[GitHub] wangliye00 opened a new issue #13949: Error: shape inconsistent while converting PyTorch model to mxnet model with onnx

[GitBox <gi...@apache.org>]

wangliye00 opened a new issue #13949: Error: shape inconsistent while converting PyTorch model to mxnet model with onnx
URL: https://github.com/apache/incubator-mxnet/issues/13949
 
 
   ## Description
   I tried to convert a pretrained pytorch resnet18 to onnx format, then to mxnet model. But shape inconsistent error occurred when I load the onnx file with mxnet.contrib.onnx module.
   
   ## Environment info (Required)
   - PyTorch 1.0, mxnet 1.3.0 (for windows)
   
   ```
   ## Error Message:
   
   File “D:\software\Anaconda3\lib\site-packages\mxnet\contrib\onnx\onnx2mx\import_model.py”, line 53, in import_model
   sym, arg_params, aux_params = graph.from_onnx(model_proto.graph)
   
   File “D:\software\Anaconda3\lib\site-packages\mxnet\contrib\onnx\onnx2mx\import_onnx.py”, line 96, in from_onnx
   self._params[init_tensor.name] = self._parse_array(init_tensor)
   
   File “D:\software\Anaconda3\lib\site-packages\mxnet\contrib\onnx\onnx2mx\import_onnx.py”, line 200, in _parse_array
   return nd.array(np_array)
   
   File “D:\software\Anaconda3\lib\site-packages\mxnet\ndarray\utils.py”, line 146, in array
   return _array(source_array, ctx=ctx, dtype=dtype)
   
   File “D:\software\Anaconda3\lib\site-packages\mxnet\ndarray\ndarray.py”, line 2435, in array
   arr[:] = source_array
   
   File “D:\software\Anaconda3\lib\site-packages\mxnet\ndarray\ndarray.py”, line 444, in setitem
   self._set_nd_basic_indexing(key, value)
   
   File “D:\software\Anaconda3\lib\site-packages\mxnet\ndarray\ndarray.py”, line 710, in _set_nd_basic_indexing
   self._sync_copyfrom(value)
   
   File “D:\software\Anaconda3\lib\site-packages\mxnet\ndarray\ndarray.py”, line 872, in _sync_copyfrom
   str(self.shape), str(source_array.shape)))
   
   ValueError: Shape inconsistent: expected () vs got (1,)
   
   ## Minimum reproducible example
   import torch
   import torchvision
   
   dummy_input = torch.randn(1, 3, 224, 224)
   model = torchvision.models.resnet18(pretrained=True)
   input_names = [ “input_1" ]
   output_names = [ “output1” ]
   
   torch.onnx.export(model, dummy_input, “resnet.onnx”, verbose=True, input_names=input_names, output_names=output_names)
   
   from mxnet.contrib import onnx as onnx_mxnet
   sym, arg, aux = onnx_mxnet.import_model(“resnet.onnx”)

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services