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")

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

请问 大大这是什么错误呀
环境：
python3.8
tensorflow 1.5.1
keras2.3.1
bert4keras-0.10.6

出错过程：
2021-07-21 23:23:30.850241: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10
2021-07-21 23:23:30.850272: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10
2021-07-21 23:23:30.850304: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.11
2021-07-21 23:23:30.850334: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11
2021-07-21 23:23:30.850365: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8
2021-07-21 23:23:30.855468: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1794] Adding visible gpu devices: 0
2021-07-21 23:23:30.855547: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0
2021-07-21 23:23:31.146373: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1206] Device interconnect StreamExecutor with strength 1 edge matrix:
2021-07-21 23:23:31.146412: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1212] 0
2021-07-21 23:23:31.146418: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1225] 0: N
2021-07-21 23:23:31.147960: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1351] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 21829 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3090, pci bus id: 0000:d5:00.0, compute capability: 8.6)
WARNING:tensorflow:From /home/xhw205/.environments/venv/lib/python3.8/site-packages/bert4keras/layers.py:1199: The name tf.matrix_band_part is deprecated. Please use tf.linalg.band_part instead.

WARNING:tensorflow:From /home/xhw205/.environments/venv/lib/python3.8/site-packages/bert4keras/layers.py:1199: The name tf.matrix_band_part is deprecated. Please use tf.linalg.band_part instead.

Traceback (most recent call last):
File "glp.py", line 103, in
model = Model(model.input, output)
File "/home/xhw205/.environments/venv/lib/python3.8/site-packages/tensorflow_core/python/keras/engine/training.py", line 147, in init
super(Model, self).init(*args, **kwargs)
File "/home/xhw205/.environments/venv/lib/python3.8/site-packages/tensorflow_core/python/keras/engine/network.py", line 164, in init
self._init_graph_network(*args, **kwargs)
File "/home/xhw205/.environments/venv/lib/python3.8/site-packages/tensorflow_core/python/training/tracking/base.py", line 457, in _method_wrapper
result = method(self, *args, **kwargs)
File "/home/xhw205/.environments/venv/lib/python3.8/site-packages/tensorflow_core/python/keras/engine/network.py", line 270, in _init_graph_network
self._validate_graph_inputs_and_outputs()
File "/home/xhw205/.environments/venv/lib/python3.8/site-packages/tensorflow_core/python/keras/engine/network.py", line 1482, in _validate_graph_inputs_and_outputs
layer = x._keras_history.layer
AttributeError: 'tuple' object has no attribute 'layer'

数据可以提供一下吗

请问是哪方面的问题

苏神好！请问有没有做预测的代码？