h5py==3.0.0 causes model save issues in tf 2.1.0

[zanshuxun]

Describe the bug(问题描述)
model.save_weights causes TypeError: a bytes-like object is required, not 'str'.

Related issue: h5py==3.0.0 causes issues with keras model loads in tensorflow 2.1.0 (tensorflow/tensorflow#44467).

Suggested sulotion:

pin to h5py==2.10.0 this for all TF >= 2.1

Traceback (most recent call last):
  File "/Users/zsx/Code/DeepCTR/examples/run_classification_criteo.py", line 53, in <module>
    model.save_weights('deepfm_weights.h5')
  File "/Users/zsx/anaconda3/envs/tf2.3/lib/python3.6/site-packages/tensorflow_core/python/keras/engine/network.py", line 1107, in save_weights
    hdf5_format.save_weights_to_hdf5_group(f, self.layers)
  File "/Users/zsx/anaconda3/envs/tf2.3/lib/python3.6/site-packages/tensorflow_core/python/keras/saving/hdf5_format.py", line 631, in save_weights_to_hdf5_group
    param_dset = g.create_dataset(name, val.shape, dtype=val.dtype)
  File "/Users/zsx/anaconda3/envs/tf2.3/lib/python3.6/site-packages/h5py/_hl/group.py", line 143, in create_dataset
    if '/' in name:
TypeError: a bytes-like object is required, not 'str'

To Reproduce(复现步骤)
There are two ways to reproduce the behavior:

• add model.save_weights('deepfm_weights.h5') in the end of run_classification_criteo.py

or

• run the DeepFM_test.py

Operating environment(运行环境):

• python version 3.6.5
• tensorflow version 2.1.0
• deepctr version 0.8.2
• h5py version 3.0.0

Additional context
Add any other context about the problem here.

[zanshuxun]

Resolved in DeepCTR 0.8.3 by setting h5py==2.10.0:

DeepCTR/setup.py

Line 7 in 9564e05

'h5py==2.10.0', 'requests'