訓練完成後獲得了模型文件,下一次想接着上次的基礎繼續進行訓練,這可怎麼辦?框架
小周來支招,打一頓就行了3d
第一次訓練模型獲得的h5文件:code
model = Sequential() model.add(LSTM(150, input_shape=(train_x.shape[1], train_x.shape[2]),return_sequences=False)) model.add(Dense(30)) model.summary() model.compile(loss=["mse"], optimizer='adam',metrics=['mape','mae','mse']) history = model.fit(train_x, train_y, epochs=epochs, batch_size=batch_size, validation_data=[test_x, test_y],verbose=2, shuffle=True) #save LeNet_model_files after train model.save('lstm_model.h5')
利用h5文件接着上次的基礎繼續進行訓練,只需:blog
#加載模型文件 model=load_model('lstm_model.h5') model._make_predict_function() #繼續用數據訓練 history = model.fit(train_x, train_y, epochs=epochs, batch_size=batch_size, validation_data=[test_x, test_y],verbose=2, shuffle=True) model.save('lstm_model2.h5')
圖上是加載模型進行預測,把相應代碼改爲下面的訓練便可實現繼續訓練!input
train_step.run(feed_dict={x: images, y_: batch[1],keep_prob1:prob1,keep_prob2:prob2,keep_prob3:prob3})