 149/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 6.0927 - mae: 6.0927\n",
            "Epoch 150/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 6.0523 - mae: 6.0523\n",
            "Epoch 151/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 6.0198 - mae: 6.0198\n",
            "Epoch 152/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 5.9809 - mae: 5.9809\n",
            "Epoch 153/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 5.9466 - mae: 5.9466\n",
            "Epoch 154/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 5.9132 - mae: 5.9132\n",
            "Epoch 155/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 5.8769 - mae: 5.8769\n",
            "Epoch 156/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 5.8488 - mae: 5.8488\n",
            "Epoch 157/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 5.8124 - mae: 5.8124\n",
            "Epoch 158/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 5.7821 - mae: 5.7821\n",
            "Epoch 159/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 5.7703 - mae: 5.7703\n",
            "Epoch 160/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 5.7309 - mae: 5.7309\n",
            "Epoch 161/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 5.6955 - mae: 5.6955\n",
            "Epoch 162/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 5.6616 - mae: 5.6616\n",
            "Epoch 163/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 5.6312 - mae: 5.6312\n",
            "Epoch 164/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 5.5859 - mae: 5.5859\n",
            "Epoch 165/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 5.5450 - mae: 5.5450\n",
            "Epoch 166/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 5.5374 - mae: 5.5374\n",
            "Epoch 167/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 5.5250 - mae: 5.5250\n",
            "Epoch 168/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 5.5546 - mae: 5.5546\n",
            "Epoch 169/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 5.4897 - mae: 5.4897\n",
            "Epoch 170/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 5.4166 - mae: 5.4166\n",
            "Epoch 171/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 5.3690 - mae: 5.3690\n",
            "Epoch 172/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 5.3360 - mae: 5.3360\n",
            "Epoch 173/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 5.2905 - mae: 5.2905\n",
            "Epoch 174/500\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 5.2491 - mae: 5.2491\n",
            "Epoch 175/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 5.2219 - mae: 5.2219\n",
            "Epoch 176/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 5.1812 - mae: 5.1812\n",
            "Epoch 177/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 5.1431 - mae: 5.1431\n",
            "Epoch 178/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 5.1084 - mae: 5.1084\n",
            "Epoch 179/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 5.0691 - mae: 5.0691\n",
            "Epoch 180/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 5.0256 - mae: 5.0256\n",
            "Epoch 181/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 4.9895 - mae: 4.9895\n",
            "Epoch 182/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 4.9532 - mae: 4.9532\n",
            "Epoch 183/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 4.9101 - mae: 4.9101\n",
            "Epoch 184/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 4.8697 - mae: 4.8697\n",
            "Epoch 185/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 4.8244 - mae: 4.8244\n",
            "Epoch 186/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 4.7908 - mae: 4.7908\n",
            "Epoch 187/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 4.7452 - mae: 4.7452\n",
            "Epoch 188/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 4.7038 - mae: 4.7038\n",
            "Epoch 189/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 4.6633 - mae: 4.6633\n",
            "Epoch 190/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 4.6230 - mae: 4.6230\n",
            "Epoch 191/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 4.5850 - mae: 4.5850\n",
            "Epoch 192/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 4.5477 - mae: 4.5477\n",
            "Epoch 193/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 4.5017 - mae: 4.5017\n",
            "Epoch 194/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 4.4597 - mae: 4.4597\n",
            "Epoch 195/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 4.4220 - mae: 4.4220\n",
            "Epoch 196/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 4.3781 - mae: 4.3781\n",
            "Epoch 197/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 4.3373 - mae: 4.3373\n",
            "Epoch 198/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 4.2917 - mae: 4.2917\n",
            "Epoch 199/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 4.2379 - mae: 4.2379\n",
            "Epoch 200/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 4.2377 - mae: 4.2377\n",
            "Epoch 201/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 4.1798 - mae: 4.1798\n",
            "Epoch 202/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 4.1335 - mae: 4.1335\n",
            "Epoch 203/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 4.0921 - mae: 4.0921\n",
            "Epoch 204/500\n",
            "3/3 [==============================] - 0s 9ms/step - loss: 4.0503 - mae: 4.0503\n",
            "Epoch 205/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 4.0133 - mae: 4.0133\n",
            "Epoch 206/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 3.9532 - mae: 3.9532\n",
            "Epoch 207/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 3.9053 - mae: 3.9053\n",
            "Epoch 208/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 3.8534 - mae: 3.8534\n",
            "Epoch 209/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 3.8254 - mae: 3.8254\n",
            "Epoch 210/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 3.7587 - mae: 3.7587\n",
            "Epoch 211/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 3.7087 - mae: 3.7087\n",
            "Epoch 212/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 3.6696 - mae: 3.6696\n",
            "Epoch 213/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 3.6006 - mae: 3.6006\n",
            "Epoch 214/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 3.5496 - mae: 3.5496\n",
            "Epoch 215/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 3.5296 - mae: 3.5296\n",
            "Epoch 216/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 3.4624 - mae: 3.4624\n",
            "Epoch 217/500\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 3.4243 - mae: 3.4243\n",
            "Epoch 218/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 3.3903 - mae: 3.3903\n",
            "Epoch 219/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 3.3535 - mae: 3.3535\n",
            "Epoch 220/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 3.3118 - mae: 3.3118\n",
            "Epoch 221/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 3.2748 - mae: 3.2748\n",
            "Epoch 222/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 3.2191 - mae: 3.2191\n",
            "Epoch 223/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 3.1756 - mae: 3.1756\n",
            "Epoch 224/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 3.1313 - mae: 3.1313\n",
            "Epoch 225/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 3.0870 - mae: 3.0870\n",
            "Epoch 226/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 3.0354 - mae: 3.0354\n",
            "Epoch 227/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 2.9615 - mae: 2.9615\n",
            "Epoch 228/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 2.9143 - mae: 2.9143\n",
            "Epoch 229/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 2.8593 - mae: 2.8593\n",
            "Epoch 230/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 2.8889 - mae: 2.8889\n",
            "Epoch 231/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 2.8284 - mae: 2.8284\n",
            "Epoch 232/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 2.7242 - mae: 2.7242\n",
            "Epoch 233/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 2.6817 - mae: 2.6817\n",
            "Epoch 234/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 2.6333 - mae: 2.6333\n",
            "Epoch 235/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 2.5796 - mae: 2.5796\n",
            "Epoch 236/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 2.5289 - mae: 2.5289\n",
            "Epoch 237/500\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 2.4870 - mae: 2.4870\n",
            "Epoch 238/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 2.4556 - mae: 2.4556\n",
            "Epoch 239/500\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 2.3795 - mae: 2.3795\n",
            "Epoch 240/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 2.3314 - mae: 2.3314\n",
            "Epoch 241/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 2.2832 - mae: 2.2832\n",
            "Epoch 242/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 2.2314 - mae: 2.2314\n",
            "Epoch 243/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 2.1851 - mae: 2.1851\n",
            "Epoch 244/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 2.1346 - mae: 2.1346\n",
            "Epoch 245/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 2.0684 - mae: 2.0684\n",
            "Epoch 246/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 2.0033 - mae: 2.0033\n",
            "Epoch 247/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 1.9715 - mae: 1.9715\n",
            "Epoch 248/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 1.8927 - mae: 1.8927\n",
            "Epoch 249/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 1.8515 - mae: 1.8515\n",
            "Epoch 250/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 1.8603 - mae: 1.8603\n",
            "Epoch 251/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 1.8415 - mae: 1.8415\n",
            "Epoch 252/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 1.7157 - mae: 1.7157\n",
            "Epoch 253/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 1.6656 - mae: 1.6656\n",
            "Epoch 254/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 1.6453 - mae: 1.6453\n",
            "Epoch 255/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 1.5627 - mae: 1.5627\n",
            "Epoch 256/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 1.4981 - mae: 1.4981\n",
            "Epoch 257/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 1.4320 - mae: 1.4320\n",
            "Epoch 258/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 1.4378 - mae: 1.4378\n",
            "Epoch 259/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 1.3689 - mae: 1.3689\n",
            "Epoch 260/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 1.2806 - mae: 1.2806\n",
            "Epoch 261/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 1.2290 - mae: 1.2290\n",
            "Epoch 262/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 1.1744 - mae: 1.1744\n",
            "Epoch 263/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 1.1196 - mae: 1.1196\n",
            "Epoch 264/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 1.0559 - mae: 1.0559\n",
            "Epoch 265/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 1.0383 - mae: 1.0383\n",
            "Epoch 266/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 1.0252 - mae: 1.0252\n",
            "Epoch 267/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.8726 - mae: 0.8726\n",
            "Epoch 268/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.8642 - mae: 0.8642\n",
            "Epoch 269/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 0.8137 - mae: 0.8137\n",
            "Epoch 270/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.7512 - mae: 0.7512\n",
            "Epoch 271/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.6721 - mae: 0.6721\n",
            "Epoch 272/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 0.6497 - mae: 0.6497\n",
            "Epoch 273/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.5573 - mae: 0.5573\n",
            "Epoch 274/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.5030 - mae: 0.5030\n",
            "Epoch 275/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.4525 - mae: 0.4525\n",
            "Epoch 276/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.4307 - mae: 0.4307\n",
            "Epoch 277/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.3267 - mae: 0.3267\n",
            "Epoch 278/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.2694 - mae: 0.2694\n",
            "Epoch 279/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.2044 - mae: 0.2044\n",
            "Epoch 280/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.1432 - mae: 0.1432\n",
            "Epoch 281/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.0899 - mae: 0.0899\n",
            "Epoch 282/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.0531 - mae: 0.0531\n",
            "Epoch 283/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.1884 - mae: 0.1884\n",
            "Epoch 284/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.2196 - mae: 0.2196\n",
            "Epoch 285/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.3041 - mae: 0.3041\n",
            "Epoch 286/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.3750 - mae: 0.3750\n",
            "Epoch 287/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.2105 - mae: 0.2105\n",
            "Epoch 288/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.5283 - mae: 0.5283\n",
            "Epoch 289/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.4136 - mae: 0.4136\n",
            "Epoch 290/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.2515 - mae: 0.2515\n",
            "Epoch 291/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.5594 - mae: 0.5594\n",
            "Epoch 292/500\n",
            "3/3 [==============================] - 0s 3ms/step - loss: 0.2695 - mae: 0.2695\n",
            "Epoch 293/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.5876 - mae: 0.5876\n",
            "Epoch 294/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.3061 - mae: 0.3061\n",
            "Epoch 295/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 0.1794 - mae: 0.1794\n",
            "Epoch 296/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.2032 - mae: 0.2032\n",
            "Epoch 297/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.4147 - mae: 0.4147\n",
            "Epoch 298/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.2804 - mae: 0.2804\n",
            "Epoch 299/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.1899 - mae: 0.1899\n",
            "Epoch 300/500\n",
            "3/3 [==============================] - 0s 9ms/step - loss: 0.5703 - mae: 0.5703\n",
            "Epoch 301/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.1625 - mae: 0.1625\n",
            "Epoch 302/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 0.1262 - mae: 0.1262\n",
            "Epoch 303/500\n",
            "3/3 [==============================] - 0s 9ms/step - loss: 0.1270 - mae: 0.1270\n",
            "Epoch 304/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.1092 - mae: 0.1092\n",
            "Epoch 305/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 0.1155 - mae: 0.1155\n",
            "Epoch 306/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 0.1055 - mae: 0.1055\n",
            "Epoch 307/500\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 0.0609 - mae: 0.0609\n",
            "Epoch 308/500\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 0.2145 - mae: 0.2145\n",
            "Epoch 309/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.2207 - mae: 0.2207\n",
            "Epoch 310/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.1359 - mae: 0.1359\n",
            "Epoch 311/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.1466 - mae: 0.1466\n",
            "Epoch 312/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.1609 - mae: 0.1609\n",
            "Epoch 313/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.2896 - mae: 0.2896\n",
            "Epoch 314/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.1023 - mae: 0.1023\n",
            "Epoch 315/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.2521 - mae: 0.2521\n",
            "Epoch 316/500\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 0.4629 - mae: 0.4629\n",
            "Epoch 317/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.3404 - mae: 0.3404\n",
            "Epoch 318/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 0.3864 - mae: 0.3864\n",
            "Epoch 319/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.1566 - mae: 0.1566\n",
            "Epoch 320/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.2626 - mae: 0.2626\n",
            "Epoch 321/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.1417 - mae: 0.1417\n",
            "Epoch 322/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 0.0390 - mae: 0.0390\n",
            "Epoch 323/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 0.0775 - mae: 0.0775\n",
            "Epoch 324/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.0974 - mae: 0.0974\n",
            "Epoch 325/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.1228 - mae: 0.1228\n",
            "Epoch 326/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.1473 - mae: 0.1473\n",
            "Epoch 327/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.1697 - mae: 0.1697\n",
            "Epoch 328/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 0.2608 - mae: 0.2608\n",
            "Epoch 329/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.2971 - mae: 0.2971\n",
            "Epoch 330/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.1233 - mae: 0.1233\n",
            "Epoch 331/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.1209 - mae: 0.1209\n",
            "Epoch 332/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 0.1324 - mae: 0.1324\n",
            "Epoch 333/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.1120 - mae: 0.1120\n",
            "Epoch 334/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 0.1111 - mae: 0.1111\n",
            "Epoch 335/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.1267 - mae: 0.1267\n",
            "Epoch 336/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.2072 - mae: 0.2072\n",
            "Epoch 337/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.0751 - mae: 0.0751\n",
            "Epoch 338/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.2655 - mae: 0.2655\n",
            "Epoch 339/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.2816 - mae: 0.2816\n",
            "Epoch 340/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.1674 - mae: 0.1674\n",
            "Epoch 341/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.2713 - mae: 0.2713\n",
            "Epoch 342/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.0626 - mae: 0.0626\n",
            "Epoch 343/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.0909 - mae: 0.0909\n",
            "Epoch 344/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.0755 - mae: 0.0755\n",
            "Epoch 345/500\n",
            "3/3 [==============================] - 0s 11ms/step - loss: 0.1124 - mae: 0.1124\n",
            "Epoch 346/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 0.1397 - mae: 0.1397\n",
            "Epoch 347/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.2964 - mae: 0.2964\n",
            "Epoch 348/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.1546 - mae: 0.1546\n",
            "Epoch 349/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.3048 - mae: 0.3048\n",
            "Epoch 350/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.1708 - mae: 0.1708\n",
            "Epoch 351/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.2868 - mae: 0.2868\n",
            "Epoch 352/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.2890 - mae: 0.2890\n",
            "Epoch 353/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.1419 - mae: 0.1419\n",
            "Epoch 354/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.2049 - mae: 0.2049\n",
            "Epoch 355/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.0602 - mae: 0.0602\n",
            "Epoch 356/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.2395 - mae: 0.2395\n",
            "Epoch 357/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.3374 - mae: 0.3374\n",
            "Epoch 358/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.1878 - mae: 0.1878\n",
            "Epoch 359/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.5096 - mae: 0.5096\n",
            "Epoch 360/500\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 0.1721 - mae: 0.1721\n",
            "Epoch 361/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.3017 - mae: 0.3017\n",
            "Epoch 362/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.2151 - mae: 0.2151\n",
            "Epoch 363/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.1139 - mae: 0.1139\n",
            "Epoch 364/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.0878 - mae: 0.0878\n",
            "Epoch 365/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.0790 - mae: 0.0790\n",
            "Epoch 366/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.0678 - mae: 0.0678\n",
            "Epoch 367/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.0393 - mae: 0.0393\n",
            "Epoch 368/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.0759 - mae: 0.0759\n",
            "Epoch 369/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.0934 - mae: 0.0934\n",
            "Epoch 370/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.1489 - mae: 0.1489\n",
            "Epoch 371/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.2343 - mae: 0.2343\n",
            "Epoch 372/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.2220 - mae: 0.2220\n",
            "Epoch 373/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.1092 - mae: 0.1092\n",
            "Epoch 374/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.1062 - mae: 0.1062\n",
            "Epoch 375/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.0867 - mae: 0.0867\n",
            "Epoch 376/500\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 0.0534 - mae: 0.0534\n",
            "Epoch 377/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.1485 - mae: 0.1485\n",
            "Epoch 378/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.2551 - mae: 0.2551\n",
            "Epoch 379/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.1253 - mae: 0.1253\n",
            "Epoch 380/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.1635 - mae: 0.1635\n",
            "Epoch 381/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.1953 - mae: 0.1953\n",
            "Epoch 382/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.2397 - mae: 0.2397\n",
            "Epoch 383/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 0.1003 - mae: 0.1003\n",
            "Epoch 384/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.0939 - mae: 0.0939\n",
            "Epoch 385/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.0903 - mae: 0.0903\n",
            "Epoch 386/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.0472 - mae: 0.0472\n",
            "Epoch 387/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 0.0826 - mae: 0.0826\n",
            "Epoch 388/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 0.0252 - mae: 0.0252\n",
            "Epoch 389/500\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 0.0860 - mae: 0.0860\n",
            "Epoch 390/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.0920 - mae: 0.0920\n",
            "Epoch 391/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.0886 - mae: 0.0886\n",
            "Epoch 392/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.0793 - mae: 0.0793\n",
            "Epoch 393/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.0326 - mae: 0.0326\n",
            "Epoch 394/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 0.1067 - mae: 0.1067\n",
            "Epoch 395/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.0411 - mae: 0.0411\n",
            "Epoch 396/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.1049 - mae: 0.1049\n",
            "Epoch 397/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.0162 - mae: 0.0162\n",
            "Epoch 398/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.0788 - mae: 0.0788\n",
            "Epoch 399/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.0770 - mae: 0.0770\n",
            "Epoch 400/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.0751 - mae: 0.0751\n",
            "Epoch 401/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.0087 - mae: 0.0087\n",
            "Epoch 402/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.0301 - mae: 0.0301\n",
            "Epoch 403/500\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 0.0784 - mae: 0.0784\n",
            "Epoch 404/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 0.0464 - mae: 0.0464\n",
            "Epoch 405/500\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 0.1339 - mae: 0.1339\n",
            "Epoch 406/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.1682 - mae: 0.1682\n",
            "Epoch 407/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.0545 - mae: 0.0545\n",
            "Epoch 408/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.2602 - mae: 0.2602\n",
            "Epoch 409/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.1238 - mae: 0.1238\n",
            "Epoch 410/500\n",
            "3/3 [==============================] - 0s 10ms/step - loss: 0.1505 - mae: 0.1505\n",
            "Epoch 411/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.1504 - mae: 0.1504\n",
            "Epoch 412/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.0575 - mae: 0.0575\n",
            "Epoch 413/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.2037 - mae: 0.2037\n",
            "Epoch 414/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.3073 - mae: 0.3073\n",
            "Epoch 415/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.1347 - mae: 0.1347\n",
            "Epoch 416/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.3645 - mae: 0.3645\n",
            "Epoch 417/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.1427 - mae: 0.1427\n",
            "Epoch 418/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.2613 - mae: 0.2613\n",
            "Epoch 419/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.1317 - mae: 0.1317\n",
            "Epoch 420/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.2833 - mae: 0.2833\n",
            "Epoch 421/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.1665 - mae: 0.1665\n",
            "Epoch 422/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.3015 - mae: 0.3015\n",
            "Epoch 423/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.3422 - mae: 0.3422\n",
            "Epoch 424/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.2030 - mae: 0.2030\n",
            "Epoch 425/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.4742 - mae: 0.4742\n",
            "Epoch 426/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.1676 - mae: 0.1676\n",
            "Epoch 427/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.1691 - mae: 0.1691\n",
            "Epoch 428/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.0721 - mae: 0.0721\n",
            "Epoch 429/500\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 0.1449 - mae: 0.1449\n",
            "Epoch 430/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 0.0905 - mae: 0.0905\n",
            "Epoch 431/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 0.1615 - mae: 0.1615\n",
            "Epoch 432/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.3174 - mae: 0.3174\n",
            "Epoch 433/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.1688 - mae: 0.1688\n",
            "Epoch 434/500\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 0.3033 - mae: 0.3033\n",
            "Epoch 435/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.1221 - mae: 0.1221\n",
            "Epoch 436/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.4242 - mae: 0.4242\n",
            "Epoch 437/500\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 0.1722 - mae: 0.1722\n",
            "Epoch 438/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.3979 - mae: 0.3979\n",
            "Epoch 439/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 0.1712 - mae: 0.1712\n",
            "Epoch 440/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.1691 - mae: 0.1691\n",
            "Epoch 441/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.2320 - mae: 0.2320\n",
            "Epoch 442/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.1695 - mae: 0.1695\n",
            "Epoch 443/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.1052 - mae: 0.1052\n",
            "Epoch 444/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.0818 - mae: 0.0818\n",
            "Epoch 445/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.0755 - mae: 0.0755\n",
            "Epoch 446/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.0674 - mae: 0.0674\n",
            "Epoch 447/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.0892 - mae: 0.0892\n",
            "Epoch 448/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.1022 - mae: 0.1022\n",
            "Epoch 449/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.2816 - mae: 0.2816\n",
            "Epoch 450/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.0850 - mae: 0.0850\n",
            "Epoch 451/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.1622 - mae: 0.1622\n",
            "Epoch 452/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 0.1380 - mae: 0.1380\n",
            "Epoch 453/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 0.3059 - mae: 0.3059\n",
            "Epoch 454/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.2213 - mae: 0.2213\n",
            "Epoch 455/500\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 0.1041 - mae: 0.1041\n",
            "Epoch 456/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.2156 - mae: 0.2156\n",
            "Epoch 457/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.2950 - mae: 0.2950\n",
            "Epoch 458/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.2662 - mae: 0.2662\n",
            "Epoch 459/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.0969 - mae: 0.0969\n",
            "Epoch 460/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.0927 - mae: 0.0927\n",
            "Epoch 461/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.0695 - mae: 0.0695\n",
            "Epoch 462/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.1083 - mae: 0.1083\n",
            "Epoch 463/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.1928 - mae: 0.1928\n",
            "Epoch 464/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.1071 - mae: 0.1071\n",
            "Epoch 465/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.1645 - mae: 0.1645\n",
            "Epoch 466/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.1433 - mae: 0.1433\n",
            "Epoch 467/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.0694 - mae: 0.0694\n",
            "Epoch 468/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.0354 - mae: 0.0354\n",
            "Epoch 469/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.0399 - mae: 0.0399\n",
            "Epoch 470/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.0949 - mae: 0.0949\n",
            "Epoch 471/500\n",
            "3/3 [==============================] - 0s 9ms/step - loss: 0.0786 - mae: 0.0786\n",
            "Epoch 472/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.0289 - mae: 0.0289\n",
            "Epoch 473/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.0472 - mae: 0.0472\n",
            "Epoch 474/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.0764 - mae: 0.0764\n",
            "Epoch 475/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.0361 - mae: 0.0361\n",
            "Epoch 476/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.0258 - mae: 0.0258\n",
            "Epoch 477/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.0954 - mae: 0.0954\n",
            "Epoch 478/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.0580 - mae: 0.0580\n",
            "Epoch 479/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.0887 - mae: 0.0887\n",
            "Epoch 480/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.0983 - mae: 0.0983\n",
            "Epoch 481/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.1662 - mae: 0.1662\n",
            "Epoch 482/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.1322 - mae: 0.1322\n",
            "Epoch 483/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.0475 - mae: 0.0475\n",
            "Epoch 484/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.1577 - mae: 0.1577\n",
            "Epoch 485/500\n",
            "3/3 [==============================] - 0s 8ms/step - loss: 0.0656 - mae: 0.0656\n",
            "Epoch 486/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.1306 - mae: 0.1306\n",
            "Epoch 487/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.0838 - mae: 0.0838\n",
            "Epoch 488/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.1107 - mae: 0.1107\n",
            "Epoch 489/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.1897 - mae: 0.1897\n",
            "Epoch 490/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.1428 - mae: 0.1428\n",
            "Epoch 491/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.0294 - mae: 0.0294\n",
            "Epoch 492/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.0423 - mae: 0.0423\n",
            "Epoch 493/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.1098 - mae: 0.1098\n",
            "Epoch 494/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.1231 - mae: 0.1231\n",
            "Epoch 495/500\n",
            "3/3 [==============================] - 0s 7ms/step - loss: 0.0661 - mae: 0.0661\n",
            "Epoch 496/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.0842 - mae: 0.0842\n",
            "Epoch 497/500\n",
            "3/3 [==============================] - 0s 4ms/step - loss: 0.1126 - mae: 0.1126\n",
            "Epoch 498/500\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.3558 - mae: 0.3558\n",
            "Epoch 499/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.1336 - mae: 0.1336\n",
            "Epoch 500/500\n",
            "3/3 [==============================] - 0s 5ms/step - loss: 0.2339 - mae: 0.2339\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7fef557ec810>"
            ]
          },
          "metadata": {},
          "execution_count": 93
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model.predict([300.0])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "G6a4pGpvzBEW",
        "outputId": "06467e4d-2cd9-41c7-b644-19279d3b9de9"
      },
      "execution_count": 96,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[309.09464]], dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 96
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Evaluating a model\n",
        "\n",
        "In practice, a typical workflow you'll go through when building neural networks is:\n",
        "\n",
        "Build a model -> fit it -> evaluate -> tweak -> fit..."
      ],
      "metadata": {
        "id": "-moBNVBQzyP0"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "X = tf.range(-10000,10000,4)\n",
        "y = X +10\n",
        "\n",
        "X = tf.cast(X, dtype = tf.float16)\n",
        "y = tf.cast(y, dtype = tf.float16)"
      ],
      "metadata": {
        "id": "EHXy5ZB10IkM"
      },
      "execution_count": 196,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model = tf.keras.Sequential([tf.keras.layers.Dense(30,activation =None),\n",
        "                             tf.keras.layers.Dense(1)])\n",
        "\n",
        "model.compile(loss =tf.keras.losses.mae,\n",
        "              optimizer = tf.keras.optimizers.Adam(lr = 0.001),\n",
        "              metrics = ['mae']\n",
        "              )\n",
        "\n",
        "model.fit(tf.expand_dims(X, axis=-1), y, epochs=500)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-jvUolks0lGn",
        "outputId": "e7fa401f-a94f-44ea-b76c-ddbeed8bc192"
      },
      "execution_count": 103,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/keras/optimizer_v2/adam.py:105: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead.\n",
            "  super(Adam, self).__init__(name, **kwargs)\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 138.8708 - mae: 138.8708\n",
            "Epoch 2/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 33.7970 - mae: 33.7970\n",
            "Epoch 3/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 14.9875 - mae: 14.9875\n",
            "Epoch 4/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 10.2409 - mae: 10.2409\n",
            "Epoch 5/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 9.4417 - mae: 9.4417\n",
            "Epoch 6/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 9.2962 - mae: 9.2962\n",
            "Epoch 7/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 9.1194 - mae: 9.1194\n",
            "Epoch 8/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 8.9571 - mae: 8.9571\n",
            "Epoch 9/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 8.7855 - mae: 8.7855\n",
            "Epoch 10/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 8.8495 - mae: 8.8495\n",
            "Epoch 11/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 8.4439 - mae: 8.4439\n",
            "Epoch 12/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 8.4119 - mae: 8.4119\n",
            "Epoch 13/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 8.1559 - mae: 8.1559\n",
            "Epoch 14/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 7.9652 - mae: 7.9652\n",
            "Epoch 15/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 7.8934 - mae: 7.8934\n",
            "Epoch 16/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 7.6751 - mae: 7.6751\n",
            "Epoch 17/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 7.5692 - mae: 7.5692\n",
            "Epoch 18/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 7.5177 - mae: 7.5177\n",
            "Epoch 19/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 7.2552 - mae: 7.2552\n",
            "Epoch 20/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 7.0544 - mae: 7.0544\n",
            "Epoch 21/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 7.1018 - mae: 7.1018\n",
            "Epoch 22/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 6.7378 - mae: 6.7378\n",
            "Epoch 23/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 6.8769 - mae: 6.8769\n",
            "Epoch 24/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 6.5474 - mae: 6.5474\n",
            "Epoch 25/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 6.3267 - mae: 6.3267\n",
            "Epoch 26/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 6.4102 - mae: 6.4102\n",
            "Epoch 27/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 6.4870 - mae: 6.4870\n",
            "Epoch 28/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 5.8733 - mae: 5.8733\n",
            "Epoch 29/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 5.8020 - mae: 5.8020\n",
            "Epoch 30/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 5.9775 - mae: 5.9775\n",
            "Epoch 31/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 5.6735 - mae: 5.6735\n",
            "Epoch 32/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 6.1687 - mae: 6.1687\n",
            "Epoch 33/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 5.2425 - mae: 5.2425\n",
            "Epoch 34/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 5.2679 - mae: 5.2679\n",
            "Epoch 35/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 4.8605 - mae: 4.8605\n",
            "Epoch 36/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 4.8550 - mae: 4.8550\n",
            "Epoch 37/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 4.6117 - mae: 4.6117\n",
            "Epoch 38/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 4.3559 - mae: 4.3559\n",
            "Epoch 39/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 4.6651 - mae: 4.6651\n",
            "Epoch 40/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 4.2237 - mae: 4.2237\n",
