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

Stable audio model example #2188

Merged
merged 12 commits into from
Jul 18, 2024

Conversation

aleksandr-mokrov
Copy link
Contributor

CVS-144663

Copy link

Check out this pull request on  ReviewNB

See visual diffs & provide feedback on Jupyter Notebooks.


Powered by ReviewNB

@aleksandr-mokrov aleksandr-mokrov marked this pull request as ready for review July 11, 2024 11:18
@aleksandr-mokrov aleksandr-mokrov requested review from a team and eaidova and removed request for a team July 11, 2024 11:18
Copy link

review-notebook-app bot commented Jul 11, 2024

View / edit / reply to this conversation on ReviewNB

eaidova commented on 2024-07-11T13:48:14Z
----------------------------------------------------------------

is it possible to use less detiled title? in case if we want to replace it by new model version it will required to rewrite title


aleksandr-mokrov commented on 2024-07-15T23:34:56Z
----------------------------------------------------------------

Renamed

Copy link

review-notebook-app bot commented Jul 11, 2024

View / edit / reply to this conversation on ReviewNB

eaidova commented on 2024-07-11T13:48:15Z
----------------------------------------------------------------

judging by described architecture, model is fully transformed. Is it possible to try to apply int8/int4 weight compression for it?


aleksandr-mokrov commented on 2024-07-15T23:34:59Z
----------------------------------------------------------------

Weight Compression is added. But only with mode=CompressWeightsMode.INT8_SYM it generates good results.

Copy link

review-notebook-app bot commented Jul 11, 2024

View / edit / reply to this conversation on ReviewNB

eaidova commented on 2024-07-11T13:48:16Z
----------------------------------------------------------------

please turns model parts to headers for simplification of notebook navigation


aleksandr-mokrov commented on 2024-07-15T23:35:04Z
----------------------------------------------------------------

Extended

Copy link

review-notebook-app bot commented Jul 11, 2024

View / edit / reply to this conversation on ReviewNB

eaidova commented on 2024-07-11T13:48:17Z
----------------------------------------------------------------

from which this print comes from?


aleksandr-mokrov commented on 2024-07-15T23:35:09Z
----------------------------------------------------------------

https://github.com/openvinotoolkit/openvino/blob/master/tools/ovc/openvino/tools/ovc/moc_frontend/pytorch_frontend_utils.py#L137C5-L137C23

Non informational outputs was removed

Copy link

review-notebook-app bot commented Jul 11, 2024

View / edit / reply to this conversation on ReviewNB

eaidova commented on 2024-07-11T13:54:43Z
----------------------------------------------------------------

Line #1.    %pip install -q "torch>=2.1" "torchaudio" einops "stable-audio-tools" "gradio>=4.19" --extra-index-url https://download.pytorch.org/whl/cpu

I met some issues with the latest stable-audio-tools release, when worked on model export, did you check that it is working?


Copy link
Contributor Author

Renamed


View entire conversation on ReviewNB

Copy link
Contributor Author

Weight Compression is added. But only with mode=CompressWeightsMode.INT8_SYM it generates good results.


View entire conversation on ReviewNB

Copy link
Contributor Author

Extended


View entire conversation on ReviewNB

Copy link
Contributor Author

@eaidova eaidova merged commit b0fddfb into openvinotoolkit:latest Jul 18, 2024
15 of 17 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.

None yet

2 participants