tf2.4 GlobalPointer 保存 pb 报错

[liuliuNLP]

提问时请尽可能提供如下信息：

基本信息

Tensorflow2.4 Keras2.3.1 bert4keras0.10.9 tf.keras

核心代码

加载预训练模型

model = build_transformer_model( config_path=config_path, checkpoint_path=checkpoint_path, return_keras_model=False )

output = GlobalPointer(len(categories), 64)(model.output) model = keras.models.Model(model.input, output) model.summary()

model.compile( loss=global_pointer_crossentropy, optimizer=Adam(learning_rate), metrics=[global_pointer_f1_score] )

evaluator = Evaluator() train_generator = data_generator(train_data, batch_size) model.fit( train_generator.forfit(), steps_per_epoch=len(train_generator), epochs=epochs, callbacks=[evaluator] ) export_path = 'model' version = "1" model.save(export_path + version, save_format="tf")

输出信息

AttributeError: 'Dropout' object has no attribute '_saved_model_inputs_spec'

[bojone]

我也不熟悉这一块。但看这个错误信息，似乎跟GlobalPointer没关系？