            "Epoch 41/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 4.1273 - mae: 4.1273\n",
            "Epoch 42/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 3.8106 - mae: 3.8106\n",
            "Epoch 43/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 3.8418 - mae: 3.8418\n",
            "Epoch 44/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 3.6057 - mae: 3.6057\n",
            "Epoch 45/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 3.4015 - mae: 3.4015\n",
            "Epoch 46/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 3.4572 - mae: 3.4572\n",
            "Epoch 47/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 3.1768 - mae: 3.1768\n",
            "Epoch 48/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 3.3201 - mae: 3.3201\n",
            "Epoch 49/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 2.8095 - mae: 2.8095\n",
            "Epoch 50/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 2.5295 - mae: 2.5295\n",
            "Epoch 51/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 2.4646 - mae: 2.4646\n",
            "Epoch 52/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 2.6022 - mae: 2.6022\n",
            "Epoch 53/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 2.6379 - mae: 2.6379\n",
            "Epoch 54/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 2.4906 - mae: 2.4906\n",
            "Epoch 55/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 2.2815 - mae: 2.2815\n",
            "Epoch 56/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.8350 - mae: 1.8350\n",
            "Epoch 57/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.7018 - mae: 1.7018\n",
            "Epoch 58/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.5176 - mae: 1.5176\n",
            "Epoch 59/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.3170 - mae: 1.3170\n",
            "Epoch 60/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.3210 - mae: 1.3210\n",
            "Epoch 61/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.3560 - mae: 1.3560\n",
            "Epoch 62/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9546 - mae: 0.9546\n",
            "Epoch 63/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7558 - mae: 0.7558\n",
            "Epoch 64/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6294 - mae: 0.6294\n",
            "Epoch 65/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7184 - mae: 0.7184\n",
            "Epoch 66/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6413 - mae: 0.6413\n",
            "Epoch 67/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.0493 - mae: 1.0493\n",
            "Epoch 68/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9943 - mae: 0.9943\n",
            "Epoch 69/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.3312 - mae: 1.3312\n",
            "Epoch 70/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.8174 - mae: 1.8174\n",
            "Epoch 71/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7409 - mae: 0.7409\n",
            "Epoch 72/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.3146 - mae: 1.3146\n",
            "Epoch 73/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.7235 - mae: 1.7235\n",
            "Epoch 74/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7367 - mae: 0.7367\n",
            "Epoch 75/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7834 - mae: 0.7834\n",
            "Epoch 76/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.3065 - mae: 1.3065\n",
            "Epoch 77/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.1626 - mae: 1.1626\n",
            "Epoch 78/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7362 - mae: 0.7362\n",
            "Epoch 79/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.0259 - mae: 1.0259\n",
            "Epoch 80/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.2463 - mae: 1.2463\n",
            "Epoch 81/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.4932 - mae: 1.4932\n",
            "Epoch 82/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7554 - mae: 0.7554\n",
            "Epoch 83/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9698 - mae: 0.9698\n",
            "Epoch 84/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.2507 - mae: 1.2507\n",
            "Epoch 85/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.0679 - mae: 1.0679\n",
            "Epoch 86/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9718 - mae: 0.9718\n",
            "Epoch 87/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8813 - mae: 0.8813\n",
            "Epoch 88/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.4028 - mae: 1.4028\n",
            "Epoch 89/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 2.6858 - mae: 2.6858\n",
            "Epoch 90/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.3292 - mae: 0.3292\n",
            "Epoch 91/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.7380 - mae: 0.7380\n",
            "Epoch 92/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.7989 - mae: 1.7989\n",
            "Epoch 93/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 2.3183 - mae: 2.3183\n",
            "Epoch 94/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 2.9182 - mae: 2.9182\n",
            "Epoch 95/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 3.3371 - mae: 3.3371\n",
            "Epoch 96/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 2.6334 - mae: 2.6334\n",
            "Epoch 97/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.5560 - mae: 1.5560\n",
            "Epoch 98/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.2149 - mae: 1.2149\n",
            "Epoch 99/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.3893 - mae: 1.3893\n",
            "Epoch 100/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.3939 - mae: 1.3939\n",
            "Epoch 101/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.9939 - mae: 1.9939\n",
            "Epoch 102/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8832 - mae: 0.8832\n",
            "Epoch 103/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6137 - mae: 0.6137\n",
            "Epoch 104/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.8379 - mae: 1.8379\n",
            "Epoch 105/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.0095 - mae: 1.0095\n",
            "Epoch 106/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7358 - mae: 0.7358\n",
            "Epoch 107/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.0513 - mae: 1.0513\n",
            "Epoch 108/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.1117 - mae: 1.1117\n",
            "Epoch 109/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7616 - mae: 0.7616\n",
            "Epoch 110/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.4117 - mae: 0.4117\n",
            "Epoch 111/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.4655 - mae: 0.4655\n",
            "Epoch 112/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.1558 - mae: 1.1558\n",
            "Epoch 113/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.4274 - mae: 1.4274\n",
            "Epoch 114/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5981 - mae: 0.5981\n",
            "Epoch 115/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.4235 - mae: 0.4235\n",
            "Epoch 116/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.0000 - mae: 1.0000\n",
            "Epoch 117/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8552 - mae: 0.8552\n",
            "Epoch 118/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.0520 - mae: 1.0520\n",
            "Epoch 119/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.8453 - mae: 0.8453\n",
            "Epoch 120/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.0583 - mae: 1.0583\n",
            "Epoch 121/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8800 - mae: 0.8800\n",
            "Epoch 122/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.1198 - mae: 1.1198\n",
            "Epoch 123/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.4143 - mae: 0.4143\n",
            "Epoch 124/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7027 - mae: 0.7027\n",
            "Epoch 125/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.7446 - mae: 1.7446\n",
            "Epoch 126/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5451 - mae: 0.5451\n",
            "Epoch 127/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.3203 - mae: 1.3203\n",
            "Epoch 128/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.3191 - mae: 1.3191\n",
            "Epoch 129/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7202 - mae: 0.7202\n",
            "Epoch 130/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.3333 - mae: 1.3333\n",
            "Epoch 131/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5272 - mae: 0.5272\n",
            "Epoch 132/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.6240 - mae: 0.6240\n",
            "Epoch 133/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8790 - mae: 0.8790\n",
            "Epoch 134/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 1.1711 - mae: 1.1711\n",
            "Epoch 135/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.0954 - mae: 1.0954\n",
            "Epoch 136/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5823 - mae: 0.5823\n",
            "Epoch 137/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.4341 - mae: 0.4341\n",
            "Epoch 138/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.2712 - mae: 1.2712\n",
            "Epoch 139/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 1.6474 - mae: 1.6474\n",
            "Epoch 140/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.0207 - mae: 1.0207\n",
            "Epoch 141/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6952 - mae: 0.6952\n",
            "Epoch 142/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8589 - mae: 0.8589\n",
            "Epoch 143/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.4496 - mae: 1.4496\n",
            "Epoch 144/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 1.4689 - mae: 1.4689\n",
            "Epoch 145/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9427 - mae: 0.9427\n",
            "Epoch 146/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.3709 - mae: 0.3709\n",
            "Epoch 147/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7934 - mae: 0.7934\n",
            "Epoch 148/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5047 - mae: 0.5047\n",
            "Epoch 149/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7866 - mae: 0.7866\n",
            "Epoch 150/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.7853 - mae: 1.7853\n",
            "Epoch 151/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6803 - mae: 0.6803\n",
            "Epoch 152/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8900 - mae: 0.8900\n",
            "Epoch 153/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5232 - mae: 0.5232\n",
            "Epoch 154/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5575 - mae: 0.5575\n",
            "Epoch 155/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5168 - mae: 0.5168\n",
            "Epoch 156/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.3850 - mae: 0.3850\n",
            "Epoch 157/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.0747 - mae: 1.0747\n",
            "Epoch 158/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.3538 - mae: 1.3538\n",
            "Epoch 159/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5969 - mae: 0.5969\n",
            "Epoch 160/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7177 - mae: 0.7177\n",
            "Epoch 161/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7326 - mae: 0.7326\n",
            "Epoch 162/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.2440 - mae: 0.2440\n",
            "Epoch 163/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7492 - mae: 0.7492\n",
            "Epoch 164/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6759 - mae: 0.6759\n",
            "Epoch 165/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6217 - mae: 0.6217\n",
            "Epoch 166/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.3432 - mae: 1.3432\n",
            "Epoch 167/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.8705 - mae: 1.8705\n",
            "Epoch 168/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.1504 - mae: 1.1504\n",
            "Epoch 169/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 1.5488 - mae: 1.5488\n",
            "Epoch 170/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8484 - mae: 0.8484\n",
            "Epoch 171/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.7419 - mae: 1.7419\n",
            "Epoch 172/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.8143 - mae: 1.8143\n",
            "Epoch 173/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.5633 - mae: 1.5633\n",
            "Epoch 174/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6392 - mae: 0.6392\n",
            "Epoch 175/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8778 - mae: 0.8778\n",
            "Epoch 176/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.1310 - mae: 1.1310\n",
            "Epoch 177/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.4935 - mae: 1.4935\n",
            "Epoch 178/500\n",
            "16/16 [==============================] - 0s 4ms/step - loss: 0.9102 - mae: 0.9102\n",
            "Epoch 179/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5584 - mae: 0.5584\n",
            "Epoch 180/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.4182 - mae: 1.4182\n",
            "Epoch 181/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 1.2922 - mae: 1.2922\n",
            "Epoch 182/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.2965 - mae: 1.2965\n",
            "Epoch 183/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8720 - mae: 0.8720\n",
            "Epoch 184/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.5484 - mae: 1.5484\n",
            "Epoch 185/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.7327 - mae: 1.7327\n",
            "Epoch 186/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.6155 - mae: 1.6155\n",
            "Epoch 187/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.7963 - mae: 1.7963\n",
            "Epoch 188/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.0154 - mae: 1.0154\n",
            "Epoch 189/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.6737 - mae: 1.6737\n",
            "Epoch 190/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.0299 - mae: 1.0299\n",
            "Epoch 191/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8510 - mae: 0.8510\n",
            "Epoch 192/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6950 - mae: 0.6950\n",
            "Epoch 193/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.6750 - mae: 1.6750\n",
            "Epoch 194/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 2.1719 - mae: 2.1719\n",
            "Epoch 195/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9149 - mae: 0.9149\n",
            "Epoch 196/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.3689 - mae: 1.3689\n",
            "Epoch 197/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.1880 - mae: 1.1880\n",
            "Epoch 198/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7639 - mae: 0.7639\n",
            "Epoch 199/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.1804 - mae: 1.1804\n",
            "Epoch 200/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.8205 - mae: 1.8205\n",
            "Epoch 201/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7029 - mae: 0.7029\n",
            "Epoch 202/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9582 - mae: 0.9582\n",
            "Epoch 203/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7938 - mae: 0.7938\n",
            "Epoch 204/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.4679 - mae: 0.4679\n",
            "Epoch 205/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.5280 - mae: 1.5280\n",
            "Epoch 206/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.7850 - mae: 1.7850\n",
            "Epoch 207/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6364 - mae: 0.6364\n",
            "Epoch 208/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9982 - mae: 0.9982\n",
            "Epoch 209/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6369 - mae: 0.6369\n",
            "Epoch 210/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9655 - mae: 0.9655\n",
            "Epoch 211/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.2502 - mae: 1.2502\n",
            "Epoch 212/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.7310 - mae: 1.7310\n",
            "Epoch 213/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.3917 - mae: 1.3917\n",
            "Epoch 214/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5643 - mae: 0.5643\n",
            "Epoch 215/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7435 - mae: 0.7435\n",
            "Epoch 216/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.1402 - mae: 1.1402\n",
            "Epoch 217/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.2090 - mae: 1.2090\n",
            "Epoch 218/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.1018 - mae: 1.1018\n",
            "Epoch 219/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.2047 - mae: 1.2047\n",
            "Epoch 220/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7613 - mae: 0.7613\n",
            "Epoch 221/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7113 - mae: 0.7113\n",
            "Epoch 222/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.1692 - mae: 1.1692\n",
            "Epoch 223/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.1810 - mae: 1.1810\n",
            "Epoch 224/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.2934 - mae: 1.2934\n",
            "Epoch 225/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.4503 - mae: 1.4503\n",
            "Epoch 226/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7729 - mae: 0.7729\n",
            "Epoch 227/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9275 - mae: 0.9275\n",
            "Epoch 228/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.4516 - mae: 0.4516\n",
            "Epoch 229/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7394 - mae: 0.7394\n",
            "Epoch 230/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.6250 - mae: 0.6250\n",
            "Epoch 231/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5091 - mae: 0.5091\n",
            "Epoch 232/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.3597 - mae: 0.3597\n",
            "Epoch 233/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5161 - mae: 0.5161\n",
            "Epoch 234/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5579 - mae: 0.5579\n",
            "Epoch 235/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.4435 - mae: 0.4435\n",
            "Epoch 236/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5409 - mae: 0.5409\n",
            "Epoch 237/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.4576 - mae: 0.4576\n",
            "Epoch 238/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6795 - mae: 0.6795\n",
            "Epoch 239/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9605 - mae: 0.9605\n",
            "Epoch 240/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.5623 - mae: 1.5623\n",
            "Epoch 241/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5289 - mae: 0.5289\n",
            "Epoch 242/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6679 - mae: 0.6679\n",
            "Epoch 243/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.3612 - mae: 0.3612\n",
            "Epoch 244/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 1.5306 - mae: 1.5306\n",
            "Epoch 245/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7756 - mae: 0.7756\n",
            "Epoch 246/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.8135 - mae: 0.8135\n",
            "Epoch 247/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.7386 - mae: 0.7386\n",
            "Epoch 248/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9234 - mae: 0.9234\n",
            "Epoch 249/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9621 - mae: 0.9621\n",
            "Epoch 250/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.3059 - mae: 1.3059\n",
            "Epoch 251/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.0175 - mae: 1.0175\n",
            "Epoch 252/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.3136 - mae: 0.3136\n",
            "Epoch 253/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6833 - mae: 0.6833\n",
            "Epoch 254/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.7611 - mae: 1.7611\n",
            "Epoch 255/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.1388 - mae: 1.1388\n",
            "Epoch 256/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.4712 - mae: 0.4712\n",
            "Epoch 257/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.6248 - mae: 0.6248\n",
            "Epoch 258/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.0532 - mae: 1.0532\n",
            "Epoch 259/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.1445 - mae: 1.1445\n",
            "Epoch 260/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.3774 - mae: 0.3774\n",
            "Epoch 261/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6478 - mae: 0.6478\n",
            "Epoch 262/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.6266 - mae: 1.6266\n",
            "Epoch 263/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.9380 - mae: 1.9380\n",
            "Epoch 264/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.6460 - mae: 1.6460\n",
            "Epoch 265/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.5537 - mae: 1.5537\n",
            "Epoch 266/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8756 - mae: 0.8756\n",
            "Epoch 267/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6926 - mae: 0.6926\n",
            "Epoch 268/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.7429 - mae: 0.7429\n",
            "Epoch 269/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.1060 - mae: 1.1060\n",
            "Epoch 270/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.5650 - mae: 1.5650\n",
            "Epoch 271/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.0857 - mae: 1.0857\n",
            "Epoch 272/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8304 - mae: 0.8304\n",
            "Epoch 273/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7979 - mae: 0.7979\n",
            "Epoch 274/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 1.3002 - mae: 1.3002\n",
            "Epoch 275/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.4132 - mae: 1.4132\n",
            "Epoch 276/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8542 - mae: 0.8542\n",
            "Epoch 277/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.4931 - mae: 0.4931\n",
            "Epoch 278/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5572 - mae: 0.5572\n",
            "Epoch 279/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.5046 - mae: 0.5046\n",
            "Epoch 280/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.3988 - mae: 0.3988\n",
            "Epoch 281/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6397 - mae: 0.6397\n",
            "Epoch 282/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.7148 - mae: 1.7148\n",
            "Epoch 283/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 1.1101 - mae: 1.1101\n",
            "Epoch 284/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.0927 - mae: 1.0927\n",
            "Epoch 285/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.6513 - mae: 1.6513\n",
            "Epoch 286/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.2161 - mae: 1.2161\n",
            "Epoch 287/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.6853 - mae: 1.6853\n",
            "Epoch 288/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.8198 - mae: 1.8198\n",
            "Epoch 289/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7713 - mae: 0.7713\n",
            "Epoch 290/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.7206 - mae: 0.7206\n",
            "Epoch 291/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6615 - mae: 0.6615\n",
            "Epoch 292/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 1.0538 - mae: 1.0538\n",
            "Epoch 293/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.6955 - mae: 1.6955\n",
            "Epoch 294/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.3449 - mae: 1.3449\n",
            "Epoch 295/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 1.3613 - mae: 1.3613\n",
            "Epoch 296/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.5329 - mae: 1.5329\n",
            "Epoch 297/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7663 - mae: 0.7663\n",
            "Epoch 298/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 2.3605 - mae: 2.3605\n",
            "Epoch 299/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.3699 - mae: 1.3699\n",
            "Epoch 300/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.7423 - mae: 1.7423\n",
            "Epoch 301/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.5997 - mae: 1.5997\n",
            "Epoch 302/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.6180 - mae: 1.6180\n",
            "Epoch 303/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.5120 - mae: 1.5120\n",
            "Epoch 304/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9012 - mae: 0.9012\n",
            "Epoch 305/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.1102 - mae: 1.1102\n",
            "Epoch 306/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 1.3225 - mae: 1.3225\n",
            "Epoch 307/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.4050 - mae: 0.4050\n",
            "Epoch 308/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8224 - mae: 0.8224\n",
            "Epoch 309/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.6992 - mae: 0.6992\n",
            "Epoch 310/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6360 - mae: 0.6360\n",
            "Epoch 311/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.0279 - mae: 1.0279\n",
            "Epoch 312/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9833 - mae: 0.9833\n",
            "Epoch 313/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.1504 - mae: 1.1504\n",
            "Epoch 314/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6791 - mae: 0.6791\n",
            "Epoch 315/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6426 - mae: 0.6426\n",
            "Epoch 316/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.4997 - mae: 0.4997\n",
            "Epoch 317/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5179 - mae: 0.5179\n",
            "Epoch 318/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 2.0884 - mae: 2.0884\n",
            "Epoch 319/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.9255 - mae: 1.9255\n",
            "Epoch 320/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8872 - mae: 0.8872\n",
            "Epoch 321/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7180 - mae: 0.7180\n",
            "Epoch 322/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9655 - mae: 0.9655\n",
            "Epoch 323/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9039 - mae: 0.9039\n",
            "Epoch 324/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8916 - mae: 0.8916\n",
            "Epoch 325/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6579 - mae: 0.6579\n",
            "Epoch 326/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.6623 - mae: 0.6623\n",
            "Epoch 327/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9464 - mae: 0.9464\n",
            "Epoch 328/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 1.6307 - mae: 1.6307\n",
            "Epoch 329/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.1295 - mae: 1.1295\n",
            "Epoch 330/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7208 - mae: 0.7208\n",
            "Epoch 331/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.3626 - mae: 0.3626\n",
            "Epoch 332/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5356 - mae: 0.5356\n",
            "Epoch 333/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6774 - mae: 0.6774\n",
            "Epoch 334/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8047 - mae: 0.8047\n",
            "Epoch 335/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8103 - mae: 0.8103\n",
            "Epoch 336/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.3488 - mae: 0.3488\n",
            "Epoch 337/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8434 - mae: 0.8434\n",
            "Epoch 338/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 1.1264 - mae: 1.1264\n",
            "Epoch 339/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9069 - mae: 0.9069\n",
            "Epoch 340/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.8525 - mae: 0.8525\n",
            "Epoch 341/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7993 - mae: 0.7993\n",
            "Epoch 342/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6870 - mae: 0.6870\n",
            "Epoch 343/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9538 - mae: 0.9538\n",
            "Epoch 344/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.2615 - mae: 1.2615\n",
            "Epoch 345/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.6632 - mae: 1.6632\n",
            "Epoch 346/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 2.2518 - mae: 2.2518\n",
            "Epoch 347/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.4529 - mae: 1.4529\n",
            "Epoch 348/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.4105 - mae: 0.4105\n",
            "Epoch 349/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.4927 - mae: 0.4927\n",
            "Epoch 350/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.1606 - mae: 1.1606\n",
            "Epoch 351/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.4982 - mae: 0.4982\n",
            "Epoch 352/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.3243 - mae: 1.3243\n",
            "Epoch 353/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.7816 - mae: 1.7816\n",
            "Epoch 354/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8269 - mae: 0.8269\n",
            "Epoch 355/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9229 - mae: 0.9229\n",
            "Epoch 356/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8253 - mae: 0.8253\n",
            "Epoch 357/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.0803 - mae: 1.0803\n",
            "Epoch 358/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7727 - mae: 0.7727\n",
            "Epoch 359/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 2.1863 - mae: 2.1863\n",
            "Epoch 360/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.8679 - mae: 1.8679\n",
            "Epoch 361/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.2107 - mae: 1.2107\n",
            "Epoch 362/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6583 - mae: 0.6583\n",
            "Epoch 363/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5103 - mae: 0.5103\n",
            "Epoch 364/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.4312 - mae: 0.4312\n",
            "Epoch 365/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 1.0128 - mae: 1.0128\n",
            "Epoch 366/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.1135 - mae: 1.1135\n",
            "Epoch 367/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6272 - mae: 0.6272\n",
            "Epoch 368/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.0564 - mae: 1.0564\n",
            "Epoch 369/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.4237 - mae: 0.4237\n",
            "Epoch 370/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.2487 - mae: 0.2487\n",
            "Epoch 371/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9388 - mae: 0.9388\n",
            "Epoch 372/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.0707 - mae: 1.0707\n",
            "Epoch 373/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5509 - mae: 0.5509\n",
            "Epoch 374/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.4490 - mae: 0.4490\n",
            "Epoch 375/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7457 - mae: 0.7457\n",
            "Epoch 376/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.6154 - mae: 1.6154\n",
            "Epoch 377/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.4611 - mae: 1.4611\n",
            "Epoch 378/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.2630 - mae: 1.2630\n",
            "Epoch 379/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7200 - mae: 0.7200\n",
            "Epoch 380/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.2454 - mae: 1.2454\n",
            "Epoch 381/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9757 - mae: 0.9757\n",
            "Epoch 382/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5296 - mae: 0.5296\n",
            "Epoch 383/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6211 - mae: 0.6211\n",
            "Epoch 384/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9736 - mae: 0.9736\n",
            "Epoch 385/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8797 - mae: 0.8797\n",
            "Epoch 386/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6165 - mae: 0.6165\n",
            "Epoch 387/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5433 - mae: 0.5433\n",
            "Epoch 388/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.3804 - mae: 0.3804\n",
            "Epoch 389/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5871 - mae: 0.5871\n",
            "Epoch 390/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.7330 - mae: 0.7330\n",
            "Epoch 391/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.4969 - mae: 1.4969\n",
            "Epoch 392/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.5048 - mae: 1.5048\n",
            "Epoch 393/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6718 - mae: 0.6718\n",
            "Epoch 394/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.4886 - mae: 0.4886\n",
            "Epoch 395/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7209 - mae: 0.7209\n",
            "Epoch 396/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.3310 - mae: 0.3310\n",
            "Epoch 397/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.3836 - mae: 0.3836\n",
            "Epoch 398/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5702 - mae: 0.5702\n",
            "Epoch 399/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8215 - mae: 0.8215\n",
            "Epoch 400/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.4320 - mae: 1.4320\n",
            "Epoch 401/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6228 - mae: 0.6228\n",
            "Epoch 402/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8793 - mae: 0.8793\n",
            "Epoch 403/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.7904 - mae: 1.7904\n",
            "Epoch 404/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.6741 - mae: 1.6741\n",
            "Epoch 405/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9114 - mae: 0.9114\n",
            "Epoch 406/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8643 - mae: 0.8643\n",
            "Epoch 407/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.3670 - mae: 1.3670\n",
            "Epoch 408/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.3299 - mae: 1.3299\n",
            "Epoch 409/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.4526 - mae: 1.4526\n",
            "Epoch 410/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.5037 - mae: 1.5037\n",
            "Epoch 411/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.0847 - mae: 1.0847\n",
            "Epoch 412/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.4471 - mae: 0.4471\n",
            "Epoch 413/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5099 - mae: 0.5099\n",
            "Epoch 414/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.2571 - mae: 0.2571\n",
            "Epoch 415/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.3236 - mae: 0.3236\n",
            "Epoch 416/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7369 - mae: 0.7369\n",
            "Epoch 417/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9124 - mae: 0.9124\n",
            "Epoch 418/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6486 - mae: 0.6486\n",
            "Epoch 419/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5205 - mae: 0.5205\n",
            "Epoch 420/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.3864 - mae: 1.3864\n",
            "Epoch 421/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.4968 - mae: 0.4968\n",
            "Epoch 422/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.3995 - mae: 0.3995\n",
            "Epoch 423/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.3447 - mae: 0.3447\n",
            "Epoch 424/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.3614 - mae: 0.3614\n",
            "Epoch 425/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.3518 - mae: 0.3518\n",
            "Epoch 426/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.0216 - mae: 1.0216\n",
            "Epoch 427/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.8521 - mae: 1.8521\n",
            "Epoch 428/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 1.8210 - mae: 1.8210\n",
            "Epoch 429/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6650 - mae: 0.6650\n",
            "Epoch 430/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.1348 - mae: 1.1348\n",
            "Epoch 431/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.4283 - mae: 1.4283\n",
            "Epoch 432/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.4035 - mae: 1.4035\n",
            "Epoch 433/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.1441 - mae: 1.1441\n",
            "Epoch 434/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.6566 - mae: 1.6566\n",
            "Epoch 435/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 2.4154 - mae: 2.4154\n",
            "Epoch 436/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5525 - mae: 0.5525\n",
            "Epoch 437/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5299 - mae: 0.5299\n",
            "Epoch 438/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.6620 - mae: 1.6620\n",
            "Epoch 439/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.0805 - mae: 1.0805\n",
            "Epoch 440/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9399 - mae: 0.9399\n",
            "Epoch 441/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.1249 - mae: 1.1249\n",
            "Epoch 442/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.3596 - mae: 1.3596\n",
            "Epoch 443/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9732 - mae: 0.9732\n",
            "Epoch 444/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.3262 - mae: 0.3262\n",
            "Epoch 445/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9514 - mae: 0.9514\n",
            "Epoch 446/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5672 - mae: 0.5672\n",
            "Epoch 447/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7293 - mae: 0.7293\n",
            "Epoch 448/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9255 - mae: 0.9255\n",
            "Epoch 449/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6140 - mae: 0.6140\n",
            "Epoch 450/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6410 - mae: 0.6410\n",
            "Epoch 451/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.1258 - mae: 1.1258\n",
            "Epoch 452/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.3800 - mae: 0.3800\n",
            "Epoch 453/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5449 - mae: 0.5449\n",
            "Epoch 454/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8596 - mae: 0.8596\n",
            "Epoch 455/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5760 - mae: 0.5760\n",
            "Epoch 456/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8289 - mae: 0.8289\n",
            "Epoch 457/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.4303 - mae: 0.4303\n",
            "Epoch 458/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.5973 - mae: 0.5973\n",
            "Epoch 459/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.4004 - mae: 0.4004\n",
            "Epoch 460/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.1202 - mae: 1.1202\n",
