model = Model(inputs=[src, tgt], outputs=[y, flow])  
#定義網(wǎng)絡的時候會給出輸入和輸出
model.compile(optimizer=Adam(lr=lr), loss=[
           losses.cc3D(), losses.gradientLoss('l2')], loss_weights=[1.0, reg_param]) 
#訓練網(wǎng)絡的時候指定loss，如果是多l(xiāng)oss，
loss weights分別對應前面的每個loss的權重，最后輸出loss的和
train_loss = model.train_on_batch(
      [X, atlas_vol], [atlas_vol, zero_flow]) 
 #開始訓練，loss中y_pred 和y_true的對應關系是：
 #輸出y與atlas_vol算cc3Dloss，輸出flow與zero_flow算gradientloss

import matplotlib.pyplot as plt

...  //數(shù)據(jù)處理代碼 省略

history = model.fit_generator(
  image_generator, steps_per_epoch=2000 // 32 ,
  epochs=16, verbose=1,
  validation_data=image_generator_TEST, validation_steps=20
)

print(history.history.keys())
plt.switch_backend('agg')  #服務器上面保存圖片 需要設置這個
//acc
plt.plot(history.history['acc'])
plt.plot(history.history['val_acc'])
plt.title('model accuracy')
plt.ylabel('accuracy')
plt.xlabel('epoch')
plt.legend(['train', 'test'], loc='upper left')
plt.savefig('acc.jpg')
//loss
plt.plot(history.history['loss'])
plt.plot(history.history['val_loss'])
plt.ylabel('loss')
plt.xlabel('epoch')
plt.legend(['train', 'test'], loc='upper left')
plt.savefig('loss.jpg')