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

update readme

Browse Source
This commit is contained in:
Kohya S
2023-07-16 15:14:49 +09:00
parent 516f64f4d9
commit 3db97f8897
1 changed files with 6 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
README.md
View File

@@ -24,12 +24,15 @@ The feature of SDXL training is now available in sdxl branch as an experimental
Summary of the feature: Summary of the feature:
- `tools/cache_latents.py` is added. This script can be used to cache the latents in advance. - `tools/cache_latents.py` is added. This script can be used to cache the latents to disk in advance.
- The options are almost the same as `sdxl_train.py'. See the help message for the usage. - The options are almost the same as `sdxl_train.py'. See the help message for the usage.
- Please launch the script as follows: - Please launch the script as follows:
`accelerate launch --num_cpu_threads_per_process 1 tools/cache_latents.py ...` `accelerate launch --num_cpu_threads_per_process 1 tools/cache_latents.py ...`
- This script should work with multi-GPU, but it is not tested in my environment. - This script should work with multi-GPU, but it is not tested in my environment.
- `tools/cache_text_encoder_outputs.py` is added. This script can be used to cache the text encoder outputs to disk in advance.
- The options are almost the same as `cache_latents.py' and `sdxl_train.py'. See the help message for the usage.
- `sdxl_train.py` is a script for SDXL fine-tuning. The usage is almost the same as `fine_tune.py`, but it also supports DreamBooth dataset. - `sdxl_train.py` is a script for SDXL fine-tuning. The usage is almost the same as `fine_tune.py`, but it also supports DreamBooth dataset.
- `--full_bf16` option is added. Thanks to KohakuBlueleaf! - `--full_bf16` option is added. Thanks to KohakuBlueleaf!
- This option enables the full bfloat16 training (includes gradients). This option is useful to reduce the GPU memory usage. - This option enables the full bfloat16 training (includes gradients). This option is useful to reduce the GPU memory usage.
@@ -39,9 +42,9 @@ Summary of the feature:
- `prepare_buckets_latents.py` now supports SDXL fine-tuning. - `prepare_buckets_latents.py` now supports SDXL fine-tuning.
- `sdxl_train_network.py` is a script for LoRA training for SDXL. The usage is almost the same as `train_network.py`. - `sdxl_train_network.py` is a script for LoRA training for SDXL. The usage is almost the same as `train_network.py`.
- Both scripts has following additional options: - Both scripts has following additional options:
- `--cache_text_encoder_outputs`: Cache the outputs of the text encoders. This option is useful to reduce the GPU memory usage. This option cannot be used with options for shuffling or dropping the captions. - `--cache_text_encoder_outputs` and `--cache_text_encoder_outputs_to_disk`: Cache the outputs of the text encoders. This option is useful to reduce the GPU memory usage. This option cannot be used with options for shuffling or dropping the captions.
- `--no_half_vae`: Disable the half-precision (mixed-precision) VAE. VAE for SDXL seems to produce NaNs in some cases. This option is useful to avoid the NaNs. - `--no_half_vae`: Disable the half-precision (mixed-precision) VAE. VAE for SDXL seems to produce NaNs in some cases. This option is useful to avoid the NaNs.
- The image generation during training is now available. However, the VAE for SDXL seems to produce NaNs in some cases when using `fp16`. The images will be black. Currently, the NaNs cannot be avoided even with `--no_half_vae` option. It works with `bf16` or without mixed precision. - The image generation during training is now available. `--no_half_vae` option also works to avoid black images.
- `--weighted_captions` option is not supported yet for both scripts. - `--weighted_captions` option is not supported yet for both scripts.
- `--min_timestep` and `--max_timestep` options are added to each training script. These options can be used to train U-Net with different timesteps. The default values are 0 and 1000. - `--min_timestep` and `--max_timestep` options are added to each training script. These options can be used to train U-Net with different timesteps. The default values are 0 and 1000.