tensorflow2.0 保存和加载模型

1、保存全模型

model.save('the_save_model.h5')
new_model = keras.models.load_model('the_save_model.h5')

2、保持为SavedModel文件

keras.experimental.export_saved_model(model, 'saved_model')
new_model = keras.experimental.load_from_saved_model('saved_model')

3、仅保持网络结构，这样导出的模型并未包含训练好的参数

config = model.get_config()
reinitialized_model = keras.Model.from_config(config)
new_prediction = reinitialized_model.predict(x_test)
assert abs(np.sum(predictions-new_prediction)) >0

4、仅保存网络参数

weights = model.get_weights()
model.set_weights(weights)
# 可以把结构和参数保存结合起来
config = model.get_config()
weights = model.get_weights()
new_model = keras.Model.from_config(config) # config只能用keras.Model的这个api
new_model.set_weights(weights)
new_predictions = new_model.predict(x_test)
np.testing.assert_allclose(predictions, new_predictions, atol=1e-6)

6、保存网络权重为SavedModel格式

model.save_weights('weight_tf_savedmodel')
model.save_weights('weight_tf_savedmodel_h5', save_format='h5')