Transfer Learning Example 02 : InceptionResnetV2

<h2> Inception Resnet V2 </h2>

# define input shape
INPUT_SHAPE = (298, 298, 3)

# get the Resnet model
resnet_layers = tf.keras.applications.InceptionResNetV2(weights='imagenet', include_top=False,
                                                        input_shape=INPUT_SHAPE)

resnet_layers.summary()


# Fine-tune all the layers
for layer in resnet_layers.layers:
    layer.trainable = True


# define sequential model
model = tf.keras.models.Sequential()
weight_decay = 0.0005
# Add the vgg convolutional base model
model.add(resnet_layers)

model.add(Conv2D(512, (3, 3), padding='same',
                 kernel_regularizer=regularizers.l2(weight_decay)))
model.add(Activation('relu'))
model.add(BatchNormalization())
model.add(Dropout(0.4))

model.add(Conv2D(512, (3, 3), padding='same',
                 kernel_regularizer=regularizers.l2(weight_decay)))
model.add(Activation('relu'))
model.add(BatchNormalization())
model.add(Dropout(0.4))

model.add(Conv2D(512, (3, 3), padding='same',
                 kernel_regularizer=regularizers.l2(weight_decay)))
model.add(Activation('relu'))
model.add(BatchNormalization())

model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.5))
model.add(Flatten())
model.add(Dense(512, kernel_regularizer=regularizers.l2(weight_decay)))
model.add(Activation('relu'))
model.add(BatchNormalization())

model.add(Dropout(0.5))
model.add(Dense(2))
model.add(Activation('softmax'))

model.compile(optimizer=tf.keras.optimizers.Adam(learning_rate=2e-5),
              loss='categorical_crossentropy',
              metrics=['accuracy'])

# view model layers
model.summary()

es_callback = tf.keras.callbacks.EarlyStopping(monitor='val_loss', patience=4,
                                               restore_best_weights=True,
                                               verbose=1)

STEP_SIZE_TRAIN = train_generator.n//train_generator.batch_size
STEP_SIZE_VALID = valid_generator.n//valid_generator.batch_size
history = model.fit_generator(generator=train_generator,
                              steps_per_epoch=STEP_SIZE_TRAIN,
                              validation_data=valid_generator,
                              callbacks=[es_callback],
                              validation_steps=STEP_SIZE_VALID,
                              epochs=1)

Load test images from csv file

test_file = pd.read_csv('xyz.csv')
test_file.head()

test_datagen = ImageDataGenerator(rescale=1./255.)

test generator from dataframe

test_generator = test_datagen.flow_from_dataframe(
dataframe=test_file,
directory="face_mask_detection/test",
x_col="filename",
y_col=None,
batch_size=32,
seed=42,
shuffle=False,
class_mode=None,
target_size=(298, 298))

evaluate the model

STEP_SIZE_TEST = test_generator.n//test_generator.batch_size
model.evaluate_generator(generator=valid_generator,
steps=STEP_SIZE_TEST)

test_generator.reset()

Making the model predictions on the testing dataset.

pred = model.predict_generator(test_generator,
steps=STEP_SIZE_TEST,
verbose=1)

predicted_class_indices = np.argmax(pred, axis=1)

labels = (train_generator.class_indices)
labels = dict((v, k) for k, v in labels.items())
predictions = [labels[k] for k in predicted_class_indices]

filenames = test_generator.filenames
results = pd.DataFrame({"Filename": filenames,
"Predictions": predictions})
results.to_csv("results.csv", index=False)