pytorch2caffe
pytorch2caffe copied to clipboard
IndexBackward error
KeyError: 'IndexBackward'
How could we fix this error?
File "/home/zq/PROJECTS/projfog/pytorch2caffe/pytorch2caffe.py", line 166, in add_layer
layer['type'] = layer_dict[parent_type]
KeyError: 'IndexBackward'
[1] 27580 segmentation fault (core dumped) python resnetdemo.py
For now, this project doesn't support Index
operation.
To support it, you have to find how caffe performs index selection.
builtins.KeyError: 'NoneType'
File "/home/dyt/ssd.pytorch/pytorch2caffe/pytorch2caffe.py", line 165, in add_layer layer['type'] = layer_dict[parent_type]
builtins.KeyError: 'NoneType'
I met broadcastBackward error
I met keyError['type']: 'ExpandBackward', and on windows,how can I solve this probelm???
I also face same problem.
what does "IndexBackward" mean?which layer's problem?
In my case, 'IndexBackward' (pytorch0.4) is corresponding to 'Slice' in caffe , you can modify your layer_dict in pytorch2caffe.py, and add the params in this layer, hope this can help you.
In my case, 'IndexBackward' (pytorch0.4) is corresponding to 'Slice' in caffe , you can modify your layer_dict in pytorch2caffe.py, and add the params in this layer, hope this can help you.
I have added 'IndexBackward': 'Slice'
in the layer_dict, how can I add the params in this layer? What's the operation in this layer? When I try 'Slice', it reports errors.
elif parent_type =='IndexBackward': eltwise_param = { 'operation': 'Slice', } layer['eltwise_param'] = eltwise_param