VictorMylle/Kohya-ss-sd-scripts
1
0
Fork 1
mirror of https://github.com/kohya-ss/sd-scripts.git synced 2026-04-09 06:45:09 +00:00
Code Issues Packages Projects Releases Wiki Activity

Fix error in applying mask in Attention and add LoRA converter script

Browse Source
This commit is contained in:
Kohya S
2024-08-21 12:30:23 +09:00
parent e17c42cb0d
commit 2b07a92c8d
3 changed files with 10 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
README.md
View File

@@ -9,6 +9,12 @@ __Please update PyTorch to 2.4.0. We have tested with `torch==2.4.0` and `torchv
The command to install PyTorch is as follows:
`pip3 install torch==2.4.0 torchvision==0.19.0 --index-url https://download.pytorch.org/whl/cu124`
Aug 21, 2024 (update 2):
Fixed an error in applying mask in Attention. The attention mask was float, but it should be bool.
Added a script `convert_flux_lora.py` to convert LoRA between sd-scripts format (BFL-based) and AI-toolkit format (Diffusers-based). See `--help` for details. BFL-based LoRA has a large module, so converting it to Diffusers format may reduce temporary memory usage in the inference environment. Note that re-conversion will increase the size of LoRA.
Aug 21, 2024:
The specification of `--apply_t5_attn_mask` has been changed. Previously, the T5 output was zero-padded, but now, two steps are taken: "1. Apply mask when encoding T5" and "2. Apply mask in the attention of Double Block". Fine tuning, LoRA training, and inference in `flux_mini_inference.py` have been changed.