search

microsoft / MMdnn

MMdnn is a set of tools to help users inter-operate among different deep learning frameworks. E.g. model conversion and visualization. Convert models between Caffe, Keras, MXNet, Tensorflow, CNTK, PyTorch Onnx and CoreML.
MIT License
5.8k stars 965 forks source link

weight_data get None during convert MXnet model to TF #616

Open why702 opened 5 years ago

why702 commented 5 years ago

Hi

I have a similar problem of #231

I used the model: https://www.dropbox.com/s/2xq8mcao6z14e3u/gamodel-r50.zip?dl=0

python -m mmdnn.conversion._script.convertToIR -f mxnet -n model-symbol.json -w model-0000.params -d model-r50 --inputShape 3,112,112

Traceback (most recent call last):
  File "C:\Users\hsiaoyuh_wang\AppData\Local\Programs\Python\Python36\lib\runpy.py", line 193, in _run_module_as_main
    "__main__", mod_spec)
  File "C:\Users\hsiaoyuh_wang\AppData\Local\Programs\Python\Python36\lib\runpy.py", line 85, in _run_code
    exec(code, run_globals)
  File "C:\Users\hsiaoyuh_wang\AppData\Local\Programs\Python\Python36\lib\site-packages\mmdnn\conversion\_script\convertToIR.py", line 197, in <module>
    _main()
  File "C:\Users\hsiaoyuh_wang\AppData\Local\Programs\Python\Python36\lib\site-packages\mmdnn\conversion\_script\convertToIR.py", line 192, in _main
    ret = _convert(args)
  File "C:\Users\hsiaoyuh_wang\AppData\Local\Programs\Python\Python36\lib\site-packages\mmdnn\conversion\_script\convertToIR.py", line 81, in _convert
    from mmdnn.conversion.mxnet.mxnet_parser import MXNetParser
  File "C:\Users\hsiaoyuh_wang\AppData\Local\Programs\Python\Python36\lib\site-packages\mmdnn\conversion\mxnet\mxnet_parser.py", line 538
    self.set_weight(source_node.name, "bias", self.weight_data.get(source_node.name + "_beta").asnumpy())
       ^
SyntaxError: invalid syntax

D:\svnML\ditSRC\modules\AIOT\div_algo\model\face\insightface\gamodel-r50>python -m mmdnn.conversion._script.convertToIR -f mxnet -n model-symbol.json -w model-0000.params -d model-r50 --inputShape 3,112,112
[10:23:54] C:\Jenkins\workspace\mxnet-tag\mxnet\src\nnvm\legacy_json_util.cc:209: Loading symbol saved by previous version v1.2.0. Attempting to upgrade...
[10:23:54] C:\Jenkins\workspace\mxnet-tag\mxnet\src\nnvm\legacy_json_util.cc:217: Symbol successfully upgraded!
C:\Users\hsiaoyuh_wang\AppData\Roaming\Python\Python36\site-packages\mxnet\module\base_module.py:56: UserWarning: You created Module with Module(..., label_names=['softmax_label']) but input with name 'softmax_label' is not found in symbol.list_arguments(). Did you mean one of:
        data
  warnings.warn(msg)
Warning: MXNet Parser has not supported operator null with name data.
Warning: convert the null operator with name [data] into input layer.
Traceback (most recent call last):
  File "C:\Users\hsiaoyuh_wang\AppData\Local\Programs\Python\Python36\lib\runpy.py", line 193, in _run_module_as_main
    "__main__", mod_spec)
  File "C:\Users\hsiaoyuh_wang\AppData\Local\Programs\Python\Python36\lib\runpy.py", line 85, in _run_code
    exec(code, run_globals)
  File "C:\Users\hsiaoyuh_wang\AppData\Local\Programs\Python\Python36\lib\site-packages\mmdnn\conversion\_script\convertToIR.py", line 197, in <module>
    _main()
  File "C:\Users\hsiaoyuh_wang\AppData\Local\Programs\Python\Python36\lib\site-packages\mmdnn\conversion\_script\convertToIR.py", line 192, in _main
    ret = _convert(args)
  File "C:\Users\hsiaoyuh_wang\AppData\Local\Programs\Python\Python36\lib\site-packages\mmdnn\conversion\_script\convertToIR.py", line 115, in _convert
    parser.run(args.dstPath)
  File "C:\Users\hsiaoyuh_wang\AppData\Local\Programs\Python\Python36\lib\site-packages\mmdnn\conversion\common\DataStructure\parser.py", line 22, in run
    self.gen_IR()
  File "C:\Users\hsiaoyuh_wang\AppData\Local\Programs\Python\Python36\lib\site-packages\mmdnn\conversion\mxnet\mxnet_parser.py", line 262, in gen_IR
    func(current_node)
  File "C:\Users\hsiaoyuh_wang\AppData\Local\Programs\Python\Python36\lib\site-packages\mmdnn\conversion\mxnet\mxnet_parser.py", line 538, in rename_BatchNorm
    self.set_weight(source_node.name, "bias", self.weight_data.get(source_node.name + "_beta").asnumpy())
AttributeError: 'NoneType' object has no attribute 'asnumpy'

I think this issue caused by the BatchNorm op "name": "gender_resnet0_batchnorm0_fwd" don't have the beta, so "self.weight_data.get(source_node.name + "_beta")" get None.

Do you have any suggestion?

vgg4resnet commented 5 years ago

I get the same issue