Pre-trained shufflenetv2_x1.5 and shufflenetv2_x2.0 raise "...not supported as of now". · Issue #3257 · pytorch/vision (original) (raw)

🐛 Bug

Pre-trained models of shufflenetv2_x1.5 and shufflenetv2_x2.0 raise the following NotImplementedError: pretrained {} is not supported as of now. Important: pre-trained models of shufflenet_v2_x0_5 and shufflenet_v2_x1_0 are working perfectly.

To Reproduce

Steps to reproduce the behavior:

1. from torchvision import models
2. model = models.shufflenet_v2_x1_5(pretrained=pretrained)
3. model = models.shufflenet_v2_x2_0(pretrained=pretrained)
4. Raises the following:

  NotImplementedError                       Traceback (most recent call last)
  <ipython-input-15-788c60a16542> in <module>
        1 # Build and train your network
        2 # Transfer Learning
  ----> 3 model = models.shufflenet_v2_x1_5(pretrained=pretrained)
  
  ~/venvs/pytorch_gpu/lib/python3.8/site-packages/torchvision/models/shufflenetv2.py in shufflenet_v2_x1_5(pretrained, progress, **kwargs)
      191         progress (bool): If True, displays a progress bar of the download to stderr
      192     """
  --> 193     return _shufflenetv2('shufflenetv2_x1.5', pretrained, progress,
      194                          [4, 8, 4], [24, 176, 352, 704, 1024], **kwargs)
      195 
  
  ~/venvs/pytorch_gpu/lib/python3.8/site-packages/torchvision/models/shufflenetv2.py in _shufflenetv2(arch, pretrained, progress, *args, **kwargs)
      145         model_url = model_urls[arch]
      146         if model_url is None:
  --> 147             raise NotImplementedError('pretrained {} is not supported as of now'.format(arch))
      148         else:
      149             state_dict = load_state_dict_from_url(model_url, progress=progress)
  
  NotImplementedError: pretrained shufflenetv2_x1.5 is not supported as of now

Expected behavior

Get the pre-trained model of Shufflenet_v2_x1_5 and shufflenet_v2_x2.0.

Environment

• PyTorch version: 1.5.1
• Is debug build: False
• CUDA used to build PyTorch: 10.2
• ROCM used to build PyTorch: N/A
• OS: Ubuntu 20.04.1 LTS (x86_64)
• GCC version: (Ubuntu 9.3.0-17ubuntu1~20.04) 9.3.0
• Clang version: Could not collect
• CMake version: Could not collect
• Python version: 3.8 (64-bit runtime)
• Is CUDA available: True
• CUDA runtime version: Could not collect
• GPU models and configuration: GPU 0: GeForce GTX 1660 Ti with Max-Q Design
• Nvidia driver version: 450.102.04
• cuDNN version: Could not collect
• HIP runtime version: N/A
• MIOpen runtime version: N/A
• Versions of relevant libraries:
• [pip3] numpy==1.19.1
• [pip3] torch==1.5.1
• [pip3] torchvision==0.6.1
• [conda] Could not collect

Additional context

Also tried to get the pre-trained model with:
model = torch.hub.load('pytorch/vision:v0.6.0', 'shufflenet_v2_x2_0', pretrained=True)
and it raised:
RuntimeError: Cannot find callable shufflenet_v2_x2_0 in hubconf