            "Epoch 461/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.4883 - mae: 0.4883\n",
            "Epoch 462/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.6413 - mae: 0.6413\n",
            "Epoch 463/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.2177 - mae: 1.2177\n",
            "Epoch 464/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 1.1949 - mae: 1.1949\n",
            "Epoch 465/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.4645 - mae: 0.4645\n",
            "Epoch 466/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.5532 - mae: 0.5532\n",
            "Epoch 467/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.6249 - mae: 0.6249\n",
            "Epoch 468/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7884 - mae: 0.7884\n",
            "Epoch 469/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8651 - mae: 0.8651\n",
            "Epoch 470/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 1.7020 - mae: 1.7020\n",
            "Epoch 471/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.5033 - mae: 1.5033\n",
            "Epoch 472/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.3690 - mae: 1.3690\n",
            "Epoch 473/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 1.8262 - mae: 1.8262\n",
            "Epoch 474/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.6697 - mae: 1.6697\n",
            "Epoch 475/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9391 - mae: 0.9391\n",
            "Epoch 476/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.7051 - mae: 0.7051\n",
            "Epoch 477/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.6033 - mae: 0.6033\n",
            "Epoch 478/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.6932 - mae: 0.6932\n",
            "Epoch 479/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.0749 - mae: 1.0749\n",
            "Epoch 480/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.6130 - mae: 0.6130\n",
            "Epoch 481/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 1.0032 - mae: 1.0032\n",
            "Epoch 482/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.4505 - mae: 1.4505\n",
            "Epoch 483/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.3857 - mae: 0.3857\n",
            "Epoch 484/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.9947 - mae: 0.9947\n",
            "Epoch 485/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.7193 - mae: 0.7193\n",
            "Epoch 486/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.4565 - mae: 0.4565\n",
            "Epoch 487/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.1971 - mae: 0.1971\n",
            "Epoch 488/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.3717 - mae: 0.3717\n",
            "Epoch 489/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9874 - mae: 0.9874\n",
            "Epoch 490/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.2534 - mae: 1.2534\n",
            "Epoch 491/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.2919 - mae: 1.2919\n",
            "Epoch 492/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.1442 - mae: 1.1442\n",
            "Epoch 493/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.4382 - mae: 1.4382\n",
            "Epoch 494/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.9113 - mae: 0.9113\n",
            "Epoch 495/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.6282 - mae: 0.6282\n",
            "Epoch 496/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 0.8873 - mae: 0.8873\n",
            "Epoch 497/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.4448 - mae: 1.4448\n",
            "Epoch 498/500\n",
            "16/16 [==============================] - 0s 2ms/step - loss: 1.6002 - mae: 1.6002\n",
            "Epoch 499/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.9710 - mae: 0.9710\n",
            "Epoch 500/500\n",
            "16/16 [==============================] - 0s 3ms/step - loss: 0.7066 - mae: 0.7066\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7fef5548e690>"
            ]
          },
          "metadata": {},
          "execution_count": 103
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model.predict([20])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UkE2tDbB0xgr",
        "outputId": "f449236e-995d-49d1-f35c-07bd55188133"
      },
      "execution_count": 104,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[29.979399]], dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 104
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#Split the data in to train and test sets\n",
        "\n",
        "X_train =X[:int(len(X)*0.8)]\n",
        "y_train =y[:int(len(y)*0.8)]\n",
        "\n",
        "X_test =X[int(len(X)*0.8):]\n",
        "y_test =y[int(len(y)*0.8):]"
      ],
      "metadata": {
        "id": "63l87mZf1U1p"
      },
      "execution_count": 199,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X_train[10],y_train[10]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5iDYZtxH9_oj",
        "outputId": "b7cc13c3-fe1f-4ce8-fdac-051eb8bba178"
      },
      "execution_count": 200,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(<tf.Tensor: shape=(), dtype=float16, numpy=-9960.0>,\n",
              " <tf.Tensor: shape=(), dtype=float16, numpy=-9950.0>)"
            ]
          },
          "metadata": {},
          "execution_count": 200
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.figure(figsize = (16,9))\n",
        "plt.scatter(X_train,y_train, c = 'blue' , label = 'training data')\n",
        "plt.scatter(X_test,y_test, c = 'red' , label = 'testing data')\n",
        "plt.legend();"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 533
        },
        "id": "PrHrZ_zf1VlM",
        "outputId": "95365baa-a1eb-4913-913c-4b7b70464a54"
      },
      "execution_count": 201,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA7kAAAIICAYAAACrRimkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdfZSdV2Ee+mdbSCAlbkYxvsCMAbupk+UZkwo8fGQR8sFMwJDbQLKSAJFuoDexwwzuum3v8sVcumJKLys0pKVh3Uipc0MhzDR8huI2tBgpQL6AWAYnRDKJBRjQyAHhSAlUAtnWvn+cc8TY1uc575k5H7/fWrN0Zp/3fWd7nWalT/bezym11gAAAMAouGi9JwAAAABNEXIBAAAYGUIuAAAAI0PIBQAAYGQIuQAAAIwMIRcAAICR8aj1nkC/PPaxj62XX375ek8DAACAPrjjjju+Vmu99OHjIxtyL7/88uzdu3e9pwEAAEAflFK+eLpx25UBAAAYGUIuAAAAI0PIBQAAYGSM7Jnc07n//vtz8ODBfPOb31zvqZDkMY95TC677LJs3LhxvacCAACMiLEKuQcPHszFF1+cyy+/PKWU9Z7OWKu15r777svBgwdzxRVXrPd0AACAETFW25W/+c1v5pJLLhFwB0ApJZdccolVdQAAoFFjFXKTCLgDxGcBAAA0bexC7no6evRodu7c2dW9L3zhC3P06NGzXvPLv/zL2b17d1fPP5u3ve1tueGGG856zUc/+tH86Z/+aeN/GwAA4EIIuWvobCH3gQceOOu9H/zgBzMxMXHWa17/+tdnfn6+6/n1QsgFAAAGgZB7FsvLyeWXJxdd1Pp3ebm3591000353Oc+l23btuXGG2/MRz/60TznOc/JT/zET2R6ejpJ8uIXvzjXXHNNZmZmcsstt5y69/LLL8/Xvva13HPPPbnqqqty3XXXZWZmJs973vNy/PjxJMkrXvGKvPe97z11/c0335ynPe1pecpTnpLPfvazSZLDhw/nx37sxzIzM5Nf/MVfzJOf/OR87Wtfe8Rc/9N/+k/53u/93jzjGc/In/zJn5wa/6//9b/mmc98Zp761Kdmfn4+X/nKV3LPPffkN3/zN/PmN78527Ztyx/90R+d9joAAIB+E3LPYHk5uf765ItfTGpt/Xv99b0F3Te+8Y35nu/5ntx5551505velCT51Kc+lV//9V/PX//1XydJ3vrWt+aOO+7I3r1785a3vCX33XffI55z991351WvelX27duXiYmJvO997zvt33vsYx+bT33qU1lYWMiv/dqvJUn+9b/+13nuc5+bffv25ad/+qfzpS996RH33Xvvvbn55pvzJ3/yJ/njP/7j7N+//9R7P/iDP5hPfOIT+fSnP52XvvSl+dVf/dVcfvnleeUrX5l/8S/+Re6888485znPOe11AAAA/TZWXyF0IV772uTYsYeOHTvWGt++vbm/84xnPOMhX6Hzlre8Je9///uTJF/+8pdz991355JLLnnIPVdccUW2bduWJLnmmmtyzz33nPbZP/VTP3Xqmt/7vd9LkvzxH//xqedfe+212bp16yPu++QnP5kf+ZEfyaWXXpokeclLXnIqhB88eDAveclLcu+99+bEiRNn/Pqf870OAACgSVZyz+A0C5xnHe/Wd3zHd5x6/dGPfjS7d+/Oxz/+8fz5n/95nvrUp572K3Ye/ehHn3q9YcOGM57n7Vx3tmsu1D/7Z/8sN9xwQz7zmc/kP/7H/3jGrwA63+sAAACaJOSewZOedGHj5+Piiy/O17/+9TO+/3d/93fZunVrtmzZks9+9rP5xCc+0f0fO4NnP/vZefe7350kue2223LkyJFHXPPMZz4zH/vYx3Lffffl/vvvz3ve856HzHFqaipJ8va3v/3U+MP/2850HQAAQD8JuWfwhjckW7Y8dGzLltZ4ty655JI8+9nPztVXX50bb7zxEe9fe+21eeCBB3LVVVflpptuyrOe9azu/9gZ3Hzzzbntttty9dVX5z3veU8e//jH5+KLL37INU94whPyute9Lj/wAz+QZz/72bnqqqtOvfe6170uP/MzP5Nrrrkmj33sY0+N/5N/8k/y/ve//1Tx1JmuAwAA6KdSa13vOfTF7Oxs3bt370PG7rrrrocEtnNZXm6dwf3Sl1oruG94Q7PncdfDt771rWzYsCGPetSj8vGPfzwLCwu58847120+F/qZAAAAJEkp5Y5a6+zDxxVPncX27cMfah/uS1/6Un72Z382J0+ezKZNm/Jbv/Vb6z0lAACAxjSyXbmU8tZSyldLKX+5auy7SykfLqXc3f53a3u8lFLeUko5UEr5i1LK01bd8/L29XeXUl6+avyaUspn2ve8pZRSmpj3OLryyivz6U9/On/+53+e22+/PU9/+tPXe0oAAMAgWFxMSnnoT7tnZ5g0dSb3bUmufdjYTUn21FqvTLKn/XuSvCDJle2f65PsSlqhOMnNSZ6Z5BlJbu4E4/Y116267+F/CwAAgG50wu2uXY9879ChoQu6jYTcWusfJvnbhw2/KEmnVvftSV68avx3assnkkyUUp6Q5PlJPlxr/dta65EkH05ybfu9f1Br/URtHSD+nVXPAgAAoBtnC7erHTq0NvNpSD/P5D6u1npv+/XfJHlc+/VUki+vuu5ge+xs4wdPM/4IpZTr01odzpN6+a4fAACAUbZhQ3Ly5HrPoi/W5CuE2iuwfa9xrrXeUmudrbXOXnrppf3+cwAAAMNly5bW6u2IBtykvyH3K+2txmn/+9X2+EqSJ6667rL22NnGLzvN+NA5evRodu7c2fX9/+E//IccO3bs1O8vfOELc/To0Sam9hCveMUr8t73vves17ztbW/LoSHbtgAAAGOrE26PH7/weycnm59PH/Uz5N6apNOQ/PIkH1g1/vPtluVnJfm79rbmDyV5Xilla7tw6nlJPtR+7+9LKc9qtyr//KpnDZWmQ+4HP/jBTExMNDG1CybkAgDAEJia6j7cJsnGjcnKcK0xNvUVQr+b5ONJvq+UcrCU8gtJ3pjkx0opdyeZb/+eJB9M8vkkB5L8VpLFJKm1/m2Sf5Pk9vbP69tjaV/z/7Xv+VyS/97EvM9peTm5/PLkoota/y4v9/S4m266KZ/73Oeybdu23HjjjUmSN73pTXn605+e7//+78/NN9+cJPmf//N/5sd//Mfzj//xP87VV1+dd73rXXnLW96SQ4cO5Ud/9Efzoz/6o0mSyy+/PF/72tdyzz335Kqrrsp1112XmZmZPO95z8vx9v8jvv322/P93//9p/7m1Vdf/Yh51Vpzww035Pu+7/syPz+fr371q6fee/3rX5+nP/3pufrqq3P99den1pr3vve92bt3b7Zv355t27bl+PHjp70OAABYJ51SqV4WpubmkhMnmpvTWqm1juTPNddcUx9u//79jxg7o6WlWrdsqTX59s+WLa3xLn3hC1+oMzMzp37/0Ic+VK+77rp68uTJ+uCDD9Yf//Efrx/72Mfqe9/73vqLv/iLp647evRorbXWJz/5yfXw4cOnxju/f+ELX6gbNmyon/70p2uttf7Mz/xMfcc73lFrrXVmZqb+6Z/+aa211le/+tUP+fsd73vf++r8/Hx94IEH6srKSv2u7/qu+p73vKfWWut999136rodO3bUW2+9tdZa6w//8A/X22+//dR7Z7ruXC7oMwEAAM7toosemmMu9Gdubr3/C85Lkr31NFlwTYqnhtJrX5us2hqcpPX7a1/b2J+47bbbctttt+WpT31qnva0p+Wzn/1s7r777jzlKU/Jhz/84bz61a/OH/3RH+W7vuu7zvmsK664Itu2bUuSXHPNNbnnnnty9OjRfP3rX88P/MAPJEl+7ud+7rT3/uEf/mFe9rKXZcOGDZmcnMxzn/vcU+995CMfyTOf+cw85SlPyR/8wR9k3759p33G+V4HAAD0Sa+lUnNzrZi7e3ez81pj/fwKoeH2pS9d2HgXaq15zWtek1/6pV96xHuf+tSn8sEPfjD/6l/9q8zNzeWXf/mXz/qsRz/60adeb9iw4dR25V5885vfzOLiYvbu3ZsnPvGJed3rXpdvfvObXV8HAAD0wZYt3Z+5TVrFUkN27vZsrOSeyZm+Z7eH79+9+OKL8/Wvf/3U789//vPz1re+Nd/4xjeSJCsrK/nqV7+aQ4cOZcuWLdmxY0duvPHGfOpTnzrt/ecyMTGRiy++OJ/85CeTJO985ztPe90P/dAP5V3velcefPDB3HvvvfnIRz6SJKeC6mMf+9h84xvfeEjj8uq5nO06AACgT3otlZqcbK3cjlDATazkntkb3pBcf/1Dtyxv2dIa79Ill1ySZz/72bn66qvzghe8IG9605ty1113ndpO/J3f+Z1ZWlrKgQMHcuONN+aiiy7Kxo0bs2vXriTJ9ddfn2uvvTaTk5Ongui5/PZv/3auu+66XHTRRfnhH/7h0259/smf/Mn8wR/8Qaanp/OkJz3p1HwmJiZy3XXX5eqrr87jH//4PP3pTz91zyte8Yq88pWvzObNm/Pxj3/8jNcBAAANm5rqrVBq48bhLJQ6T6WOaAvu7Oxs3bt370PG7rrrrlx11VXn/5Dl5dYZ3C99qbWC+4Y3JNu3NzzT/vrGN76R7/zO70ySvPGNb8y9996bX//1X1/nWX3bBX8mAAAwrnoNt0nr3O2Qn7ntKKXcUWudffi4ldyz2b596ELtw/3+7/9+fuVXfiUPPPBAnvzkJ+dtb3vbek8JAAC4UKX0dv8IhdtzEXJH3Ete8pK85CUvWe9pAAAA3VAqdcEUTwEAAAwapVJdG7uV3FprSq9L/TRiVM+DAwBA13o9d7t580PLc8fQWK3kPuYxj8l9990nXA2AWmvuu+++POYxj1nvqQAAwPrrrNz2EnAXFsY+4CZjtpJ72WWX5eDBgzl8+PB6T4W0/o8Ol1122XpPAwAA1s/8fLJnT2/PWFhIdu5sZj4jYKxC7saNG3PFFVes9zQAAAA0JvfJWIVcAACAddfrudsxbEy+EEIuAADAWlAqtSbGqngKAABgzSmVWlNCLgAAQD/MzzcTbmtVLHUBbFcGAABoksbkdSXkAgAANKGJcDs9nezb18x8xpSQCwAA0KsNG5KTJ7u/X6lUY5zJBQAA6FanVKqXgKtUqlFCLgAAwIVSKjWwbFcGAAA4X0qlBp6QCwAAcC5KpYaG7coAAABnsrzc2pbcS8Cdnm5tSxZw14SVXAAAgIdbXk527OjtGRMTyZEjzcyH82YlFwAAYLWtW3sPuAsLAu46EXIBAACSbzcmHz3a/TM0Jq8725UBAIDxplRqpFjJBQAAxpNSqZFkJRcAABgvTZRKWbkdWEIuAAAwHjQmjwUhFwAAGH1TU8mhQ93ff9FFyYMPNjcf+saZXAAAYHR1GpN7CbgLCwLuELGSCwAAjB6NyWPLSi4AADA6NCaPPSu5AADAaNi0Kbn//u7vVyo1EqzkAgAAw21mprV6223AnZhordwKuCPBSi4AADCcZmaS/fu7v3/jxuTEiebmw0CwkgsAAAyXzsptLwF3YUHAHVFCLgAAMBw6pVK9hNu5udbW5J07m5sXA8V2ZQAAYLAtLyc7dvT2jLm5ZPfuZubDQBNyAQCAwbV1a3L0aPf3T04mKyvNzYeBZ7syAAAweDrnbrsNuBs3trYlC7hjR8gFAAAGh1IpeiTkAgAA629xsZlwq1Rq7DmTCwAArJ/FxWTXrt6eoVSKVfq6kltK+b5Syp2rfv6+lPLPSymvK6WsrBp/4ap7XlNKOVBK+atSyvNXjV/bHjtQSrmpn/MGAAD6rPN1QL0E3MnJ1sqtgMsqfV3JrbX+VZJtSVJK2ZBkJcn7k/zTJG+utf7a6utLKdNJXppkJslkkt2llO9tv/0bSX4sycEkt5dSbq219rCXAQAAWBe9NiZPTCRHjjQ3H0bKWp7JnUvyuVrrF89yzYuSvLPW+q1a6xeSHEjyjPbPgVrr52utJ5K8s30tAAAwLKamemtMTpKlJQGXs1rLkPvSJL+76vcbSil/UUp5aylla3tsKsmXV11zsD12pnEAAGDQdUqlDh3q/hmdUqnt25ubFyNpTUJuKWVTkp9I8p720K4k35PWVuZ7k/y7hv7O9aWUvaWUvYcPH27ikQAAQLc64baXc7cak7lAa9Wu/IIkn6q1fiVJOv8mSSnlt5L8t/avK0meuOq+y9pjOcv4KbXWW5LckiSzs7O1qckDAAAXoInG5OnpZN++ZubDWFmrkPuyrNqqXEp5Qq313vavP5nkL9uvb03yn0sp/z6t4qkrk/xZkpLkylLKFWmF25cm+bk1mjsAAHC+Nm1K7r+/+/snJ5OVR6xnwXnre8gtpXxHWq3Iv7Rq+FdLKduS1CT3dN6rte4rpbw7yf4kDyR5Va31wfZzbkjyoSQbkry11ur/rAMAAINiaqq3M7cak2lIqXU0d/XOzs7WvXv3rvc0AABgtPUabktJ3vEOhVJcsFLKHbXW2YePr2W7MgAAMCrm53tvTF5aSk6eFHBp1FqdyQUAAEaBUikGnJALAACcn15LpYRb1oCQCwAAnN3WrcnRo93frzGZNeRMLgAAcHpTU61zt90G3FJa524FXNaQkAsAADxUr6VSnXCrVIp1YLsyAADQMj+f7NnT2zOWlgRb1pWQCwAA466JcKtUigEh5AIAwDjbsKG1rbhbwi0DRsgFAIBx1Gtj8sREcuRIc/OBhiieAgCAcTIz01tj8saNSa0CLgNLyAUAgHHQaUzev7/7ZywtJSdONDcn6APblQEAYJQ1USo1N5fs3t3MfKDPrOQCAMAoWl5urdz2EnDn5lpbkwVchoiVXAAAGDWbNiX339/9/RqTGWJCLgAAjAqNyWC7MgAADD2NyXCKkAsAAMOqE257aUxeWNCYzEgRcgEAYNh0SqV6CbedUqmdO5ubFwwAZ3IBAGBYLC8nO3b09gxfB8SIE3IBAGDQNRFuJyeTlZVm5gMDTMgFAIBBpjEZLogzuQAAMIg0JkNXhFwAABgki4sak6EHtisDAMAgWFxMdu3q7RlKpUDIBQCA9bS8nBzZsZhXZVdKtw9RKgWnCLkAALBOOp1SD3YbcIVbeAQhFwAA1tjUVHLo0Ld/v+CAu3lzcuxYk1OCkaF4CgAA1kinMHl1wL1gCwsCLpyFkAsAAH3WRGFyFhZaXwm0c2dj84JRJOQCAECfdMLtuUqTb8tc6sPGavsn09PCLVwAZ3IBAKAPNm1K7r///K69NrvzPzKf52XPqbG7JucyveLrgOBCCbkAANCgh5dKna9r0wq0ExPJkSPJdMPzgnFhuzIAADRgaqq3UqlSkqWlVsAFuifkAgBAD+bne29MXlpKTp5Mtm9vbl4wroRcAADoQqdUas+ec197Jp3CZOEWmuNMLgAAXIDFxXO3JZ/L9HSyb18z8wEeSsgFAIDzdCGNyaczOZmsrDQ3H+CRbFcGAIBz2Lq1tTW524DbKZUScKH/hFwAADiDTqnU0aPd3d8Jt0qlYO3YrgwAAA8zP99boVTSCreCLaw9IRcAANqaCLdzc8nu3c3MB7hwQi4AACTZsKG1rbhbGpNhMDiTCwDAWOuUSnUbcKenW991K+DCYLCSCwDAWNq6tftCqSTZvDk5dqy5+QDNsJILAMBY6bUxOWmVSgm4MJis5AIAMBaUSsF46PtKbinlnlLKZ0opd5ZS9rbHvruU8uFSyt3tf7e2x0sp5S2llAOllL8opTxt1XNe3r7+7lLKy/s9bwAARsPycmvltpeAOzfXOncr4MLgW6vtyj9aa91Wa51t/35Tkj211iuT7Gn/niQvSHJl++f6JLuSVihOcnOSZyZ5RpKbO8EYAABOpxNud+zo/hmdUinhFobHem1XflGSH2m/fnuSjyZ5dXv8d2qtNcknSikTpZQntK/9cK31b5OklPLhJNcm+d21nTYAAMNgy5bk+PHu75+YSI4caW4+wNpZi5XcmuS2UsodpZTr22OPq7Xe2379N0ke1349leTLq+492B470/hDlFKuL6XsLaXsPXz4cJP/DQAADIGZmdbqbbcBd2KitXIr4MLwWouV3B+sta6UUv6XJB8upXx29Zu11lpKqU38oVrrLUluSZLZ2dlGngkAwOCbmUn27+/tGdX/7xFGQt9XcmutK+1/v5rk/Wmdqf1Kexty2v9+tX35SpInrrr9svbYmcYBABhjnXO3vQTcTqkUMBr6GnJLKd9RSrm48zrJ85L8ZZJbk3Qakl+e5APt17cm+fl2y/Kzkvxde1vzh5I8r5SytV049bz2GAAAY6iJUimNyTCa+r1d+XFJ3l9K6fyt/1xr/R+llNuTvLuU8gtJvpjkZ9vXfzDJC5McSHIsyT9Nklrr35ZS/k2S29vXvb5TQgUAwPhYXu4t2CbJ5GSyYk8gjKxSR3RvxuzsbN27d+96TwMAgIZs3ZocPdr9/RqTYbSUUu5Y9TW1p6zV9+QCAEBXOo3J3QbcjRs1JsM4EXIBABhInXDbS6nUwkJy4kRzcwIGn5ALAMBAWVxsJtzWmuzc2dy8gOGwFt+TCwAA56RUCmiCkAsAwLrrtVRKuAU6hFwAANbN1FRy6FD392/enBw71tx8gOHnTC4AAGuuUyrVS8BdWBBwgUcScgEAWDNKpYB+s10ZAIC+W1xMdu3q7RkLC4ItcG5CLgAAfdNEuFUqBVwIIRcAgL7YsiU5frz7+ycmkiNHmpsPMB6cyQUAoFFTU61zt90G3FKSpSUBF+iOkAsAQCOaaExeWkpOnky2b29uXsB4EXIBAOhJk43Jwi3QK2dyAQDoyvJysmNHb8+Ynk727WtmPgCJkAsAwAUSboFBJuQCAHDeNCYDg86ZXAAAzqlTKqUxGRh0Qi4AAGc0P997qZTGZGAtCbkAADxCJ9zu2dP9MzQmA+vBmVwAAE5RKgUMOyEXAIAkvZdKTU4mKyvNzQegG7YrAwCMuamp3kqlJiZa25IFXGAQCLkAAGOq05h86FD3z9CYDAwaIRcAYMw00ZisVAoYVM7kAgCMicXFZNeu3p4xN5fs3t3MfAD6QcgFABhxGpOBcSLkAgCMsK1bk6NHu79fYzIwbIRcAIARNDXVW6HU5s3JsWPNzQdgrSieAgAYIU01Jgu4wLAScgEARsDiYu+NyXNzGpOB4We7MgDAEGuiVEpjMjBKhFwAgCGkMRng9IRcAIAhs2VLcvx49/crlQJGmTO5AABDolMq1W3ALUWpFDD6hFwAgAE3P997qdTSUnLypFIpYPQJuQAAA6oTbvfs6f4ZCwsak4Hx4kwuAMCAUSoF0D0hFwBggPRaKjU5maysNDcfgGFjuzIAwACYmuqtVGpiorUtWcAFxp2QCwCwjjrh9tCh7p+xtJQcOdLcnACGmZALALAOOqVSvYRbpVIAj+RMLgDAGlpcTHbt6u0Zc3PJ7t3NzAdg1Ai5AABroInGZOEW4NyEXACAPtOYDLB2hFwAgD6ZmurtzO3mzcmxY83NB2Ac9K14qpTyxFLKR0op+0sp+0op/0d7/HWllJVSyp3tnxeuuuc1pZQDpZS/KqU8f9X4te2xA6WUm/o1ZwCAJszMNNOYLOACXLh+tis/kOT/rLVOJ3lWkleVUqbb77251rqt/fPBJGm/99IkM0muTbKzlLKhlLIhyW8keUGS6SQvW/UcAICBsbjYCrf793f/DI3JAL3p23blWuu9Se5tv/56KeWuJFNnueVFSd5Za/1Wki+UUg4keUb7vQO11s8nSSnlne1re/hfHwAAzdGYDDA41uR7cksplyd5apJPtoduKKX8RSnlraWUre2xqSRfXnXbwfbYmcYBANbV8nJr5baXgDs93Vq5FXABmtH3kFtK+c4k70vyz2utf59kV5LvSbItrZXef9fg37q+lLK3lLL38OHDTT0WAOARtmzp7SuBJiZa4XbfvubmBECfQ24pZWNaAXe51vp7SVJr/Uqt9cFa68kkv5Vvb0leSfLEVbdf1h470/gj1FpvqbXO1lpnL7300mb/YwAA8u1SqW6/Emjz5la4PXKk2XkB0NLPduWS5LeT3FVr/ferxp+w6rKfTPKX7de3JnlpKeXRpZQrklyZ5M+S3J7kylLKFaWUTWmVU93ar3kDAJxOJ9z2UiqlMRmg//r5PbnPTvK/JflMKeXO9tj/nVY78rYkNck9SX4pSWqt+0op706rUOqBJK+qtT6YJKWUG5J8KMmGJG+ttdrYAwCsifn5ZM+e3p6xsJDs3NnMfAA4u1JrXe859MXs7Gzdu3fvek8DABhSy8u9nblNWqVSztwC9Ecp5Y5a6+zDx/u5kgsAMJS2bOn+zG0i3AKsJyEXAKBtaio5dKj7+ycnk5XT1mMCsFaEXABg7PUabi+6KHnwwebmA0D3+v49uQAAg2p+vtWY3EvAXVgQcAEGiZALAIydxcVWuO2lNXlhofV9t1qTAQaL7coAwNhoojF5bi7ZvbuZ+QDQPCEXABgLvTYmK5UCGA5CLgAw0notldq8OTl2rLn5ANBfzuQCACNpZqb3UqmlJQEXYNgIuQDASOmUSu3f3/0zOqVS27c3Ny8A1obtygDASFhcTHbt6u0ZSqUAhp+QCwAMtSYak6enk337mpkPAOtLyAUAhlavjckTE8mRI83NB4D150wuADB0OqVS3QbczZtbZ24FXIDRI+QCAEOjE257KZXSmAww2oRcAGDgzc9rTAbg/DiTCwAMLKVSAFwoIRcAGEi9lkoJtwDjScgFAAbK1q3J0aPd3z85maysNDcfAIaLkAsADISpqeTQoe7vv+ii5MEHm5sPAMNJ8RQAsK46pVK9BNyFBQEXgBYhFwBYF4uLrXC7Z0/3z+g0Ju/c2dy8ABhutisDAGtqcTHZtau3Z8zNJbt3NzMfAEaLkAsArJlNm5L77+/+fqVSAJyL7coAQN9NTbW2JncbcCcmWtuSBVwAzkXIBQD6Zmamt1KpjRtb4fbIkWbnBcDoEnIBgMZ1wu3+/d0/Y2EhOXGiuTkBMB6EXACgMZ3G5F7C7dycxmQAuqd4CgDo2fJysmNHb8+Ynk727WtmPgCMLyEXAOjJli3J8ePd368xGYAm2QitOFIAACAASURBVK4MAHSlc+6224C7ebPGZACaJ+QCABekiVKppaXk2LHm5gQAHUIuAHBe5uebaUyuNdm+vbl5AcBqzuQCAGfVRKnU3Fyye3cz8wGAsxFyAYAz6rVUSmMyAGtNyAUAHmHr1uTo0e7v15gMwHoRcgGAU6amkkOHur//oouSBx9sbj4AcKEUTwEAp0qlegm4CwsCLgDrT8gFgDG2uNgKt3v2dP+MTmPyzp3NzQsAumW7MgCMocXFZNeu3p6hMRmAQSTkAsCY2bAhOXmy+/uVSgEwyGxXBoAxMTXV2prcbcCdmGhtSxZwARhkQi4AjLiZmd5KpTZubIXbI0eanRcA9IOQCwAjqhNu9+/v/hkLC8mJE83NCQD6TcgFgBHTaUzuJdzOzWlMBmA4KZ4CgBGxvJzs2NHbM6ank337mpkPAKwHIRcARsCWLcnx493fL9wCMCqGZrtyKeXaUspflVIOlFJuWu/5AMAg6DQmdxtwN29ubUsWcAEYFUMRckspG5L8RpIXJJlO8rJSyvT6zgoA1k+vjclJsrSUHDvW3JwAYBAMRchN8owkB2qtn6+1nkjyziQvWuc5AcCam59vpjG51mT79ubmBQCDYlhC7lSSL6/6/WB77CFKKdeXUvaWUvYePnx4zSYHAP22vNwKt3v2dP8MjckAjINhCbnnpdZ6S611ttY6e+mll673dACgZ51w20tr8vR0K9zu3t3cvABgUA1Lu/JKkieu+v2y9hgAjKxeG5MnJ5MV/9sSgDEzLCu5tye5spRyRSllU5KXJrl1necEAH3Ra2Pyxo2tlVsBF4BxNBQht9b6QJIbknwoyV1J3l1r9WUHAIyUTqlUr43JJ040NycAGDbDsl05tdYPJvnges8DAJo2P99boVTSakxWKAUAQxRyAWDULC4mu3b19oy5OYVSALCakAsA62DDhuTkye7vVyoFAKc3FGdyAWBUdEqlug24k5NKpQDgbKzkAsAamJrqrVBq40aFUgBwPqzkAkAfzcz03pi8sCDgAsD5EnIBoA8WF1vhdv/+7p8xN9famqw1GQDOn+3KANCg5eVkx47enqExGQC6J+QCQEO2bEmOH+/+/unpZN++5uYDAONIyAWAHvVaKrV5c3LsWHPzAYBx5kwuAHSp11KpUpKlJQEXAJok5ALABZqf771Uammp9V2527c3Ny8AQMgFgPO2vNwKt3v2dP+MTmOycAsA/eFMLgCcQxONyUqlAGBtCLkAcBa9NiZPTiYrK83NBwA4OyEXAE6j18bkjRuTEyeamw8AcH6cyQWAVTqlUr0E3KUlARcA1ouQCwD5drjtpVRqYUGpFACsN9uVARhri4vJrl29PWNuLtm9u5n5AAC9EXIBGFsbNrS+q7ZbSqUAYPDYrgzA2Jmaam1N7jbgTk62tiULuAAweKzkAjA2em1M3rw5OXasufkAAM2zkgvAyJuZaaYxWcAFgMEn5AIwshYXW+F2//7unzE3pzEZAIaJ7coAjJzl5WTHjt6eoTEZAIaTkAvAyGgi3E5PJ/v2NTMfAGDtCbkAjIStW5OjR7u/X6kUAIwGZ3IBGGqdUqluA24pSqUAYJQIuQAMpfn53kullpZa35WrVAoARoftygAMlfn5ZM+e3p6hVAoARpeQC8BQUCoFAJwPIReAgbdlS3L8ePf3T04mKyvNzQcAGFxCLgADa2oqOXSo+/s1JgPA+FE8BcDA6TQm9xJwNSYDwHgScgEYGE00Ji8sJLVqTAaAcWW7MgDrbnEx2bWrt2doTAYAEiEXgHW2YUPru2q7pTEZAFjNdmUA1sXUVGtrcrcBd3KytS1ZwAUAVrOSC8Ca0pgMAPSTlVwA1oTGZABgLQi5APTV4qLGZABg7diuDEBfLC8nO3b09gyNyQDAhRJyAWhUE+FWYzIA0C0hF4DGbNmSHD/e/f1KpQCAXjmTC0DPOqVS3QbcUpRKAQDNEHIB6Nr8fO+lUktLre/KVSoFADTBdmUALtj8fLJnT2/PUCoFAPSDkAvAeVMqBQAMur5sVy6lvKmU8tlSyl+UUt5fSploj19eSjleSrmz/fObq+65ppTymVLKgVLKW0oppT3+3aWUD5dS7m7/u7Ufcwbg7LZs6S3gTk62vutWwAUA+qlfZ3I/nOTqWuv3J/nrJK9Z9d7naq3b2j+vXDW+K8l1Sa5s/1zbHr8pyZ5a65VJ9rR/B2CNTE31Vio1MdEKtysrzc4LAOB0+hJya6231VofaP/6iSSXne36UsoTkvyDWusnaq01ye8keXH77RcleXv79dtXjQPQR53G5EOHun/G0lJy5EhzcwIAOJe1aFf+35P891W/X1FK+XQp5WOllOe0x6aSHFx1zcH2WJI8rtZ6b/v13yR53Jn+UCnl+lLK3lLK3sOHDzc0fYDx0kRj8sJCa/VWYzIAsNa6Lp4qpexO8vjTvPXaWusH2te8NskDSZbb792b5Em11vtKKdck+S+llJnz/Zu11lpKqWd5/5YktyTJ7OzsGa8D4JEWF5Ndu3p7hsZkAGC9dR1ya63zZ3u/lPKKJP9rkrn2FuTUWr+V5Fvt13eUUj6X5HuTrOShW5ova48lyVdKKU+otd7b3tb81W7nDMDpbdjQ+q7abmlMBgAGRb/ala9N8n8l+Yla67FV45eWUja0X//DtAqmPt/ejvz3pZRntVuVfz7JB9q33Zrk5e3XL181DkCPtm5tbU3uNuBqTAYABk2/vif3/03y6CQfbn8T0CfaTco/lOT1pZT7k5xM8spa69+271lM8rYkm9M6w9s5x/vGJO8upfxCki8m+dk+zRlgbExN9VYotXlzcuzYua8DAFhrfQm5tdZ/dIbx9yV53xne25vk6tOM35dkrtEJAoypmZneCqWSVmOyQikAYFCtRbsyAOtscVFjMgAwHvq1XRmAAaAxGQAYN0IuwAhaXk527OjtGRqTAYBhJOQCjJgtW5Ljx7u/f2IiOXKkufkAAKwlZ3IBRsTMTOvcbbcBd/Pm1plbARcAGGZCLsCQ64TbXkqllpZ8JRAAMBqEXIAhNT/fe7idm9OYDACMFmdyAYaMUikAgDMTcgGGSK+lUsItADDqhFyAITA1lRw61P39GpMBgHHhTC7AAOuUSvUScJeWBFwAYHwIuQADqIlSqYUFpVIAwPgRcgEGyOJiK9zu2dP9MzrhdufO5uYFADAsnMkFGBAbNiQnT3Z/v1IpAAAhF2Ddbd2aHD3a/f2Tk8nKSnPzAQAYZkIuwDrptTF58+bk2LHm5gMAMAqcyQVYY001Jgu4AACPJOQCrJFOqZTGZACA/rFdGaDPFheTXbt6e8bcXLJ7dzPzAQAYZUIuQJ8sLyc7dvT2DI3JAAAXRsgF6IMtW5Ljx7u/f2IiOXKkufkAAIwLZ3IBGtQpleo24G7e3DpzK+ACAHRHyAVoQCfc9lIqpTEZAKB3Qi5AD+bnNSYDAAwSZ3IBuqBUCgBgMAm5ABeo11Ip4RYAoH+EXIDztHVrcvRo9/drTAYA6D9ncgHOoXPutpeAu7Qk4AIArAUhF+AMOuF2z57un6FUCgBgbdmuDPAw8/O9BdukFW537mxmPgAAnD8hF6BNYzIAwPATcgHSe2Py5GSystLcfAAA6I6QC4w1jckAAKNF8RQwlmZmem9MrlXABQAYNEIuMFY6jcn793f/jLm5VsAFAGDwCLnAWFhe7v3rgDrhdvfu5uYFAECznMkFRloTjclzc4ItAMCwEHKBkbVpU3L//d3fr1QKAGD42K4MjJxOqVS3AXdiQqkUAMCwspILjIyZmd4KpTZuTE6caG4+AACsPSu5wNBbXOy9MXlhQcAFABgFVnKBoaVUCgCAhxNygaHTRLidnExWVpqZDwAAg0PIBYbK1q3J0aPd3y/cAgCMNiEXGApTU8mhQ93fr1QKAGA8KJ4CBlrn64B6CbhKpQAAxkffQm4p5XWllJVSyp3tnxeueu81pZQDpZS/KqU8f9X4te2xA6WUm1aNX1FK+WR7/F2llE39mjcwGJpqTK412bmzuXkBADDY+r1d+c211l9bPVBKmU7y0iQzSSaT7C6lfG/77d9I8mNJDia5vZRya611f5J/237WO0spv5nkF5Ls6vPcgXWwuJjs6vF/up27BQAYX+txJvdFSd5Za/1Wki+UUg4keUb7vQO11s8nSSnlnUleVEq5K8lzk/xc+5q3J3ldhFwYOVu2JMePd3+/cAsAQL/P5N5QSvmLUspbSylb22NTSb686pqD7bEzjV+S5Git9YGHjQMjYmqqtTW524A7MdHalizgAgDQU8gtpewupfzlaX5elNZK6/ck2Zbk3iT/roH5nms+15dS9pZS9h4+fLjffw7oUSfc9lIqtbSUHDnS3JwAABhuPW1XrrXOn891pZTfSvLf2r+uJHniqrcva4/lDOP3JZkopTyqvZq7+vqHz+eWJLckyezsbD3P/wxgjTVx7nZhQaEUAACP1M925Ses+vUnk/xl+/WtSV5aSnl0KeWKJFcm+bMktye5st2kvCmtcqpba601yUeS/HT7/pcn+UC/5g30T6cxuZeAqzEZAICz6Wfx1K+WUrYlqUnuSfJLSVJr3VdKeXeS/UkeSPKqWuuDSVJKuSHJh5JsSPLWWuu+9rNeneSdpZT/J8mnk/x2H+cNNKyJldvp6WTfvnNfBwDAeCuthdLRMzs7W/fu3bve04Cxt2lTcv/93d8/MeHMLQAAj1RKuaPWOvvw8X63KwNjqlMq1W3ALUWpFAAAF07IBRo1P99bY3In3J48mWzf3uzcAAAYff08kwuMkfn5ZM+e3p6xtCTYAgDQGyEX6IlSKQAABomQC3St11Ip4RYAgKYJucAF27o1OXq0+/snJ5OVlebmAwAAHYqngPO2dWurGKrbgNsplRJwAQDoFyEXOKdOY3Ivq7cakwEAWAu2KwNn1ERj8txcsnt3M/MBAIBzEXKBR1heTnbs6O0ZSqUAAFgPQi7wEFu2JMePd3+/cAsAwHoScoEkvTcmT0wkR440Nx8AAOiG4ikYczMzvZVKbd6c1CrgAgAwGIRcGFOdxuT9+7t/xtJScuxYc3MCAIBe2a4MY0ZjMgAAo8xKLoyJ5eXWym0vAXdurrU1WcAFAGBQWcmFMbBpU3L//d3fr1QKAIBhYSUXRlinVKrbgDsxoVQKAIDhYiUXRtDMTG+FUhs3JidONDcfAABYK1ZyYYR0Vm57CbgLCwIuAADDS8iFEdApleol3HZKpXbubG5eAACw1mxXhiG2vJzs2NHbM3wdEAAAo0TIhSG1dWty9Gj3909OJisrzc0HAAAGgZALQ2ZqKjl0qPv7lUoBADDKnMmFIdEpleol4CqVAgBg1Am5MOAWF5tpTFYqBQDAOLBdGQbU4mKya1dvz1hYEGwBABgvQi4MmCYak5VKAQAwroRcGCC9NiZPTCRHjjQ3HwAAGDbO5MIAmJpqnbvtJeAuLQm4AAAg5MI66pRK9dqYXGuyfXtz8wIAgGFluzKsA6VSAADQH0IurKEmwu30dLJvXzPzAQCAUSPkwhrZtCm5//7u79eYDAAA5+ZMLvRZp1Sq24A7MdE6cyvgAgDAuQm50CedcNttqVQpGpMBAOBCCbnQsPn53huTl5aSkyc1JgMAwIVyJhcaolQKAADWn5ALDei1VEq4BQCAZtiuDD3YurW3Uqnp6VaplIALAADNsJILXdi6NTl6tPv7N25MTpxobj4AAECLlVy4AJ1SqW4DbqcxWcAFAID+sJIL52F+Ptmzp7dnLC1pSwYAgH4TcuEsmgi3c3PJ7t3NzAcAADg7IRfOYMOG1nfVdktjMgAArD0hFx6m11KpiYnkyJHm5gMAAJw/xVPQNjPTW6nU5s2trwMScAEAYP30JeSWUt5VSrmz/XNPKeXO9vjlpZTjq977zVX3XFNK+Uwp5UAp5S2llNIe/+5SyodLKXe3/93ajzkzvjrhdv/+7p+xtJQcO9bcnAAAgO70JeTWWl9Sa91Wa92W5H1Jfm/V25/rvFdrfeWq8V1JrktyZfvn2vb4TUn21FqvTLKn/Tv0rPN1QL2E27m51uqt1mQAABgMfd2u3F6N/dkkv3uO656Q5B/UWj9Ra61JfifJi9tvvyjJ29uv375qHLqyvNwKt720JnfCrdZkAAAYLP0+k/ucJF+ptd69auyKUsqnSykfK6U8pz02leTgqmsOtseS5HG11nvbr/8myePO9MdKKdeXUvaWUvYePny4of8ERkUn3O7Y0f0zpqeFWwAAGGRdtyuXUnYnefxp3nptrfUD7dcvy0NXce9N8qRa632llGuS/JdSysz5/s1aay2l1LO8f0uSW5Jkdnb2jNcxfqamkkOHur9fYzIAAAyHrkNurXX+bO+XUh6V5KeSXLPqnm8l+Vb79R2llM8l+d4kK0kuW3X7Ze2xJPlKKeUJtdZ729uav9rtnBk/MzO9nbnduDE5caK5+QAAAP3Vz+3K80k+W2s9tQ25lHJpKWVD+/U/TKtg6vPt7ch/X0p5Vvsc788n6awG35rk5e3XL181DmfURGPywoKACwAAw6brldzz8NI8snDqh5K8vpRyf5KTSV5Za/3b9nuLSd6WZHOS/97+SZI3Jnl3KeUXknwxrSIrOK3l5d7O3CatUilnbgEAYDiVVpnx6Jmdna179+5d72mwRoRbAAAYL6WUO2qtsw8f7+dKLvRdE+F2cjJZWTn3dQAAwOATchlaW7cmR492f79SKQAAGD39/p5caFynVKqXgKtUCgAARpOQy9BYXGymMbnWZOfO5uYFAAAMDtuVGXiLi8muXb09Y2FBsAUAgHEg5DKwlEoBAAAXSshlIPVaKiXcAgDAeHIml4HSa6nU5s2tM7cCLgAAjCchl4HQCbe9lkodO9bcnAAAgOEj5LKuNCYDAABNciaXddFEY/L0dLJvXzPzAQAARoOQy5rbsCE5ebL7+5VKAQAAZ2K7Mmumc+6224A7MaFUCgAAODshl75rolRqaSk5cqS5OQEAAKNJyKVvmiiVWlpqrd5u397cvAAAgNHlTC6Na6JUamlJsAUAAC6ckEtjmgi3c3PJ7t3NzAcAABg/Qi6N6LUxWbgFAACaIOTSky1bkuPHu7//oouSBx9sbj4AAMB4UzxFVzqlUr0E3KUlARcAAGiWlVwuiFIpAABgkAm5nBelUgAAwDAQcjknpVIAAMCwcCaXM9qypXXuttuAOzeX1CrgAgAAa8dKLo8wNZUcOtT9/RMTyZEjzc0HAADgfFnJ5ZSpqdbKbS8Bd2lJwAUAANaPkMuprwPqJdx2tiZrTQYAANaT7cpjTqkUAAAwSqzkjimlUgAAwCiykjtmtmxJjh/v/v7JyWRlpbn5AAAANEnIHRO9NiZv3pwcO9bcfAAAAPrBduUR10Rj8sKCgAsAAAwHK7kjrJTe7lcqBQAADBsruSOoUyrVLaVSAADAsLKSO0KUSgEAAOPOSu4I6Jy77TbgTk62Vm4FXAAAYNhZyR1iGpMBAAAeykruENKYDAAAcHpC7hCZn28m3Naa7NzZ3LwAAAAGhe3KQ6LXrwNSKgUAAIwDK7kDrrM1uVtKpQAAgHFiJXdAKZUCAAC4cFZyB4xSKQAAgO4JuQNCqRQAAEDvbFdeZ/PzyZ49vT1jYUGwBQAASITcddNEuJ2eTvbta2Y+AAAAo6Cn7cqllJ8ppewrpZwspcw+7L3XlFIOlFL+qpTy/FXj17bHDpRSblo1fkUp5ZPt8XeVUja1xx/d/v1A+/3Le5nzINi0qbeAu3lza1uygAsAAPBQvZ7J/cskP5XkD1cPllKmk7w0yUySa5PsLKVsKKVsSPIbSV6QZDrJy9rXJsm/TfLmWus/SnIkyS+0x38hyZH2+Jvb1w2tqank/vu7v1+pFAAAwJn1FHJrrXfVWv/qNG+9KMk7a63fqrV+IcmBJM9o/xyotX6+1noiyTuTvKiUUpI8N8l72/e/PcmLVz3r7e3X700y175+KHVbLKVUCgAA4Nz61a48leTLq34/2B470/glSY7WWh942PhDntV+/+/a148F4RYAAOD8nbN4qpSyO8njT/PWa2utH2h+St0rpVyf5PokedKTnrTOs+mNUikAAIALd86QW2ud7+K5K0meuOr3y9pjOcP4fUkmSimPaq/Wrr6+86yDpZRHJfmu9vWnm+stSW5JktnZ2drFvPuulNbK7JkItwAAAN3r13blW5O8tN2MfEWSK5P8WZLbk1zZblLelFY51a211prkI0l+un3/y5N8YNWzXt5+/dNJ/qB9/VB6xztOPz4xoTEZAACgV71+hdBPllIOJvmBJL9fSvlQktRa9yV5d5L9Sf5HklfVWh9sr9LekORDSe5K8u72tUny6iT/spRyIK0zt7/dHv/tJJe0x/9lklNfOzSMtm9PlpaSJz+5tar75Ce3fj9yZL1nBgAAMPzKEC+KntXs7Gzdu3fvek8DAACAPiil3FFrnX34eL+2KwMAAMCaE3IBAAAYGUIuAAAAI0PIBQAAYGQIuQAAAIwMIRcAAICRIeQCAAAwMoRcAAAARoaQCwAAwMgQcgEAABgZQi4AAAAjQ8gFAABgZAi5AAAAjAwhFwAAgJEh5AIAADAyhFwAAABGhpALAADAyCi11vWeQ1+UUg4n+eJ6z+McHpvka+s9CR7CZzJ4fCaDyecyeHwmg8nnMnh8JoPJ5zJ4huEzeXKt9dKHD45syB0GpZS9tdbZ9Z4H3+YzGTw+k8Hkcxk8PpPB5HMZPD6TweRzGTzD/JnYrgwAAMDIEHIBAAAYGULu+rplvSfAI/hMBo/PZDD5XAaPz2Qw+VwGj89kMPlcBs/QfibO5PL/t3d3oXJdZRjH/w+NiVD7kVQpIa0mR2qxVzYNkou2iClpErTxo5V402gF8QssIhIJSG+r1IuiGBCLrdS2Wi0GQWyqrV4lrY2nSWqaZk6smJAmYLURlGjx9WK9I+scZs6Xe86emTw/WJw17957svde867Ze2bNipmZmZmZ2djwN7lmZmZmZmY2NnyT2yBJd0h6UdJ/JG2YseyrkjqSjkm6tYpvyVhH0q4qvk7SgYw/Jml5xlfk404uX7tUxzfq8rxNZnlF0mTG10r6Z7VsT7XNDZIO5/m+X5IyvkrSPknH8+/Kto5r1Em6R9Kp6vxvq5Y1kje2MJK+IeklSYckPSHp8ow7V4ZUv5yw5km6WtLTkv6Q7/lfzHhjfZktTr63H87z/7uM9eyDVNyf5/6QpPXV8+zM9Y9L2tnW8Yw6SddW+TAp6Zyku50rS0/SA5LOSjpSxRrLjX7XAK2KCJeGCvBu4FrgGWBDFb8OeAFYAawDpoCLskwBE8DyXOe63OZHwI6s7wE+m/XPAXuyvgN4rO3jHsUC3Ad8LetrgSN91nsW2AgI+AWwNeNfB3ZlfRdwb9vHNKoFuAf4co94Y3njsuA22Qwsy/q93de3c2U4y2w54TKQ870aWJ/1S4CXs79qrC9zWXTbvAK8dUasZx8EbMu+Stl3Hcj4KuBE/l2Z9ZVtH9uol3y9vwq8w7nSyvm/GVhfv4c3mRv9rgHaLP4mt0ERcTQijvVYtB14NCLOR8QfgQ7w3iydiDgREf8CHgW256cf7wcez+0fBD5UPdeDWX8c2DQUn5aMkDxfHwMemWO91cClEbE/SgY/RO92qNvHmtNk3tgCRMSTEfFGPtwPXDXb+s6V1vXMiZb3aWxFxOmIOJj1vwNHgTWzbLKgvmywe39B6tcHbQceimI/cHn2ZbcC+yLitYj4K7AP2LLUOz2GNgFTEfGnWdZxrgxIRPwWeG1GuJHcmOMaoDW+yV0aa4A/V49PZqxf/Argb9VFZjc+7bly+eu5vs3fTcCZiDhexdZJ+r2k30i6KWNrKOe+q26HKyPidNZfBa4c6B6Pvy/kkJgHquGsTeaNLd5dlE9lu5wrw6dfTtiAqfxk6HrgQIaa6Mts8QJ4UtLzkj6dsX59kNtlae1g+pcLzpX2NZUbs10DtMY3uQsk6SlJR3oUf6I0BObZPh9nekd7Gnh7RFwPfAn4oaRL5/tv5qdWnqZ8FnO0y3eAdwLvobTFfa3u7AViPrkiaTfwBvBwhpwrZknSW4CfAHdHxDnclw2DGyNiPbAV+Lykm+uF7oPaoTI/xm3AjzPkXBky45gby9regVETEbcsYrNTwNXV46syRp/4XyhDA5blt1L1+t3nOilpGXBZrm/M3T55zj4C3FBtcx44n/XnJU0B76Kc63qYZt0OZyStjojTOUzjbHNHMX7mmzeSvgv8PB82mTc2wzxy5RPAB4BN+ebnXBles+WKDYCkN1FucB+OiJ8CRMSZavn/05fZIkXEqfx7VtITlGGu/fqgfu1yCnjfjPgzA971cbcVONjNEefK0GgqN2a7BmiNv8ldGnuBHSozI68DrqH8QPs54BqVGWGXU4Zy7M0LyqeB23P7ncDPqufqzmZ2O/Dr7gWozcstwEsR8b9hFZLeJumirE9Q2udEDuE4J2lj/t7zTnq3Q90+tkDZsXZ9GOjO/Ndk3tgCSNoCfAW4LSL+UcWdK8OpZ060vE9jK1/j3wOORsQ3q3gjfdlSHMM4knSxpEu6dcoEekfo3wftBe5UsRF4PfuyXwKbJa3MYbSbM2aLN20EnXNlaDSSG3NcA7RnUDNaXYiFkqgnKd90nKE0fHfZbsrMcMeoZhyjzGD2ci7bXcUnKIndoQzvWJHxN+fjTi6faPu4R6kA3wc+MyP2UeBFYBI4CHywWraB0vlOAd8ClPErgF8Bx4GngFVtH9uoFuAHwGHgEKVjXV0tayRvXBbcJh3K724ms3RndHeuDGnplxMuAznXN1KG9R2qcmRbk32Zy6LaZYIy6+4L2U/tznjPPogyC+y389wfZvr/inFX9oMd4JNtH9soF+Biykiry6qYc2Xp2+ERytDwf1PuVT7VZG70uwZos3QvQszMzMzMzMxGLvzdggAAAFtJREFUnocrm5mZmZmZ2djwTa6ZmZmZmZmNDd/kmpmZmZmZ2djwTa6ZmZmZmZmNDd/kmpmZmZmZ2djwTa6ZmZmZmZmNDd/kmpmZmZmZ2djwTa6ZmZmZmZmNjf8C3fLyqRxKgF4AAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 1152x648 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model = tf.keras.Sequential([\n",
        "                             tf.keras.layers.Dense(50,activation =None),\n",
        "                             tf.keras.layers.Dense(1)\n",
        "])\n",
        "\n",
        "model.compile(loss = 'mae',\n",
        "              optimizer = tf.keras.optimizers.Adam(learning_rate = 0.001),\n",
        "              metrics = ['mae']\n",
        "              )\n",
        "\n",
        "model.fit(tf.expand_dims(X_train, axis=-1),y_train,epochs = 10)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rx0qMlbX11LH",
        "outputId": "df344472-3a8b-48b6-90d9-bdb7dc8349c5"
      },
      "execution_count": 202,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n",
            "125/125 [==============================] - 1s 2ms/step - loss: 289.2575 - mae: 289.2575\n",
            "Epoch 2/10\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 9.9023 - mae: 9.9023\n",
            "Epoch 3/10\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 9.5011 - mae: 9.5011\n",
            "Epoch 4/10\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.1969 - mae: 7.1969\n",
            "Epoch 5/10\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 10.5358 - mae: 10.5358\n",
            "Epoch 6/10\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.7656 - mae: 8.7656\n",
            "Epoch 7/10\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 6.1543 - mae: 6.1543\n",
            "Epoch 8/10\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 9.9963 - mae: 9.9963\n",
            "Epoch 9/10\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 4.9184 - mae: 4.9184\n",
            "Epoch 10/10\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 5.6307 - mae: 5.6307\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7fef59a5add0>"
            ]
          },
          "metadata": {},
          "execution_count": 202
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "n8H6thHE2yAq",
        "outputId": "c6041499-6a7a-4495-a3c5-069d2cf6fc1f"
      },
      "execution_count": 203,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_62\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " dense_103 (Dense)           (32, 50)                  100       \n",
            "                                                                 \n",
            " dense_104 (Dense)           (32, 1)                   51        \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 151\n",
            "Trainable params: 151\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "tf.random.set_seed(42)\n",
        "model = tf.keras.Sequential([tf.keras.layers.Dense(50,activation =None , name = 'input_layer'),\n",
        "                             tf.keras.layers.Dense(1,name = 'output_layer')],name = 'Hello_world')\n",
        "model.compile(loss ='mae',\n",
        "              optimizer = tf.keras.optimizers.Adam(learning_rate = 0.0001),\n",
        "              metrics = ['mae'])"
      ],
      "metadata": {
        "id": "K7f-pAFB3abd"
      },
      "execution_count": 234,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Total params -  total number of parameters in the mode.\n",
        "\n",
        "Trainable parameters - these are the parameters(patterns) the model can update as it trains.\n",
        "\n",
        "Non-trainable params - these parameters aren't updated during training (e.g. transfer learning)"
      ],
      "metadata": {
        "id": "c9ku2ImA4Fq4"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model.fit(tf.expand_dims(X_train, axis=-1),y_train,epochs = 10)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Ak4wjwyp4dHY",
        "outputId": "51c103c3-cca4-4df7-d8bf-4e3db3ac96f2"
      },
      "execution_count": 235,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n",
            "125/125 [==============================] - 1s 2ms/step - loss: 3424.1064 - mae: 3424.1064\n",
            "Epoch 2/10\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 2535.8328 - mae: 2535.8328\n",
            "Epoch 3/10\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1614.9121 - mae: 1614.9121\n",
            "Epoch 4/10\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 647.4970 - mae: 647.4970\n",
            "Epoch 5/10\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 19.3082 - mae: 19.3082\n",
            "Epoch 6/10\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.4917 - mae: 8.4917\n",
            "Epoch 7/10\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.2585 - mae: 8.2585\n",
            "Epoch 8/10\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.2284 - mae: 8.2284\n",
            "Epoch 9/10\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.0881 - mae: 8.0881\n",
            "Epoch 10/10\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.9561 - mae: 7.9561\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7fef58e3b7d0>"
            ]
          },
          "metadata": {},
          "execution_count": 235
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TJRhd4e4ADSb",
        "outputId": "2b37662c-f030-4863-877d-b00631a65c2d"
      },
      "execution_count": 236,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"Hello_world\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_layer (Dense)         (32, 50)                  100       \n",
            "                                                                 \n",
            " output_layer (Dense)        (32, 1)                   51        \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 151\n",
            "Trainable params: 151\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "y_pred = model.predict(X_test)\n",
        "\n",
        "from sklearn.metrics import mean_squared_error\n",
        "mean_squared_error(y_test,y_pred)**0.5"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lIM7VwkF5WOC",
        "outputId": "f3e0ba45-e9f5-4991-da31-6b586198c1b6"
      },
      "execution_count": 225,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "2.8831075708576535"
            ]
          },
          "metadata": {},
          "execution_count": 225
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from tensorflow.keras.utils import plot_model"
      ],
      "metadata": {
        "id": "RupnNVqH5pHI"
      },
      "execution_count": 226,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plot_model(model, show_shapes = True)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 312
        },
        "id": "URztHaID8XU_",
        "outputId": "372e35c5-5b06-4f42-cb33-06d9358f635e"
      },
      "execution_count": 237,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "execution_count": 237
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Visualizing our model's predictions"
      ],
      "metadata": {
        "id": "-e3d8HZY--Qk"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# make some predictions\n",
        "\n",
        "y_pred = model.predict(X_test)"
      ],
      "metadata": {
        "id": "tZPmts_SAYK7"
      },
      "execution_count": 239,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def plot_predictions(train_data = X_train,\n",
        "                     train_labels = y_train,\n",
        "                     test_data = X_test,\n",
        "                     test_labels = y_test,\n",
        "                     predictions = y_pred):\n",
        "  \n",
        "  plt.figure(figsize =(16,9))\n",
        "  plt.scatter(train_data , train_labels , c = 'b' , label = 'training data')\n",
        "  plt.scatter(test_data , test_labels ,  c = 'g', label = 'test data')\n",
        "  plt.scatter(test_data,predictions,c = 'r' , label = 'predictions')\n",
        "  plt.legend();"
      ],
      "metadata": {
        "id": "XBXFbBLVAcg6"
      },
      "execution_count": 247,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plot_predictions()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 533
        },
        "id": "ndPXIBZ7BExP",
        "outputId": "b1fedb6d-a5df-41d0-af76-01e572b4a6d8"
      },
