dl_model = Sequential()

dl_model.add(Conv2D(32, (3, 3), input_shape=(64, 48, 3), activation='relu'))
dl_model.add(Conv2D(32, (3, 3), activation='relu'))
dl_model.add(MaxPooling2D(pool_size=(2, 2)))

dl_model.add(Flatten())
dl_model.add(Dense(128, activation='relu'))
dl_model.add(Dropout(0.5))
dl_model.add(Dense(1, activation='sigmoid'))

adam = Adam(lr=0.00001)

dl_model.compile(loss='binary_crossentropy', optimizer=adam, metrics=['accuracy'])


train_data_generator = ImageDataGenerator(rescale=1. / 255,
                                          shear_range=0.2,
                                          zoom_range=0.2,
                                          rotation_range=40,
                                          width_shift_range=0.2,
                                          height_shift_range=0.2,
                                          horizontal_flip=True,
                                          fill_mode='nearest')

test_data_generator = ImageDataGenerator(rescale=1. / 255)

training_set = train_data_generator.flow_from_directory(path_training,
                                                        target_size=(64, 48),
                                                        batch_size=32,
                                                        class_mode='binary')

validation_set = test_data_generator.flow_from_directory(path_validation,
                                                         target_size=(64, 48),
                                                         batch_size=32,
                                                         class_mode='binary')


dl_model.fit_generator(training_set, 8000, args.epochs, validation_data=validation_set, validation_steps=2000, verbose=1)