lllyasviel / lllyasviel/ControlNet

About "Sudden Converge Phenomenon"

Open
#589 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
34.1k
Forks
3k
PR merge metrics
No merged PRs in 30d

Description

Regarding the number of steps in the "Sudden Converge Phenomenon" you mentioned, if this phenomenon occurs in step 6000, and my dataset has 10000 images with a batch_size of 1, will this phenomenon occur at epoch 0? Isn't it because there are still some images that haven't been sent to training?

Because I seem to have seen mention somewhere that training one epoch is enough, but if I look for it again, I can't find where it was mentioned

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

The issue names no files, tests, or entry points. First trace the training-step and epoch terminology in the project's documentation and surrounding training guidance, then establish whether the stated dataset and batch size affect the phenomenon. Done means documenting a clear, unambiguous answer with the relevant training context.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Documentation
Difficulty
3/5
Estimated time
1-2 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.