modelscope / modelscope/DiffSynth-Studio
微调过程的显存消耗
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 13.1k
- Forks
- 1.3k
- Avg merge
- 13h 12m
- Merged PRs (30d)
- 45
Description
您好，非常感谢您的工作！它对我帮助很大！
但是，我近期在使用lora微调LongCat-Video模型时，所消耗的显存过大，大概为80多G，就算开启了--use_gradient_checkpointing_offload \也没有太多变化。我想要通过模型并行等方式减小其显存消耗，请问项目中是否有一些集成的方法可以实现这一点，或者是否有适配DiffSynth库的显存降低的代码供参考呢？
如果您能够抽时间解答我的问题，将会对我帮助很大，再次感谢您！
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
The issue concerns LoRA fine-tuning of the LongCat-Video model and the --use_gradient_checkpointing_offload option. Start by locating the fine-tuning entry point and checking whether model parallelism or other memory-reduction integrations are already documented or implemented. Done would require a defined, working approach that reduces the reported GPU memory use.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100