      "execution_count": 248,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA7kAAAIICAYAAACrRimkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdf5iddX0n/PeXkEjGUjMCWjNIgxqFBDRAirCUrXZGS5Vqa21RkqfYqmxnxH12ry6Kj91i3Xqt/WFt3ZJYfHTBa6b+1i2trmKi9Fl/YA0YhUQU0IBJKAZMUEwggdzPH+ecOEASknPuMzPnzOt1XXPNme+573s+IXrJ2/t7v0+pqioAAADQD46Y7gEAAACgLkIuAAAAfUPIBQAAoG8IuQAAAPQNIRcAAIC+IeQCAADQN46c7gG65dhjj60WLVo03WMAAADQBTfccMM9VVUd9+j1vg25ixYtyrp166Z7DAAAALqglHLH/tZtVwYAAKBvCLkAAAD0DSEXAACAvtG3z+Tuz549e7J58+Y88MAD0z0KSY466qgcf/zxmTt37nSPAgAA9IlZFXI3b96co48+OosWLUopZbrHmdWqqsq9996bzZs358QTT5zucQAAgD4xq7YrP/DAAznmmGME3BmglJJjjjnGXXUAAKBWsyrkJhFwZxB/FwAAQN1mXcidTjt27MiqVavaOvclL3lJduzYcdBj/uRP/iRr1qxp6/oHc9VVV+WSSy456DHXXXddvvKVr9T+uwEAAA6HkDuFDhZyH3rooYOe+5nPfCYLFiw46DFvf/vbMzIy0vZ8nRByAQCAmUDIPYiJiWTRouSIIxrfJyY6u95ll12W22+/PcuWLcull16a6667Lueee25e9rKXZcmSJUmS3/zN38wZZ5yRpUuX5sorr9x37qJFi3LPPfdk06ZNOfnkk/P6178+S5cuzYtf/OLs2rUrSfKa17wmH//4x/cdf/nll+f000/PqaeemltuuSVJsm3btrzoRS/K0qVL87rXvS6/+Iu/mHvuuecxs/7P//k/8+xnPztnnnlmvvzlL+9b/6d/+qc8//nPz2mnnZaRkZHcfffd2bRpU9773vfm3e9+d5YtW5b/83/+z36PAwAA6DYh9wAmJpKLL07uuCOpqsb3iy/uLOi+853vzDOf+cysX78+f/mXf5kkufHGG/O3f/u3+e53v5sk+cAHPpAbbrgh69aty3ve857ce++9j7nOrbfemje84Q3ZsGFDFixYkE984hP7/X3HHntsbrzxxoyOjuav/uqvkiR/+qd/ml/91V/Nhg0b8spXvjJ33nnnY8676667cvnll+fLX/5yvvSlL2Xjxo373vvlX/7lXH/99fnGN76RV73qVfmLv/iLLFq0KH/4h3+Y//yf/3PWr1+fc889d7/HAQAAdNus+gihw/HWtyY7dz5ybefOxvqKFfX9njPPPPMRH6Hznve8J5/61KeSJD/4wQ9y66235phjjnnEOSeeeGKWLVuWJDnjjDOyadOm/V77Fa94xb5jPvnJTyZJvvSlL+27/nnnnZfBwcHHnPe1r30tL3jBC3LcccclSS644IJ9IXzz5s254IILctddd2X37t0H/PifQz0OAACgTu7kHsB+bnAedL1dT3ziE/e9vu6667JmzZp89atfzTe/+c2cdtpp+/2InSc84Qn7Xs+ZM+eAz/O2jjvYMYfrjW98Yy655JLcdNNN+fu///sDfgTQoR4HAABQJyH3AE444fDWD8XRRx+dn/zkJwd8/7777svg4GAGBgZyyy235Prrr2//lx3AOeeck49+9KNJkmuvvTbbt29/zDHPf/7z8y//8i+59957s2fPnnzsYx97xIxDQ0NJkquvvnrf+qP/bAc6DgAAoJuE3AN4xzuSgYFHrg0MNNbbdcwxx+Scc87JKaeckksvvfQx75933nl56KGHcvLJJ+eyyy7LWWed1f4vO4DLL7881157bU455ZR87GMfyy/8wi/k6KOPfsQxT3va0/K2t70tZ599ds4555ycfPLJ+95729velt/5nd/JGWeckWOPPXbf+m/8xm/kU5/61L7iqQMdBwAA0E2lqqrpnqErli9fXq1bt+4Ra9/+9rcfEdgez8RE4xncO+9s3MF9xzvqfR53Ojz44IOZM2dOjjzyyHz1q1/N6Oho1q9fP23zHO7fCQAAQJKUUm6oqmr5o9cVTx3EihW9H2of7c4778zv/u7vZu/evZk3b17e9773TfdIAAAAtallu3Ip5QOllB+WUm6etPbkUsrnSym3Nr8PNtdLKeU9pZTbSinfKqWcPumci5rH31pKuWjS+hmllJua57ynlFLqmHs2Wrx4cb7xjW/km9/8Zr7+9a/nl37pl6Z7JAAAYAYY+/RY7ji6pCo/+9r8870Xvep6JveqJOc9au2yJGurqlqcZG3z5yT59SSLm18XJ1mdNEJxksuTPD/JmUkubwXj5jGvn3Teo38XAAAAbRj79Fi+eVzJFeevzgn3JyU/+xr6SXou6NYScquq+v+S/OhRyy9P0qrVvTrJb05a/2DVcH2SBaWUpyX5tSSfr6rqR1VVbU/y+STnNd/7+aqqrq8aDxB/cNK1AAAAaMPkcPvce34WbCdrBd1e0s1ncp9aVdVdzdf/luSpzddDSX4w6bjNzbWDrW/ez/pjlFIuTuPucE7o5LN+AAAA+ti3jiu54p7G6966T/v4puQjhJp3YLte41xV1ZVVVS2vqmr5cccd1+1fBwAA0FPWPLPxrO2pB7hz2w+6GXLvbm41TvP7D5vrW5I8fdJxxzfXDrZ+/H7We86OHTuyatWqts//m7/5m+zcufNxj7vuuuty/vnnH/SY9evX5zOf+UzbswAAAL2jFW6Hv3d44bZKsuXoLg7WBd0MudckaTUkX5TkHyet/16zZfmsJPc1tzV/LsmLSymDzcKpFyf5XPO9H5dSzmq2Kv/epGv1lKkKuYdCyAUAgP439K6h7Dzi8MNt0gi4DyU5/sdd35Rbq7o+QuhDSb6a5DmllM2llNcmeWeSF5VSbk0y0vw5ST6T5HtJbkvyviRjSVJV1Y+S/LckX29+vb25luYx/2/znNuT/O865n48EzdNZNHfLMoRf3pEFv3NokzcNNHR9S677LLcfvvtWbZsWS699NIkyV/+5V/ml37pl/Lc5z43l19+eZLkpz/9aV760pfmec97Xk455ZR85CMfyXve855s3bo1L3zhC/PCF77wMdf+7Gc/m5NOOimnn356PvnJT+5b/9d//decffbZOe200/Lv/t2/y3e+853s3r07f/Inf5KPfOQjWbZsWT7ykY/s9zgAAKA3tUqlNv+XrZlfHf625CrJ7iOSuVVvBdwkKVUPDn0oli9fXq1bt+4Ra9/+9rdz8sknH9L5EzdN5OJ/ujg79/zszunA3IFc+RtXZsWpK9qaadOmTTn//PNz882NjxO+9tpr8/GPfzx///d/n6qq8rKXvSxvetObsm3btnz2s5/N+973viTJfffdlyc96UlZtGhR1q1bl2OPPfYR133ggQeyePHifOELX8iznvWsXHDBBdm5c2f++Z//OT/+8Y8zMDCQI488MmvWrMnq1avziU98IldddVXWrVuXv/u7v0uSAx7XbYfzdwIAADy+bx3XeOY2aS/cJkkZHk7WrKlzrNqVUm6oqmr5o9e72a7c09669q2PCLhJsnPPzrx17VvbDrmPdu211+baa6/NaaedliS5//77c+utt+bcc8/NH/3RH+XNb35zzj///Jx77rkHvc4tt9ySE088MYsXL06SrFy5MldeeWWSRkC+6KKLcuutt6aUkj179uz3God6HAAAMDOteWZjW/Kp6e9w+3impF25F915352Htd6Oqqrylre8JevXr8/69etz22235bWvfW2e/exn58Ybb8ypp56aP/7jP87b3/72tn/Hf/2v/zUvfOELc/PNN+ef/umf8sADD3R0HAAAMLO0WyqVNMJtlUa4LVXV8wE3EXIP6IQn7f9zdg+0fiiOPvro/OQnP/sk5V/7tV/LBz7wgdx///1Jki1btuSHP/xhtm7dmoGBgaxcuTKXXnppbrzxxv2e33LSSSdl06ZNuf3225MkH/rQh/a9d99992VoqPGxwlddddUBZznQcQAAwMx03eJ5nYfbJUv6Jty2CLkH8I7hd2Rg7sAj1gbmDuQdw+9o+5rHHHNMzjnnnJxyyim59NJL8+IXvzgXXnhhzj777Jx66ql55StfmZ/85Ce56aabcuaZZ2bZsmX50z/90/zxH/9xkuTiiy/Oeeed95jiqaOOOipXXnllXvrSl+b000/PU57ylH3vvelNb8pb3vKWnHbaaXnooYf2rb/whS/Mxo0b9xVPHeg4AABgZmk1Jv/KbXs6D7cbNnRnyGmkeOogJm6ayFvXvjV33ndnTnjSCXnH8Dtqex6XBsVTAABwaIbeNZSvvG1rTmhsBG3rudu9Seb0SQZUPNWGFaeuEGoBAIBp97lnlGz+fuN1J6VSc/poW/KB2K4MAAAwQ7VKpV78faVSh0rIBQAAmGGUSrVPyAUAAJghlEp1TsgFAACYZkPvGsodR5ds/i9bM79qL9xuefqCWR1uW4RcAACAaTLywZF94faE+w9va3Ir3D4w74iUqsrxd27v3qA9RMjtYdddd13OP//8JMk111yTd77znQc8dseOHVm1atW+n7du3ZpXvvKVXZ8RAADYv28eV/L5i9YedrhNJm1NrqrMf/Dh7gzYo4TcGejhhw//P6Qve9nLctlllx3w/UeH3IULF+bjH/94W/MBAADta5VKPfeezkqljuiTz7utm5B7MBMTyaJFyRFHNL5PTHR8yU2bNuWkk07KihUrcvLJJ+eVr3xldu7cmUWLFuXNb35zTj/99HzsYx/Ltddem7PPPjunn356fud3fif339/4xOfPfvazOemkk3L66afnk5/85L7rXnXVVbnkkkuSJHfffXd+67d+K8973vPyvOc9L1/5yldy2WWX5fbbb8+yZcty6aWXZtOmTTnllFOSJA888EB+//d/P6eeempOO+20fPGLX9x3zVe84hU577zzsnjx4rzpTW9K0gjhr3nNa3LKKafk1FNPzbvf/e6O/7kAAEC/Uyo1NY6c7gFmrImJ5OKLk507Gz/fcUfj5yRZsaKjS3/nO9/J+9///pxzzjn5gz/4g313WI855pjceOONueeee/KKV7wia9asyROf+MT8+Z//ef76r/86b3rTm/L6178+X/jCF/KsZz0rF1xwwX6v/x//43/Mr/zKr+RTn/pUHn744dx///155zvfmZtvvjnr169P0gjbLVdccUVKKbnppptyyy235MUvfnG++93vJknWr1+fb3zjG3nCE56Q5zznOXnjG9+YH/7wh9myZUtuvvnmJI27xAAAwP4NvWsoX3nb1mxu3Lc67HCbNEqlPHN7aNzJPZC3vvVnAbdl587Geoee/vSn55xzzkmSrFy5Ml/60peSZF9ovf7667Nx48acc845WbZsWa6++urccccdueWWW3LiiSdm8eLFKaVk5cqV+73+F77whYyOjiZJ5syZkyc96UkHnedLX/rSvmuddNJJ+cVf/MV9IXd4eDhPetKTctRRR2XJkiW544478oxnPCPf+9738sY3vjGf/exn8/M///Md/zMBAIB+o1Rqegi5B3LnnYe3fhhKKfv9+YlPfGKSpKqqvOhFL8r69euzfv36bNy4Me9///s7/r3teMITnrDv9Zw5c/LQQw9lcHAw3/zmN/OCF7wg733ve/O6171uWmYDAICZaOSDI22XSj063CqVOnxC7oGccMLhrR+GO++8M1/96leTJP/wD/+QX/7lX37E+2eddVa+/OUv57bbbkuS/PSnP813v/vdnHTSSdm0aVNuv/32JMmHPvSh/V5/eHg4q1evTtJ4fva+++7L0UcfnZ/85Cf7Pf7cc8/NRPN54+9+97u5884785znPOeA899zzz3Zu3dvfvu3fzt/9md/lhtvvPEw/vQAANCfJofbdkul9kZjcqeE3AN5xzuSgYFHrg0MNNY79JznPCdXXHFFTj755Gzfvn3f1uKW4447LldddVVe/epX57nPfW7OPvvs3HLLLTnqqKNy5ZVX5qUvfWlOP/30POUpT9nv9f/2b/82X/ziF3PqqafmjDPOyMaNG3PMMcfknHPOySmnnJJLL730EcePjY1l7969OfXUU3PBBRfkqquuesQd3EfbsmVLXvCCF2TZsmVZuXJl/vt//+8d/zMBAIBe9q0Ow22rVGqOxuSOlapP/yEuX768Wrdu3SPWvv3tb+fkk08+9ItMTDSewb3zzsYd3He8o+PSqU2bNuX888/fV9o02x323wkAAMwgQ+8ayvf+y9bMy+EF20SpVKdKKTdUVbX80evalQ9mxYqOQy0AANB/Rj44kve/YW1HjckPzDsi8x98OMfXPdwsZ7vyFFu0aJG7uAAA0KOUSs18Qi4AAMDjqKNU6sGiVGoqzLqQ26/PIPcifxcAAMx0EzdN1NOYPD6eo/b699+pMKtC7lFHHZV7771XuJoBqqrKvffem6OOOmq6RwEAgMeYuGkiF/52yaufu7KzxuTx8RxRVbp+ptCsKp46/vjjs3nz5mzbtm26RyGN/9Ph+OM9Zg8AwMwy+M7B/NtbduTCaEzuRbMq5M6dOzcnnnjidI8BAADMQK3G5B9pTO5ps2q7MgAAwKN10picKJWaaYRcAABgVlIq1Z9m1XZlAACAiZsm8um3rcz4Jw8/2CY/K5U6Ynw8RaHUjCPkAgAAs8LkcNtJqVQRbmc0IRcAAOh7b7hwMP/jQ501JpclS5ING2qejLp5JhcAAOhbIx8cyY+PLPm7D+3IEWnvs24fmHdESlUJuD1CyAUAAPrO5Mbkox9u7+6txuTeJOQCAAB9Q2MynskFAAD6wrXPKLnw+43XSqVmL3dyAQCAnnbpa4ayt5S86Pvt3bmt0iiVKlWVCLg9T8gFAAB60tIrlubHR5b8xdVb2y6V2hdulUr1DduVAQCAnrL0iqW57o825uYHGz+3szX5/ifOzdH37657NGYAd3IBAICe0CqVuvmSjTn2wQ62Jo+PC7h9zJ1cAABgRpu4aSLHvnxl56VSo6PJqlV1jsYMJOQCAAAz1vvOmpfXfm3PYd+1TYTb2cp2ZQAAYMZpNSa/7mt7Oi+VEnBnFSEXAACYMepoTP7Rk+drTJ7FhFwAAGDajX16LHcc3SiVOvrh9sLt/U+cm1JVOebenV2akl4g5AIAANNm7NNj+eZxJVecvzon3N9eY/JDSUpVaUwmSZdDbinlOaWU9ZO+flxK+U+llLeVUrZMWn/JpHPeUkq5rZTynVLKr01aP6+5dlsp5bJuzg0AAHTXxE0T+dwzGuH2ufd08HFAo6OZW1WPdzizSFfblauq+k6SZUlSSpmTZEuSTyX5/STvrqrqryYfX0pZkuRVSZYmWZhkTSnl2c23r0jyoiSbk3y9lHJNVVUbuzk/AABQv1oak4eHkzVrap6MfjCV25WHk9xeVdUdBznm5Uk+XFXVg1VVfT/JbUnObH7dVlXV96qq2p3kw81jAQCAHvGGCwc7bkze/syFjVIpAZcDmMqQ+6okH5r08yWllG+VUj5QShlsrg0l+cGkYzY31w60DgAAzHBjnx7Lj48s+bsP7ei4MfnJt23pzpD0jSkJuaWUeUleluRjzaXVSZ6Zxlbmu5K8q6bfc3EpZV0pZd22bdvquCQAANCmVmPyFeevPuzG5KQRbh+Yd4TGZA7LVN3J/fUkN1ZVdXeSVFV1d1VVD1dVtTfJ+9LYjpw0ntl9+qTzjm+uHWj9EaqqurKqquVVVS0/7rjjuvDHAAAAHk8djcl7k5Tx8cx/8OHuDEnf6mrx1CSvzqStyqWUp1VVdVfzx99KcnPz9TVJ/qGU8tdpFE8tTvKvafx3YnEp5cQ0wu2rklw4RbMDAACH6NpnlFzx/cbrtkulRkdTVq2qcyxmka6H3FLKE9NoRf4Pk5b/opSyLI3/HG9qvVdV1YZSykeTbEzj467eUFXVw83rXJLkc0nmJPlAVVUbuj07AABwaN5w4WD+x4d25EXRmMz0KlWffqbU8uXLq3Xr1k33GAAA0NfecOFg3tNGoVTys3C7/ZkLFUpx2EopN1RVtfzR61PZrgwAAPSJkQ+O7GtMnpP2Au6DR2tMpn5CLgAAcMhapVKfv2ht243JrVKpo36sMZn6TVXxFAAA0OM6LZWqkhwxPp6yYkXNk8HPuJMLAAAc1BeeNSdVKXnR99v7OKAqjcbkI6oqEXDpMiEXAADYrzdcOJi9peSFt+9tO9xuf+bClKpKfCQQU0TIBQAAHmFyqdThtia3wu1Pn7JAqRTTQsgFAACSNMLtHUe3Vyr16HD7c3dv79KUcHBCLgAAzHKTw+0J97f33O2eRLhlRhByAQBgFrv2GZ2F21ap1Lyq6s6AcJiEXAAAmIXed9a87O20MXl4WKkUM46QCwAAs8ilrxnK3lLyuq/tabtUal+4XbOmO0NCB4RcAACYBUY+OJKdR5T8xdVbO25MFm6ZyYRcAADoY5NLpeZXhxduk0a43TVXqRS9Q8gFAIA+NHHTRL55XGelUnuTlPHxDOxWKkXvOHK6BwAAAOp17TNKLvx+43U7d26TRrgtK1bUORZMCXdyAQCgT9TamCzg0qOEXAAA6HG1NCYvWaJUir4g5AIAQI9aesXS/PjIzhqTf/Tk+Y1wu2FDd4aEKeaZXAAA6DETN03klF9dmZvvafzcbmPywO4qx9Q9HEwzd3IBAKBHtBqTL3zuyjz3ng6eu9WYTB9zJxcAAGa4iZsmcuzLV3bemDw6mqxaVedoMOMIuQAAMIO976x5ee3X9hz2XdtkUrhdssQzt8watisDAMAMtPSKpdl5RE2NyQIus4iQCwAAM8jYp8ey7aiSmy/ZmPmVxmQ4XEIuAADMAGOfHssdR5dccf7qHPtge6VSDyUpVZVj7t3ZnSGhBwi5AAAwjSYmkvKSsfzt+atzwv0dNCaPjmZupTEZFE8BAMA0GRxMduxINv3c6hyZww+3icZkeDQhFwAAptjQULJ1689+bt3BPRT7wu3wcLJmTd2jQc8TcgEAYIosXZps3Njeua1w+6Mnz/fMLRyEZ3IBAKDLxsaSUtoLuK1nbu9/4lylUnAIhFwAAOiSVrhdvfrgx9151NzsrzKqSvLTIxuNyUffv7sbI0LfEXIBAKAL5s17/HDbsuiB3fuCbuvroZKU8fH83B6NyXA4PJMLAAA1enSp1KFa9EDjTu2CBcn27f5FHdrlTi4AANRgaKixNbmdgJs0zh0fbwRcoH1CLgAAdGBkpLNwmzTC7d69yYoV9c0Fs5WQCwAAbWiVSq1d2/41RkeTqhJuoU62+gMAwGEYGzv0QqkDWbIk2bChnnmARxJyAQDgEM2bl+zZ0/75CxcmW7bUNw/wWLYrAwDA4xgcbGxNbjfgtkqlBFzoPiEXAAAOoFUqtWNHe+e3wq1SKZg6tisDAMCjjIx0ViiVNMKtYAtTT8gFAICmOsLt8HCyZk098wCHT8gFAIAkc+Y0thW3S2MyzAyeyQUAYFZrlUq1G3CXLGl81q2ACzODO7kAAMxKg4PtF0olyfz5yc6d9c0D1MOdXAAAZpVOG5OTRqmUgAszkzu5AADMCkqlYHbo+p3cUsqmUspNpZT1pZR1zbUnl1I+X0q5tfl9sLleSinvKaXcVkr5Vinl9EnXuah5/K2llIu6PTcAAP1hYqJx57aTgDs83HjuVsCFmW+qtiu/sKqqZVVVLW/+fFmStVVVLU6ytvlzkvx6ksXNr4uTrE4aoTjJ5Umen+TMJJe3gjEAAOxPK9yuXNn+NVqlUsIt9I7p2q788iQvaL6+Osl1Sd7cXP9gVVVVkutLKQtKKU9rHvv5qqp+lCSllM8nOS/Jh6Z2bAAAesHAQLJrV/vnL1iQbN9e3zzA1JmKO7lVkmtLKTeUUi5urj21qqq7mq//LclTm6+Hkvxg0rmbm2sHWn+EUsrFpZR1pZR127Ztq/PPAABAD1i6tHH3tt2Au2BB486tgAu9ayru5P5yVVVbSilPSfL5Usotk9+sqqoqpVR1/KKqqq5McmWSLF++vJZrAgAw8y1dmmzc2Nk1Kv/2CH2h63dyq6ra0vz+wySfSuOZ2rub25DT/P7D5uFbkjx90unHN9cOtA4AwCzWeu62k4DbKpUC+kNXQ24p5YmllKNbr5O8OMnNSa5J0mpIvijJPzZfX5Pk95oty2clua+5rflzSV5cShlsFk69uLkGAMAsVEeplMZk6E/d3q781CSfKqW0ftc/VFX12VLK15N8tJTy2iR3JPnd5vGfSfKSJLcl2Znk95OkqqoflVL+W5KvN497e6uECgCA2WNiorNgmyQLFyZb7AmEvlWqPt2bsXz58mrdunXTPQYAADUZHEx27Gj/fI3J0F9KKTdM+pjafabqc3IBAKAtrcbkdgPu3Lkak2E2EXIBAJiRWuG2k1Kp0dFk9+76ZgJmPiEXAIAZZWysnnBbVcmqVfXNBfSGqficXAAAeFxKpYA6CLkAAEy7TkulhFugRcgFAGDaDA0lW7e2f/78+cnOnfXNA/Q+z+QCADDlWqVSnQTc0VEBF3gsIRcAgCmjVAroNtuVAQDourGxZPXqzq4xOirYAo9PyAUAoGvqCLdKpYDDIeQCANAVAwPJrl3tn79gQbJ9e33zALODZ3IBAKjV0FDjudt2A24pyfi4gAu0R8gFAKAWdTQmj48ne/cmK1bUNxcwuwi5AAB0pM7GZOEW6JRncgEAaMvERLJyZWfXWLIk2bChnnkAEiEXAIDDJNwCM5mQCwDAIdOYDMx0nskFAOBxtUqlNCYDM52QCwDAAY2MdF4qpTEZmEpCLgAAj9EKt2vXtn8NjcnAdPBMLgAA+yiVAnqdkAsAQJLOS6UWLky2bKlvHoB22K4MADDLDQ11Viq1YEFjW7KAC8wEQi4AwCzVakzeurX9a2hMBmYaIRcAYJapozFZqRQwU3kmFwBglhgbS1av7uwaw8PJmjX1zAPQDUIuAECf05gMzCZCLgBAHxscTHbsaEw/PqUAACAASURBVP98jclArxFyAQD60NBQZ4VS8+cnO3fWNw/AVFE8BQDQR+pqTBZwgV4l5AIA9IGxsc4bk4eHNSYDvc92ZQCAHlZHqZTGZKCfCLkAAD1IYzLA/gm5AAA9ZmAg2bWr/fOVSgH9zDO5AAA9olUq1W7ALUWpFND/hFwAgBluZKTzUqnx8WTvXqVSQP8TcgEAZqhWuF27tv1rjI5qTAZmF8/kAgDMMEqlANon5AIAzCCdlkotXJhs2VLfPAC9xnZlAIAZYGios1KpBQsa25IFXGC2E3IBAKZRK9xu3dr+NcbHk+3b65sJoJcJuQAA06BVKtVJuFUqBfBYnskFAJhCY2PJ6tWdXWN4OFmzpp55APqNkAsAMAXqaEwWbgEen5ALANBlGpMBpo6QCwDQJUNDnT1zO39+snNnffMAzAZdK54qpTy9lPLFUsrGUsqGUsr/3Vx/WyllSyllffPrJZPOeUsp5bZSyndKKb82af285tptpZTLujUzAEAdli6tpzFZwAU4fN1sV34oyR9VVbUkyVlJ3lBKWdJ8791VVS1rfn0mSZrvvSrJ0iTnJVlVSplTSpmT5Iokv55kSZJXT7oOAMCMMTbWCLcbN7Z/DY3JAJ3p2nblqqruSnJX8/VPSinfTjJ0kFNenuTDVVU9mOT7pZTbkpzZfO+2qqq+lySllA83j+3gfz4AAOqjMRlg5piSz8ktpSxKclqSrzWXLimlfKuU8oFSymBzbSjJDyadtrm5dqB1AIBpNTHRuHPbScBdsqRx51bABahH10NuKeXnknwiyX+qqurHSVYneWaSZWnc6X1Xjb/r4lLKulLKum3bttV1WQCAxxgY6OwjgRYsaITbDRvqmwmALofcUsrcNALuRFVVn0ySqqrurqrq4aqq9iZ5X362JXlLkqdPOv345tqB1h+jqqorq6paXlXV8uOOO67ePwwAQH5WKtXuRwLNn98It9u31zsXAA3dbFcuSd6f5NtVVf31pPWnTTrst5Lc3Hx9TZJXlVKeUEo5McniJP+a5OtJFpdSTiylzEujnOqabs0NALA/rXDbSamUxmSA7uvm5+Sek+T/SnJTKWV9c+3/SaMdeVmSKsmmJP8hSaqq2lBK+WgahVIPJXlDVVUPJ0kp5ZIkn0syJ8kHqqqysQcAmBIjI8natZ1dY3Q0WbWqnnkAOLhSVdV0z9AVy5cvr9atWzfdYwAAPWpiorNnbpNGqZRnbgG6o5RyQ1VVyx+93s07uQAAPWlgoP1nbhPhFmA6CbkAAE1DQ8nWre2fv3BhsmW/9ZgATBUhFwCY9ToNt0cckTz8cH3zANC+rn9OLgDATDUy0mhM7iTgjo4KuAAziZALAMw6Y2ONcNtJa/LoaOPzbrUmA8wstisDALNGHY3Jw8PJmjX1zANA/YRcAGBW6LQxWakUQG8QcgGAvtZpqdT8+cnOnfXNA0B3eSYXAOhLS5d2Xio1Pi7gAvQaIRcA6CutUqmNG9u/RqtUasWK+uYCYGrYrgwA9IWxsWT16s6uoVQKoPcJuQBAT6ujMXnJkmTDhnrmAWB6CbkAQM/qtDF5wYJk+/b65gFg+nkmFwDoOa1SqXYD7vz5jWduBVyA/iPkAgA9oxVuOymV0pgM0N+EXABgxhsZ0ZgMwKHxTC4AMGMplQLgcAm5AMCM1GmplHALMDsJuQDAjDI4mOzY0f75CxcmW7bUNw8AvUXIBQBmhKGhZOvW9s8/4ojk4YfrmweA3qR4CgCYVq1SqU4C7uiogAtAg5ALAEyLsbFGuF27tv1rtBqTV62qby4AepvtygDAlBobS1av7uwaw8PJmjX1zANAfxFyAYApM29esmdP++crlQLg8diuDAB03dBQY2tyuwF3wYLGtmQBF4DHI+QCAF2zdGlnpVJz5zbC7fbt9c4FQP8ScgGA2rXC7caN7V9jdDTZvbu+mQCYHYRcAKA2rcbkTsLt8LDGZADap3gKAOjYxESycmVn11iyJNmwoZ55AJi9hFwAoCMDA8muXe2frzEZgDrZrgwAtKX13G27AXf+fI3JANRPyAUADksdpVLj48nOnfXNBAAtQi4AcEhGRuppTK6qZMWK+uYCgMk8kwsAHFQdpVLDw8maNfXMAwAHI+QCAAfUaamUxmQAppqQCwA8xuBgsmNH++drTAZgugi5AMA+Q0PJ1q3tn3/EEcnDD9c3DwAcLsVTAMC+UqlOAu7oqIALwPQTcgFgFhsba4TbtWvbv0arMXnVqvrmAoB22a4MALPQ2FiyenVn19CYDMBMJOQCwCwzZ06yd2/75yuVAmAms10ZAGaJoaHG1uR2A+6CBY1tyQIuADOZkAsAfW7p0s5KpebObYTb7dvrnQsAukHIBYA+1Qq3Gze2f43R0WT37vpmAoBuE3IBoM+0GpM7CbfDwxqTAehNiqcAoE9MTCQrV3Z2jSVLkg0b6pkHAKaDkAsAfWBgINm1q/3zhVsA+kXPbFcupZxXSvlOKeW2Uspl0z0PAMwErcbkdgPu/PmNbckCLgD9oidCbillTpIrkvx6kiVJXl1KWTK9UwHA9Om0MTlJxseTnTvrmwkAZoKeCLlJzkxyW1VV36uqaneSDyd5+TTPBABTbmSknsbkqkpWrKhvLgCYKXol5A4l+cGknzc31x6hlHJxKWVdKWXdtm3bpmw4AOi2iYlGuF27tv1raEwGYDbolZB7SKqqurKqquVVVS0/7rjjpnscAOhYK9x20pq8ZEkj3K5ZU99cADBT9Uq78pYkT5/08/HNNQDoW502Ji9cmGzxv5YAzDK9cif360kWl1JOLKXMS/KqJNdM80wA0BWdNibPndu4cyvgAjAb9UTIrarqoSSXJPlckm8n+WhVVT7sAIC+0iqV6rQxeffu+mYCgF7TK9uVU1XVZ5J8ZrrnAIC6jYx0ViiVNBqTFUoBQA+FXADoN2NjyerVnV1jeFihFABMJuQCwDSYMyfZu7f985VKAcD+9cQzuQDQL1qlUu0G3IULlUoBwMG4kwsAU2BoqLNCqblzFUoBwKFwJxcAumjp0s4bk0dHBVwAOFRCLgB0wdhYI9xu3Nj+NYaHG1uTtSYDwKGzXRkAajQxkaxc2dk1NCYDQPuEXACoycBAsmtX++cvWZJs2FDfPAAwGwm5ANChTkul5s9Pdu6sbx4AmM08kwsAbeq0VKqUZHxcwAWAOgm5AHCYRkY6L5UaH298Vu6KFfXNBQAIuQBwyCYmGuF27dr2r9FqTBZuAaA7PJMLAI+jjsZkpVIAMDWEXAA4iE4bkxcuTLZsqW8eAODghFwA2I9OG5Pnzk12765vHgDg0HgmFwAmaZVKdRJwx8cFXACYLkIuAORn4baTUqnRUaVSADDdbFcGYFYbG0tWr+7sGsPDyZo19cwDAHRGyAVg1pozp/FZte1SKgUAM4/tygDMOkNDja3J7QbchQsb25IFXACYedzJBWDW6LQxef78ZOfO+uYBAOrnTi4AfW/p0noakwVcAJj5hFwA+tbYWCPcbtzY/jWGhzUmA0AvsV0ZgL4zMZGsXNnZNTQmA0BvEnIB6Bt1hNslS5ING+qZBwCYekIuAH1hcDDZsaP985VKAUB/8EwuAD2tVSrVbsAtRakUAPQTIReAnjQy0nmp1Ph447NylUoBQP+wXRmAnjIykqxd29k1lEoBQP8ScgHoCUqlAIBDIeQCMOMNDCS7drV//sKFyZYt9c0DAMxcQi4AM9bQULJ1a/vna0wGgNlH8RQAM06rMbmTgKsxGQBmJyEXgBmjjsbk0dGkqjQmA8BsZbsyANNubCxZvbqza2hMBgASIReAaTZnTuOzatulMRkAmMx2ZQCmxdBQY2tyuwF34cLGtmQBFwCYzJ1cAKaUxmQAoJvcyQVgSmhMBgCmgpALQFeNjWlMBgCmju3KAHTFxESycmVn19CYDAAcLiEXgFrVEW41JgMA7RJyAajNwECya1f75yuVAgA65ZlcADrWKpVqN+CWolQKAKiHkAtA20ZGOi+VGh9vfFauUikAoA62KwNw2EZGkrVrO7uGUikAoBuEXAAOmVIpAGCm68p25VLKX5ZSbimlfKuU8qlSyoLm+qJSyq5Syvrm13snnXNGKeWmUsptpZT3lFJKc/3JpZTPl1JubX4f7MbMABzcwEBnAXfhwsZn3Qq4AEA3deuZ3M8nOaWqqucm+W6St0x67/aqqpY1v/5w0vrqJK9Psrj5dV5z/bIka6uqWpxkbfNnAKbI0FBnpVILFjTC7ZYt9c4FALA/XQm5VVVdW1XVQ80fr09y/MGOL6U8LcnPV1V1fVVVVZIPJvnN5tsvT3J18/XVk9YB6KJWY/LWre1fY3w82b69vpkAAB7PVLQr/0GS/z3p5xNLKd8opfxLKeXc5tpQks2TjtncXEuSp1ZVdVfz9b8leeqBflEp5eJSyrpSyrpt27bVND7A7FJHY/LoaOPurcZkAGCqtV08VUpZk+QX9vPWW6uq+sfmMW9N8lCSieZ7dyU5oaqqe0spZyT5X6WUpYf6O6uqqkop1UHevzLJlUmyfPnyAx4HwGONjSWrV3d2DY3JAMB0azvkVlU1crD3SymvSXJ+kuHmFuRUVfVgkgebr28opdye5NlJtuSRW5qPb64lyd2llKdVVXVXc1vzD9udGYD9mzOn8Vm17dKYDADMFN1qVz4vyZuSvKyqqp2T1o8rpcxpvn5GGgVT32tuR/5xKeWsZqvy7yX5x+Zp1yS5qPn6oknrAHRocLCxNbndgKsxGQCYabr1Obl/l+QJST7f/CSg65tNyv8+ydtLKXuS7E3yh1VV/ah5zliSq5LMT+MZ3tZzvO9M8tFSymuT3JHkd7s0M8CsMTTUWaHU/PnJzp2PfxwAwFTrSsitqupZB1j/RJJPHOC9dUlO2c/6vUmGax0QYJZaurSzQqmk0ZisUAoAmKmmol0ZgGk2NqYxGQCYHbq1XRmAGUBjMgAw2wi5AH1oYiJZubKza2hMBgB6kZAL0GcGBpJdu9o/f8GCZPv2+uYBAJhKnskF6BNLlzaeu2034M6f33jmVsAFAHqZkAvQ41rhtpNSqfFxHwkEAPQHIRegR42MdB5uh4c1JgMA/cUzuQA9RqkUAMCBCbkAPaTTUinhFgDod0IuQA8YGkq2bm3/fI3JAMBs4ZlcgBmsVSrVScAdHxdwAYDZQ8gFmIHqKJUaHVUqBQDMPkIuwAwyNtYIt2vXtn+NVrhdtaq+uQAAeoVncgFmiDlzkr172z9fqRQAgJALMO0GB5MdO9o/f+HCZMuW+uYBAOhlQi7ANOm0MXn+/GTnzvrmAQDoB57JBZhidTUmC7gAAI8l5AJMkVaplMZkAIDusV0ZoMvGxpLVqzu7xvBwsmZNPfMAAPQzIRegSyYmkpUrO7uGxmQAgMMj5AJ0wcBAsmtX++cvWJBs317fPAAAs4VncgFq1CqVajfgzp/feOZWwAUAaI+QC1CDVrjtpFRKYzIAQOeEXIAOjIxoTAYAmEk8kwvQBqVSAAAzk5ALcJg6LZUSbgEAukfIBThEg4PJjh3tn68xGQCg+zyTC/A4Ws/ddhJwx8cFXACAqSDkAhxAK9yuXdv+NZRKAQBMLduVAR5lZKSzYJs0wu2qVfXMAwDAoRNyAZo0JgMA9D4hFyCdNyYvXJhs2VLfPAAAtEfIBWY1jckAAP1F8RQwKy1d2nljclUJuAAAM42QC8wqrcbkjRvbv8bwcCPgAgAw8wi5wKwwMdH5xwG1wu2aNfXNBQBAvTyTC/S1OhqTh4cFWwCAXiHkAn1r3rxkz572z1cqBQDQe2xXBvpOq1Sq3YC7YIFSKQCAXuVOLtA3li7trFBq7txk9+765gEAYOq5kwv0vLGxzhuTR0cFXACAfuBOLtCzlEoBAPBoQi7Qc+oItwsXJlu21DMPAAAzh5AL9JTBwWTHjvbPF24BAPqbkAv0hKGhZOvW9s9XKgUAMDsongJmtNbHAXUScJVKAQDMHl0LuaWUt5VStpRS1je/XjLpvbeUUm4rpXynlPJrk9bPa67dVkq5bNL6iaWUrzXXP1JKmdetuYGZoa7G5KpKVq2qby4AAGa2bm9XfndVVX81eaGUsiTJq5IsTbIwyZpSyrObb1+R5EVJNif5einlmqqqNib58+a1PlxKeW+S1yZZ3eXZgWkwNpas7vC/3Z67BQCYvabjmdyXJ/lwVVUPJvl+KeW2JGc237utqqrvJUkp5cNJXl5K+XaSX01yYfOYq5O8LUIu9J2BgWTXrvbPF24BAOj2M7mXlFK+VUr5QCllsLk2lOQHk47Z3Fw70PoxSXZUVfXQo9aBPjE01Nia3G7AXbCgsS1ZwAUAoKOQW0pZU0q5eT9fL0/jTuszkyxLcleSd9Uw7+PNc3EpZV0pZd22bdu6/euADrXCbSelUuPjyfbt9c0EAEBv62i7clVVI4dyXCnlfUn+ufnjliRPn/T28c21HGD93iQLSilHNu/mTj7+0fNcmeTKJFm+fHl1iH8MYIrV8dzt6KhCKQAAHqub7cpPm/TjbyW5ufn6miSvKqU8oZRyYpLFSf41ydeTLG42Kc9Lo5zqmqqqqiRfTPLK5vkXJfnHbs0NdE+rMbmTgKsxGQCAg+lm8dRflFKWJamSbEryH5KkqqoNpZSPJtmY5KEkb6iq6uEkKaVckuRzSeYk+UBVVRua13pzkg+XUv4syTeSvL+LcwM1q+PO7ZIlyYYNj38cAACzW2ncKO0/y5cvr9atWzfdY8CsN29esmdP++cvWOCZWwAAHquUckNVVcsfvd7tdmVglmqVSrUbcEtRKgUAwOETcoFajYx01pjcCrd79yYrVtQ7GwAA/a+bz+QCs8jISLJ2bWfXGB8XbAEA6IyQC3REqRQAADOJkAu0rdNSKeEWAIC6CbnAYRscTHbsaP/8hQuTLVvqmwcAAFoUTwGHbHCwUQzVbsBtlUoJuAAAdIuQCzyuVmNyJ3dvNSYDADAVbFcGDqiOxuTh4WTNmnrmAQCAxyPkAo8xMZGsXNnZNZRKAQAwHYRc4BEGBpJdu9o/X7gFAGA6CblAks4bkxcsSLZvr28eAABoh+IpmOWWLu2sVGr+/KSqBFwAAGYGIRdmqVZj8saN7V9jfDzZubO+mQAAoFO2K8MsozEZAIB+5k4uzBITE407t50E3OHhxtZkARcAgJnKnVyYBebNS/bsaf98pVIAAPQKd3Khj7VKpdoNuAsWKJUCAKC3uJMLfWjp0s4KpebOTXbvrm8eAACYKu7kQh9p3bntJOCOjgq4AAD0LiEX+kCrVKqTcNsqlVq1qr65AABgqtmuDD1sYiJZubKza/g4IAAA+omQCz1qcDDZsaP98xcuTLZsqW8eAACYCYRc6DFDQ8nWre2fr1QKAIB+5plc6BGtUqlOAq5SKQAA+p2QCzPc2Fg9jclKpQAAmA1sV4YZamwsWb26s2uMjgq2AADMLkIuzDB1NCYrlQIAYLYScmEG6bQxecGCZPv2+uYBAIBe45lcmAGGhhrP3XYScMfHBVwAABByYRq1SqU6bUyuqmTFivrmAgCAXmW7MkwDpVIAANAdQi5MoTrC7ZIlyYYN9cwDAAD9RsiFKTJvXrJnT/vna0wGAIDH55lc6LJWqVS7AXfBgsYztwIuAAA8PiEXuqQVbtstlSpFYzIAABwuIRdqNjLSeWPy+Hiyd6/GZAAAOFyeyYWaKJUCAIDpJ+RCDTotlRJuAQCgHrYrQwcGBzsrlVqypFEqJeACAEA93MmFNgwOJjt2tH/+3LnJ7t31zQMAADS4kwuHoVUq1W7AbTUmC7gAANAd7uTCIRgZSdau7ewa4+PakgEAoNuEXDiIOsLt8HCyZk098wAAAAcn5MIBzJnT+KzadmlMBgCAqSfkwqN0Wiq1YEGyfXt98wAAAIdO8RQ0LV3aWanU/PmNjwMScAEAYPp0JeSWUj5SSlnf/NpUSlnfXF9UStk16b33TjrnjFLKTaWU20op7ymllOb6k0spny+l3Nr8PtiNmZm9WuF248b2rzE+nuzcWd9MAABAe7oScququqCqqmVVVS1L8okkn5z09u2t96qq+sNJ66uTvD7J4ubXec31y5KsrapqcZK1zZ+hY62PA+ok3A4PN+7eak0GAICZoavblZt3Y383yYce57inJfn5qqqur6qqSvLBJL/ZfPvlSa5uvr560jq0ZWKiEW47aU1uhVutyQAAMLN0+5ncc5PcXVXVrZPWTiylfKOU8i+llHOba0NJNk86ZnNzLUmeWlXVXc3X/5bkqQf6ZaWUi0sp60op67Zt21bTH4F+0Qq3K1e2f40lS4RbAACYydpuVy6lrEnyC/t5661VVf1j8/Wr88i7uHclOaGqqntLKWck+V+llKWH+jurqqpKKdVB3r8yyZVJsnz58gMex+wzNJRs3dr++RqTAQCgN7QdcquqGjnY+6WUI5O8IskZk855MMmDzdc3lFJuT/LsJFuSHD/p9OOba0lydynlaVVV3dXc1vzDdmdm9lm6tLNnbufOTXbvrm8eAACgu7q5XXkkyS1VVe3bhlxKOa6UMqf5+hlpFEx9r7kd+cellLOaz/H+XpLW3eBrklzUfH3RpHU4oDoak0dHBVwAAOg1bd/JPQSvymMLp/59kreXUvYk2ZvkD6uq+lHzvbEkVyWZn+R/N7+S5J1JPlpKeW2SO9IosoL9mpjo7JnbpFEq5ZlbAADoTaVRZtx/li9fXq1bt266x2CKCLcAADC7lFJuqKpq+aPXu3knF7qujnC7cGGyZcvjHwcAAMx8Qi49a3Aw2bGj/fOVSgEAQP/p9ufkQu1apVKdBFylUgAA0J+EXHrG2Fg9jclVlaxaVd9cAADAzGG7MjPe2FiyenVn1xgdFWwBAGA2EHKZsZRKAQAAh0vIZUbqtFRKuAUAgNnJM7nMKJ2WSs2f33jmVsAFAIDZSchlRmiF205LpXburG8mAACg9wi5TCuNyQAAQJ08k8u0qKMxecmSZMOGeuYBAAD6g5DLlJszJ9m7t/3zlUoBAAAHYrsyU6b13G27AXfBAqVSAADAwQm5dF0dpVLj48n27fXNBAAA9Cchl66po1RqfLxx93bFivrmAgAA+pdncqldHaVS4+OCLQAAcPiEXGpTR7gdHk7WrKlnHgAAYPYRcqlFp43Jwi0AAFAHIZeODAwku3a1f/4RRyQPP1zfPAAAwOymeIq2tEqlOgm44+MCLgAAUC93cjksSqUAAICZTMjlkCiVAgAAeoGQy+NSKgUAAPQKz+RyQAMDjedu2w24w8NJVQm4AADA1HEnl8cYGkq2bm3//AULku3b65sHAADgULmTyz5DQ407t50E3PFxARcAAJg+Qi77Pg6ok3Db2pqsNRkAAJhOtivPckqlAACAfuJO7iylVAoAAOhH7uTOMgMDya5d7Z+/cGGyZUt98wAAANRJyJ0lOm1Mnj8/2bmzvnkAAAC6wXblPldHY/LoqIALAP9/e3cbY0dVBnD8/4RaNIq0FIJQoLRajfWL4MZggsRAA4WoRa2mhISqJESEREOM1jQxfkWjJESQYDAivgCihEajCIjyiULBCi21suUltkHwpYKJBkEeP8y5OLvu3e3evbszd/r/JSed+8xL587Z5+ycO+eelSSNBp/kdljE3PZ3UilJkiRJo8YnuR3Um1RqUE4qJUmSJGlU+SS3Q5xUSpIkSdKhzie5HdD73u2gHdzjj6+e3NrBlSRJkjTqfJI7wpwxWZIkSZIm8knuCHLGZEmSJEmamp3cEbJ27XA6t5lw7bXDOy9JkiRJaguHK4+Iuf45ICeVkiRJknQo8Eluy/WGJg/KSaUkSZIkHUp8kttSTiolSZIkSbPnk9yWcVIpSZIkSRqcndyWcFIpSZIkSZo7hys3bO1auOeeuR3j0kvt2EqSJEkS2MltzDA6t2vWwK5dwzkfSZIkSeqCOQ1XjoiPRsSuiHglIsYmrftiRIxHxJ6IOKcWX1di4xGxuRZfGRHbSvyWiFhc4oeX1+Nl/clzOec2WLx4bh3c172uGpZsB1eSJEmSJprrd3J3Ah8G7qsHI2INsBF4B7AOuDYiDouIw4BrgHOBNcAFZVuAK4GrMvMtwAHg4hK/GDhQ4leV7UbW8uXw0kuD7++kUpIkSZLU35w6uZm5OzP3TLFqPXBzZr6YmU8C48C7SxnPzCcy89/AzcD6iAjgTOC2sv+NwPm1Y91Ylm8Dzirbj6RBJ5ZyUilJkiRJmtl8za68HPhj7fW+EusXXwb8PTNfnhSfcKyy/vmy/SHBzq0kSZIkHbwZJ56KiLuBN02xaktm3jH8UxpcRFwCXAJw0kknNXw2c+OkUpIkSZI0ezN2cjNz7QDH3Q+cWHt9QonRJ/5XYElELCpPa+vb9461LyIWAUeW7ac61+uB6wHGxsZygPOedxHVk9l+7NxKkiRJ0uDma7jyVmBjmRl5JbAaeAB4EFhdZlJeTDU51dbMTOBeYEPZfxNwR+1Ym8ryBuBXZfuRdNNNU8eXLHHGZEmSJEmaq7n+CaEPRcQ+4D3AzyLiToDM3AXcCjwG/AK4LDP/U57SXg7cCewGbi3bAnwBuCIixqm+c3tDid8ALCvxK4BX/+zQKLrwQvje92DFiuqp7ooV1esDB5o+M0mSJEkafTHCD0WnNTY2ltu3b2/6NCRJkiRJ8yAiHsrMscnx+RquLEmSJEnSgrOTK0mSJEnqDDu5kiRJkqTOsJMrSZIkSeoMO7mSJEmSpM6wkytJkiRJ6gw7uZIkSZKkzrCTK0mSJEnqDDu5kiRJkqTOsJMrSZIkSeoMO7mSJEmSpM6wkytJkiRJ6gw7uZIkSZKkzrCTK0mSJEnqDDu5kiRJkqTOsJMrSZIkSeoMO7mSJEmSpM6IzGz6HOZFRPwZeLrp85jB0cBfmj4JTWCdtI910k7WS/tYJ+1kvbSPddJO1kv7jEKdrMjMYyYHO9vJYe/ftQAABpFJREFUHQURsT0zx5o+D/2PddI+1kk7WS/tY520k/XSPtZJO1kv7TPKdeJwZUmSJElSZ9jJlSRJkiR1hp3cZl3f9Ano/1gn7WOdtJP10j7WSTtZL+1jnbST9dI+I1snfidXkiRJktQZPsmVJEmSJHWGndwhioiPRsSuiHglIsYmrftiRIxHxJ6IOKcWX1di4xGxuRZfGRHbSvyWiFhc4oeX1+Nl/ckL9f5GXbluO0p5KiJ2lPjJEfGv2rrravu8KyIeLdf76oiIEj8qIu6KiMfLv0ubel+jLiK+HBH7a9f/vNq6oeSNZicivhoRv4+IRyLi9ohYUuLmSkv1ywkNX0ScGBH3RsRj5Xf+Z0p8aG2ZBlN+tz9arv/2EpuyDYrK1eXaPxIRp9aOs6ls/3hEbGrq/Yy6iHhbLR92RMQLEfFZc2XhRcS3I+K5iNhZiw0tN/rdAzQqMy1DKsDbgbcBvwbGavE1wO+Aw4GVwF7gsFL2AquAxWWbNWWfW4GNZfk64NKy/GngurK8Ebil6fc9igX4GvClsnwysLPPdg8ApwEB/Bw4t8S/Amwuy5uBK5t+T6NagC8Dn5siPrS8scy6Ts4GFpXlK3s/3+ZKO8t0OWGZl+t9HHBqWT4C+ENpr4bWllkGrpungKMnxaZsg4DzSlsVpe3aVuJHAU+Uf5eW5aVNv7dRL+Xn/U/ACnOlket/BnBq/Xf4MHOj3z1Ak8UnuUOUmbszc88Uq9YDN2fmi5n5JDAOvLuU8cx8IjP/DdwMrC+ffpwJ3Fb2vxE4v3asG8vybcBZrfi0ZISU6/Ux4IczbHcc8MbMvD+rDP4uU9dDvX40PMPMG81CZv4yM18uL+8HTphue3OlcVPmRMPn1FmZ+UxmPlyW/wHsBpZPs8us2rL5PftDUr82aD3w3azcDywpbdk5wF2Z+bfMPADcBaxb6JPuoLOAvZn59DTbmCvzJDPvA/42KTyU3JjhHqAxdnIXxnLgj7XX+0qsX3wZ8PfaTWYvPuFYZf3zZXsdvPcCz2bm47XYyoj4bUT8JiLeW2LLqa59T70ejs3MZ8ryn4Bj5/WMu+/yMiTm27XhrMPMGw3uk1SfyvaYK+3TLyc0z6L6ytApwLYSGkZbpsEl8MuIeCgiLimxfm2Q9bKwNjLx4YK50rxh5cZ09wCNsZM7SxFxd0TsnKL4iVILHGT9XMDEhvYZ4KTMPAW4AvhBRLzxYP/P8qmV05RPY4Z6+SbwZuCdVHXxtUZP9hBxMLkSEVuAl4Hvl5C5IhUR8Qbgx8BnM/MFbMva4PTMPBU4F7gsIs6or7QNakZU82N8EPhRCZkrLdPF3FjU9AmMmsxcO8Bu+4ETa69PKDH6xP9KNTRgUXkqVd++d6x9EbEIOLJsL2aun3LNPgy8q7bPi8CLZfmhiNgLvJXqWteHadbr4dmIOC4znynDNJ4b3rvonoPNm4j4FvDT8nKYeaNJDiJXPg68Hzir/PIzV9prulzRPIiI11B1cL+fmT8ByMxna+vn0pZpQJm5v/z7XETcTjXMtV8b1K9e9gPvmxT/9TyfetedCzzcyxFzpTWGlRvT3QM0xie5C2MrsDGqmZFXAqupvqD9ILA6qhlhF1MN5dhabijvBTaU/TcBd9SO1ZvNbAPwq94NqA7KWuD3mfnqsIqIOCYiDivLq6jq54kyhOOFiDitfN/zIqauh3r9aJZKw9rzIaA3898w80azEBHrgM8DH8zMf9bi5ko7TZkTDZ9TZ5Wf8RuA3Zn59Vp8KG3ZQryHLoqI10fEEb1lqgn0dtK/DdoKXBSV04DnS1t2J3B2RCwtw2jPLjENbsIIOnOlNYaSGzPcAzRnvma0OhQLVaLuo3rS8SxVxffWbaGaGW4PtRnHqGYw+0NZt6UWX0WV2ONUwzsOL/HXltfjZf2qpt/3KBXgO8CnJsU+AuwCdgAPAx+orRujanz3At8AosSXAfcAjwN3A0c1/d5GtQA3AY8Cj1A1rMfV1g0lbyyzrpNxqu/d7CilN6O7udLS0i8nLPNyrU+nGtb3SC1HzhtmW2YZqF5WUc26+7vSTm0p8SnbIKpZYK8p1/5RJv5VjE+WdnAc+ETT722UC/B6qpFWR9Zi5srC18MPqYaGv0TVV7l4mLnR7x6gydK7CZEkSZIkaeQ5XFmSJEmS1Bl2ciVJkiRJnWEnV5IkSZLUGXZyJUmSJEmdYSdXkiRJktQZdnIlSZIkSZ1hJ1eSJEmS1Bl2ciVJkiRJnfFfSfsHxT27PYcAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 1152x648 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Evaluating our model's predictions with regression evaluation metrics"
      ],
      "metadata": {
        "id": "zZFDWszyBTtE"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Evaluate the model on the test \n",
        "model.evaluate(X_test,y_test)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0M9-Rtw8BIvx",
        "outputId": "02803f92-2964-4ba2-ca46-7d53b5765ba9"
      },
      "execution_count": 249,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "32/32 [==============================] - 0s 1ms/step - loss: 14.3982 - mae: 14.1531\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[14.39822769165039, 14.153050422668457]"
            ]
          },
          "metadata": {},
          "execution_count": 249
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#calculate the mae\n",
        "mae = tf.keras.losses.mae(y_test,tf.squeeze(y_pred))\n",
        "mae"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "L0GcN_gRBw-i",
        "outputId": "ce691914-cd06-4282-cdf7-50c9fcf0bb16"
      },
      "execution_count": 257,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tf.Tensor: shape=(), dtype=float32, numpy=14.492734>"
            ]
          },
          "metadata": {},
          "execution_count": 257
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "tf.keras.losses.huber(y_test,tf.squeeze(y_pred))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zHQu6I7WCBOE",
        "outputId": "7bcc9aec-4f42-46f2-cbbe-eaa34edcca1e"
      },
      "execution_count": 258,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tf.Tensor: shape=(), dtype=float32, numpy=13.992734>"
            ]
          },
          "metadata": {},
          "execution_count": 258
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "tf.keras.losses.mean_squared_error(y_test,tf.squeeze(y_pred))**0.5"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5JlNa2RTCGZN",
        "outputId": "1100b3b8-e4de-4529-b7ab-f3f336ac6cf0"
      },
      "execution_count": 260,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tf.Tensor: shape=(), dtype=float32, numpy=14.771125>"
            ]
          },
          "metadata": {},
          "execution_count": 260
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def mae(y_true,y_pred):\n",
        "  return tf.keras.losses.mae(y_true=y_true,y_pred=tf.squeeze(y_pred))\n",
        "def mse(y_true,y_pred):\n",
        "  return tf.keras.losses.mean_squared_error(y_true=y_true,y_pred=tf.squeeze(y_pred))"
      ],
      "metadata": {
        "id": "-r6gHeQkCudz"
      },
      "execution_count": 281,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "tf.random.set_seed(42)\n",
        "\n",
        "model_1 = tf.keras.Sequential([tf.keras.layers.Dense(1)])\n",
        "model_1.compile(loss = tf.keras.losses.mae,\n",
        "                optimizer = tf.keras.optimizers.Adam(learning_rate=0.0001),\n",
        "                metrics = ['mae'])\n",
        "model_1.fit(tf.expand_dims(X_train, axis=-1),y_train , epochs = 100)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2y-Nz4s-DEdM",
        "outputId": "67fb5153-2bd0-4661-83e1-a90ac2457409"
      },
      "execution_count": 265,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/100\n",
            "125/125 [==============================] - 1s 2ms/step - loss: 1798.5562 - mae: 1798.5562\n",
            "Epoch 2/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 1745.7394 - mae: 1745.7394\n",
            "Epoch 3/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 1692.8746 - mae: 1692.8746\n",
            "Epoch 4/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1640.0171 - mae: 1640.0171\n",
            "Epoch 5/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 1587.2333 - mae: 1587.2333\n",
            "Epoch 6/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1534.4362 - mae: 1534.4362\n",
            "Epoch 7/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 1481.6509 - mae: 1481.6509\n",
            "Epoch 8/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 1428.8228 - mae: 1428.8228\n",
            "Epoch 9/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 1376.0140 - mae: 1376.0140\n",
            "Epoch 10/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 1323.2362 - mae: 1323.2362\n",
            "Epoch 11/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1270.4260 - mae: 1270.4260\n",
            "Epoch 12/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 1217.6523 - mae: 1217.6523\n",
            "Epoch 13/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1164.8407 - mae: 1164.8407\n",
            "Epoch 14/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 1112.0177 - mae: 1112.0177\n",
            "Epoch 15/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1059.1979 - mae: 1059.1979\n",
            "Epoch 16/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1006.3981 - mae: 1006.3981\n",
            "Epoch 17/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 953.5740 - mae: 953.5740\n",
            "Epoch 18/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 900.7652 - mae: 900.7652\n",
            "Epoch 19/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 847.9304 - mae: 847.9304\n",
            "Epoch 20/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 795.0850 - mae: 795.0850\n",
            "Epoch 21/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 742.2966 - mae: 742.2966\n",
            "Epoch 22/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 689.4682 - mae: 689.4682\n",
            "Epoch 23/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 636.6749 - mae: 636.6749\n",
            "Epoch 24/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 583.8888 - mae: 583.8888\n",
            "Epoch 25/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 531.0484 - mae: 531.0484\n",
            "Epoch 26/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 478.2477 - mae: 478.2477\n",
            "Epoch 27/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 425.4261 - mae: 425.4261\n",
            "Epoch 28/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 372.6020 - mae: 372.6020\n",
            "Epoch 29/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 319.8245 - mae: 319.8245\n",
            "Epoch 30/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 267.0456 - mae: 267.0456\n",
            "Epoch 31/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 214.2823 - mae: 214.2823\n",
            "Epoch 32/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 161.5495 - mae: 161.5495\n",
            "Epoch 33/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 108.9085 - mae: 108.9085\n",
            "Epoch 34/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 56.5619 - mae: 56.5619\n",
            "Epoch 35/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 13.7739 - mae: 13.7739\n",
            "Epoch 36/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.4150 - mae: 8.4150\n",
            "Epoch 37/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.3960 - mae: 8.3960\n",
            "Epoch 38/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.3816 - mae: 8.3816\n",
            "Epoch 39/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.3619 - mae: 8.3619\n",
            "Epoch 40/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.3517 - mae: 8.3517\n",
            "Epoch 41/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.3372 - mae: 8.3372\n",
            "Epoch 42/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.3289 - mae: 8.3289\n",
            "Epoch 43/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.3153 - mae: 8.3153\n",
            "Epoch 44/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.3010 - mae: 8.3010\n",
            "Epoch 45/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.2957 - mae: 8.2957\n",
            "Epoch 46/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.2823 - mae: 8.2823\n",
            "Epoch 47/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.2770 - mae: 8.2770\n",
            "Epoch 48/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.2597 - mae: 8.2597\n",
            "Epoch 49/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.2565 - mae: 8.2565\n",
            "Epoch 50/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.2467 - mae: 8.2467\n",
            "Epoch 51/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.2435 - mae: 8.2435\n",
            "Epoch 52/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.2252 - mae: 8.2252\n",
            "Epoch 53/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.2100 - mae: 8.2100\n",
            "Epoch 54/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.2114 - mae: 8.2114\n",
            "Epoch 55/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.1940 - mae: 8.1940\n",
            "Epoch 56/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.1862 - mae: 8.1862\n",
            "Epoch 57/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.1662 - mae: 8.1662\n",
            "Epoch 58/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.1821 - mae: 8.1821\n",
            "Epoch 59/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.1572 - mae: 8.1572\n",
            "Epoch 60/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.1435 - mae: 8.1435\n",
            "Epoch 61/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.1491 - mae: 8.1491\n",
            "Epoch 62/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.1436 - mae: 8.1436\n",
            "Epoch 63/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.1257 - mae: 8.1257\n",
            "Epoch 64/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.1120 - mae: 8.1120\n",
            "Epoch 65/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.0950 - mae: 8.0950\n",
            "Epoch 66/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.0890 - mae: 8.0890\n",
            "Epoch 67/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.0764 - mae: 8.0764\n",
            "Epoch 68/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.0729 - mae: 8.0729\n",
            "Epoch 69/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.0714 - mae: 8.0714\n",
            "Epoch 70/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.0567 - mae: 8.0567\n",
            "Epoch 71/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.0550 - mae: 8.0550\n",
            "Epoch 72/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.0607 - mae: 8.0607\n",
            "Epoch 73/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.0361 - mae: 8.0361\n",
            "Epoch 74/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.0247 - mae: 8.0247\n",
            "Epoch 75/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.9969 - mae: 7.9969\n",
            "Epoch 76/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 7.9990 - mae: 7.9990\n",
            "Epoch 77/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 7.9915 - mae: 7.9915\n",
            "Epoch 78/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 7.9822 - mae: 7.9822\n",
            "Epoch 79/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.9766 - mae: 7.9766\n",
            "Epoch 80/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 7.9594 - mae: 7.9594\n",
            "Epoch 81/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.9545 - mae: 7.9545\n",
            "Epoch 82/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.9462 - mae: 7.9462\n",
            "Epoch 83/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.9434 - mae: 7.9434\n",
            "Epoch 84/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 7.9229 - mae: 7.9229\n",
            "Epoch 85/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 7.9035 - mae: 7.9035\n",
            "Epoch 86/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.9128 - mae: 7.9128\n",
            "Epoch 87/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 7.9085 - mae: 7.9085\n",
            "Epoch 88/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.8791 - mae: 7.8791\n",
            "Epoch 89/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.9048 - mae: 7.9048\n",
            "Epoch 90/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.8647 - mae: 7.8647\n",
            "Epoch 91/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.8484 - mae: 7.8484\n",
            "Epoch 92/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.8676 - mae: 7.8676\n",
            "Epoch 93/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.8423 - mae: 7.8423\n",
            "Epoch 94/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.8270 - mae: 7.8270\n",
            "Epoch 95/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.8174 - mae: 7.8174\n",
            "Epoch 96/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 7.8137 - mae: 7.8137\n",
            "Epoch 97/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 7.8163 - mae: 7.8163\n",
            "Epoch 98/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 7.8005 - mae: 7.8005\n",
            "Epoch 99/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 7.7861 - mae: 7.7861\n",
            "Epoch 100/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.7886 - mae: 7.7886\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7fef4def1b10>"
            ]
          },
          "metadata": {},
          "execution_count": 265
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "y_preds_1 = model_1.predict(X_test)"
      ],
      "metadata": {
        "id": "lnWuv5G0Dx3r"
      },
      "execution_count": 266,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plot_predictions(predictions = y_preds_1)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 533
        },
        "id": "K5gp1HHQEH2B",
        "outputId": "1034f878-39e2-43fe-8a50-0ea3ed220b5f"
      },
      "execution_count": 267,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1152x648 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "mae_1 = mae(y_test,tf.squeeze(y_preds_1))\n",
        "mse_1 = mse(y_test,tf.squeeze(y_preds_1))**0.5"
      ],
      "metadata": {
        "id": "9OowM24uEeWy"
      },
      "execution_count": 299,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "tf.random.set_seed(42)\n",
        "\n",
        "model_2 = tf.keras.Sequential([tf.keras.layers.Dense(50),\n",
        "                               tf.keras.layers.Dense(1)])\n",
        "model_2.compile(loss = tf.keras.losses.mae,\n",
        "                optimizer = tf.keras.optimizers.Adam(learning_rate=0.0001),\n",
        "                metrics = ['mae'])\n",
        "model_2.fit(tf.expand_dims(X_train, axis=-1),y_train , epochs = 100)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KCNqw4cLEN2X",
        "outputId": "bc678987-1654-438f-e94d-f1cc6c7b39c6"
      },
      "execution_count": 274,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/100\n",
            "125/125 [==============================] - 1s 2ms/step - loss: 3424.1064 - mae: 3424.1064\n",
            "Epoch 2/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2535.8328 - mae: 2535.8328\n",
            "Epoch 3/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1614.9121 - mae: 1614.9121\n",
            "Epoch 4/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 647.4970 - mae: 647.4970\n",
            "Epoch 5/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 19.3082 - mae: 19.3082\n",
            "Epoch 6/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.4917 - mae: 8.4917\n",
            "Epoch 7/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.2585 - mae: 8.2585\n",
            "Epoch 8/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.2284 - mae: 8.2284\n",
            "Epoch 9/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.0881 - mae: 8.0881\n",
            "Epoch 10/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.9561 - mae: 7.9561\n",
            "Epoch 11/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.7665 - mae: 7.7665\n",
            "Epoch 12/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.9294 - mae: 7.9294\n",
            "Epoch 13/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.6346 - mae: 7.6346\n",
            "Epoch 14/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.5539 - mae: 7.5539\n",
            "Epoch 15/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.4221 - mae: 7.4221\n",
            "Epoch 16/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.2959 - mae: 7.2959\n",
            "Epoch 17/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.2296 - mae: 7.2296\n",
            "Epoch 18/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.1808 - mae: 7.1808\n",
            "Epoch 19/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 7.1401 - mae: 7.1401\n",
            "Epoch 20/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.0548 - mae: 7.0548\n",
            "Epoch 21/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 6.8150 - mae: 6.8150\n",
            "Epoch 22/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 6.9073 - mae: 6.9073\n",
            "Epoch 23/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 6.7349 - mae: 6.7349\n",
            "Epoch 24/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 6.7792 - mae: 6.7792\n",
            "Epoch 25/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 6.7463 - mae: 6.7463\n",
            "Epoch 26/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 6.3191 - mae: 6.3191\n",
            "Epoch 27/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 6.2421 - mae: 6.2421\n",
            "Epoch 28/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 6.2251 - mae: 6.2251\n",
            "Epoch 29/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 6.1517 - mae: 6.1517\n",
            "Epoch 30/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 6.0856 - mae: 6.0856\n",
            "Epoch 31/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 5.9489 - mae: 5.9489\n",
            "Epoch 32/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 5.6714 - mae: 5.6714\n",
            "Epoch 33/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 5.8314 - mae: 5.8314\n",
            "Epoch 34/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 5.5979 - mae: 5.5979\n",
            "Epoch 35/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 5.5214 - mae: 5.5214\n",
            "Epoch 36/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 5.6788 - mae: 5.6788\n",
            "Epoch 37/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 5.4502 - mae: 5.4502\n",
            "Epoch 38/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 5.2335 - mae: 5.2335\n",
            "Epoch 39/100\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 5.0410 - mae: 5.0410\n",
            "Epoch 40/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 5.0894 - mae: 5.0894\n",
            "Epoch 41/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 5.0800 - mae: 5.0800\n",
            "Epoch 42/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 4.9866 - mae: 4.9866\n",
            "Epoch 43/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 4.8417 - mae: 4.8417\n",
            "Epoch 44/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 4.6637 - mae: 4.6637\n",
            "Epoch 45/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 4.5307 - mae: 4.5307\n",
            "Epoch 46/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 4.6425 - mae: 4.6425\n",
            "Epoch 47/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 4.4513 - mae: 4.4513\n",
            "Epoch 48/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 4.2882 - mae: 4.2882\n",
            "Epoch 49/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 4.2336 - mae: 4.2336\n",
            "Epoch 50/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 4.2172 - mae: 4.2172\n",
            "Epoch 51/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 4.1850 - mae: 4.1850\n",
            "Epoch 52/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 3.9473 - mae: 3.9473\n",
            "Epoch 53/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 3.9777 - mae: 3.9777\n",
            "Epoch 54/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 3.7659 - mae: 3.7659\n",
            "Epoch 55/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 4.0578 - mae: 4.0578\n",
            "Epoch 56/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 3.6186 - mae: 3.6186\n",
            "Epoch 57/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 3.5697 - mae: 3.5697\n",
            "Epoch 58/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 3.5668 - mae: 3.5668\n",
            "Epoch 59/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 3.4126 - mae: 3.4126\n",
            "Epoch 60/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 3.4229 - mae: 3.4229\n",
            "Epoch 61/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2.9995 - mae: 2.9995\n",
            "Epoch 62/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2.9957 - mae: 2.9957\n",
            "Epoch 63/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 3.0615 - mae: 3.0615\n",
            "Epoch 64/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 3.0183 - mae: 3.0183\n",
            "Epoch 65/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2.8421 - mae: 2.8421\n",
            "Epoch 66/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2.5174 - mae: 2.5174\n",
            "Epoch 67/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2.6350 - mae: 2.6350\n",
            "Epoch 68/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2.6438 - mae: 2.6438\n",
            "Epoch 69/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2.4821 - mae: 2.4821\n",
            "Epoch 70/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2.3678 - mae: 2.3678\n",
            "Epoch 71/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2.4409 - mae: 2.4409\n",
            "Epoch 72/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2.3280 - mae: 2.3280\n",
            "Epoch 73/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2.0888 - mae: 2.0888\n",
            "Epoch 74/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2.0051 - mae: 2.0051\n",
            "Epoch 75/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2.1919 - mae: 2.1919\n",
            "Epoch 76/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2.0000 - mae: 2.0000\n",
            "Epoch 77/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.9873 - mae: 1.9873\n",
            "Epoch 78/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.6976 - mae: 1.6976\n",
            "Epoch 79/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.7704 - mae: 1.7704\n",
            "Epoch 80/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.7037 - mae: 1.7037\n",
            "Epoch 81/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.5740 - mae: 1.5740\n",
