Updated Micrograd engine for Softmax function

Demos/demo.py

@@ -0,0 +1,103 @@
+import random
+import numpy as np
+import matplotlib.pyplot as plt
+from micrograd.engine import Value
+from micrograd.nn import Neuron, Layer, MLP
+np.random.seed(1337)
+random.seed(1337)
+# make up a dataset
+
+from sklearn.datasets import make_moons, make_blobs
+X, y = make_moons(n_samples=100, noise=0.1)
+
+y = y*2 - 1 # make y be -1 or 1
+# visualize in 2D
+plt.figure(figsize=(5,5))
+plt.scatter(X[:,0], X[:,1], c=y, s=20, cmap='jet')
+# initialize a model 
+model = MLP(2, [16, 16, 1],["tanh", "tanh", "linear"]) # 2-layer neural network
+print(model)
+print("number of parameters", len(model.parameters()))
+
+
+# loss function
+def loss(batch_size=None):
+
+    # inline DataLoader :)
+    if batch_size is None:
+        Xb, yb = X, y
+    else:
+        ri = np.random.permutation(X.shape[0])[:batch_size]
+        Xb, yb = X[ri], y[ri]
+    inputs = [list(map(Value, xrow)) for xrow in Xb]
+
+    # forward the model to get scores
+    scores = list(map(model, inputs))
+
+    # svm "max-margin" loss
+    losses = [(1 + -yi*scorei).relu() for yi, scorei in zip(yb, scores)]
+    data_loss = sum(losses) * (1.0 / len(losses))
+    # L2 regularization
+    alpha = 1e-4
+    reg_loss = alpha * sum((p*p for p in model.parameters()))
+    total_loss = data_loss + reg_loss
+
+    # also get accuracy
+    accuracy = [(yi > 0) == (scorei.data > 0) for yi, scorei in zip(yb, scores)]
+    return total_loss, sum(accuracy) / len(accuracy)
+
+total_loss, acc = loss()
+print(total_loss, acc)
+
+
+
+import numpy as np
+import matplotlib.pyplot as plt
+from matplotlib.animation import FuncAnimation
+
+# Define the mesh grid
+h = 0.25
+x_min, x_max = X[:, 0].min() - 1, X[:, 0].max() + 1
+y_min, y_max = X[:, 1].min() - 1, X[:, 1].max() + 1
+xx, yy = np.meshgrid(np.arange(x_min, x_max, h),
+                     np.arange(y_min, y_max, h))
+Xmesh = np.c_[xx.ravel(), yy.ravel()]
+
+fig, ax = plt.subplots(figsize=(6, 6))
+
+def update(k):
+    """Update function for animation"""
+    global model  # Ensure we modify the model each step
+
+    # Forward pass
+    total_loss, acc = loss()
+
+    # Backward pass
+    model.zero_grad()
+    total_loss.backward()
+
+    # Update weights using SGD
+    learning_rate = 1.0 - 0.9 * k / 100
+    for p in model.parameters():
+        p.data -= learning_rate * p.grad
+
+    # Compute new decision boundary
+    inputs = [list(map(Value, xrow)) for xrow in Xmesh]
+    scores = list(map(model, inputs))
+    Z = np.array([s.data > 0 for s in scores])
+    Z = Z.reshape(xx.shape)
+
+    # Clear and re-plot only necessary parts
+    ax.clear()
+    ax.contourf(xx, yy, Z, cmap=plt.cm.Spectral, alpha=0.8)
+    ax.scatter(X[:, 0], X[:, 1], c=y, s=40, cmap=plt.cm.Spectral)
+    ax.set_xlim(xx.min(), xx.max())
+    ax.set_ylim(yy.min(), yy.max())
+    ax.set_xticks([])
+    ax.set_yticks([])
+    ax.set_title(f"Iteration {k}")
+
+# Run the animation
+ani = FuncAnimation(fig, update, frames=100, interval=50, repeat=False)
+
+plt.show()  # Display animation