AttributeError: 'str' object has no attribute 'decode'

jaried commented 2 years ago

New Issue Checklist

I have read the Contribution Guidelines
I searched for existing GitHub issues

Issue Description

Reproducible Code

Which OS are you using ? windows 10 professional tensorlayer: 2.2.5 tensorflow : 2.8 python:3.7
Please provide a reproducible code of your issue. Without any reproducible code, you will probably not receive any help. tutorial_PG

I use the trained network to test, and the following error is prompted:

Traceback (most recent call last):
  File "D:\Program Files\JetBrains\PyCharm Community Edition 2021.2.2\plugins\python-ce\helpers\pydev\pydevd.py", line 1483, in _exec
    pydev_imports.execfile(file, globals, locals)  # execute the script
  File "D:\Program Files\JetBrains\PyCharm Community Edition 2021.2.2\plugins\python-ce\helpers\pydev\_pydev_imps\_pydev_execfile.py", line 18, in execfile
    exec(compile(contents+"\n", file, 'exec'), glob, loc)
  File "D:/Tony/Documents/yunpan/invest/2022/quant/gym/study PG/PG/pg.py", line 220, in <module>
    agent.load()
  File "D:/Tony/Documents/yunpan/invest/2022/quant/gym/study PG/PG/pg.py", line 162, in load
    tl.files.load_hdf5_to_weights_in_order(os.path.join(path, 'pg_policy.hdf5'), self.model)
  File "D:\Anaconda3\envs\gym_env\lib\site-packages\tensorlayer\files\utils.py", line 2730, in load_hdf5_to_weights_in_order
    _load_weights_from_hdf5_group_in_order(f, network.all_layers)
  File "D:\Anaconda3\envs\gym_env\lib\site-packages\tensorlayer\files\utils.py", line 2605, in _load_weights_from_hdf5_group_in_order
    layer_names = [n.decode('utf8') for n in f.attrs["layer_names"]]
  File "D:\Anaconda3\envs\gym_env\lib\site-packages\tensorlayer\files\utils.py", line 2605, in <listcomp>
    layer_names = [n.decode('utf8') for n in f.attrs["layer_names"]]
AttributeError: 'str' object has no attribute 'decode'
python-BaseException

Laicheng0830 commented 2 years ago

Modify code layer_names = [n.decode('utf8') for n in f.attrs["layer_names"]] to layer_names = [n if isinstance(n, str) else n.decode('utf8') for n in f.attrs["layer_names"]]

jaried commented 2 years ago

tensorlayer\files\utils.py, line 2617, the same question, I modified。

Traceback (most recent call last):
  File "D:/Tony/Documents/yunpan/invest/2022/quant/gym/study PG/PG/pg.py", line 219, in <module>
    agent.load()
  File "D:/Tony/Documents/yunpan/invest/2022/quant/gym/study PG/PG/pg.py", line 161, in load
    tl.files.load_hdf5_to_weights_in_order(os.path.join(path, 'pg_policy.hdf5'), self.model)
  File "D:\Anaconda3\envs\gym_env\lib\site-packages\tensorlayer\files\utils.py", line 2730, in load_hdf5_to_weights_in_order
    _load_weights_from_hdf5_group_in_order(f, network.all_layers)
  File "D:\Anaconda3\envs\gym_env\lib\site-packages\tensorlayer\files\utils.py", line 2617, in _load_weights_from_hdf5_group_in_order
    weight_names = [n.decode('utf8') for n in g.attrs['weight_names']]
  File "D:\Anaconda3\envs\gym_env\lib\site-packages\tensorlayer\files\utils.py", line 2617, in <listcomp>
    weight_names = [n.decode('utf8') for n in g.attrs['weight_names']]
AttributeError: 'str' object has no attribute 'decode'

深圳？ Keep running, new questions:

Traceback (most recent call last):
  File "D:/Tony/Documents/yunpan/invest/2022/quant/gym/study PG/PG/pg.py", line 219, in <module>
    agent.load()
  File "D:/Tony/Documents/yunpan/invest/2022/quant/gym/study PG/PG/pg.py", line 161, in load
    tl.files.load_hdf5_to_weights_in_order(os.path.join(path, 'pg_policy.hdf5'), self.model)
  File "D:\Anaconda3\envs\gym_env\lib\site-packages\tensorlayer\files\utils.py", line 2729, in load_hdf5_to_weights_in_order
    _load_weights_from_hdf5_group_in_order(f, network.all_layers)
  File "D:\Anaconda3\envs\gym_env\lib\site-packages\tensorlayer\files\utils.py", line 2618, in _load_weights_from_hdf5_group_in_order
    assign_tf_variable(layer.all_weights[iid], np.asarray(g[w_name]))
IndexError: list index out of range

Laicheng0830 commented 2 years ago

Line 2617 of the code also needs to be modified in the same way. Now this is a problem you may need to check if the parameters are imported correctly.This error occurs when the parameter shapes are inconsistent.

jaried commented 2 years ago

原来的代码是tensorlayer提供的，问题也是tensorlayer产生的。我在上面的回答中，也提到，2617行我已经改好了。新的问题是2618行。

我是继续在这个issue提问还是新开一个issue？

The original code is provided by tensorlayer, and the problem is also generated by tensorlayer. In my answer above, I also mentioned that I have changed line 2617. The new problem is line 2618.

Do I continue to ask questions in this issue or open a new one?

waqar3411 commented 2 years ago

Hello @jaried... i resolved this issue by downgrading the h5py version to 2.10.0 pip install h5py==2.10.0

hope this helps...

jaried commented 2 years ago

Thanks for your answer.

tensorlayer / TensorLayer