lllyasviel / lllyasviel/ControlNet
training is so slow
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 34.1k
- Forks
- 3k
- PR merge metrics
- No merged PRs in 30d
Description
Hi, You really did a great job. thanks so much. When I train it myself. It is so slow, I do not think it is the right speed.
Image gives the trainning process. How do you think that?
Because I use the pytorch-lightning==2.2.3, I can run only when I modified from "pl.Trainer(gpus=1, precision=32, callbacks=[logger])" to "pl.Trainer(precision=32, callbacks=[logger])". Dose it matter?
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
Start with the reported training setup and the pytorch-lightning==2.2.3 Trainer configuration, comparing `pl.Trainer(gpus=1, precision=32, callbacks=[logger])` with the configuration that runs. Use the attached training-process image and establish whether the observed speed is expected or caused by the changed GPU configuration; document the reproducible cause and verification.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- machine-learning, performance
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100