1,保存模型:
my_model = create_model_function( ...... )
my_model.compile( ...... )
my_model.fit( ...... )
model_name . save( filepath, overwrite: bool=True, include_optimizer: bool=True )
filepath:保存的路徑
overwrite:如果存在源文件,是否覆蓋
include_optimizer:是否保存優化器狀態
ex : mymodel.save(filepath="p402/my_model.h5", includeoptimizer=False)
2, 載入模型:
my_model = keras . models . load_model( filepath )
載入后可以繼續訓練:
my_model . fit( X_train_2,Y_train_2 )
也可以直接評估:
preds = my_model . evaluate( X_test, Y_test )
print ( "Loss = " + str( preds[0] ) )
print ( "Test Accuracy = " + str( preds[1] ) )
————————————————
原文鏈接:https://blog.csdn.net/wslkd0123/article/details/80647041