lllyasviel / lllyasviel/ControlNet
Should I adjust more parameters during training phase?
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 34.1k
- Forks
- 3k
- PR merge metrics
- No merged PRs in 30d
Description
Hi,
I'm training this network using my own dataset, following your instructions. I start with the example, i.e., tutorial_train.py, and then modify the dataset to fit my own. I find that the training script presented is very simple, with very little paramters to adjust. However, I find many more parameters presented in inference files, for example, **model.control_scales** presented in gradio_seg2image.py. I'm wondering if these parameters also need to be adjusted during the training phase? what are the impacts of these parameters?
Thank you.
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 by reading tutorial_train.py and compare its training parameters with model.control_scales in gradio_seg2image.py. Determine whether the issue calls for documentation of these parameters or a training workflow change; the work is complete only when their training-time use and effects are clearly documented.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Documentation
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 15/100