Potential Conflict Between ParlAI and PyTorch 1.13 in parlai.core.torch_generator_agent.py

[Guest400123064]

Bug description
I installed ParlAI from source (v1.7.1), Fairseq (v0.12.2), and (with pip) torch (v1.13). When running from the command line parlai interactive -m bart, it complaints:

File "/home/wangy49/Projects/empathic-bot/packages/ParlAI/parlai/core/torch_generator_agent.py", line 1599, in advance self.partial_hyps[path_selection.hypothesis_ids.long()], RuntimeError: indices should be either on cpu or on the same device as the indexed tensor (cpu)

Reproduction steps
With the version specified above, run parlai interactive -m bart

Expected behavior
Expected to start an interactive session with BART.

Logs

(Some model downloading and loading logs are not shown here)

File "/home/wangy49/Projects/empathic-bot/packages/ParlAI/parlai/core/torch_generator_agent.py", line 1599, in advance self.partial_hyps[path_selection.hypothesis_ids.long()], RuntimeError: indices should be either on cpu or on the same device as the indexed tensor (cpu)

Additional context
Running the code with T4 GPU with cuda 11.6

When using torch (v1.12), the problem was resolved. First mentioned in the followups in #4031

[github-actions]

This issue has not had activity in 30 days. Please feel free to reopen if you have more issues. You may apply the "never-stale" tag to prevent this from happening.

[klshuster]

This is resolved on latest ParlAI