train_lmks.py

import keras
import datetime
from keras.layers import Input, Dense
from keras.models import Model
from keras.callbacks import TensorBoard, ModelCheckpoint, ReduceLROnPlateau
from keras.applications import mobilenetv2
import numpy as np

img_size = 224

mode = 'lmks'  # [bbs, lmks]
if mode is 'bbs':
    output_size = 4
elif mode is 'lmks':
    output_size = 10

start_time = datetime.datetime.now().strftime('%Y_%m_%d_%H_%M_%S')

data_training = np.load('dataset/lmks_training.npy', allow_pickle=True)
data_validation = np.load('dataset/lmks_validation.npy', allow_pickle=True)

x_train = np.array(data_training.item().get('imgs'))
y_train = np.array(data_training.item().get(mode))

x_test = np.array(data_validation.item().get('imgs'))
y_test = np.array(data_validation.item().get(mode))

x_train = x_train.astype('float32') / 255.
x_test = x_test.astype('float32') / 255.
x_train = np.reshape(x_train, (-1, img_size, img_size, 3))
x_test = np.reshape(x_test, (-1, img_size, img_size, 3))

y_train = np.reshape(y_train, (-1, output_size))
y_test = np.reshape(y_test, (-1, output_size))

inputs = Input(shape=(img_size, img_size, 3))

mobilenetv2_model = mobilenetv2.MobileNetV2(input_shape=(img_size, img_size, 3), alpha=1.0, include_top=False,
                                            weights='imagenet', input_tensor=inputs, pooling='max')

net = Dense(128, activation='relu')(mobilenetv2_model.layers[-1].output)
net = Dense(64, activation='relu')(net)
net = Dense(output_size, activation='linear')(net)

model = Model(inputs=inputs, outputs=net)

model.summary()

# training
model.compile(optimizer=keras.optimizers.Adam(), loss='mse')

model.fit(x_train, y_train, epochs=50, batch_size=32, shuffle=True,
          validation_data=(x_test, y_test), verbose=1,
          callbacks=[
              TensorBoard(log_dir='logs/%s' % (start_time)),
              ModelCheckpoint('./models/%s.h5' % (start_time), monitor='val_loss', verbose=1, save_best_only=True,
                              mode='auto'),
              ReduceLROnPlateau(monitor='val_loss', factor=0.2, patience=5, verbose=1, mode='auto')
          ]
          )