Lista2_Exercício1.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Lista2-Exercício1.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyOiGG9XiKdNZGpm1Mio2kYu",
      "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/nascimento-luciano/Artificial-Intelligence/blob/master/Lista2_Exerc%C3%ADcio1.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PYsHzNntkiGM"
      },
      "source": [
        "*Exercício 1*"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 952
        },
        "id": "t0smjK-wYaR8",
        "outputId": "22830b6c-c903-4bc1-b00c-dfb9b35e36c0"
      },
      "source": [
        "#rede completa para regressão linear\n",
        "import tensorflow as tf\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import keras\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Dense, Activation\n",
        "\n",
        "#Dtrain (data set)\n",
        "xs = np.array([-4.0, -3.0, -2.0, -1.0, 0.0, 1.0, 2.0], dtype=float)\n",
        "ys =2*xs+10\n",
        "\n",
        "#criando a rede neural\n",
        "model=tf.keras.Sequential(name='Regressao_Linear')\n",
        "#configurando a primeira camada da rede\n",
        "model.add(tf.keras.layers.Dense(1, input_dim=1, use_bias=1, activation='linear'))\n",
        "opt=tf.keras.optimizers.SGD(learning_rate=0.01, momentum=0.0)\n",
        "model.compile(optimizer=opt, loss='mean_squared_error', metrics='mean_absolute_error')\n",
        "\n",
        "#verificar quanto os coeficientes angular e linear estão próximos dos esperados\n",
        "first_layer_weights = model.layers[0].get_weights()[0]                           #get_weight() returns a list of weights\n",
        "print('first_layer_weights:', model.layers[0].get_weights()[0].shape)\n",
        "print(model.summary())\n",
        "#treinando a rede\n",
        "history=model.fit(xs, ys, epochs=1000, verbose=0)\n",
        "#testando a rede\n",
        "xs_test=np.linspace(-2,2,11);\n",
        "ys_test=model.predict(xs_test)\n",
        "\n",
        "#Pesos e tendências das camadas após treinar o modelo\n",
        "print(\"Pesos e tendências das camadas após treinar o modelo: \\n\")\n",
        "for layer in model.layers:\n",
        "  print(layer.name)\n",
        "  print(\"weights\")\n",
        "  print(\"shape: \",layer.get_weights()[0].shape,'\\n',layer.get_weights()[0])\n",
        "  print(\"bias\")\n",
        "  print(\"shape: \",layer.get_weights()[1].shape,'\\n',layer.get_weights()[1],'\\n')\n",
        "\n",
        "#fazendo os gráficos\n",
        "plt.rcParams.update({'font.size': 12})\n",
        "plt.figure()\n",
        "plt.plot(xs_test, ys_test, 'r.')\n",
        "plt.plot(xs_test, (2*xs_test+10), 'b-')\n",
        "plt.legend(['previsões', 'reta original'])\n",
        "plt.xlabel('x');plt.ylabel('y')\n",
        "#mostra os dados em history\n",
        "print(history.history.keys())\n",
        "#plota o loss e a metrica\n",
        "plt.figure()\n",
        "plt.subplot(2,1,1)\n",
        "plt.plot(history.history['loss'],'k')\n",
        "plt.legend(['loss'])\n",
        "plt.subplot(2,1,2)\n",
        "plt.plot(history.history['mean_absolute_error'], 'k--')\n",
        "plt.legend(['métrica'])\n",
        "plt.tight_layout()\n",
        "plt.show()\n",
        "\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "first_layer_weights: (1, 1)\n",
            "Model: \"Regressao_Linear\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "dense (Dense)                (None, 1)                 2         \n",
            "=================================================================\n",
            "Total params: 2\n",
            "Trainable params: 2\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "None\n",
            "Pesos e tendências das camadas após treinar o modelo: \n",
            "\n",
            "dense\n",
            "weights\n",
            "shape:  (1, 1) \n",
            " [[1.9999933]]\n",
            "bias\n",
            "shape:  (1,) \n",
            " [9.9999695] \n",
            "\n",
            "dict_keys(['loss', 'mean_absolute_error'])\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}