Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix gguf-convert-endian script #4037

Merged
merged 2 commits into from
Nov 11, 2023
Merged

Fix gguf-convert-endian script #4037

merged 2 commits into from
Nov 11, 2023

Conversation

monatis
Copy link
Collaborator

@monatis monatis commented Nov 11, 2023

Reported in this issue comment

@monatis monatis requested a review from KerfuffleV2 November 11, 2023 10:21
@monatis
Copy link
Collaborator Author

monatis commented Nov 11, 2023

cc @chenqiny

@chenqiny
Copy link
Contributor

It's resolved. Thank you.

Copy link
Collaborator

@KerfuffleV2 KerfuffleV2 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Oops, thanks for fixing that! I guess I messed it up when refactored those scripts to work with the Poetry package stuff.

All those linters and type checkers can catch the tiniest discrepancy in whitespace or whatever but can't warn me about something that simple and dumb? :(

@KerfuffleV2 KerfuffleV2 merged commit e86fc56 into master Nov 11, 2023
7 checks passed
@monatis monatis deleted the fix-gguf-convert-endian branch November 11, 2023 16:14
@cebtenzzre
Copy link
Collaborator

cebtenzzre commented Nov 11, 2023

All those linters and type checkers can catch the tiniest discrepancy in whitespace or whatever but can't warn me about something that simple and dumb? :(

mypy and pytype aren't so good with unbound variables, but pyright can catch them (and there may be a flake8 plugin that can do it too). It's not my favorite because it has a lot of false positives.

/home/jared/src/forks/llama.cpp/gguf-py/scripts/gguf-convert-endian.py
  /home/jared/src/forks/llama.cpp/gguf-py/scripts/gguf-convert-endian.py:34:23 - warning: "order" is possibly unbound (reportUnboundVariable)
  /home/jared/src/forks/llama.cpp/gguf-py/scripts/gguf-convert-endian.py:35:36 - warning: "order" is possibly unbound (reportUnboundVariable)
  /home/jared/src/forks/llama.cpp/gguf-py/scripts/gguf-convert-endian.py:45:67 - warning: "order" is possibly unbound (reportUnboundVariable)
  /home/jared/src/forks/llama.cpp/gguf-py/scripts/gguf-convert-endian.py:50:8 - warning: "order" is possibly unbound (reportUnboundVariable)
0 errors, 4 warnings, 0 informations

@chenqiny
Copy link
Contributor

I tested it with new generated llama2 7b model with convert.py and llama2 7b converted from convert script.

@KerfuffleV2
Copy link
Collaborator

Are you letting the suspense build before you tell us whether the result of testing was success or failure? :) I'm on the edge of my seat! 或者说我都迫不及待啦!

@chenqiny
Copy link
Contributor

Are you letting the suspense build before you tell us whether the result of testing was success or failure? :) I'm on the edge of my seat! 或者说我都迫不及待啦!

I just built it with the pull request.

olexiyb pushed a commit to Sanctum-AI/llama.cpp that referenced this pull request Nov 23, 2023
* Fix gguf-convert-endian script

* Bump version and update description
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

5 participants