vgg16_cifar100_train.py

import tensorflow.keras as keras
import tensorflow as tf
import tensorflow.keras.layers as layers
from tensorflow.keras.preprocessing.image import ImageDataGenerator

batch_size = 32

AUTOTUNE = tf.data.experimental.AUTOTUNE

from tensorflow.keras.applications import VGG16
learning_rates = [ .00001]
for lr in learning_rates:
    model = VGG16(weights=None, classes=100, input_shape=(32,32, 3), include_top=False)
    last = model.output
    x = keras.layers.Flatten(name='flatten')(last)
    x = keras.layers.Dense(4096, name='fc1', activation='relu')(x)
    x = keras.layers.Dropout(.5)(x)
    x = keras.layers.Dense(4096, name='fc2', activation='relu')(x)
    x = keras.layers.Dropout(.5)(x)
    x = keras.layers.Dense(100, name='predictions', activation='softmax')(x)
    model = keras.models.Model(inputs=model.input, outputs=x)
    model.summary()

    from keras.datasets import cifar100
    from keras.preprocessing.image import ImageDataGenerator

    model.load_weights(f'./vgg16_cifar100_lf_{lr}.h5')

    batch_size = 32
    num_classes = 100
    epochs = 200
    data_augmentation = True
    num_predictions = 20


    # The data, split between train and test sets:
    (x_train, y_train), (x_test, y_test) = cifar100.load_data(label_mode='fine')
    print('x_train shape:', x_train.shape)
    print(x_train.shape[0], 'train samples')
    print(x_test.shape[0], 'test samples')

    # Convert class vectors to binary class matrices.
    y_train = keras.utils.to_categorical(y_train, num_classes)
    y_test = keras.utils.to_categorical(y_test, num_classes)

    x_train = x_train.astype('float32')
    x_test = x_test.astype('float32')
    x_train /= 255
    x_test /= 255

    opt = keras.optimizers.RMSprop(lr=lr, decay=1.0e-6)

    # Let's train the model using RMSprop
    model.compile(loss='categorical_crossentropy',
                optimizer=opt,
                metrics=['accuracy'])

    if not data_augmentation:
        print('Not using data augmentation.')
        model.fit(x_train, y_train,
                batch_size=batch_size,
                epochs=epochs,
                validation_data=(x_test, y_test),
                shuffle=True)
    else:
        print('Using real-time data augmentation.')
        # This will do preprocessing and realtime data augmentation:
        datagen = ImageDataGenerator(
            featurewise_center=False,  # set input mean to 0 over the dataset
            samplewise_center=False,  # set each sample mean to 0
            featurewise_std_normalization=False,  # divide inputs by std of the dataset
            samplewise_std_normalization=False,  # divide each input by its std
            zca_whitening=False,  # apply ZCA whitening
            zca_epsilon=1e-06,  # epsilon for ZCA whitening
            rotation_range=0,  # randomly rotate images in the range (degrees, 0 to 180)
            # randomly shift images horizontally (fraction of total width)
            width_shift_range=0.1,
            # randomly shift images vertically (fraction of total height)
            height_shift_range=0.1,
            shear_range=0.1,  # set range for random shear
            zoom_range=0.1,  # set range for random zoom
            channel_shift_range=0.,  # set range for random channel shifts
            # set mode for filling points outside the input boundaries
            fill_mode='nearest',
            cval=0.,  # value used for fill_mode = "constant"
            horizontal_flip=True,  # randomly flip images
            vertical_flip=False,  # randomly flip images
            # set rescaling factor (applied before any other transformation)
            rescale=None,
            # set function that will be applied on each input
            preprocessing_function=None,
            # image data format, either "channels_first" or "channels_last"
            data_format=None,
            # fraction of images reserved for validation (strictly between 0 and 1)
            validation_split=0.0)

        # Compute quantities required for feature-wise normalization
        # (std, mean, and principal components if ZCA whitening is applied).
        datagen.fit(x_train)
        check = tf.keras.callbacks.ModelCheckpoint(f'./vgg16_cifar100_lf_{lr}.h5', verbose=1, save_best_only=True)
        # Fit the model on the batches generated by datagen.flow().
        model.fit_generator(datagen.flow(x_train, y_train,
                                        batch_size=batch_size),
                            epochs=epochs,
                            #use_multiprocessing=True,
                            callbacks=[check],
                            validation_data=(x_test, y_test),
                            verbose=1)

        model.load_weights(f'./vgg16_cifar100_lf_{lr}.h5')

        scores = model.evaluate(x_test, y_test, verbose=1)
        print('Test loss:', scores[0])
        print('Test accuracy:', scores[1])