run_dien.py failed

[pingsutw]

Describe the bug(问题描述)

  File "run_dien.py", line 53, in <module>
    dnn_hidden_units=[4, 4, 4], dnn_dropout=0.6, gru_type="AUGRU", use_negsampling=True)
  File "/home/dslab/anaconda3/lib/python3.7/site-packages/deepctr/models/dien.py", line 285, in DIEN
    model.add_loss(alpha * aux_loss_1)
  File "/home/dslab/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/keras/engine/base_layer.py", line 1132, in add_loss
    self._graph_network_add_loss(symbolic_loss)
  File "/home/dslab/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/keras/engine/network.py", line 1426, in _graph_network_add_loss
    new_nodes, new_layers = _map_subgraph_network(self.inputs, [symbolic_loss])
  File "/home/dslab/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/keras/engine/network.py", line 1653, in _map_subgraph_network
    _, nodes_by_depth, layers, _ = _map_graph_network(inputs, outputs)
  File "/home/dslab/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/keras/engine/network.py", line 1636, in _map_graph_network
    str(all_names.count(name)) + ' times in the model. '
ValueError: The name "bn1gru" is used 2 times in the model. All layer names should be unique.

To Reproduce(复现步骤) python run_dien.py

Operating environment(运行环境):

• python version [3.7]
• tensorflow version [2.0]
• deepctr version [0.6.3]

Additional context I clone latest deepctr code, and try to run dien example,I got above error

[pingsutw]

update : I change Tensorflow version to 1.14.0, it works maybe we need change some code to let it compatible with TF 2.0