ff68020f31663a0c0023a47dc347531f1f4250af,ann_architectures/cifar10/alexnet.py,,,#,24

Before Change

model.add(Dense(10))
model.add(Activation("softmax"))

sgd = SGD(lr=0.0001, decay=1e-6, momentum=0.9, nesterov=True)
model.compile(loss="categorical_crossentropy", optimizer=sgd,
              metrics=["accuracy"])

// Whether to apply global contrast normalization and ZCA whitening

After Change

testflow = testgen.flow(X_test, Y_test, batch_size=batch_size)

checkpointer = ModelCheckpoint(filepath="alexnet2.{epoch:02d}-{val_acc:.2f}.h5",
                               verbose=1, save_best_only=True)

// Fit the model on the batches generated by datagen.flow()
history = model.fit_generator(trainflow, nb_epoch=nb_epoch,
                              samples_per_epoch=X_train.shape[0],
                              validation_data=testflow,
                              nb_val_samples=len(X_test),
                              callbacks=[checkpointer])
plot_history(history)

score = model.evaluate_generator(testflow, val_samples=len(X_test))

Instances