microsoft / microsoft/microxcaling
Enquire on Matmul Layer diagram mode_config
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 364
- Forks
- 53
- PR merge metrics
- No merged PRs in 30d
Description
Hello, I just saw the overall diagram of Matmul Layer in your github website(https://github.com/microsoft/microxcaling/blob/main/images/rounding_and_quantizing_optns_matmul.png)
But after I checked the matmul.py,
I saw some inconsistencies between the diagram and the code.
In line number 33 of 'matmul.py'(https://github.com/microsoft/microxcaling/blob/main/mx/matmul.py#L33),
I saw that
'assert mode_config in ["aa", "aw", "wa"]'
but the diagram shows mode_config of "aa", "ww", "wa".
I want to make sure which one is right.
Thank you in advance.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by comparing the mode_config assertion at line 33 of mx/matmul.py with the labels in images/rounding_and_quantizing_optns_matmul.png. Confirm with maintainers which naming is intended, then update the inconsistent source or diagram. Done means the documented modes and the implementation agree.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Documentation
- Difficulty
- 2/5
- Estimated time
- 1-3 hours
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100