You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository was archived by the owner on Mar 20, 2026. It is now read-only.
I am currently experiencing issues running fairseq version 0.12.2 on Python 3.11.8. Despite fairseq being compatible with Python versions 3.6 and above, I've encountered problems that seem to be specific to Python 3.11.
Steps to Reproduce
Install fairseq version 0.12.2 on Python 3.11.8.
Run a standard fairseq training command or script.
Observe any errors or unexpected behavior.
Expected Behavior
fairseq should operate without any compatibility issues, allowing for smooth training and execution of models.
Actual Behavior
(Describe the errors or issues encountered, including any error messages or logs)
Environment
OS: Windows 11 CoreSingleLanguage
Python version: 3.11.8
fairseq version: 0.12.2
Additional Context
(Include any other relevant information such as modifications to config.py, custom settings, or related dependencies)
Any insights or solutions to address these compatibility issues would be greatly appreciated.
Description
I am currently experiencing issues running
fairseqversion 0.12.2 on Python 3.11.8. Despitefairseqbeing compatible with Python versions 3.6 and above, I've encountered problems that seem to be specific to Python 3.11.Steps to Reproduce
fairseqversion 0.12.2 on Python 3.11.8.fairseqtraining command or script.Expected Behavior
fairseqshould operate without any compatibility issues, allowing for smooth training and execution of models.Actual Behavior
(Describe the errors or issues encountered, including any error messages or logs)
Environment
fairseqversion: 0.12.2Additional Context
(Include any other relevant information such as modifications to
config.py, custom settings, or related dependencies)Any insights or solutions to address these compatibility issues would be greatly appreciated.