Traceback (most recent call last): File "/builds/software-machine-learning-infra-frameworks-workspaces-robhar02/pytorch/test/jit/test_freezing.py", line 1475, in test_module_with_shared_type_instances m = _static_quant(m) File "/builds/software-machine-learning-infra-frameworks-workspaces-robhar02/pytorch/test/jit/test_freezing.py", line 1469, in _static_quant torch.ao.quantization.convert(qModel, inplace=True) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/typing_extensions.py", line 2853, in wrapper return arg(*args, **kwargs) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/ao/quantization/quantize.py", line 669, in convert _convert( File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/ao/quantization/quantize.py", line 726, in _convert _convert( File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/ao/quantization/quantize.py", line 734, in _convert reassign[name] = swap_module( File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/ao/quantization/quantize.py", line 779, in swap_module new_mod = qmod.from_float( File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/ao/nn/quantized/modules/conv.py", line 606, in from_float return _ConvNd.from_float( File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/ao/nn/quantized/modules/conv.py", line 320, in from_float return cls.get_qconv(mod, activation_post_process, weight_post_process) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/ao/nn/quantized/modules/conv.py", line 258, in get_qconv qconv = cls( File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/ao/nn/quantized/modules/conv.py", line 547, in init super()._init( File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/ao/nn/quantized/modules/conv.py", line 118, in _init self.set_weight_bias(qweight, bias_float) File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/ao/nn/quantized/modules/conv.py", line 567, in set_weight_bias self._packed_params = torch.ops.quantized.conv2d_prepack( File "/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/_ops.py", line 1275, in call return self._op(*args, **kwargs) RuntimeError: unknown architecure

To execute this test, run the following from the base repo dir: python test/jit/test_freezing.py TestFreezing.test_module_with_shared_type_instances

Fix Plan

The fix involves updating the PyTorch version to a newer one where the issue is resolved.

• Update PyTorch to the latest version using pip:

pip install --upgrade torch

* Alternatively, if you are using a conda environment, update PyTorch using conda:
  ```bash
conda update pytorch

• If the issue persists, try updating the torchvision package as well:

pip install --upgrade torchvision

  or
  ```bash
conda update torchvision

• Verify that the torch and torchvision versions are up-to-date by checking the versions:

import torch import torchvision print(torch.version) print(torchvision.version)

### Verification
To verify that the fix worked, re-run the test that was failing:
```bash
python test/jit/test_freezing.py TestFreezing.test_module_with_shared_type_instances

If the test passes, the issue is resolved.

Extra Tips

• Make sure to update all dependent packages to ensure compatibility with the new PyTorch version.
• If you are using a specific PyTorch version for a project, consider creating a virtual environment to isolate the dependencies and avoid version conflicts.
• Keep an eye on the PyTorch issue tracker for any updates related to this issue: https://github.com/pytorch/pytorch/issues/177254