kohya-ss / kohya-ss/sd-scripts

Random captions

Open
#2,120 1 comment 0 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
7.2k
Forks
1.2k
Avg merge
11m
Merged PRs (30d)
2

Description

How can I use a random caption from several choices on different lines in the caption file, so that when selecting a caption for a batch, a random caption (from a random line) is used?
I'm just a little confused: the T5 embedding cache on the disk takes 8 MB. I created two-line captions. Now, in theory, the cache should take up 16 MB, or two 8 MB files should be created. But in fact, I see the same 8 MB caches.

Contributor guide

No contributing guide indexed for this repository

Research direction

No source file, test, or entry point is named. Start by tracing the caption-file handling and T5 embedding-cache creation for batch caption selection; determine whether separate caption lines should produce separate cache data and document or test the observed behavior.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
30/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.