def create_model(weights_path=None, layers_removed=3):
model = Sequential()
input_shape = (config.IMAGE_SIZE, config.IMAGE_SIZE, 3)
# Block 1
model.add(Conv2D(32, (5, 5), padding='same', activation='relu', input_shape=input_shape))
model.add(Conv2D(32, (5, 5), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2), strides=(2, 2)))
# Block 2
model.add(Conv2D(64, (3, 3), padding='same', activation='relu'))
model.add(Conv2D(64, (3, 3), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2), strides=(2, 2)))
# Block 3 - optional
model.add(Conv2D(64, (5, 5), padding='same', activation='relu'))
model.add(Conv2D(64, (5, 5), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2), strides=(2, 2)))
# Block 4
model.add(Conv2D(128, (3, 3), padding='same', activation='relu'))
model.add(Conv2D(128, (3, 3), activation='relu'))
model.add(Conv2D(128, (3, 3), padding='same', activation='relu'))
model.add(Conv2D(128, (3, 3), activation='relu'))
model.add(MaxPooling2D(pool_size=(4, 4), strides=(2, 2)))
model.add(Dropout(0.2))
# Classification block
model.add(Flatten())
model.add(Dense(4096, activation='relu'))
model.add(Dense(config.HIDDEN_LAYER_SIZES[0][0], activation='relu'))
model.add(Dense(config.CLASSES, activation='softmax'))
if weights_path:
model.load_weights(weights_path)
for _ in range(layers_removed):
model.pop()
return model
