LeNet ARCHITECTURE EX-8

 import tensorflow as tf

import matplotlib.pyplot as plt

from tensorflow.keras import datasets, layers, models, losses

(x_train, y_train), (x_test, y_test)=tf.keras.datasets.mnist.load_data()

x_train = tf.pad(x_train, [[0, 0], [2,2], [2,2]])/255

x_test = tf.pad(x_test, [[0, 0], [2,2], [2,2]])/255

x_train = tf.expand_dims(x_train, axis=3, name=None)

x_test = tf.expand_dims(x_test, axis=3, name=None)

x_val = x_train[-2000:,:,:,:]

y_val = y_train[-2000:]

x_train = x_train[:-2000,:,:,:]

y_train = y_train[:-2000]

model = models.Sequential([

 layers.Conv2D(6, 5, activation='tanh', input_shape=x_train.shape[1:]),

 layers.AveragePooling2D(2),

 layers.Activation('sigmoid'),

 layers.Conv2D(16, 5, activation='tanh'),

 layers.AveragePooling2D(2),

 layers.Activation('sigmoid'),

 layers.Conv2D(120, 5, activation='tanh'),

 layers.Flatten(),

 layers.Dense(84, activation='tanh'),

 layers.Dense(10, activation='softmax')

])

model.summary()

model.compile(optimizer='adam',

 loss=losses.sparse_categorical_crossentropy,

 metrics=['accuracy'])

history = model.fit(x_train, y_train,

 batch_size=64,

 epochs=40,

 validation_data=(x_val, y_val))

fig, axs = plt.subplots(2, 1, figsize=(15,15))

axs[0].plot(history.history['loss'])

axs[0].plot(history.history['val_loss'])

axs[0].title.set_text('Training Loss vs Validation Loss')

axs[0].legend(['Train', 'Val'])

axs[1].plot(history.history['accuracy'])

axs[1].plot(history.history['val_accuracy'])

axs[1].title.set_text('Training Accuracy vs Validation Accuracy')

axs[1].legend(['Train', 'Val'])

results = model.evaluate(x_test, y_test)

print("Loss = {}, Accuray = {}".format(results[0], results[1]))