            "Epoch 82/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.5892 - mae: 1.5892\n",
            "Epoch 83/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4959 - mae: 1.4959\n",
            "Epoch 84/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4209 - mae: 1.4209\n",
            "Epoch 85/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4883 - mae: 1.4883\n",
            "Epoch 86/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.6150 - mae: 1.6150\n",
            "Epoch 87/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4741 - mae: 1.4741\n",
            "Epoch 88/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3339 - mae: 1.3339\n",
            "Epoch 89/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.2931 - mae: 1.2931\n",
            "Epoch 90/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3926 - mae: 1.3926\n",
            "Epoch 91/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3310 - mae: 1.3310\n",
            "Epoch 92/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.2747 - mae: 1.2747\n",
            "Epoch 93/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3835 - mae: 1.3835\n",
            "Epoch 94/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4386 - mae: 1.4386\n",
            "Epoch 95/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4387 - mae: 1.4387\n",
            "Epoch 96/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4684 - mae: 1.4684\n",
            "Epoch 97/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.2981 - mae: 1.2981\n",
            "Epoch 98/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4065 - mae: 1.4065\n",
            "Epoch 99/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3920 - mae: 1.3920\n",
            "Epoch 100/100\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.2086 - mae: 1.2086\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7fef5517a990>"
            ]
          },
          "metadata": {},
          "execution_count": 274
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "y_preds_2 = model_2.predict(X_test)\n",
        "\n",
        "mse_2 = mse(y_test,y_preds_2)**0.5\n",
        "mae_2 = mae(y_test,y_preds_2)"
      ],
      "metadata": {
        "id": "5bYgDAyfEbn8"
      },
      "execution_count": 316,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "tf.random.set_seed(42)\n",
        "\n",
        "model_3 = tf.keras.Sequential([tf.keras.layers.Dense(10),\n",
        "                               tf.keras.layers.Dense(10),\n",
        "                               tf.keras.layers.Dense(1)])\n",
        "model_3.compile(loss = tf.keras.losses.mae,\n",
        "                optimizer = tf.keras.optimizers.Adam(learning_rate=0.0001),\n",
        "                metrics = ['mae'])\n",
        "model_3.fit(tf.expand_dims(X_train, axis=-1),y_train , epochs = 200)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ReGE2TcNFEyc",
        "outputId": "20c4031e-c53c-420f-fa5a-e0d074f55a5b"
      },
      "execution_count": 286,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/200\n",
            "125/125 [==============================] - 1s 2ms/step - loss: 4145.8828 - mae: 4145.8828\n",
            "Epoch 2/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 3130.8860 - mae: 3130.8860\n",
            "Epoch 3/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2085.2371 - mae: 2085.2371\n",
            "Epoch 4/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 985.7358 - mae: 985.7358\n",
            "Epoch 5/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 80.1218 - mae: 80.1218\n",
            "Epoch 6/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.5493 - mae: 8.5493\n",
            "Epoch 7/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.3076 - mae: 8.3076\n",
            "Epoch 8/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.2347 - mae: 8.2347\n",
            "Epoch 9/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.1490 - mae: 8.1490\n",
            "Epoch 10/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.9661 - mae: 7.9661\n",
            "Epoch 11/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.8472 - mae: 7.8472\n",
            "Epoch 12/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.8299 - mae: 7.8299\n",
            "Epoch 13/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.7984 - mae: 7.7984\n",
            "Epoch 14/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.7267 - mae: 7.7267\n",
            "Epoch 15/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.5223 - mae: 7.5223\n",
            "Epoch 16/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.3815 - mae: 7.3815\n",
            "Epoch 17/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.2814 - mae: 7.2814\n",
            "Epoch 18/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.2870 - mae: 7.2870\n",
            "Epoch 19/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.2553 - mae: 7.2553\n",
            "Epoch 20/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.1380 - mae: 7.1380\n",
            "Epoch 21/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 6.9191 - mae: 6.9191\n",
            "Epoch 22/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.0208 - mae: 7.0208\n",
            "Epoch 23/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 6.8123 - mae: 6.8123\n",
            "Epoch 24/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 6.7779 - mae: 6.7779\n",
            "Epoch 25/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 6.7151 - mae: 6.7151\n",
            "Epoch 26/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 6.5306 - mae: 6.5306\n",
            "Epoch 27/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 6.3454 - mae: 6.3454\n",
            "Epoch 28/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 6.3678 - mae: 6.3678\n",
            "Epoch 29/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 6.3521 - mae: 6.3521\n",
            "Epoch 30/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 6.4112 - mae: 6.4112\n",
            "Epoch 31/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 6.3171 - mae: 6.3171\n",
            "Epoch 32/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 5.9526 - mae: 5.9526\n",
            "Epoch 33/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 5.9548 - mae: 5.9548\n",
            "Epoch 34/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 5.7253 - mae: 5.7253\n",
            "Epoch 35/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 5.7837 - mae: 5.7837\n",
            "Epoch 36/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 5.4399 - mae: 5.4399\n",
            "Epoch 37/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 5.5126 - mae: 5.5126\n",
            "Epoch 38/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 5.6538 - mae: 5.6538\n",
            "Epoch 39/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 5.2901 - mae: 5.2901\n",
            "Epoch 40/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 5.2515 - mae: 5.2515\n",
            "Epoch 41/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 5.0397 - mae: 5.0397\n",
            "Epoch 42/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 5.3904 - mae: 5.3904\n",
            "Epoch 43/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 5.1387 - mae: 5.1387\n",
            "Epoch 44/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 5.0538 - mae: 5.0538\n",
            "Epoch 45/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 4.8369 - mae: 4.8369\n",
            "Epoch 46/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 4.6794 - mae: 4.6794\n",
            "Epoch 47/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 4.7731 - mae: 4.7731\n",
            "Epoch 48/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 4.8384 - mae: 4.8384\n",
            "Epoch 49/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 4.4165 - mae: 4.4165\n",
            "Epoch 50/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 4.3113 - mae: 4.3113\n",
            "Epoch 51/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 4.2579 - mae: 4.2579\n",
            "Epoch 52/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 4.0877 - mae: 4.0877\n",
            "Epoch 53/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 4.0868 - mae: 4.0868\n",
            "Epoch 54/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 3.8628 - mae: 3.8628\n",
            "Epoch 55/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 3.8306 - mae: 3.8306\n",
            "Epoch 56/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 3.9407 - mae: 3.9407\n",
            "Epoch 57/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 3.8041 - mae: 3.8041\n",
            "Epoch 58/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 3.7152 - mae: 3.7152\n",
            "Epoch 59/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 3.5779 - mae: 3.5779\n",
            "Epoch 60/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 3.7183 - mae: 3.7183\n",
            "Epoch 61/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 3.4034 - mae: 3.4034\n",
            "Epoch 62/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 3.3611 - mae: 3.3611\n",
            "Epoch 63/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 3.1165 - mae: 3.1165\n",
            "Epoch 64/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 3.2381 - mae: 3.2381\n",
            "Epoch 65/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 3.1961 - mae: 3.1961\n",
            "Epoch 66/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2.8281 - mae: 2.8281\n",
            "Epoch 67/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2.8377 - mae: 2.8377\n",
            "Epoch 68/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 3.0040 - mae: 3.0040\n",
            "Epoch 69/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2.6730 - mae: 2.6730\n",
            "Epoch 70/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2.6091 - mae: 2.6091\n",
            "Epoch 71/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2.4469 - mae: 2.4469\n",
            "Epoch 72/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2.6149 - mae: 2.6149\n",
            "Epoch 73/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2.4868 - mae: 2.4868\n",
            "Epoch 74/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2.4197 - mae: 2.4197\n",
            "Epoch 75/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2.2991 - mae: 2.2991\n",
            "Epoch 76/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2.1886 - mae: 2.1886\n",
            "Epoch 77/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2.0878 - mae: 2.0878\n",
            "Epoch 78/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2.1658 - mae: 2.1658\n",
            "Epoch 79/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2.2241 - mae: 2.2241\n",
            "Epoch 80/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.8605 - mae: 1.8605\n",
            "Epoch 81/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.6953 - mae: 1.6953\n",
            "Epoch 82/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.8041 - mae: 1.8041\n",
            "Epoch 83/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.7580 - mae: 1.7580\n",
            "Epoch 84/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.6012 - mae: 1.6012\n",
            "Epoch 85/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.5637 - mae: 1.5637\n",
            "Epoch 86/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2.0013 - mae: 2.0013\n",
            "Epoch 87/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.5469 - mae: 1.5469\n",
            "Epoch 88/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4341 - mae: 1.4341\n",
            "Epoch 89/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3932 - mae: 1.3932\n",
            "Epoch 90/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4503 - mae: 1.4503\n",
            "Epoch 91/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3691 - mae: 1.3691\n",
            "Epoch 92/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.2909 - mae: 1.2909\n",
            "Epoch 93/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.2558 - mae: 1.2558\n",
            "Epoch 94/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4598 - mae: 1.4598\n",
            "Epoch 95/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.5341 - mae: 1.5341\n",
            "Epoch 96/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4127 - mae: 1.4127\n",
            "Epoch 97/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.6254 - mae: 1.6254\n",
            "Epoch 98/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3680 - mae: 1.3680\n",
            "Epoch 99/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.6582 - mae: 1.6582\n",
            "Epoch 100/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3010 - mae: 1.3010\n",
            "Epoch 101/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3150 - mae: 1.3150\n",
            "Epoch 102/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4740 - mae: 1.4740\n",
            "Epoch 103/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3556 - mae: 1.3556\n",
            "Epoch 104/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3685 - mae: 1.3685\n",
            "Epoch 105/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.2838 - mae: 1.2838\n",
            "Epoch 106/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3144 - mae: 1.3144\n",
            "Epoch 107/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.5318 - mae: 1.5318\n",
            "Epoch 108/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3736 - mae: 1.3736\n",
            "Epoch 109/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3591 - mae: 1.3591\n",
            "Epoch 110/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4267 - mae: 1.4267\n",
            "Epoch 111/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4250 - mae: 1.4250\n",
            "Epoch 112/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4296 - mae: 1.4296\n",
            "Epoch 113/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.5318 - mae: 1.5318\n",
            "Epoch 114/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.5286 - mae: 1.5286\n",
            "Epoch 115/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3490 - mae: 1.3490\n",
            "Epoch 116/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4545 - mae: 1.4545\n",
            "Epoch 117/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3481 - mae: 1.3481\n",
            "Epoch 118/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4942 - mae: 1.4942\n",
            "Epoch 119/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.6339 - mae: 1.6339\n",
            "Epoch 120/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.5235 - mae: 1.5235\n",
            "Epoch 121/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.2827 - mae: 1.2827\n",
            "Epoch 122/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4899 - mae: 1.4899\n",
            "Epoch 123/200\n",
            "125/125 [==============================] - 0s 3ms/step - loss: 1.3977 - mae: 1.3977\n",
            "Epoch 124/200\n",
            "125/125 [==============================] - 0s 3ms/step - loss: 1.2720 - mae: 1.2720\n",
            "Epoch 125/200\n",
            "125/125 [==============================] - 0s 3ms/step - loss: 1.2757 - mae: 1.2757\n",
            "Epoch 126/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3040 - mae: 1.3040\n",
            "Epoch 127/200\n",
            "125/125 [==============================] - 0s 3ms/step - loss: 1.3660 - mae: 1.3660\n",
            "Epoch 128/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.2804 - mae: 1.2804\n",
            "Epoch 129/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.2903 - mae: 1.2903\n",
            "Epoch 130/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.2558 - mae: 1.2558\n",
            "Epoch 131/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.5907 - mae: 1.5907\n",
            "Epoch 132/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3620 - mae: 1.3620\n",
            "Epoch 133/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3204 - mae: 1.3204\n",
            "Epoch 134/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.2904 - mae: 1.2904\n",
            "Epoch 135/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3255 - mae: 1.3255\n",
            "Epoch 136/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.2508 - mae: 1.2508\n",
            "Epoch 137/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3814 - mae: 1.3814\n",
            "Epoch 138/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3750 - mae: 1.3750\n",
            "Epoch 139/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4752 - mae: 1.4752\n",
            "Epoch 140/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3701 - mae: 1.3701\n",
            "Epoch 141/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.5187 - mae: 1.5187\n",
            "Epoch 142/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.8710 - mae: 1.8710\n",
            "Epoch 143/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3946 - mae: 1.3946\n",
            "Epoch 144/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4950 - mae: 1.4950\n",
            "Epoch 145/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3504 - mae: 1.3504\n",
            "Epoch 146/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4732 - mae: 1.4732\n",
            "Epoch 147/200\n",
            "125/125 [==============================] - 0s 3ms/step - loss: 1.4121 - mae: 1.4121\n",
            "Epoch 148/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.7484 - mae: 1.7484\n",
            "Epoch 149/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3590 - mae: 1.3590\n",
            "Epoch 150/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4714 - mae: 1.4714\n",
            "Epoch 151/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.5152 - mae: 1.5152\n",
            "Epoch 152/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3490 - mae: 1.3490\n",
            "Epoch 153/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.2701 - mae: 1.2701\n",
            "Epoch 154/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.2875 - mae: 1.2875\n",
            "Epoch 155/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4249 - mae: 1.4249\n",
            "Epoch 156/200\n",
            "125/125 [==============================] - 0s 3ms/step - loss: 1.3629 - mae: 1.3629\n",
            "Epoch 157/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.2556 - mae: 1.2556\n",
            "Epoch 158/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3435 - mae: 1.3435\n",
            "Epoch 159/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4660 - mae: 1.4660\n",
            "Epoch 160/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3977 - mae: 1.3977\n",
            "Epoch 161/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.5240 - mae: 1.5240\n",
            "Epoch 162/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3861 - mae: 1.3861\n",
            "Epoch 163/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.6246 - mae: 1.6246\n",
            "Epoch 164/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4771 - mae: 1.4771\n",
            "Epoch 165/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3596 - mae: 1.3596\n",
            "Epoch 166/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.2794 - mae: 1.2794\n",
            "Epoch 167/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.6516 - mae: 1.6516\n",
            "Epoch 168/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4527 - mae: 1.4527\n",
            "Epoch 169/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4423 - mae: 1.4423\n",
            "Epoch 170/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.2763 - mae: 1.2763\n",
            "Epoch 171/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4343 - mae: 1.4343\n",
            "Epoch 172/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4145 - mae: 1.4145\n",
            "Epoch 173/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3713 - mae: 1.3713\n",
            "Epoch 174/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4406 - mae: 1.4406\n",
            "Epoch 175/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4207 - mae: 1.4207\n",
            "Epoch 176/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3820 - mae: 1.3820\n",
            "Epoch 177/200\n",
            "125/125 [==============================] - 0s 3ms/step - loss: 1.4212 - mae: 1.4212\n",
            "Epoch 178/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3603 - mae: 1.3603\n",
            "Epoch 179/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3538 - mae: 1.3538\n",
            "Epoch 180/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.6966 - mae: 1.6966\n",
            "Epoch 181/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3231 - mae: 1.3231\n",
            "Epoch 182/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4061 - mae: 1.4061\n",
            "Epoch 183/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.5221 - mae: 1.5221\n",
            "Epoch 184/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.6706 - mae: 1.6706\n",
            "Epoch 185/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3477 - mae: 1.3477\n",
            "Epoch 186/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3991 - mae: 1.3991\n",
            "Epoch 187/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.2464 - mae: 1.2464\n",
            "Epoch 188/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.2937 - mae: 1.2937\n",
            "Epoch 189/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3265 - mae: 1.3265\n",
            "Epoch 190/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3880 - mae: 1.3880\n",
            "Epoch 191/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4529 - mae: 1.4529\n",
            "Epoch 192/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.2730 - mae: 1.2730\n",
            "Epoch 193/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.2504 - mae: 1.2504\n",
            "Epoch 194/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3755 - mae: 1.3755\n",
            "Epoch 195/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3826 - mae: 1.3826\n",
            "Epoch 196/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4119 - mae: 1.4119\n",
            "Epoch 197/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.4729 - mae: 1.4729\n",
            "Epoch 198/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3055 - mae: 1.3055\n",
            "Epoch 199/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.2634 - mae: 1.2634\n",
            "Epoch 200/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1.3136 - mae: 1.3136\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7fef58ad0310>"
            ]
          },
          "metadata": {},
          "execution_count": 286
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "y_preds_3 = model_3.predict(X_test)\n",
        "\n",
        "mse_3 = mse(y_test,y_preds_3)**0.5\n",
        "mae_3 = mae(y_test,y_preds_3)"
      ],
      "metadata": {
        "id": "z-2IFeB6FkYU"
      },
      "execution_count": 320,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model_3.evaluate(X_test,y_test)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zgvl6gtmFrIN",
        "outputId": "27882a71-993d-4df1-8f87-46bd0450a1a7"
      },
      "execution_count": 289,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "32/32 [==============================] - 0s 1ms/step - loss: 2.4368 - mae: 2.3973\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[2.4367806911468506, 2.397286891937256]"
            ]
          },
          "metadata": {},
          "execution_count": 289
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "tf.random.set_seed(42)\n",
        "\n",
        "model_4 =tf.keras.Sequential([\n",
        "                              tf.keras.layers.Dense(10),\n",
        "                              tf.keras.layers.Dense(1)\n",
        "])\n",
        "\n",
        "model_4.compile(loss = tf.keras.losses.mae,\n",
        "                optimizer = tf.keras.optimizers.Adam(learning_rate = 0.00001),\n",
        "                metrics = ['mae'])\n",
        "\n",
        "model_4.fit(tf.expand_dims(X_train, axis=-1),y_train , epochs = 200)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "kaZpnXQIGYPn",
        "outputId": "1c8688d2-0d12-4da9-acb2-cdad638ca982"
      },
      "execution_count": 292,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/200\n",
            "125/125 [==============================] - 1s 2ms/step - loss: 2651.0085 - mae: 2651.0085\n",
            "Epoch 2/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2614.8899 - mae: 2614.8899\n",
            "Epoch 3/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2578.6584 - mae: 2578.6584\n",
            "Epoch 4/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2542.3523 - mae: 2542.3523\n",
            "Epoch 5/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2506.0200 - mae: 2506.0200\n",
            "Epoch 6/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2469.5991 - mae: 2469.5991\n",
            "Epoch 7/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2433.1091 - mae: 2433.1091\n",
            "Epoch 8/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2396.5122 - mae: 2396.5122\n",
            "Epoch 9/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2359.8501 - mae: 2359.8501\n",
            "Epoch 10/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2323.1362 - mae: 2323.1362\n",
            "Epoch 11/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2286.3210 - mae: 2286.3210\n",
            "Epoch 12/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2249.4565 - mae: 2249.4565\n",
            "Epoch 13/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2212.4907 - mae: 2212.4907\n",
            "Epoch 14/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 2175.4441 - mae: 2175.4441\n",
            "Epoch 15/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 2138.3257 - mae: 2138.3257\n",
            "Epoch 16/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2101.1497 - mae: 2101.1497\n",
            "Epoch 17/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 2063.8850 - mae: 2063.8850\n",
            "Epoch 18/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 2026.5580 - mae: 2026.5580\n",
            "Epoch 19/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1989.1417 - mae: 1989.1417\n",
            "Epoch 20/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1951.6509 - mae: 1951.6509\n",
            "Epoch 21/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1914.1312 - mae: 1914.1312\n",
            "Epoch 22/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1876.5137 - mae: 1876.5137\n",
            "Epoch 23/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1838.8529 - mae: 1838.8529\n",
            "Epoch 24/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 1801.1302 - mae: 1801.1302\n",
            "Epoch 25/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1763.3013 - mae: 1763.3013\n",
            "Epoch 26/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1725.4321 - mae: 1725.4321\n",
            "Epoch 27/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1687.4807 - mae: 1687.4807\n",
            "Epoch 28/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1649.4601 - mae: 1649.4601\n",
            "Epoch 29/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 1611.4049 - mae: 1611.4049\n",
            "Epoch 30/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 1573.2795 - mae: 1573.2795\n",
            "Epoch 31/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 1535.0920 - mae: 1535.0920\n",
            "Epoch 32/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 1496.8407 - mae: 1496.8407\n",
            "Epoch 33/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1458.5371 - mae: 1458.5371\n",
            "Epoch 34/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1420.1440 - mae: 1420.1440\n",
            "Epoch 35/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1381.7203 - mae: 1381.7203\n",
            "Epoch 36/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1343.1940 - mae: 1343.1940\n",
            "Epoch 37/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 1304.6688 - mae: 1304.6688\n",
            "Epoch 38/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1266.0369 - mae: 1266.0369\n",
            "Epoch 39/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 1227.3772 - mae: 1227.3772\n",
            "Epoch 40/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1188.6230 - mae: 1188.6230\n",
            "Epoch 41/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1149.8596 - mae: 1149.8596\n",
            "Epoch 42/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 1110.9630 - mae: 1110.9630\n",
            "Epoch 43/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 1072.0315 - mae: 1072.0315\n",
            "Epoch 44/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 1033.0278 - mae: 1033.0278\n",
            "Epoch 45/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 993.9941 - mae: 993.9941\n",
            "Epoch 46/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 954.8455 - mae: 954.8455\n",
            "Epoch 47/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 915.6643 - mae: 915.6643\n",
            "Epoch 48/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 876.4048 - mae: 876.4048\n",
            "Epoch 49/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 837.0891 - mae: 837.0891\n",
            "Epoch 50/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 797.7017 - mae: 797.7017\n",
            "Epoch 51/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 758.2466 - mae: 758.2466\n",
            "Epoch 52/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 718.7407 - mae: 718.7407\n",
            "Epoch 53/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 679.1516 - mae: 679.1516\n",
            "Epoch 54/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 639.5218 - mae: 639.5218\n",
            "Epoch 55/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 599.8110 - mae: 599.8110\n",
            "Epoch 56/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 560.0422 - mae: 560.0422\n",
            "Epoch 57/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 520.1816 - mae: 520.1816\n",
            "Epoch 58/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 480.2857 - mae: 480.2857\n",
            "Epoch 59/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 440.2988 - mae: 440.2988\n",
            "Epoch 60/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 400.2343 - mae: 400.2343\n",
            "Epoch 61/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 360.0829 - mae: 360.0829\n",
            "Epoch 62/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 319.8643 - mae: 319.8643\n",
            "Epoch 63/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 279.5811 - mae: 279.5811\n",
            "Epoch 64/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 239.2062 - mae: 239.2062\n",
            "Epoch 65/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 198.7586 - mae: 198.7586\n",
            "Epoch 66/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 158.2274 - mae: 158.2274\n",
            "Epoch 67/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 117.6477 - mae: 117.6477\n",
            "Epoch 68/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 77.0882 - mae: 77.0882\n",
            "Epoch 69/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 37.0280 - mae: 37.0280\n",
            "Epoch 70/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 9.8490 - mae: 9.8490\n",
            "Epoch 71/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.4229 - mae: 8.4229\n",
            "Epoch 72/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.4241 - mae: 8.4241\n",
            "Epoch 73/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.4108 - mae: 8.4108\n",
            "Epoch 74/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.4020 - mae: 8.4020\n",
            "Epoch 75/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.3968 - mae: 8.3968\n",
            "Epoch 76/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.3864 - mae: 8.3864\n",
            "Epoch 77/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.3886 - mae: 8.3886\n",
            "Epoch 78/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.3774 - mae: 8.3774\n",
            "Epoch 79/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.3716 - mae: 8.3716\n",
            "Epoch 80/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.3693 - mae: 8.3693\n",
            "Epoch 81/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.3599 - mae: 8.3599\n",
            "Epoch 82/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.3589 - mae: 8.3589\n",
            "Epoch 83/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.3559 - mae: 8.3559\n",
            "Epoch 84/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.3471 - mae: 8.3471\n",
            "Epoch 85/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.3485 - mae: 8.3485\n",
            "Epoch 86/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.3418 - mae: 8.3418\n",
            "Epoch 87/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.3351 - mae: 8.3351\n",
            "Epoch 88/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.3216 - mae: 8.3216\n",
            "Epoch 89/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.3351 - mae: 8.3351\n",
            "Epoch 90/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.3185 - mae: 8.3185\n",
            "Epoch 91/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.3153 - mae: 8.3153\n",
            "Epoch 92/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.3160 - mae: 8.3160\n",
            "Epoch 93/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.3040 - mae: 8.3040\n",
            "Epoch 94/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.3101 - mae: 8.3101\n",
            "Epoch 95/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.2999 - mae: 8.2999\n",
            "Epoch 96/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.2918 - mae: 8.2918\n",
            "Epoch 97/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.2929 - mae: 8.2929\n",
            "Epoch 98/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.2906 - mae: 8.2906\n",
            "Epoch 99/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.2789 - mae: 8.2789\n",
            "Epoch 100/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.2744 - mae: 8.2744\n",
            "Epoch 101/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.2652 - mae: 8.2652\n",
            "Epoch 102/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.2775 - mae: 8.2775\n",
            "Epoch 103/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.2716 - mae: 8.2716\n",
            "Epoch 104/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.2649 - mae: 8.2649\n",
            "Epoch 105/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.2535 - mae: 8.2535\n",
            "Epoch 106/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.2434 - mae: 8.2434\n",
            "Epoch 107/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.2412 - mae: 8.2412\n",
            "Epoch 108/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.2414 - mae: 8.2414\n",
            "Epoch 109/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.2371 - mae: 8.2371\n",
            "Epoch 110/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.2242 - mae: 8.2242\n",
            "Epoch 111/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.2267 - mae: 8.2267\n",
            "Epoch 112/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.2246 - mae: 8.2246\n",
            "Epoch 113/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.2055 - mae: 8.2055\n",
            "Epoch 114/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.2045 - mae: 8.2045\n",
            "Epoch 115/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.2090 - mae: 8.2090\n",
            "Epoch 116/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.1998 - mae: 8.1998\n",
            "Epoch 117/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.1943 - mae: 8.1943\n",
            "Epoch 118/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.2002 - mae: 8.2002\n",
            "Epoch 119/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.1963 - mae: 8.1963\n",
            "Epoch 120/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.1788 - mae: 8.1788\n",
            "Epoch 121/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.1862 - mae: 8.1862\n",
            "Epoch 122/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.1730 - mae: 8.1730\n",
            "Epoch 123/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.1730 - mae: 8.1730\n",
            "Epoch 124/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.1835 - mae: 8.1835\n",
            "Epoch 125/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.1551 - mae: 8.1551\n",
            "Epoch 126/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.1474 - mae: 8.1474\n",
            "Epoch 127/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.1613 - mae: 8.1613\n",
            "Epoch 128/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.1401 - mae: 8.1401\n",
            "Epoch 129/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.1409 - mae: 8.1409\n",
            "Epoch 130/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.1330 - mae: 8.1330\n",
            "Epoch 131/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.1424 - mae: 8.1424\n",
            "Epoch 132/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.1185 - mae: 8.1185\n",
            "Epoch 133/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.1260 - mae: 8.1260\n",
            "Epoch 134/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.1099 - mae: 8.1099\n",
            "Epoch 135/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.1025 - mae: 8.1025\n",
            "Epoch 136/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.1199 - mae: 8.1199\n",
            "Epoch 137/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.1051 - mae: 8.1051\n",
            "Epoch 138/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.0941 - mae: 8.0941\n",
            "Epoch 139/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.0985 - mae: 8.0985\n",
            "Epoch 140/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.0877 - mae: 8.0877\n",
            "Epoch 141/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.0839 - mae: 8.0839\n",
            "Epoch 142/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.0731 - mae: 8.0731\n",
            "Epoch 143/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.0730 - mae: 8.0730\n",
            "Epoch 144/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.0672 - mae: 8.0672\n",
            "Epoch 145/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.0666 - mae: 8.0666\n",
            "Epoch 146/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.0630 - mae: 8.0630\n",
            "Epoch 147/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.0599 - mae: 8.0599\n",
            "Epoch 148/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.0501 - mae: 8.0501\n",
            "Epoch 149/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.0681 - mae: 8.0681\n",
            "Epoch 150/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.0326 - mae: 8.0326\n",
            "Epoch 151/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.0337 - mae: 8.0337\n",
            "Epoch 152/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.0289 - mae: 8.0289\n",
            "Epoch 153/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.0291 - mae: 8.0291\n",
            "Epoch 154/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.0139 - mae: 8.0139\n",
            "Epoch 155/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.0210 - mae: 8.0210\n",
            "Epoch 156/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 8.0120 - mae: 8.0120\n",
            "Epoch 157/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.0108 - mae: 8.0108\n",
            "Epoch 158/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 8.0273 - mae: 8.0273\n",
            "Epoch 159/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 7.9852 - mae: 7.9852\n",
            "Epoch 160/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.9969 - mae: 7.9969\n",
            "Epoch 161/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 7.9917 - mae: 7.9917\n",
            "Epoch 162/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 7.9912 - mae: 7.9912\n",
            "Epoch 163/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.9862 - mae: 7.9862\n",
            "Epoch 164/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.9744 - mae: 7.9744\n",
