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

chore: update run.sh script #11

Merged
merged 4 commits into from
Apr 11, 2024

Conversation

axel7083
Copy link
Contributor

What does this PR do?

The current script in this repository is out of date compared to the one in https://github.com/containers/ai-lab-recipes/blob/main/model_servers/llamacpp_python/src/run.sh

Screenshot / video of UI

What issues does this PR fix or reference?

Related to containers/podman-desktop-extension-ai-lab#825

How to test this PR?

  • podman build ./chat

@axel7083 axel7083 requested a review from a team as a code owner April 10, 2024 13:16
@axel7083 axel7083 requested review from benoitf and lstocchi April 10, 2024 13:16
Copy link
Contributor

@benoitf benoitf left a comment

Choose a reason for hiding this comment

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

I'm not sure about adding stuff around CONFIG_PATH as it's never used in the extension
and for the other env variables as well (except MODEL_PATH, HOST and PORT)

https://github.com/containers/podman-desktop-extension-ai-lab/blob/3e6f29ed112c2fc2e1696b632c82d6d6e66f3a7c/packages/backend/src/utils/inferenceUtils.ts#L150

so should we not only add

--clip_model_path "None" --chat_format "llama-2"

@axel7083
Copy link
Contributor Author

I'm not sure about adding stuff around CONFIG_PATH as it's never used in the extension and for the other env variables as well (except MODEL_PATH, HOST and PORT)

  • GPU_LAYERS will be required at some point (I would prefer to have it now)
  • CLIP_MODEL_PATH I don't know
  • CHAT_FORMAT is necessary to fix the issue

I removed the code related to the config file

chat/run.sh Outdated Show resolved Hide resolved
chat/run.sh Outdated Show resolved Hide resolved
chat/requirements.txt Outdated Show resolved Hide resolved
chat/run.sh Outdated Show resolved Hide resolved
chat/run.sh Outdated Show resolved Hide resolved
axel7083 and others added 4 commits April 11, 2024 14:11
Co-authored-by: Florent BENOIT <[email protected]>
Signed-off-by: axel7083 <[email protected]>
Signed-off-by: axel7083 <[email protected]>
Copy link
Contributor

@jeffmaury jeffmaury left a comment

Choose a reason for hiding this comment

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

LGTM

@benoitf
Copy link
Contributor

benoitf commented Apr 11, 2024

I'm still wondering, if ai lab is the only thing using these images if we should not remove env var that are not yet provided by Ai lab

like GPU_LAYERS and CLIP_MODEL_PATH

and add them the day it'll be handled

but then side question, how ai lab is ensuring that the image is matching the precondition.

if I have the extension being installed for a while and update the extension, will it update the image or notify that the service is no longer compliant ?

@axel7083
Copy link
Contributor Author

I'm still wondering, if ai lab is the only thing using these images if we should not remove env var that are not yet provided by Ai lab

like GPU_LAYERS and CLIP_MODEL_PATH

I am in favour of keeping them

and add them the day it'll be handled

but then side question, how ai lab is ensuring that the image is matching the precondition.

if I have the extension being installed for a while and update the extension, will it update the image or notify that the service is no longer compliant ?

We have nothing keeping track of the version

@axel7083 axel7083 merged commit 91431f5 into containers:main Apr 11, 2024
2 checks passed
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.

Merlinite model needs "chat_format" set to "openchat"
3 participants