kohya-ss / kohya-ss/sd-scripts

Influence of seed

Open
#1,358 0 comments 0 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
7.2k
Forks
1.2k
Avg merge
11m
Merged PRs (30d)
2

Description

Hi @kohya-ss, I've noticed that the result of a dreambooth training can deviate a lot when changing the seed. What exactly does the seed influence when passing it to `sdxl_train.py` and what happens if I don't pass the seed?

Contributor guide

No contributing guide indexed for this repository

Research direction

Start by reading the seed handling and training entry point in sdxl_train.py. Trace what the seed affects and the behavior when it is omitted, then document both cases with enough context to explain why DreamBooth results can differ.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
documentation, machine-learning
Issue type
Documentation
Difficulty
2/5
Estimated time
1-3 hours
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.