Conversation
monsieuricon

K. Ryabitsev

Playing with using Gemini to summarize long bug reports before bugbot sends them to the mailing list. The results are hit-and-miss.

Here's a more successful result for bug #219224
1
0
2
Man2Dev@fosstodon.org

Man2Dev idle

Reply to @monsieuricon
Edited 2 months ago

@monsieuricon hello, I'm in the fedora ai-ml Sig and was wondering if there is a git repo for this project to get get involved with or make suggestions for? If it's not open is it ok if I sent suggestions though email?

1
0
0
monsieuricon

K. Ryabitsev

Reply to @Man2Dev@fosstodon.org
@Man2Dev Hi! The project is bugspray [1], but the summarization parts are just rapid prototyping using langchain, not any real committed code.

.. [1] https://git.kernel.org/pub/scm/utils/bugspray/bugspray.git/
1
0
1
monsieuricon

K. Ryabitsev

Reply to @monsieuricon
@Man2Dev This is the janky code I have right now: https://paste.centos.org/view/df328fff

It requires a GOOGLE_API_KEY env var, but you should be able to replace that with any other llm.
1
0
1
Man2Dev@fosstodon.org

Man2Dev idle

Reply to @monsieuricon

@monsieuricon You might have better luck using llama.cpp or its downstream version, Ollama (if you need additional features), with the Llama 3.1 model for your PoC. Llama.cpp is relatively minimal. And Llama 3.1 model, offers a long context window. Additionally, both llama.cpp and Ollama expose an API that's compatible with OpenAI's API, allowing for easy fine tuning. Furthermore, both the tool and model support quantization, which can lower resource usage.

1
0
0
monsieuricon

K. Ryabitsev

Reply to @Man2Dev@fosstodon.org
@Man2Dev I'll take a look at that for sure. I have access to a Google Dev project that made it easy to get started, but the ultimate goal is to have a locally running LLM instance.
1
0
1
Man2Dev@fosstodon.org

Man2Dev idle

Reply to @monsieuricon

@monsieuricon If you're looking for better suggestions, there are people with more experience and knowledge than me in the AI-ML SIG.
We're mostly active in the Matrix channel:
https://matrix.to/#/#ai-ml:fedoraproject.org

0
0
0
About social.kernel.org

Terms of service

  1. Please do not use this service in violation of the Linux Kernel Code of Conduct. Doing so will result in your account suspension with the referral of the matter to the CoC committee.

  2. "Repeating"/"boosting" someone else's status on this platform will be treated as endorsement and will fall under rule #1.

  3. You are encouraged to use this platform to promote your work on the Linux Kernel, but there is no restriction on permitted topics (with the exception of anything covered by #1 above).

  4. There is no requirement to post in English, but it should be considered the primary language of communication on this platform.

Privacy notice

The admins of this service have access to all posted statuses. They aren't looking, but if it's something they shouldn't know about, then you should not post it on this platform.

Please see the Linux Foundation Privacy Policy, which applies to this platform as well.

Getting your own account

If you would like an account on this instance, please check that the following applies to you:

  • You are listed in MAINTAINERS or CREDITS

  • OR: You have a kernel.org account or email address

  • OR: You have a long and established history of involvement with the Linux Kernel

If the above is true and you agree with the Terms of Service and Privacy Notice listed above, please use these instructions to request an account: