kohya-ss/sd-scripts

Shouldn't fix_noise_scheduler_betas_for_zero_terminal_snr be applied BEFORE prepare_scheduler_for_custom_training?

Open

#1,905 opened on Jan 27, 2025

View on GitHub
 (2 comments) (0 reactions) (0 assignees)Python (1,218 forks)batch import
help wanted

Repository metrics

Stars
 (7,198 stars)
PR merge metrics
 (Avg merge 18h 39m) (16 merged PRs in 30d)

Description

Hello,

Reference code: https://github.com/kohya-ss/sd-scripts/blob/0778dd9b1df0d6aa33287ded3ce4195f3d03251b/train_network.py#L195

I noticed that fix_noise_scheduler_betas_for_zero_terminal_snr modifies the alphas_cumprod for the noise scheduler, and prepare_scheduler_for_custom_training uses alphas_cumprod to calculate all_snr. The issue with the current order is that all_snr will reflect NON-ztsnr snrs, not ztsnr, causing anything that uses all_snr to use snrs that do not align with ztsnr.

As such, shouldn't fix_noise_scheduler_betas_for_zero_terminal_snr always be applied before prepare_scheduler_for_custom_training?

Thank you.

Contributor guide

Research direction
Investigate the order of calls in the training script: ensure fix noise scheduler betas for zero terminal snr is called before prepare scheduler for custom training so that all snr aligns with zero terminal SNR.
Tech stack
python
Domain
machine learningai
Issue type
Bug
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
2
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-3 hours
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Active
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Clear
Prerequisites
PythonStable Diffusion training scripts
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
60