Llama3 video fix#45040
Conversation
Signed-off-by: Wang, Yi <yi.a.wang@intel.com>
Signed-off-by: Wang, Yi <yi.a.wang@intel.com>
|
The current failure (on Untangle config inheritance (#41541) For this PR, it fixed the CI, but I don't know if it's the best / correct way. Could you check, taking the above context. |
Signed-off-by: Wang, Yi <yi.a.wang@intel.com>
|
this issue still exists in the latest main, will you consider merging the PR @ydshieh @zucchini-nlp ? |
zucchini-nlp
left a comment
There was a problem hiding this comment.
Thanks for the ping, notifications in gh get lost sometimes 🫠
I'll trigger a slow CI and will merge
|
run-slow: video_llama_3 |
|
This comment contains models: ["models/video_llama_3"] |
CI ResultsCommit Info
Model CI Report❌ 2 new failed tests from this PR 😭
|
|
I will fix it on your behalf by checking inside CI runners, looks like a diff due to hardware |
|
[For maintainers] Suggested jobs to run (before merge) run-slow: video_llama_3 |
|
run-slow: video_llama_3 |
|
This comment contains models: ["models/video_llama_3"] |
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
when running tests/models/video_llama_3/test_modeling_video_llama_3.py::VideoLlama3IntegrationTest all fail cause by lm_head.weight is missing.