RVC-Project / RVC-Project/Retrieval-based-Voice-Conversion-WebUI
多张GPU时显存未正确分配|GPU memory is not properly allocated when using multiple GPUs
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 38.4k
- Forks
- 5.3k
- PR merge metrics
- No merged PRs in 30d
Description
我在服务器上有8张4090，设置0-1-2-3-4-5-6-7后，batch_size仅为4就提示显存不足，而改为0-1两张卡，batch_size设置为8，正常运行，每张卡有15g显存占用，显然不合理，我3060 6g笔记本都能跑8的batch_size，
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Reproduce the report with eight GPUs selected and batch sizes 4 and 8, then trace the multi-GPU memory allocation path. Compare the observed per-GPU usage with the two-GPU case and identify why the larger setup runs out of memory. Done means the reported configurations allocate memory correctly without an unexpected out-of-memory error.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 30/100