            "Epoch 165/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.9723 - mae: 7.9723\n",
            "Epoch 166/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.9613 - mae: 7.9613\n",
            "Epoch 167/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.9658 - mae: 7.9658\n",
            "Epoch 168/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.9576 - mae: 7.9576\n",
            "Epoch 169/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.9582 - mae: 7.9582\n",
            "Epoch 170/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.9457 - mae: 7.9457\n",
            "Epoch 171/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 7.9409 - mae: 7.9409\n",
            "Epoch 172/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.9344 - mae: 7.9344\n",
            "Epoch 173/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.9287 - mae: 7.9287\n",
            "Epoch 174/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.9285 - mae: 7.9285\n",
            "Epoch 175/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.9201 - mae: 7.9201\n",
            "Epoch 176/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.9301 - mae: 7.9301\n",
            "Epoch 177/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.9081 - mae: 7.9081\n",
            "Epoch 178/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 7.9109 - mae: 7.9109\n",
            "Epoch 179/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.9107 - mae: 7.9107\n",
            "Epoch 180/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.9186 - mae: 7.9186\n",
            "Epoch 181/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.8834 - mae: 7.8834\n",
            "Epoch 182/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.8921 - mae: 7.8921\n",
            "Epoch 183/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.8832 - mae: 7.8832\n",
            "Epoch 184/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.8791 - mae: 7.8791\n",
            "Epoch 185/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 7.8762 - mae: 7.8762\n",
            "Epoch 186/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.8752 - mae: 7.8752\n",
            "Epoch 187/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.8723 - mae: 7.8723\n",
            "Epoch 188/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.8794 - mae: 7.8794\n",
            "Epoch 189/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.8601 - mae: 7.8601\n",
            "Epoch 190/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 7.8515 - mae: 7.8515\n",
            "Epoch 191/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.8531 - mae: 7.8531\n",
            "Epoch 192/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.8401 - mae: 7.8401\n",
            "Epoch 193/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.8453 - mae: 7.8453\n",
            "Epoch 194/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.8337 - mae: 7.8337\n",
            "Epoch 195/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.8261 - mae: 7.8261\n",
            "Epoch 196/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.8316 - mae: 7.8316\n",
            "Epoch 197/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.8251 - mae: 7.8251\n",
            "Epoch 198/200\n",
            "125/125 [==============================] - 0s 1ms/step - loss: 7.8263 - mae: 7.8263\n",
            "Epoch 199/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.8093 - mae: 7.8093\n",
            "Epoch 200/200\n",
            "125/125 [==============================] - 0s 2ms/step - loss: 7.8151 - mae: 7.8151\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7fefdc6eca50>"
            ]
          },
          "metadata": {},
          "execution_count": 292
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "y_preds_4 = model_4.predict(X_test)\n",
        "plot_predictions(predictions = y_preds_4)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 533
        },
        "id": "EWVDD2eQGx5u",
        "outputId": "b6387eb5-9821-430e-956e-3a47b6420ff2"
      },
      "execution_count": 294,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1152x648 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "mae_4 = mae(y_test,y_preds_4)\n",
        "mse_4 = mse(y_test,y_preds_4)**0.5"
      ],
      "metadata": {
        "id": "aBrZZdXhHIui"
      },
      "execution_count": 312,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import pandas as pd\n",
        "model_results = [['model_1',mae_1.numpy(),mse_1.numpy()],\n",
        "                 ['model_2',mae_2.numpy(),mse_2.numpy()],\n",
        "                 ['model_3',mae_3.numpy(),mse_3.numpy()],\n",
        "                 ['model_4',mae_4.numpy(),mse_4.numpy()]]\n",
        "all_results = pd.DataFrame(model_results, columns = ['model','mae','mse'])"
      ],
      "metadata": {
        "id": "VTDPL3k6HYI-"
      },
      "execution_count": 321,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "all_results"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 175
        },
        "id": "6jPeqZ4hIm2n",
        "outputId": "e2e99804-3da2-477e-b76e-a03c468e143e"
      },
      "execution_count": 322,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-49c420b5-06e5-4e8a-95de-0835783123d6\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>model</th>\n",
              "      <th>mae</th>\n",
              "      <th>mse</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>model_1</td>\n",
              "      <td>18.692965</td>\n",
              "      <td>18.942373</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>model_2</td>\n",
              "      <td>2.239893</td>\n",
              "      <td>2.883107</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>model_3</td>\n",
              "      <td>2.454816</td>\n",
              "      <td>2.759735</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>model_4</td>\n",
              "      <td>17.852034</td>\n",
              "      <td>18.104015</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-49c420b5-06e5-4e8a-95de-0835783123d6')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-49c420b5-06e5-4e8a-95de-0835783123d6 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-49c420b5-06e5-4e8a-95de-0835783123d6');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "     model        mae        mse\n",
              "0  model_1  18.692965  18.942373\n",
              "1  model_2   2.239893   2.883107\n",
              "2  model_3   2.454816   2.759735\n",
              "3  model_4  17.852034  18.104015"
            ]
          },
          "metadata": {},
          "execution_count": 322
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_2.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "YEa9tK9UI0CO",
        "outputId": "679d5ad9-f8f7-4959-ee47-d77c29769573"
      },
      "execution_count": 323,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_71\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " dense_113 (Dense)           (32, 50)                  100       \n",
            "                                                                 \n",
            " dense_114 (Dense)           (32, 1)                   51        \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 151\n",
            "Trainable params: 151\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "One of your main goals should be to minimize the time between your expirments.\n",
        "\n",
        "TensorBoard - a component of the tensorflow library to help track modelling experiments\n",
        "Weights and Biases"
      ],
      "metadata": {
        "id": "gxrbBHukI_xM"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Saving our models\n",
        "\n",
        "1. The SavedModel Format\n",
        "2. The HDF5 Format"
      ],
      "metadata": {
        "id": "c3v3qRzjJT1b"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model_2.save(\"bestmodel_savedmodel_format\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "g-Ho-hQALBJd",
        "outputId": "2841f255-1c9b-4665-a564-b90c05827c08"
      },
      "execution_count": 325,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "INFO:tensorflow:Assets written to: bestmodel_savedmodel_format/assets\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_2.save('best_model_hdf5_format.h5')"
      ],
      "metadata": {
        "id": "xYj67F8cJ947"
      },
      "execution_count": 326,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Loading in a saved model"
      ],
      "metadata": {
        "id": "ut3EG_BaLrxB"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "bestmodel = tf.keras.models.load_model(\"bestmodel_savedmodel_format\")"
      ],
      "metadata": {
        "id": "jXnOOR2gLWL4"
      },
      "execution_count": 329,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "bestmodel.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "WjAkh8znLqme",
        "outputId": "5bccb740-f1f9-41c4-b684-a80de30da47d"
      },
      "execution_count": 331,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_71\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " dense_113 (Dense)           (None, 50)                100       \n",
            "                                                                 \n",
            " dense_114 (Dense)           (None, 1)                 51        \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 151\n",
            "Trainable params: 151\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_2.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "i7y6y-26L8m3",
        "outputId": "4326b818-b9f9-467b-f4a0-7b820c2bc7d5"
      },
      "execution_count": 332,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_71\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " dense_113 (Dense)           (None, 50)                100       \n",
            "                                                                 \n",
            " dense_114 (Dense)           (None, 1)                 51        \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 151\n",
            "Trainable params: 151\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "np.allclose(bestmodel.predict(X_test),model_2.predict(X_test))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FK9L3mj3MBTy",
        "outputId": "7f418ab4-e25e-415a-8f1f-6e643f530173"
      },
      "execution_count": 333,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "True"
            ]
          },
          "metadata": {},
          "execution_count": 333
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "loaded_model = tf.keras.models.load_model('best_model_hdf5_format.h5')"
      ],
      "metadata": {
        "id": "grEOAwUjMhhC"
      },
      "execution_count": 336,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "np.allclose(loaded_model.predict(X_test),model_2.predict(X_test))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Xjp7uehfMnxb",
        "outputId": "1a86d2cb-6521-4759-eacd-9660baf1c6c8"
      },
      "execution_count": 337,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "True"
            ]
          },
          "metadata": {},
          "execution_count": 337
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "A larger example"
      ],
      "metadata": {
        "id": "biXeqIL_MwrS"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import tensorflow as tf\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "%matplotlib inline"
      ],
      "metadata": {
        "id": "H6c1Z2bONH1L"
      },
      "execution_count": 430,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df = pd.read_csv('https://raw.githubusercontent.com/stedy/Machine-Learning-with-R-datasets/master/insurance.csv')"
      ],
      "metadata": {
        "id": "19_sBMeeN21v"
      },
      "execution_count": 431,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X = df.drop(columns = 'charges', axis = 'columns')\n",
        "y = df['charges']"
      ],
      "metadata": {
        "id": "OmhwPj49PQMW"
      },
      "execution_count": 432,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import train_test_split"
      ],
      "metadata": {
        "id": "2Dcx9SWKOITR"
      },
      "execution_count": 433,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X_train,X_test,y_train,y_test = train_test_split(X,y,test_size = 0.25,shuffle = True,random_state=42)"
      ],
      "metadata": {
        "id": "pUsCx1w6OXOT"
      },
      "execution_count": 434,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "tf.random.set_seed(42)\n",
        "\n",
        "insurance_model = tf.keras.Sequential([\n",
        "                                       tf.keras.layers.Dense(30),\n",
        "                                       tf.keras.layers.Dense(1)\n",
        "])\n",
        "insurance_model.compile(loss = tf.keras.losses.mae,\n",
        "                        optimizer = tf.keras.optimizers.Adam(learning_rate = 0.0001),\n",
        "                        metrics = ['mae'])"
      ],
      "metadata": {
        "id": "iuIAs__UPrv1"
      },
      "execution_count": 389,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "insurance_model.fit(X_train,y_train,epochs = 100)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1vSIjwUOQaJT",
        "outputId": "a9d7edb9-4362-4ce6-fa73-d30a7f3b7fe2"
      },
      "execution_count": 390,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13243.9102 - mae: 13243.9102\n",
            "Epoch 2/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13241.4473 - mae: 13241.4473\n",
            "Epoch 3/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13238.9609 - mae: 13238.9609\n",
            "Epoch 4/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13236.4473 - mae: 13236.4473\n",
            "Epoch 5/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13233.8965 - mae: 13233.8965\n",
            "Epoch 6/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13231.3193 - mae: 13231.3193\n",
            "Epoch 7/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13228.6963 - mae: 13228.6963\n",
            "Epoch 8/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13226.0420 - mae: 13226.0420\n",
            "Epoch 9/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 13223.3359 - mae: 13223.3359\n",
            "Epoch 10/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 13220.5908 - mae: 13220.5908\n",
            "Epoch 11/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13217.7900 - mae: 13217.7900\n",
            "Epoch 12/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 13214.9375 - mae: 13214.9375\n",
            "Epoch 13/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 13212.0439 - mae: 13212.0439\n",
            "Epoch 14/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13209.0889 - mae: 13209.0889\n",
            "Epoch 15/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13206.0908 - mae: 13206.0908\n",
            "Epoch 16/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13203.0264 - mae: 13203.0264\n",
            "Epoch 17/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 13199.9043 - mae: 13199.9043\n",
            "Epoch 18/100\n",
            "32/32 [==============================] - 0s 4ms/step - loss: 13196.7188 - mae: 13196.7188\n",
            "Epoch 19/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13193.4775 - mae: 13193.4775\n",
            "Epoch 20/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13190.1816 - mae: 13190.1816\n",
            "Epoch 21/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13186.8271 - mae: 13186.8271\n",
            "Epoch 22/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13183.4033 - mae: 13183.4033\n",
            "Epoch 23/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13179.9170 - mae: 13179.9170\n",
            "Epoch 24/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 13176.3711 - mae: 13176.3711\n",
            "Epoch 25/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 13172.7539 - mae: 13172.7539\n",
            "Epoch 26/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13169.0820 - mae: 13169.0820\n",
            "Epoch 27/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13165.3379 - mae: 13165.3379\n",
            "Epoch 28/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13161.5254 - mae: 13161.5254\n",
            "Epoch 29/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13157.6455 - mae: 13157.6455\n",
            "Epoch 30/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 13153.6973 - mae: 13153.6973\n",
            "Epoch 31/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13149.6660 - mae: 13149.6660\n",
            "Epoch 32/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13145.5605 - mae: 13145.5605\n",
            "Epoch 33/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 13141.3896 - mae: 13141.3896\n",
            "Epoch 34/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13137.1426 - mae: 13137.1426\n",
            "Epoch 35/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13132.8135 - mae: 13132.8135\n",
            "Epoch 36/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13128.4258 - mae: 13128.4258\n",
            "Epoch 37/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 13123.9453 - mae: 13123.9453\n",
            "Epoch 38/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13119.4072 - mae: 13119.4072\n",
            "Epoch 39/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13114.7891 - mae: 13114.7891\n",
            "Epoch 40/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 13110.0859 - mae: 13110.0859\n",
            "Epoch 41/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 13105.3301 - mae: 13105.3301\n",
            "Epoch 42/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13100.4912 - mae: 13100.4912\n",
            "Epoch 43/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13095.5811 - mae: 13095.5811\n",
            "Epoch 44/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13090.5918 - mae: 13090.5918\n",
            "Epoch 45/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13085.5381 - mae: 13085.5381\n",
            "Epoch 46/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13080.4072 - mae: 13080.4072\n",
            "Epoch 47/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13075.2070 - mae: 13075.2070\n",
            "Epoch 48/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13069.9414 - mae: 13069.9414\n",
            "Epoch 49/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 13064.6191 - mae: 13064.6191\n",
            "Epoch 50/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13059.2031 - mae: 13059.2031\n",
            "Epoch 51/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13053.7314 - mae: 13053.7314\n",
            "Epoch 52/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13048.1885 - mae: 13048.1885\n",
            "Epoch 53/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13042.5742 - mae: 13042.5742\n",
            "Epoch 54/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13036.9150 - mae: 13036.9150\n",
            "Epoch 55/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13031.1680 - mae: 13031.1680\n",
            "Epoch 56/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13025.3613 - mae: 13025.3613\n",
            "Epoch 57/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13019.4863 - mae: 13019.4863\n",
            "Epoch 58/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13013.5674 - mae: 13013.5674\n",
            "Epoch 59/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13007.5557 - mae: 13007.5557\n",
            "Epoch 60/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13001.4834 - mae: 13001.4834\n",
            "Epoch 61/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12995.3301 - mae: 12995.3301\n",
            "Epoch 62/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12989.1221 - mae: 12989.1221\n",
            "Epoch 63/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 12982.8359 - mae: 12982.8359\n",
            "Epoch 64/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12976.4863 - mae: 12976.4863\n",
            "Epoch 65/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12970.0771 - mae: 12970.0771\n",
            "Epoch 66/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12963.6035 - mae: 12963.6035\n",
            "Epoch 67/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12957.0635 - mae: 12957.0635\n",
            "Epoch 68/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12950.4590 - mae: 12950.4590\n",
            "Epoch 69/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12943.7949 - mae: 12943.7949\n",
            "Epoch 70/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12937.0566 - mae: 12937.0566\n",
            "Epoch 71/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12930.2549 - mae: 12930.2549\n",
            "Epoch 72/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 12923.3906 - mae: 12923.3906\n",
            "Epoch 73/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12916.4600 - mae: 12916.4600\n",
            "Epoch 74/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12909.4590 - mae: 12909.4590\n",
            "Epoch 75/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 12902.4023 - mae: 12902.4023\n",
            "Epoch 76/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12895.2715 - mae: 12895.2715\n",
            "Epoch 77/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12888.0840 - mae: 12888.0840\n",
            "Epoch 78/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12880.8145 - mae: 12880.8145\n",
            "Epoch 79/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12873.4873 - mae: 12873.4873\n",
            "Epoch 80/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12866.1094 - mae: 12866.1094\n",
            "Epoch 81/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12858.6592 - mae: 12858.6592\n",
            "Epoch 82/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12851.1523 - mae: 12851.1523\n",
            "Epoch 83/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12843.5908 - mae: 12843.5908\n",
            "Epoch 84/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12835.9551 - mae: 12835.9551\n",
            "Epoch 85/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 12828.2559 - mae: 12828.2559\n",
            "Epoch 86/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12820.4971 - mae: 12820.4971\n",
            "Epoch 87/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12812.6914 - mae: 12812.6914\n",
            "Epoch 88/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12804.7939 - mae: 12804.7939\n",
            "Epoch 89/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12796.8564 - mae: 12796.8564\n",
            "Epoch 90/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12788.8340 - mae: 12788.8340\n",
            "Epoch 91/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12780.7520 - mae: 12780.7520\n",
            "Epoch 92/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12772.6357 - mae: 12772.6357\n",
            "Epoch 93/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12764.4248 - mae: 12764.4248\n",
            "Epoch 94/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12756.1680 - mae: 12756.1680\n",
            "Epoch 95/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12747.8496 - mae: 12747.8496\n",
            "Epoch 96/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12739.4707 - mae: 12739.4707\n",
            "Epoch 97/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12731.0293 - mae: 12731.0293\n",
            "Epoch 98/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12722.5127 - mae: 12722.5127\n",
            "Epoch 99/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12713.9424 - mae: 12713.9424\n",
            "Epoch 100/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12705.3223 - mae: 12705.3223\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7fef5790abd0>"
            ]
          },
          "metadata": {},
          "execution_count": 390
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "1. Add an extra layer with more hidden units\n",
        "2. Train for longer"
      ],
      "metadata": {
        "id": "pduCCGDAROC8"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "tf.random.set_seed(42)\n",
        "\n",
        "insurance_model_2 = tf.keras.Sequential([\n",
        "                                         tf.keras.layers.Dense(100),\n",
        "                                         tf.keras.layers.Dense(10),\n",
        "                                         tf.keras.layers.Dense(1)\n",
        "])\n",
        "\n",
        "insurance_model_2.compile(loss = tf.keras.losses.mae,\n",
        "                          optimizer = tf.keras.optimizers.Adam(learning_rate = 0.0005),\n",
        "                          metrics = ['mae']\n",
        "                          )\n",
        "\n",
        "insurance_model_2.fit(X_train,y_train,verbose = 1 , epochs = 100)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "I7GusbtpQmU1",
        "outputId": "ad7cb448-ecb7-4a8c-dffa-fe4bbe2aa892"
      },
      "execution_count": 391,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/100\n",
            "32/32 [==============================] - 1s 3ms/step - loss: 13224.9580 - mae: 13224.9580\n",
            "Epoch 2/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 13167.8271 - mae: 13167.8271\n",
            "Epoch 3/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 13087.2861 - mae: 13087.2861\n",
            "Epoch 4/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 12966.6055 - mae: 12966.6055\n",
            "Epoch 5/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12787.7979 - mae: 12787.7979\n",
            "Epoch 6/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 12533.6133 - mae: 12533.6133\n",
            "Epoch 7/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12185.2109 - mae: 12185.2109\n",
            "Epoch 8/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 11727.9121 - mae: 11727.9121\n",
            "Epoch 9/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 11158.0996 - mae: 11158.0996\n",
            "Epoch 10/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 10498.5850 - mae: 10498.5850\n",
            "Epoch 11/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 9812.5586 - mae: 9812.5586\n",
            "Epoch 12/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 9126.8271 - mae: 9126.8271\n",
            "Epoch 13/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 8502.2646 - mae: 8502.2646\n",
            "Epoch 14/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 8005.4917 - mae: 8005.4917\n",
            "Epoch 15/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7680.7051 - mae: 7680.7051\n",
            "Epoch 16/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 7513.9736 - mae: 7513.9736\n",
            "Epoch 17/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 7447.1040 - mae: 7447.1040\n",
            "Epoch 18/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7420.0430 - mae: 7420.0430\n",
            "Epoch 19/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7405.8770 - mae: 7405.8770\n",
            "Epoch 20/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 7394.0933 - mae: 7394.0933\n",
            "Epoch 21/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 7384.0220 - mae: 7384.0220\n",
            "Epoch 22/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 7375.1353 - mae: 7375.1353\n",
            "Epoch 23/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7364.3574 - mae: 7364.3574\n",
            "Epoch 24/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 7355.2920 - mae: 7355.2920\n",
            "Epoch 25/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7345.1353 - mae: 7345.1353\n",
            "Epoch 26/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7333.8687 - mae: 7333.8687\n",
            "Epoch 27/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7323.6743 - mae: 7323.6743\n",
            "Epoch 28/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7313.8149 - mae: 7313.8149\n",
            "Epoch 29/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 7302.7964 - mae: 7302.7964\n",
            "Epoch 30/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7292.7363 - mae: 7292.7363\n",
            "Epoch 31/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7282.5757 - mae: 7282.5757\n",
            "Epoch 32/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7271.1528 - mae: 7271.1528\n",
            "Epoch 33/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7260.7051 - mae: 7260.7051\n",
            "Epoch 34/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7250.2715 - mae: 7250.2715\n",
            "Epoch 35/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7238.3560 - mae: 7238.3560\n",
            "Epoch 36/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 7227.2642 - mae: 7227.2642\n",
            "Epoch 37/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 7216.9160 - mae: 7216.9160\n",
            "Epoch 38/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7206.0815 - mae: 7206.0815\n",
            "Epoch 39/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 7194.2754 - mae: 7194.2754\n",
            "Epoch 40/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7184.4253 - mae: 7184.4253\n",
            "Epoch 41/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 7171.8184 - mae: 7171.8184\n",
            "Epoch 42/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7161.1670 - mae: 7161.1670\n",
            "Epoch 43/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7147.7769 - mae: 7147.7769\n",
            "Epoch 44/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7137.5020 - mae: 7137.5020\n",
            "Epoch 45/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7125.5151 - mae: 7125.5151\n",
            "Epoch 46/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7111.9341 - mae: 7111.9341\n",
            "Epoch 47/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 7099.2769 - mae: 7099.2769\n",
            "Epoch 48/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7087.7456 - mae: 7087.7456\n",
            "Epoch 49/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7075.5679 - mae: 7075.5679\n",
            "Epoch 50/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7060.8496 - mae: 7060.8496\n",
            "Epoch 51/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7048.0991 - mae: 7048.0991\n",
            "Epoch 52/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7036.8755 - mae: 7036.8755\n",
            "Epoch 53/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7022.6440 - mae: 7022.6440\n",
            "Epoch 54/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 7010.8320 - mae: 7010.8320\n",
            "Epoch 55/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6996.1260 - mae: 6996.1260\n",
            "Epoch 56/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6982.4600 - mae: 6982.4600\n",
            "Epoch 57/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6968.2588 - mae: 6968.2588\n",
            "Epoch 58/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6954.5029 - mae: 6954.5029\n",
            "Epoch 59/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6941.4351 - mae: 6941.4351\n",
            "Epoch 60/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6926.5059 - mae: 6926.5059\n",
            "Epoch 61/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6911.5464 - mae: 6911.5464\n",
            "Epoch 62/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6896.7397 - mae: 6896.7397\n",
            "Epoch 63/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6882.1592 - mae: 6882.1592\n",
            "Epoch 64/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6867.7637 - mae: 6867.7637\n",
            "Epoch 65/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6852.1396 - mae: 6852.1396\n",
            "Epoch 66/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6835.8726 - mae: 6835.8726\n",
            "Epoch 67/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6821.6670 - mae: 6821.6670\n",
            "Epoch 68/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6805.1299 - mae: 6805.1299\n",
            "Epoch 69/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6789.9385 - mae: 6789.9385\n",
            "Epoch 70/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6773.0747 - mae: 6773.0747\n",
            "Epoch 71/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6756.0405 - mae: 6756.0405\n",
            "Epoch 72/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6738.5771 - mae: 6738.5771\n",
            "Epoch 73/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6722.1714 - mae: 6722.1714\n",
            "Epoch 74/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6705.9487 - mae: 6705.9487\n",
            "Epoch 75/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6688.3911 - mae: 6688.3911\n",
            "Epoch 76/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6672.9307 - mae: 6672.9307\n",
            "Epoch 77/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6653.9727 - mae: 6653.9727\n",
            "Epoch 78/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6636.6699 - mae: 6636.6699\n",
            "Epoch 79/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6620.9067 - mae: 6620.9067\n",
            "Epoch 80/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6604.5425 - mae: 6604.5425\n",
            "Epoch 81/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6587.9736 - mae: 6587.9736\n",
            "Epoch 82/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6576.0698 - mae: 6576.0698\n",
            "Epoch 83/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6558.6299 - mae: 6558.6299\n",
            "Epoch 84/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6547.4150 - mae: 6547.4150\n",
            "Epoch 85/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6531.2227 - mae: 6531.2227\n",
            "Epoch 86/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6518.1040 - mae: 6518.1040\n",
            "Epoch 87/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6507.3159 - mae: 6507.3159\n",
            "Epoch 88/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6495.6719 - mae: 6495.6719\n",
            "Epoch 89/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6488.7119 - mae: 6488.7119\n",
            "Epoch 90/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6477.6646 - mae: 6477.6646\n",
            "Epoch 91/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6469.0024 - mae: 6469.0024\n",
            "Epoch 92/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6459.9141 - mae: 6459.9141\n",
            "Epoch 93/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6452.1523 - mae: 6452.1523\n",
            "Epoch 94/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6446.2358 - mae: 6446.2358\n",
            "Epoch 95/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6438.7646 - mae: 6438.7646\n",
            "Epoch 96/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6431.7046 - mae: 6431.7046\n",
            "Epoch 97/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6424.0522 - mae: 6424.0522\n",
            "Epoch 98/100\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6418.6260 - mae: 6418.6260\n",
            "Epoch 99/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6413.4858 - mae: 6413.4858\n",
            "Epoch 100/100\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6407.9561 - mae: 6407.9561\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7fef57803c10>"
            ]
          },
          "metadata": {},
          "execution_count": 391
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "insurance_model_2.evaluate(X_test,y_test)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9T9LBxyARxuJ",
        "outputId": "963061f2-b342-4e5d-a2a9-f2a3183e59b8"
      },
      "execution_count": 394,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "11/11 [==============================] - 0s 2ms/step - loss: 6458.7544 - mae: 6458.7544\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[6458.75439453125, 6458.75439453125]"
            ]
          },
          "metadata": {},
          "execution_count": 394
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "insurance_model.evaluate(X_test,y_test)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1TdMi6BISTvF",
        "outputId": "d019cae0-1ca6-48a8-b790-d831a7444813"
      },
      "execution_count": 395,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "11/11 [==============================] - 0s 2ms/step - loss: 12707.6445 - mae: 12707.6445\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[12707.64453125, 12707.64453125]"
            ]
          },
          "metadata": {},
          "execution_count": 395
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "tf.random.set_seed(42)\n",
        "\n",
        "insurance_model_3 = tf.keras.Sequential([\n",
        "                                         tf.keras.layers.Dense(50),\n",
        "                                         tf.keras.layers.Dense(30),\n",
        "                                         tf.keras.layers.Dense(10),\n",
        "                                         tf.keras.layers.Dense(1)\n",
        "])\n",
        "\n",
        "insurance_model_3.compile(loss = tf.keras.losses.mae,\n",
        "                          optimizer = tf.keras.optimizers.Adam(learning_rate=0.0005),\n",
        "                          metrics = ['mae'])\n",
        "\n",
        "history = insurance_model_3.fit(X_train,y_train, epochs = 200)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "VPkVenEwSw8C",
        "outputId": "0c852c29-ee7f-4994-8e71-2f4fef36d7ff"
      },
      "execution_count": 409,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/200\n",
            "32/32 [==============================] - 1s 2ms/step - loss: 13251.7949 - mae: 13251.7949\n",
            "Epoch 2/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13186.9688 - mae: 13186.9688\n",
            "Epoch 3/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 13061.7617 - mae: 13061.7617\n",
            "Epoch 4/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 12794.7549 - mae: 12794.7549\n",
            "Epoch 5/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 12248.9297 - mae: 12248.9297\n",
            "Epoch 6/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 11246.1387 - mae: 11246.1387\n",
            "Epoch 7/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 9770.2988 - mae: 9770.2988\n",
            "Epoch 8/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 8245.8379 - mae: 8245.8379\n",
            "Epoch 9/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7511.0459 - mae: 7511.0459\n",
            "Epoch 10/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 7406.1743 - mae: 7406.1743\n",
            "Epoch 11/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 7389.3223 - mae: 7389.3223\n",
            "Epoch 12/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7370.4360 - mae: 7370.4360\n",
            "Epoch 13/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7352.6675 - mae: 7352.6675\n",
            "Epoch 14/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 7333.1294 - mae: 7333.1294\n",
            "Epoch 15/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7315.0078 - mae: 7315.0078\n",
            "Epoch 16/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 7296.0083 - mae: 7296.0083\n",
            "Epoch 17/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 7282.3672 - mae: 7282.3672\n",
            "Epoch 18/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 7261.3501 - mae: 7261.3501\n",
            "Epoch 19/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7240.3989 - mae: 7240.3989\n",
            "Epoch 20/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7221.5522 - mae: 7221.5522\n",
            "Epoch 21/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7201.2373 - mae: 7201.2373\n",
            "Epoch 22/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7183.4604 - mae: 7183.4604\n",
            "Epoch 23/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7161.4341 - mae: 7161.4341\n",
            "Epoch 24/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7145.2368 - mae: 7145.2368\n",
            "Epoch 25/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 7123.4692 - mae: 7123.4692\n",
            "Epoch 26/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7100.0244 - mae: 7100.0244\n",
            "Epoch 27/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 7077.3281 - mae: 7077.3281\n",
            "Epoch 28/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 7054.8672 - mae: 7054.8672\n",
            "Epoch 29/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7032.4990 - mae: 7032.4990\n",
            "Epoch 30/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 7008.6211 - mae: 7008.6211\n",
