lllyasviel / lllyasviel/ControlNet

Time embedding and why it is needed in the separate network?

Open
#62 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

Hi! Thanks for the great work on ControlNet, it is no less than a game changer. Upon closer look try to port it over to Draw Things, I find that because we duplicate parameters, we also have a time embedding passed into the ControlNet. This requires us to evaluate the network at each denoising step. If we can safely remove the time embedding, we don't need to do that and the evaluation can be faster.

Just wondering if you considered that and what's the thought process. Thanks!

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

No files or tests are named. Start by tracing how the ControlNet receives the time embedding during denoising and how Draw Things invokes the separate network; determine whether that input can be removed without changing behavior. Done would require a documented decision and, if feasible, implementation evidence covering the resulting evaluation behavior.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
ai, machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.