deezer / deezer/spleeter

[Discussion] cuda memory optimisation issues

Open
#893 0 comments 0 reactions 0 assignees View on GitHub
question
Dominant language
Python
Stars
28.4k
Forks
3.1k
PR merge metrics
No merged PRs in 30d

Description

Hey, I require some assistance in building a pipeline that includes `spleeter` audio splitting using the 2stems model.

GPU Spec - `Tesla V100-PCIE-16GB`

I wanted to know, how can we clear up the memory usage caused during the process.

```separator = Separator('spleeter:2stems')```

Is there any way to clear up the memory usage caused by this process so that it can be used later?

Contributor guide

Open the contributing guide

Research direction

Start with the `Separator('spleeter:2stems')` pipeline and reproduce the GPU memory behavior on the stated Tesla V100-PCIE-16GB setup. Determine whether the issue is memory retained after processing or expected model allocation; done would require a documented, reproducible resolution, but this discussion names no files or tests.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, tensorflow
Domain
machine-learning
Issue type
Bug
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.