            "Epoch 31/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6985.1553 - mae: 6985.1553\n",
            "Epoch 32/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6960.2778 - mae: 6960.2778\n",
            "Epoch 33/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6934.5312 - mae: 6934.5312\n",
            "Epoch 34/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6909.3438 - mae: 6909.3438\n",
            "Epoch 35/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6882.0654 - mae: 6882.0654\n",
            "Epoch 36/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6855.2959 - mae: 6855.2959\n",
            "Epoch 37/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6828.7246 - mae: 6828.7246\n",
            "Epoch 38/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6801.3096 - mae: 6801.3096\n",
            "Epoch 39/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6771.6040 - mae: 6771.6040\n",
            "Epoch 40/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6745.3096 - mae: 6745.3096\n",
            "Epoch 41/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6713.7109 - mae: 6713.7109\n",
            "Epoch 42/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6683.0796 - mae: 6683.0796\n",
            "Epoch 43/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6648.9346 - mae: 6648.9346\n",
            "Epoch 44/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6628.1729 - mae: 6628.1729\n",
            "Epoch 45/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6598.0259 - mae: 6598.0259\n",
            "Epoch 46/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6579.1743 - mae: 6579.1743\n",
            "Epoch 47/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6547.9336 - mae: 6547.9336\n",
            "Epoch 48/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6525.4316 - mae: 6525.4316\n",
            "Epoch 49/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6513.1064 - mae: 6513.1064\n",
            "Epoch 50/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6491.6445 - mae: 6491.6445\n",
            "Epoch 51/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6484.3486 - mae: 6484.3486\n",
            "Epoch 52/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6473.2544 - mae: 6473.2544\n",
            "Epoch 53/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6457.4893 - mae: 6457.4893\n",
            "Epoch 54/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6447.4829 - mae: 6447.4829\n",
            "Epoch 55/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6437.5728 - mae: 6437.5728\n",
            "Epoch 56/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6428.9609 - mae: 6428.9609\n",
            "Epoch 57/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6418.4087 - mae: 6418.4087\n",
            "Epoch 58/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6408.0093 - mae: 6408.0093\n",
            "Epoch 59/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6398.0386 - mae: 6398.0386\n",
            "Epoch 60/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6387.0024 - mae: 6387.0024\n",
            "Epoch 61/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6377.9678 - mae: 6377.9678\n",
            "Epoch 62/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6367.7417 - mae: 6367.7417\n",
            "Epoch 63/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6356.4683 - mae: 6356.4683\n",
            "Epoch 64/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6353.2241 - mae: 6353.2241\n",
            "Epoch 65/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6341.7593 - mae: 6341.7593\n",
            "Epoch 66/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6328.1523 - mae: 6328.1523\n",
            "Epoch 67/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6315.8706 - mae: 6315.8706\n",
            "Epoch 68/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6303.5352 - mae: 6303.5352\n",
            "Epoch 69/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6293.2363 - mae: 6293.2363\n",
            "Epoch 70/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6283.3970 - mae: 6283.3970\n",
            "Epoch 71/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6271.8433 - mae: 6271.8433\n",
            "Epoch 72/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6259.1685 - mae: 6259.1685\n",
            "Epoch 73/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6246.8730 - mae: 6246.8730\n",
            "Epoch 74/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6237.7422 - mae: 6237.7422\n",
            "Epoch 75/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6223.3423 - mae: 6223.3423\n",
            "Epoch 76/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6211.2920 - mae: 6211.2920\n",
            "Epoch 77/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6197.8223 - mae: 6197.8223\n",
            "Epoch 78/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6184.1797 - mae: 6184.1797\n",
            "Epoch 79/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6168.1157 - mae: 6168.1157\n",
            "Epoch 80/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6154.9023 - mae: 6154.9023\n",
            "Epoch 81/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6139.8994 - mae: 6139.8994\n",
            "Epoch 82/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6131.8784 - mae: 6131.8784\n",
            "Epoch 83/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6113.7236 - mae: 6113.7236\n",
            "Epoch 84/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6102.3188 - mae: 6102.3188\n",
            "Epoch 85/200\n",
            "32/32 [==============================] - 0s 4ms/step - loss: 6077.3818 - mae: 6077.3818\n",
            "Epoch 86/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6058.1143 - mae: 6058.1143\n",
            "Epoch 87/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 6043.2559 - mae: 6043.2559\n",
            "Epoch 88/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6026.3208 - mae: 6026.3208\n",
            "Epoch 89/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6006.5840 - mae: 6006.5840\n",
            "Epoch 90/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 5989.4878 - mae: 5989.4878\n",
            "Epoch 91/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 5967.1958 - mae: 5967.1958\n",
            "Epoch 92/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 5944.9473 - mae: 5944.9473\n",
            "Epoch 93/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 5926.3228 - mae: 5926.3228\n",
            "Epoch 94/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 5906.9473 - mae: 5906.9473\n",
            "Epoch 95/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 5878.0449 - mae: 5878.0449\n",
            "Epoch 96/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 5852.5981 - mae: 5852.5981\n",
            "Epoch 97/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 5828.3228 - mae: 5828.3228\n",
            "Epoch 98/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 5801.8853 - mae: 5801.8853\n",
            "Epoch 99/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 5776.7163 - mae: 5776.7163\n",
            "Epoch 100/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 5749.8159 - mae: 5749.8159\n",
            "Epoch 101/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 5713.3467 - mae: 5713.3467\n",
            "Epoch 102/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 5680.3638 - mae: 5680.3638\n",
            "Epoch 103/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 5645.6885 - mae: 5645.6885\n",
            "Epoch 104/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 5613.7612 - mae: 5613.7612\n",
            "Epoch 105/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 5570.5381 - mae: 5570.5381\n",
            "Epoch 106/200\n",
            "32/32 [==============================] - 0s 4ms/step - loss: 5538.5352 - mae: 5538.5352\n",
            "Epoch 107/200\n",
            "32/32 [==============================] - 0s 4ms/step - loss: 5485.9756 - mae: 5485.9756\n",
            "Epoch 108/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 5440.4565 - mae: 5440.4565\n",
            "Epoch 109/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 5392.8501 - mae: 5392.8501\n",
            "Epoch 110/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 5340.8374 - mae: 5340.8374\n",
            "Epoch 111/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 5282.5093 - mae: 5282.5093\n",
            "Epoch 112/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 5242.5127 - mae: 5242.5127\n",
            "Epoch 113/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 5165.4346 - mae: 5165.4346\n",
            "Epoch 114/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 5096.9600 - mae: 5096.9600\n",
            "Epoch 115/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 5025.3364 - mae: 5025.3364\n",
            "Epoch 116/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 4947.8008 - mae: 4947.8008\n",
            "Epoch 117/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 4862.2461 - mae: 4862.2461\n",
            "Epoch 118/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 4780.4517 - mae: 4780.4517\n",
            "Epoch 119/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 4678.8345 - mae: 4678.8345\n",
            "Epoch 120/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 4583.8979 - mae: 4583.8979\n",
            "Epoch 121/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 4464.5005 - mae: 4464.5005\n",
            "Epoch 122/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 4361.7383 - mae: 4361.7383\n",
            "Epoch 123/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 4250.0679 - mae: 4250.0679\n",
            "Epoch 124/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 4152.4180 - mae: 4152.4180\n",
            "Epoch 125/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 4054.7542 - mae: 4054.7542\n",
            "Epoch 126/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3997.8882 - mae: 3997.8882\n",
            "Epoch 127/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3931.6292 - mae: 3931.6292\n",
            "Epoch 128/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3888.7878 - mae: 3888.7878\n",
            "Epoch 129/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3844.9155 - mae: 3844.9155\n",
            "Epoch 130/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3830.5049 - mae: 3830.5049\n",
            "Epoch 131/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3817.2629 - mae: 3817.2629\n",
            "Epoch 132/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3801.4785 - mae: 3801.4785\n",
            "Epoch 133/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3796.3547 - mae: 3796.3547\n",
            "Epoch 134/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3792.6045 - mae: 3792.6045\n",
            "Epoch 135/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3797.2781 - mae: 3797.2781\n",
            "Epoch 136/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3777.0852 - mae: 3777.0852\n",
            "Epoch 137/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3782.3777 - mae: 3782.3777\n",
            "Epoch 138/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3768.1848 - mae: 3768.1848\n",
            "Epoch 139/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3764.9436 - mae: 3764.9436\n",
            "Epoch 140/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3766.7607 - mae: 3766.7607\n",
            "Epoch 141/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3763.0601 - mae: 3763.0601\n",
            "Epoch 142/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3761.4175 - mae: 3761.4175\n",
            "Epoch 143/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3754.2383 - mae: 3754.2383\n",
            "Epoch 144/200\n",
            "32/32 [==============================] - 0s 4ms/step - loss: 3756.7166 - mae: 3756.7166\n",
            "Epoch 145/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3748.0300 - mae: 3748.0300\n",
            "Epoch 146/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3748.5652 - mae: 3748.5652\n",
            "Epoch 147/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3741.9509 - mae: 3741.9509\n",
            "Epoch 148/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3737.9856 - mae: 3737.9856\n",
            "Epoch 149/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3739.1895 - mae: 3739.1895\n",
            "Epoch 150/200\n",
            "32/32 [==============================] - 0s 4ms/step - loss: 3737.5820 - mae: 3737.5820\n",
            "Epoch 151/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3736.7415 - mae: 3736.7415\n",
            "Epoch 152/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3735.2368 - mae: 3735.2368\n",
            "Epoch 153/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3729.6511 - mae: 3729.6511\n",
            "Epoch 154/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3734.3987 - mae: 3734.3987\n",
            "Epoch 155/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3724.8140 - mae: 3724.8140\n",
            "Epoch 156/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3743.7959 - mae: 3743.7959\n",
            "Epoch 157/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3727.6082 - mae: 3727.6082\n",
            "Epoch 158/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3722.5540 - mae: 3722.5540\n",
            "Epoch 159/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3723.1401 - mae: 3723.1401\n",
            "Epoch 160/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3721.0996 - mae: 3721.0996\n",
            "Epoch 161/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3723.2451 - mae: 3723.2451\n",
            "Epoch 162/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3717.0632 - mae: 3717.0632\n",
            "Epoch 163/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3715.8127 - mae: 3715.8127\n",
            "Epoch 164/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3712.4968 - mae: 3712.4968\n",
            "Epoch 165/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3714.6584 - mae: 3714.6584\n",
            "Epoch 166/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3718.6694 - mae: 3718.6694\n",
            "Epoch 167/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3706.8342 - mae: 3706.8342\n",
            "Epoch 168/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3705.8914 - mae: 3705.8914\n",
            "Epoch 169/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3704.7275 - mae: 3704.7275\n",
            "Epoch 170/200\n",
            "32/32 [==============================] - 0s 4ms/step - loss: 3702.3103 - mae: 3702.3103\n",
            "Epoch 171/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3708.6650 - mae: 3708.6650\n",
            "Epoch 172/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3709.8010 - mae: 3709.8010\n",
            "Epoch 173/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3699.4419 - mae: 3699.4419\n",
            "Epoch 174/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3710.4241 - mae: 3710.4241\n",
            "Epoch 175/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3704.2854 - mae: 3704.2854\n",
            "Epoch 176/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3697.4548 - mae: 3697.4548\n",
            "Epoch 177/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3695.7412 - mae: 3695.7412\n",
            "Epoch 178/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3695.7512 - mae: 3695.7512\n",
            "Epoch 179/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3690.6755 - mae: 3690.6755\n",
            "Epoch 180/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3689.1885 - mae: 3689.1885\n",
            "Epoch 181/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3686.4846 - mae: 3686.4846\n",
            "Epoch 182/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3686.6206 - mae: 3686.6206\n",
            "Epoch 183/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3683.4023 - mae: 3683.4023\n",
            "Epoch 184/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3689.1069 - mae: 3689.1069\n",
            "Epoch 185/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3692.6602 - mae: 3692.6602\n",
            "Epoch 186/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3684.4258 - mae: 3684.4258\n",
            "Epoch 187/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3689.3569 - mae: 3689.3569\n",
            "Epoch 188/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3690.2456 - mae: 3690.2456\n",
            "Epoch 189/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3683.1094 - mae: 3683.1094\n",
            "Epoch 190/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3678.0107 - mae: 3678.0107\n",
            "Epoch 191/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3680.5867 - mae: 3680.5867\n",
            "Epoch 192/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3678.3032 - mae: 3678.3032\n",
            "Epoch 193/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3675.2563 - mae: 3675.2563\n",
            "Epoch 194/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3677.2043 - mae: 3677.2043\n",
            "Epoch 195/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3671.2188 - mae: 3671.2188\n",
            "Epoch 196/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3668.3225 - mae: 3668.3225\n",
            "Epoch 197/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3670.5769 - mae: 3670.5769\n",
            "Epoch 198/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3670.1355 - mae: 3670.1355\n",
            "Epoch 199/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3675.2937 - mae: 3675.2937\n",
            "Epoch 200/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3668.8755 - mae: 3668.8755\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "pd.DataFrame(history.history).plot()\n",
        "plt.ylabel('loss')\n",
        "plt.xlabel('epochs')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 296
        },
        "id": "yVn7VdbTS2Yx",
        "outputId": "812382d7-2ff1-41ce-ba44-19687760d60a"
      },
      "execution_count": 407,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 0, 'epochs')"
            ]
          },
          "metadata": {},
          "execution_count": 407
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.preprocessing import MinMaxScaler\n",
        "from sklearn.preprocessing import StandardScaler"
      ],
      "metadata": {
        "id": "U0_swAwiTzVc"
      },
      "execution_count": 411,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "tf.random.set_seed(42)\n",
        "\n",
        "insurance_model_3 = tf.keras.Sequential([\n",
        "                                         tf.keras.layers.Dense(50),\n",
        "                                         tf.keras.layers.Dense(30),\n",
        "                                         tf.keras.layers.Dense(10),\n",
        "                                         tf.keras.layers.Dense(1)\n",
        "])\n",
        "\n",
        "insurance_model_3.compile(loss = tf.keras.losses.mae,\n",
        "                          optimizer = tf.keras.optimizers.Adam(learning_rate=0.0005),\n",
        "                          metrics = ['mae'])\n",
        "\n",
        "history = insurance_model_3.fit(X_train,y_train, epochs = 200)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "J18UL7NfWLu5",
        "outputId": "2b7d6ec2-e8b8-4086-e9f6-d606bdf79113"
      },
      "execution_count": 421,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/200\n",
            "32/32 [==============================] - 1s 3ms/step - loss: 13267.8350 - mae: 13267.8350\n",
            "Epoch 2/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13267.5439 - mae: 13267.5439\n",
            "Epoch 3/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 13267.1025 - mae: 13267.1025\n",
            "Epoch 4/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13266.2314 - mae: 13266.2314\n",
            "Epoch 5/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 13264.2705 - mae: 13264.2705\n",
            "Epoch 6/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 13259.8232 - mae: 13259.8232\n",
            "Epoch 7/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 13250.2686 - mae: 13250.2686\n",
            "Epoch 8/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 13231.6719 - mae: 13231.6719\n",
            "Epoch 9/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 13199.0781 - mae: 13199.0781\n",
            "Epoch 10/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 13146.7969 - mae: 13146.7969\n",
            "Epoch 11/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 13068.3594 - mae: 13068.3594\n",
            "Epoch 12/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 12956.8438 - mae: 12956.8438\n",
            "Epoch 13/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 12804.4717 - mae: 12804.4717\n",
            "Epoch 14/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 12602.8379 - mae: 12602.8379\n",
            "Epoch 15/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 12343.8857 - mae: 12343.8857\n",
            "Epoch 16/200\n",
            "32/32 [==============================] - 0s 4ms/step - loss: 12023.1592 - mae: 12023.1592\n",
            "Epoch 17/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 11638.8496 - mae: 11638.8496\n",
            "Epoch 18/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 11196.4238 - mae: 11196.4238\n",
            "Epoch 19/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 10687.2998 - mae: 10687.2998\n",
            "Epoch 20/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 10078.8545 - mae: 10078.8545\n",
            "Epoch 21/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 9365.5684 - mae: 9365.5684\n",
            "Epoch 22/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 8535.7637 - mae: 8535.7637\n",
            "Epoch 23/200\n",
            "32/32 [==============================] - 0s 4ms/step - loss: 7624.6201 - mae: 7624.6201\n",
            "Epoch 24/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 6639.4927 - mae: 6639.4927\n",
            "Epoch 25/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 5602.4873 - mae: 5602.4873\n",
            "Epoch 26/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 4592.4033 - mae: 4592.4033\n",
            "Epoch 27/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3928.7390 - mae: 3928.7390\n",
            "Epoch 28/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3711.7395 - mae: 3711.7395\n",
            "Epoch 29/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3678.1160 - mae: 3678.1160\n",
            "Epoch 30/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3664.9077 - mae: 3664.9077\n",
            "Epoch 31/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3655.7119 - mae: 3655.7119\n",
            "Epoch 32/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3648.7070 - mae: 3648.7070\n",
            "Epoch 33/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3640.8042 - mae: 3640.8042\n",
            "Epoch 34/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3633.5352 - mae: 3633.5352\n",
            "Epoch 35/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3629.7007 - mae: 3629.7007\n",
            "Epoch 36/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3621.5745 - mae: 3621.5745\n",
            "Epoch 37/200\n",
            "32/32 [==============================] - 0s 4ms/step - loss: 3617.3701 - mae: 3617.3701\n",
            "Epoch 38/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3610.9609 - mae: 3610.9609\n",
            "Epoch 39/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3604.8157 - mae: 3604.8157\n",
            "Epoch 40/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3598.2288 - mae: 3598.2288\n",
            "Epoch 41/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3598.8611 - mae: 3598.8611\n",
            "Epoch 42/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3588.8970 - mae: 3588.8970\n",
            "Epoch 43/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3583.9048 - mae: 3583.9048\n",
            "Epoch 44/200\n",
            "32/32 [==============================] - 0s 4ms/step - loss: 3578.4045 - mae: 3578.4045\n",
            "Epoch 45/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3573.0566 - mae: 3573.0566\n",
            "Epoch 46/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3564.6284 - mae: 3564.6284\n",
            "Epoch 47/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3561.5242 - mae: 3561.5242\n",
            "Epoch 48/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3554.6199 - mae: 3554.6199\n",
            "Epoch 49/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3548.7952 - mae: 3548.7952\n",
            "Epoch 50/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3543.4402 - mae: 3543.4402\n",
            "Epoch 51/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3535.4814 - mae: 3535.4814\n",
            "Epoch 52/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3531.1648 - mae: 3531.1648\n",
            "Epoch 53/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3528.6509 - mae: 3528.6509\n",
            "Epoch 54/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3522.2686 - mae: 3522.2686\n",
            "Epoch 55/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3521.8926 - mae: 3521.8926\n",
            "Epoch 56/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3516.4280 - mae: 3516.4280\n",
            "Epoch 57/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3513.5933 - mae: 3513.5933\n",
            "Epoch 58/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3506.0728 - mae: 3506.0728\n",
            "Epoch 59/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3503.8311 - mae: 3503.8311\n",
            "Epoch 60/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3497.7686 - mae: 3497.7686\n",
            "Epoch 61/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3497.3103 - mae: 3497.3103\n",
            "Epoch 62/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3494.3694 - mae: 3494.3694\n",
            "Epoch 63/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3492.1003 - mae: 3492.1003\n",
            "Epoch 64/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3489.6536 - mae: 3489.6536\n",
            "Epoch 65/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3487.3818 - mae: 3487.3818\n",
            "Epoch 66/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3484.4351 - mae: 3484.4351\n",
            "Epoch 67/200\n",
            "32/32 [==============================] - 0s 4ms/step - loss: 3481.0261 - mae: 3481.0261\n",
            "Epoch 68/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3475.5559 - mae: 3475.5559\n",
            "Epoch 69/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3477.8606 - mae: 3477.8606\n",
            "Epoch 70/200\n",
            "32/32 [==============================] - 0s 4ms/step - loss: 3470.9373 - mae: 3470.9373\n",
            "Epoch 71/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3465.5203 - mae: 3465.5203\n",
            "Epoch 72/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3469.7874 - mae: 3469.7874\n",
            "Epoch 73/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3469.7122 - mae: 3469.7122\n",
            "Epoch 74/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3469.1719 - mae: 3469.1719\n",
            "Epoch 75/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.5374 - mae: 3466.5374\n",
            "Epoch 76/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3468.3713 - mae: 3468.3713\n",
            "Epoch 77/200\n",
            "32/32 [==============================] - 0s 4ms/step - loss: 3466.7490 - mae: 3466.7490\n",
            "Epoch 78/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.6411 - mae: 3466.6411\n",
            "Epoch 79/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3467.7744 - mae: 3467.7744\n",
            "Epoch 80/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3465.2546 - mae: 3465.2546\n",
            "Epoch 81/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.4980 - mae: 3466.4980\n",
            "Epoch 82/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3467.3079 - mae: 3467.3079\n",
            "Epoch 83/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.7363 - mae: 3466.7363\n",
            "Epoch 84/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.6245 - mae: 3466.6245\n",
            "Epoch 85/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.1733 - mae: 3466.1733\n",
            "Epoch 86/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3468.2817 - mae: 3468.2817\n",
            "Epoch 87/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3465.4492 - mae: 3465.4492\n",
            "Epoch 88/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.8413 - mae: 3466.8413\n",
            "Epoch 89/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3465.8704 - mae: 3465.8704\n",
            "Epoch 90/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3467.2185 - mae: 3467.2185\n",
            "Epoch 91/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3468.3618 - mae: 3468.3618\n",
            "Epoch 92/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3467.0994 - mae: 3467.0994\n",
            "Epoch 93/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.9775 - mae: 3466.9775\n",
            "Epoch 94/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.8723 - mae: 3466.8723\n",
            "Epoch 95/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3465.4983 - mae: 3465.4983\n",
            "Epoch 96/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3469.4714 - mae: 3469.4714\n",
            "Epoch 97/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3470.5969 - mae: 3470.5969\n",
            "Epoch 98/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3471.7563 - mae: 3471.7563\n",
            "Epoch 99/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3465.5925 - mae: 3465.5925\n",
            "Epoch 100/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3465.8987 - mae: 3465.8987\n",
            "Epoch 101/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3464.9402 - mae: 3464.9402\n",
            "Epoch 102/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3469.5200 - mae: 3469.5200\n",
            "Epoch 103/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3467.9363 - mae: 3467.9363\n",
            "Epoch 104/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3464.6018 - mae: 3464.6018\n",
            "Epoch 105/200\n",
            "32/32 [==============================] - 0s 4ms/step - loss: 3469.0537 - mae: 3469.0537\n",
            "Epoch 106/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.4775 - mae: 3466.4775\n",
            "Epoch 107/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3465.9624 - mae: 3465.9624\n",
            "Epoch 108/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3467.4937 - mae: 3467.4937\n",
            "Epoch 109/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3464.7844 - mae: 3464.7844\n",
            "Epoch 110/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3465.9392 - mae: 3465.9392\n",
            "Epoch 111/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.3408 - mae: 3466.3408\n",
            "Epoch 112/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.7190 - mae: 3466.7190\n",
            "Epoch 113/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3467.4795 - mae: 3467.4795\n",
            "Epoch 114/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.1785 - mae: 3466.1785\n",
            "Epoch 115/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3466.7529 - mae: 3466.7529\n",
            "Epoch 116/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3469.1758 - mae: 3469.1758\n",
            "Epoch 117/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.2539 - mae: 3466.2539\n",
            "Epoch 118/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3467.0549 - mae: 3467.0549\n",
            "Epoch 119/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3464.9263 - mae: 3464.9263\n",
            "Epoch 120/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3470.6724 - mae: 3470.6724\n",
            "Epoch 121/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3469.7261 - mae: 3469.7261\n",
            "Epoch 122/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3465.3770 - mae: 3465.3770\n",
            "Epoch 123/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3465.5122 - mae: 3465.5122\n",
            "Epoch 124/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3467.2039 - mae: 3467.2039\n",
            "Epoch 125/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3468.3303 - mae: 3468.3303\n",
            "Epoch 126/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3466.8374 - mae: 3466.8374\n",
            "Epoch 127/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3467.1702 - mae: 3467.1702\n",
            "Epoch 128/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3465.6799 - mae: 3465.6799\n",
            "Epoch 129/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.0127 - mae: 3466.0127\n",
            "Epoch 130/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3467.1165 - mae: 3467.1165\n",
            "Epoch 131/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.3284 - mae: 3466.3284\n",
            "Epoch 132/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3466.4763 - mae: 3466.4763\n",
            "Epoch 133/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.3057 - mae: 3466.3057\n",
            "Epoch 134/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.3777 - mae: 3466.3777\n",
            "Epoch 135/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.6477 - mae: 3466.6477\n",
            "Epoch 136/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3467.8794 - mae: 3467.8794\n",
            "Epoch 137/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3467.0276 - mae: 3467.0276\n",
            "Epoch 138/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3468.2170 - mae: 3468.2170\n",
            "Epoch 139/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3469.5786 - mae: 3469.5786\n",
            "Epoch 140/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3468.0476 - mae: 3468.0476\n",
            "Epoch 141/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3467.0093 - mae: 3467.0093\n",
            "Epoch 142/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3465.7949 - mae: 3465.7949\n",
            "Epoch 143/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3467.2539 - mae: 3467.2539\n",
            "Epoch 144/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3465.7615 - mae: 3465.7615\n",
            "Epoch 145/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3470.2739 - mae: 3470.2739\n",
            "Epoch 146/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3469.9375 - mae: 3469.9375\n",
            "Epoch 147/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3467.2607 - mae: 3467.2607\n",
            "Epoch 148/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.7788 - mae: 3466.7788\n",
            "Epoch 149/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.2046 - mae: 3466.2046\n",
            "Epoch 150/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3466.5032 - mae: 3466.5032\n",
            "Epoch 151/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3465.8528 - mae: 3465.8528\n",
            "Epoch 152/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3465.5203 - mae: 3465.5203\n",
            "Epoch 153/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3465.8250 - mae: 3465.8250\n",
            "Epoch 154/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3467.3064 - mae: 3467.3064\n",
            "Epoch 155/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3470.2246 - mae: 3470.2246\n",
            "Epoch 156/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3465.1389 - mae: 3465.1389\n",
            "Epoch 157/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.3076 - mae: 3466.3076\n",
            "Epoch 158/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3467.2324 - mae: 3467.2324\n",
            "Epoch 159/200\n",
            "32/32 [==============================] - 0s 4ms/step - loss: 3466.5464 - mae: 3466.5464\n",
            "Epoch 160/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3467.2200 - mae: 3467.2200\n",
            "Epoch 161/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3465.9355 - mae: 3465.9355\n",
            "Epoch 162/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3471.1145 - mae: 3471.1145\n",
            "Epoch 163/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3463.7312 - mae: 3463.7312\n",
            "Epoch 164/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3465.0959 - mae: 3465.0959\n",
            "Epoch 165/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.0005 - mae: 3466.0005\n",
            "Epoch 166/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3467.2722 - mae: 3467.2722\n",
            "Epoch 167/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3465.4084 - mae: 3465.4084\n",
            "Epoch 168/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3465.4624 - mae: 3465.4624\n",
            "Epoch 169/200\n",
            "32/32 [==============================] - 0s 4ms/step - loss: 3465.6860 - mae: 3465.6860\n",
            "Epoch 170/200\n",
            "32/32 [==============================] - 0s 4ms/step - loss: 3467.0044 - mae: 3467.0044\n",
            "Epoch 171/200\n",
            "32/32 [==============================] - 0s 4ms/step - loss: 3467.8555 - mae: 3467.8555\n",
            "Epoch 172/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3467.4341 - mae: 3467.4341\n",
            "Epoch 173/200\n",
            "32/32 [==============================] - 0s 4ms/step - loss: 3464.4648 - mae: 3464.4648\n",
            "Epoch 174/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3468.4048 - mae: 3468.4048\n",
            "Epoch 175/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3467.6572 - mae: 3467.6572\n",
            "Epoch 176/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3465.5227 - mae: 3465.5227\n",
            "Epoch 177/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.1338 - mae: 3466.1338\n",
            "Epoch 178/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3470.4248 - mae: 3470.4248\n",
            "Epoch 179/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3465.8088 - mae: 3465.8088\n",
            "Epoch 180/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3465.6897 - mae: 3465.6897\n",
            "Epoch 181/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.4526 - mae: 3466.4526\n",
            "Epoch 182/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3465.5481 - mae: 3465.5481\n",
            "Epoch 183/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3464.6865 - mae: 3464.6865\n",
            "Epoch 184/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3468.5049 - mae: 3468.5049\n",
            "Epoch 185/200\n",
            "32/32 [==============================] - 0s 4ms/step - loss: 3464.6265 - mae: 3464.6265\n",
            "Epoch 186/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3471.5759 - mae: 3471.5759\n",
            "Epoch 187/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3467.9946 - mae: 3467.9946\n",
            "Epoch 188/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3463.3242 - mae: 3463.3242\n",
            "Epoch 189/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3467.5315 - mae: 3467.5315\n",
            "Epoch 190/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3472.0000 - mae: 3472.0000\n",
            "Epoch 191/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.4934 - mae: 3466.4934\n",
            "Epoch 192/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.0415 - mae: 3466.0415\n",
            "Epoch 193/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3466.6123 - mae: 3466.6123\n",
            "Epoch 194/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.2129 - mae: 3466.2129\n",
            "Epoch 195/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3465.2185 - mae: 3465.2185\n",
            "Epoch 196/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3466.3093 - mae: 3466.3093\n",
            "Epoch 197/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3464.0652 - mae: 3464.0652\n",
            "Epoch 198/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3465.9475 - mae: 3465.9475\n",
            "Epoch 199/200\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3467.5479 - mae: 3467.5479\n",
            "Epoch 200/200\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3465.0500 - mae: 3465.0500\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "pd.DataFrame(history.history).plot()\n",
        "plt.ylabel('loss')\n",
        "plt.xlabel('epochs')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 296
        },
        "id": "HDkdxFxfWOpo",
        "outputId": "c0404086-d9c7-449d-a71e-f6150929c41e"
      },
      "execution_count": 422,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 0, 'epochs')"
            ]
          },
          "metadata": {},
          "execution_count": 422
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "id": "gGTmX8QOXbXC",
        "outputId": "f9dcb13d-e1cf-48dd-bf32-8e08d61d328e"
      },
      "execution_count": 435,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-4b904936-5495-44bc-858c-7f5707fe5018\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>age</th>\n",
              "      <th>sex</th>\n",
              "      <th>bmi</th>\n",
              "      <th>children</th>\n",
              "      <th>smoker</th>\n",
              "      <th>region</th>\n",
              "      <th>charges</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>19</td>\n",
              "      <td>female</td>\n",
              "      <td>27.900</td>\n",
              "      <td>0</td>\n",
              "      <td>yes</td>\n",
              "      <td>southwest</td>\n",
              "      <td>16884.92400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>18</td>\n",
              "      <td>male</td>\n",
              "      <td>33.770</td>\n",
              "      <td>1</td>\n",
              "      <td>no</td>\n",
              "      <td>southeast</td>\n",
              "      <td>1725.55230</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>28</td>\n",
              "      <td>male</td>\n",
              "      <td>33.000</td>\n",
              "      <td>3</td>\n",
              "      <td>no</td>\n",
              "      <td>southeast</td>\n",
              "      <td>4449.46200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>33</td>\n",
              "      <td>male</td>\n",
              "      <td>22.705</td>\n",
              "      <td>0</td>\n",
              "      <td>no</td>\n",
              "      <td>northwest</td>\n",
              "      <td>21984.47061</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>32</td>\n",
              "      <td>male</td>\n",
              "      <td>28.880</td>\n",
              "      <td>0</td>\n",
              "      <td>no</td>\n",
              "      <td>northwest</td>\n",
              "      <td>3866.85520</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1333</th>\n",
              "      <td>50</td>\n",
              "      <td>male</td>\n",
              "      <td>30.970</td>\n",
              "      <td>3</td>\n",
              "      <td>no</td>\n",
              "      <td>northwest</td>\n",
              "      <td>10600.54830</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1334</th>\n",
              "      <td>18</td>\n",
              "      <td>female</td>\n",
              "      <td>31.920</td>\n",
              "      <td>0</td>\n",
              "      <td>no</td>\n",
              "      <td>northeast</td>\n",
              "      <td>2205.98080</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1335</th>\n",
              "      <td>18</td>\n",
              "      <td>female</td>\n",
              "      <td>36.850</td>\n",
              "      <td>0</td>\n",
              "      <td>no</td>\n",
              "      <td>southeast</td>\n",
              "      <td>1629.83350</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1336</th>\n",
              "      <td>21</td>\n",
              "      <td>female</td>\n",
              "      <td>25.800</td>\n",
              "      <td>0</td>\n",
              "      <td>no</td>\n",
              "      <td>southwest</td>\n",
              "      <td>2007.94500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1337</th>\n",
              "      <td>61</td>\n",
              "      <td>female</td>\n",
              "      <td>29.070</td>\n",
              "      <td>0</td>\n",
              "      <td>yes</td>\n",
              "      <td>northwest</td>\n",
              "      <td>29141.36030</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>1338 rows × 7 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-4b904936-5495-44bc-858c-7f5707fe5018')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-4b904936-5495-44bc-858c-7f5707fe5018 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-4b904936-5495-44bc-858c-7f5707fe5018');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "      age     sex     bmi  children smoker     region      charges\n",
              "0      19  female  27.900         0    yes  southwest  16884.92400\n",
              "1      18    male  33.770         1     no  southeast   1725.55230\n",
              "2      28    male  33.000         3     no  southeast   4449.46200\n",
              "3      33    male  22.705         0     no  northwest  21984.47061\n",
              "4      32    male  28.880         0     no  northwest   3866.85520\n",
              "...   ...     ...     ...       ...    ...        ...          ...\n",
              "1333   50    male  30.970         3     no  northwest  10600.54830\n",
              "1334   18  female  31.920         0     no  northeast   2205.98080\n",
              "1335   18  female  36.850         0     no  southeast   1629.83350\n",
              "1336   21  female  25.800         0     no  southwest   2007.94500\n",
              "1337   61  female  29.070         0    yes  northwest  29141.36030\n",
              "\n",
              "[1338 rows x 7 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 435
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.compose import make_column_transformer\n",
        "from sklearn.preprocessing import MinMaxScaler,OneHotEncoder\n",
        "from sklearn.model_selection import train_test_split"
      ],
      "metadata": {
        "id": "49HMaXUsWntw"
      },
      "execution_count": 447,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "ct = make_column_transformer((MinMaxScaler(),['age','bmi','children']),\n",
        "                             (OneHotEncoder(handle_unknown='ignore'),['sex','smoker','region']),\n",
        "                             remainder = 'passthrough'\n",
        "                             )"
      ],
      "metadata": {
        "id": "ZuTYW0JzXE0w"
      },
      "execution_count": 445,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X = df.drop('charges', axis = 1)\n",
        "y = df['charges']"
      ],
      "metadata": {
        "id": "dIVI0-7JXqmf"
      },
      "execution_count": 446,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X_train,X_test,y_train,y_test = train_test_split(X,y,test_size=0.25 , random_state = 42)"
      ],
      "metadata": {
        "id": "0IW_3_iPYc2h"
      },
      "execution_count": 454,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "ct.fit(X_train)\n",
        "X_train = ct.transform(X_train)\n",
        "X_test = ct.transform(X_test)"
      ],
      "metadata": {
        "id": "2QpbyFiCYoqA"
      },
      "execution_count": 455,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "insurance_model_4 = tf.keras.Sequential([\n",
        "                                        tf.keras.layers.Dense(50),\n",
        "                                        tf.keras.layers.Dense(20),\n",
        "                                        tf.keras.layers.Dense(1)\n",
        "                                    ])\n",
        "\n",
        "insurance_model_4.compile(loss = tf.keras.losses.mae,\n",
        "                          optimizer = tf.keras.optimizers.Adam(0.0005),\n",
        "                          metrics = ['mae'])\n",
        "\n",
        "history = model.fit(X_train,y_train, epochs = 300)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "avceQL7JYwz4",
        "outputId": "ec66b1c6-31a5-4824-c370-4f910a6d4f21"
      },
      "execution_count": 463,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3552.1765 - mae: 3552.1765\n",
            "Epoch 2/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3551.5322 - mae: 3551.5322\n",
            "Epoch 3/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3550.8352 - mae: 3550.8352\n",
            "Epoch 4/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3550.3904 - mae: 3550.3904\n",
            "Epoch 5/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3550.2510 - mae: 3550.2510\n",
            "Epoch 6/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3548.7986 - mae: 3548.7986\n",
            "Epoch 7/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3547.8684 - mae: 3547.8684\n",
            "Epoch 8/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3547.3838 - mae: 3547.3838\n",
            "Epoch 9/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3547.2432 - mae: 3547.2432\n",
            "Epoch 10/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3546.3101 - mae: 3546.3101\n",
            "Epoch 11/300\n",
            "32/32 [==============================] - 0s 4ms/step - loss: 3545.4189 - mae: 3545.4189\n",
            "Epoch 12/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3543.8616 - mae: 3543.8616\n",
            "Epoch 13/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3544.8972 - mae: 3544.8972\n",
            "Epoch 14/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3542.7256 - mae: 3542.7256\n",
            "Epoch 15/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3541.9731 - mae: 3541.9731\n",
            "Epoch 16/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3541.1575 - mae: 3541.1575\n",
            "Epoch 17/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3542.4158 - mae: 3542.4158\n",
            "Epoch 18/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3539.7078 - mae: 3539.7078\n",
            "Epoch 19/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3539.1201 - mae: 3539.1201\n",
            "Epoch 20/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3538.1858 - mae: 3538.1858\n",
            "Epoch 21/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3537.1716 - mae: 3537.1716\n",
            "Epoch 22/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3537.4141 - mae: 3537.4141\n",
            "Epoch 23/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3537.1936 - mae: 3537.1936\n",
            "Epoch 24/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3538.4490 - mae: 3538.4490\n",
            "Epoch 25/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3536.2522 - mae: 3536.2522\n",
            "Epoch 26/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3534.3994 - mae: 3534.3994\n",
            "Epoch 27/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3534.0210 - mae: 3534.0210\n",
            "Epoch 28/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3532.8792 - mae: 3532.8792\n",
            "Epoch 29/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3533.0967 - mae: 3533.0967\n",
            "Epoch 30/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3532.3469 - mae: 3532.3469\n",
            "Epoch 31/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3531.4824 - mae: 3531.4824\n",
            "Epoch 32/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3531.0574 - mae: 3531.0574\n",
            "Epoch 33/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3530.6799 - mae: 3530.6799\n",
            "Epoch 34/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3529.5320 - mae: 3529.5320\n",
            "Epoch 35/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3530.5549 - mae: 3530.5549\n",
            "Epoch 36/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3529.5640 - mae: 3529.5640\n",
            "Epoch 37/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3529.4561 - mae: 3529.4561\n",
            "Epoch 38/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3528.0149 - mae: 3528.0149\n",
            "Epoch 39/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3527.3455 - mae: 3527.3455\n",
            "Epoch 40/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3526.6995 - mae: 3526.6995\n",
            "Epoch 41/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3527.9231 - mae: 3527.9231\n",
            "Epoch 42/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3526.4680 - mae: 3526.4680\n",
            "Epoch 43/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3525.5513 - mae: 3525.5513\n",
            "Epoch 44/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3525.9382 - mae: 3525.9382\n",
            "Epoch 45/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3524.5996 - mae: 3524.5996\n",
            "Epoch 46/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3524.1243 - mae: 3524.1243\n",
            "Epoch 47/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3524.4404 - mae: 3524.4404\n",
            "Epoch 48/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3523.4575 - mae: 3523.4575\n",
            "Epoch 49/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3523.3679 - mae: 3523.3679\n",
            "Epoch 50/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3522.4390 - mae: 3522.4390\n",
            "Epoch 51/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3521.8645 - mae: 3521.8645\n",
            "Epoch 52/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3521.1863 - mae: 3521.1863\n",
            "Epoch 53/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3521.1816 - mae: 3521.1816\n",
            "Epoch 54/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3520.6406 - mae: 3520.6406\n",
            "Epoch 55/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3520.9971 - mae: 3520.9971\n",
            "Epoch 56/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3519.9753 - mae: 3519.9753\n",
            "Epoch 57/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3520.3003 - mae: 3520.3003\n",
            "Epoch 58/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3518.8269 - mae: 3518.8269\n",
            "Epoch 59/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3518.4607 - mae: 3518.4607\n",
            "Epoch 60/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3517.6262 - mae: 3517.6262\n",
            "Epoch 61/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3517.5908 - mae: 3517.5908\n",
            "Epoch 62/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3517.8867 - mae: 3517.8867\n",
            "Epoch 63/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3516.7168 - mae: 3516.7168\n",
            "Epoch 64/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3515.9319 - mae: 3515.9319\n",
            "Epoch 65/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3516.0811 - mae: 3516.0811\n",
            "Epoch 66/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3515.4011 - mae: 3515.4011\n",
            "Epoch 67/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3514.8413 - mae: 3514.8413\n",
            "Epoch 68/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3513.7893 - mae: 3513.7893\n",
            "Epoch 69/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3514.3699 - mae: 3514.3699\n",
            "Epoch 70/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3513.2239 - mae: 3513.2239\n",
            "Epoch 71/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3512.2051 - mae: 3512.2051\n",
            "Epoch 72/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3512.4470 - mae: 3512.4470\n",
            "Epoch 73/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3512.5593 - mae: 3512.5593\n",
            "Epoch 74/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3512.8484 - mae: 3512.8484\n",
            "Epoch 75/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3510.6082 - mae: 3510.6082\n",
            "Epoch 76/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3510.8459 - mae: 3510.8459\n",
            "Epoch 77/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3509.9617 - mae: 3509.9617\n",
            "Epoch 78/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3509.2576 - mae: 3509.2576\n",
            "Epoch 79/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3509.5933 - mae: 3509.5933\n",
            "Epoch 80/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3508.2493 - mae: 3508.2493\n",
            "Epoch 81/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3508.8440 - mae: 3508.8440\n",
            "Epoch 82/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3507.0439 - mae: 3507.0439\n",
            "Epoch 83/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3508.0681 - mae: 3508.0681\n",
            "Epoch 84/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3506.8105 - mae: 3506.8105\n",
            "Epoch 85/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3505.8621 - mae: 3505.8621\n",
            "Epoch 86/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3505.4043 - mae: 3505.4043\n",
            "Epoch 87/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3505.0032 - mae: 3505.0032\n",
            "Epoch 88/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3504.4910 - mae: 3504.4910\n",
            "Epoch 89/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3503.5718 - mae: 3503.5718\n",
            "Epoch 90/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3503.4453 - mae: 3503.4453\n",
            "Epoch 91/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3503.0562 - mae: 3503.0562\n",
            "Epoch 92/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3502.5986 - mae: 3502.5986\n",
            "Epoch 93/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3503.0579 - mae: 3503.0579\n",
            "Epoch 94/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3502.4531 - mae: 3502.4531\n",
            "Epoch 95/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3501.6145 - mae: 3501.6145\n",
            "Epoch 96/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3501.9795 - mae: 3501.9795\n",
            "Epoch 97/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3507.0032 - mae: 3507.0032\n",
            "Epoch 98/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3503.1821 - mae: 3503.1821\n",
            "Epoch 99/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3499.8733 - mae: 3499.8733\n",
            "Epoch 100/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3499.8894 - mae: 3499.8894\n",
            "Epoch 101/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3499.3923 - mae: 3499.3923\n",
            "Epoch 102/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3501.1033 - mae: 3501.1033\n",
            "Epoch 103/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3499.0220 - mae: 3499.0220\n",
            "Epoch 104/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3498.3083 - mae: 3498.3083\n",
            "Epoch 105/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3498.7004 - mae: 3498.7004\n",
            "Epoch 106/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3498.1448 - mae: 3498.1448\n",
            "Epoch 107/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3497.4243 - mae: 3497.4243\n",
            "Epoch 108/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3499.1511 - mae: 3499.1511\n",
            "Epoch 109/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3496.6006 - mae: 3496.6006\n",
            "Epoch 110/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3496.3425 - mae: 3496.3425\n",
            "Epoch 111/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3496.0137 - mae: 3496.0137\n",
            "Epoch 112/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3495.8135 - mae: 3495.8135\n",
            "Epoch 113/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3496.9663 - mae: 3496.9663\n",
            "Epoch 114/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3496.3621 - mae: 3496.3621\n",
            "Epoch 115/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3495.2993 - mae: 3495.2993\n",
            "Epoch 116/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3494.8306 - mae: 3494.8306\n",
            "Epoch 117/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3494.7876 - mae: 3494.7876\n",
            "Epoch 118/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3494.7258 - mae: 3494.7258\n",
            "Epoch 119/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3493.6145 - mae: 3493.6145\n",
            "Epoch 120/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3494.2517 - mae: 3494.2517\n",
            "Epoch 121/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3494.2546 - mae: 3494.2546\n",
            "Epoch 122/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3492.1843 - mae: 3492.1843\n",
            "Epoch 123/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3492.0291 - mae: 3492.0291\n",
            "Epoch 124/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3492.3047 - mae: 3492.3047\n",
            "Epoch 125/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3492.4153 - mae: 3492.4153\n",
            "Epoch 126/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3491.7136 - mae: 3491.7136\n",
            "Epoch 127/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3492.1050 - mae: 3492.1050\n",
            "Epoch 128/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3490.5879 - mae: 3490.5879\n",
            "Epoch 129/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3490.2104 - mae: 3490.2104\n",
            "Epoch 130/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3490.2161 - mae: 3490.2161\n",
            "Epoch 131/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3489.2534 - mae: 3489.2534\n",
            "Epoch 132/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3489.1682 - mae: 3489.1682\n",
            "Epoch 133/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3489.0405 - mae: 3489.0405\n",
            "Epoch 134/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3488.7666 - mae: 3488.7666\n",
            "Epoch 135/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3488.3079 - mae: 3488.3079\n",
            "Epoch 136/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3488.4253 - mae: 3488.4253\n",
            "Epoch 137/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3487.6479 - mae: 3487.6479\n",
            "Epoch 138/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3487.2952 - mae: 3487.2952\n",
            "Epoch 139/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3487.6816 - mae: 3487.6816\n",
            "Epoch 140/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3487.7004 - mae: 3487.7004\n",
            "Epoch 141/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3486.2415 - mae: 3486.2415\n",
            "Epoch 142/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3486.3000 - mae: 3486.3000\n",
            "Epoch 143/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3486.8701 - mae: 3486.8701\n",
            "Epoch 144/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3485.5576 - mae: 3485.5576\n",
            "Epoch 145/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3485.5884 - mae: 3485.5884\n",
            "Epoch 146/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3484.4060 - mae: 3484.4060\n",
            "Epoch 147/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3483.9551 - mae: 3483.9551\n",
            "Epoch 148/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3484.2402 - mae: 3484.2402\n",
            "Epoch 149/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3483.0496 - mae: 3483.0496\n",
            "Epoch 150/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3482.9158 - mae: 3482.9158\n",
            "Epoch 151/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3482.3538 - mae: 3482.3538\n",
            "Epoch 152/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3482.2288 - mae: 3482.2288\n",
            "Epoch 153/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3482.1047 - mae: 3482.1047\n",
            "Epoch 154/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3482.5662 - mae: 3482.5662\n",
            "Epoch 155/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3482.6907 - mae: 3482.6907\n",
            "Epoch 156/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3481.6360 - mae: 3481.6360\n",
            "Epoch 157/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3481.7029 - mae: 3481.7029\n",
            "Epoch 158/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3481.1785 - mae: 3481.1785\n",
            "Epoch 159/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3480.7397 - mae: 3480.7397\n",
            "Epoch 160/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3480.9985 - mae: 3480.9985\n",
            "Epoch 161/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3479.3159 - mae: 3479.3159\n",
            "Epoch 162/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3480.0132 - mae: 3480.0132\n",
            "Epoch 163/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3478.5117 - mae: 3478.5117\n",
            "Epoch 164/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3478.8484 - mae: 3478.8484\n",
            "Epoch 165/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3477.9922 - mae: 3477.9922\n",
            "Epoch 166/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3477.9783 - mae: 3477.9783\n",
            "Epoch 167/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3477.2854 - mae: 3477.2854\n",
            "Epoch 168/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3476.8958 - mae: 3476.8958\n",
            "Epoch 169/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3477.0164 - mae: 3477.0164\n",
            "Epoch 170/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3476.4458 - mae: 3476.4458\n",
            "Epoch 171/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3477.2314 - mae: 3477.2314\n",
            "Epoch 172/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3476.5195 - mae: 3476.5195\n",
            "Epoch 173/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3475.3120 - mae: 3475.3120\n",
            "Epoch 174/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3475.2234 - mae: 3475.2234\n",
            "Epoch 175/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3474.8640 - mae: 3474.8640\n",
            "Epoch 176/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3474.4753 - mae: 3474.4753\n",
            "Epoch 177/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3474.0864 - mae: 3474.0864\n",
            "Epoch 178/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3473.6167 - mae: 3473.6167\n",
            "Epoch 179/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3472.6648 - mae: 3472.6648\n",
            "Epoch 180/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3472.4824 - mae: 3472.4824\n",
            "Epoch 181/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3472.3621 - mae: 3472.3621\n",
            "Epoch 182/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3472.4331 - mae: 3472.4331\n",
            "Epoch 183/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3470.9392 - mae: 3470.9392\n",
            "Epoch 184/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3471.9785 - mae: 3471.9785\n",
            "Epoch 185/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3471.3250 - mae: 3471.3250\n",
            "Epoch 186/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3471.7703 - mae: 3471.7703\n",
            "Epoch 187/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3470.8560 - mae: 3470.8560\n",
            "Epoch 188/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3469.5845 - mae: 3469.5845\n",
            "Epoch 189/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3470.5366 - mae: 3470.5366\n",
            "Epoch 190/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3470.7871 - mae: 3470.7871\n",
            "Epoch 191/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3469.1196 - mae: 3469.1196\n",
            "Epoch 192/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3468.3582 - mae: 3468.3582\n",
            "Epoch 193/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3468.0181 - mae: 3468.0181\n",
            "Epoch 194/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3467.7661 - mae: 3467.7661\n",
            "Epoch 195/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3467.3914 - mae: 3467.3914\n",
            "Epoch 196/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3466.8601 - mae: 3466.8601\n",
            "Epoch 197/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3466.4131 - mae: 3466.4131\n",
            "Epoch 198/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3466.3806 - mae: 3466.3806\n",
            "Epoch 199/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3466.0588 - mae: 3466.0588\n",
            "Epoch 200/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3465.6438 - mae: 3465.6438\n",
            "Epoch 201/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3464.8953 - mae: 3464.8953\n",
            "Epoch 202/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3464.6238 - mae: 3464.6238\n",
            "Epoch 203/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3463.9875 - mae: 3463.9875\n",
            "Epoch 204/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3463.8916 - mae: 3463.8916\n",
            "Epoch 205/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3463.8843 - mae: 3463.8843\n",
            "Epoch 206/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3463.9575 - mae: 3463.9575\n",
            "Epoch 207/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3463.8787 - mae: 3463.8787\n",
            "Epoch 208/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3463.6455 - mae: 3463.6455\n",
            "Epoch 209/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3464.0156 - mae: 3464.0156\n",
            "Epoch 210/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.7585 - mae: 3462.7585\n",
            "Epoch 211/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3463.5913 - mae: 3463.5913\n",
            "Epoch 212/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.6887 - mae: 3462.6887\n",
            "Epoch 213/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3462.8562 - mae: 3462.8562\n",
            "Epoch 214/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3464.9302 - mae: 3464.9302\n",
            "Epoch 215/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3463.0049 - mae: 3463.0049\n",
            "Epoch 216/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3463.2444 - mae: 3463.2444\n",
            "Epoch 217/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3462.2361 - mae: 3462.2361\n",
            "Epoch 218/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3463.9597 - mae: 3463.9597\n",
            "Epoch 219/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.1736 - mae: 3462.1736\n",
            "Epoch 220/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.7676 - mae: 3462.7676\n",
            "Epoch 221/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3464.8799 - mae: 3464.8799\n",
            "Epoch 222/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3463.6086 - mae: 3463.6086\n",
            "Epoch 223/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3461.9148 - mae: 3461.9148\n",
            "Epoch 224/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.0981 - mae: 3462.0981\n",
            "Epoch 225/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3461.5352 - mae: 3461.5352\n",
            "Epoch 226/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.0437 - mae: 3462.0437\n",
            "Epoch 227/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3461.7427 - mae: 3461.7427\n",
            "Epoch 228/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.0903 - mae: 3462.0903\n",
            "Epoch 229/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.2949 - mae: 3462.2949\n",
            "Epoch 230/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3463.0381 - mae: 3463.0381\n",
            "Epoch 231/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3461.8491 - mae: 3461.8491\n",
            "Epoch 232/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.8027 - mae: 3462.8027\n",
            "Epoch 233/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3462.5034 - mae: 3462.5034\n",
            "Epoch 234/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.6436 - mae: 3462.6436\n",
            "Epoch 235/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3461.8267 - mae: 3461.8267\n",
            "Epoch 236/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.3250 - mae: 3462.3250\n",
            "Epoch 237/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3461.6021 - mae: 3461.6021\n",
            "Epoch 238/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.2620 - mae: 3462.2620\n",
            "Epoch 239/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3461.4653 - mae: 3461.4653\n",
            "Epoch 240/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3461.7927 - mae: 3461.7927\n",
            "Epoch 241/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.0710 - mae: 3462.0710\n",
            "Epoch 242/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3461.5833 - mae: 3461.5833\n",
            "Epoch 243/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3463.5918 - mae: 3463.5918\n",
            "Epoch 244/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3461.9277 - mae: 3461.9277\n",
            "Epoch 245/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3463.0891 - mae: 3463.0891\n",
            "Epoch 246/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.0354 - mae: 3462.0354\n",
            "Epoch 247/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.7173 - mae: 3462.7173\n",
            "Epoch 248/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.6821 - mae: 3462.6821\n",
            "Epoch 249/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3462.5195 - mae: 3462.5195\n",
            "Epoch 250/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.0383 - mae: 3462.0383\n",
            "Epoch 251/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3461.5798 - mae: 3461.5798\n",
            "Epoch 252/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3461.5786 - mae: 3461.5786\n",
            "Epoch 253/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3462.4243 - mae: 3462.4243\n",
            "Epoch 254/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.2686 - mae: 3462.2686\n",
            "Epoch 255/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.0671 - mae: 3462.0671\n",
            "Epoch 256/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3462.0698 - mae: 3462.0698\n",
            "Epoch 257/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.3889 - mae: 3462.3889\n",
            "Epoch 258/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.0154 - mae: 3462.0154\n",
            "Epoch 259/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.0852 - mae: 3462.0852\n",
            "Epoch 260/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3464.4490 - mae: 3464.4490\n",
            "Epoch 261/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3461.8386 - mae: 3461.8386\n",
            "Epoch 262/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3461.3953 - mae: 3461.3953\n",
            "Epoch 263/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3461.7551 - mae: 3461.7551\n",
            "Epoch 264/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3462.0898 - mae: 3462.0898\n",
            "Epoch 265/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3462.5293 - mae: 3462.5293\n",
            "Epoch 266/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.0986 - mae: 3462.0986\n",
            "Epoch 267/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3461.4060 - mae: 3461.4060\n",
            "Epoch 268/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3461.9390 - mae: 3461.9390\n",
            "Epoch 269/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.3293 - mae: 3462.3293\n",
            "Epoch 270/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3461.7312 - mae: 3461.7312\n",
            "Epoch 271/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3461.5969 - mae: 3461.5969\n",
            "Epoch 272/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3462.5679 - mae: 3462.5679\n",
            "Epoch 273/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3461.6936 - mae: 3461.6936\n",
            "Epoch 274/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.6785 - mae: 3462.6785\n",
            "Epoch 275/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.0212 - mae: 3462.0212\n",
            "Epoch 276/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3461.7375 - mae: 3461.7375\n",
            "Epoch 277/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.6106 - mae: 3462.6106\n",
            "Epoch 278/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.8130 - mae: 3462.8130\n",
            "Epoch 279/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3463.0371 - mae: 3463.0371\n",
            "Epoch 280/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.1462 - mae: 3462.1462\n",
            "Epoch 281/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3461.7756 - mae: 3461.7756\n",
            "Epoch 282/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3462.5068 - mae: 3462.5068\n",
            "Epoch 283/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3461.6785 - mae: 3461.6785\n",
            "Epoch 284/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3463.1985 - mae: 3463.1985\n",
            "Epoch 285/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3461.8899 - mae: 3461.8899\n",
            "Epoch 286/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3461.1763 - mae: 3461.1763\n",
            "Epoch 287/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3462.0928 - mae: 3462.0928\n",
            "Epoch 288/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3463.0916 - mae: 3463.0916\n",
            "Epoch 289/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3461.5425 - mae: 3461.5425\n",
            "Epoch 290/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3461.5850 - mae: 3461.5850\n",
            "Epoch 291/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.3066 - mae: 3462.3066\n",
            "Epoch 292/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3461.8875 - mae: 3461.8875\n",
            "Epoch 293/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3461.9077 - mae: 3461.9077\n",
            "Epoch 294/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3461.3564 - mae: 3461.3564\n",
            "Epoch 295/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3461.6982 - mae: 3461.6982\n",
            "Epoch 296/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3461.7847 - mae: 3461.7847\n",
            "Epoch 297/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3461.5503 - mae: 3461.5503\n",
            "Epoch 298/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3462.6079 - mae: 3462.6079\n",
            "Epoch 299/300\n",
            "32/32 [==============================] - 0s 2ms/step - loss: 3462.1235 - mae: 3462.1235\n",
            "Epoch 300/300\n",
            "32/32 [==============================] - 0s 3ms/step - loss: 3463.3408 - mae: 3463.3408\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "pd.DataFrame(history.history).plot()\n",
        "plt.ylabel('loss')\n",
        "plt.xlabel('epochs')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 296
        },
        "id": "nTWBlTgEZGLW",
        "outputId": "c763adf3-f6ad-4da7-bb58-f9fbc5f6230b"
      },
      "execution_count": 461,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 0, 'epochs')"
            ]
          },
          "metadata": {},
          "execution_count": 461
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "insurance_model_4.evaluate(X_test,y_test)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "duK-pFckami4",
        "outputId": "709dd5bb-d959-4f0d-963f-aac467e7e32b"
      },
      "execution_count": 473,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "11/11 [==============================] - 0s 4ms/step - loss: 13277.7520 - mae: 13277.7520\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[13277.751953125, 13277.751953125]"
            ]
          },
          "metadata": {},
          "execution_count": 473
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "o7GjPCoKZUgz",
        "outputId": "2df5b3f9-13cf-465f-91cb-34d6f04d61c6"
      },
      "execution_count": 462,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_105\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " dense_213 (Dense)           (None, 50)                600       \n",
            "                                                                 \n",
            " dense_214 (Dense)           (None, 20)                1020      \n",
            "                                                                 \n",
            " dense_215 (Dense)           (None, 1)                 21        \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 1,641\n",
            "Trainable params: 1,641\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        ""
      ],
      "metadata": {
        "id": "-HyJAcxVaHG1"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}