[dev] Improved LoRA Rank Settings & Definition Changes #853
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Same TypeError as issue d8ahazard#804, Step I do: 1. Select `Model` 2. Load Settings 3. Save Settings 4. Generate Ckpt ``` Exception compiling checkpoint: stat: path should be string, bytes, os.PathLike or integer, not list Traceback (most recent call last): File "E:\stable-diffusion-webui\extensions\sd_dreambooth_extension\dreambooth\diff_to_sd.py", line 388, in compile_checkpoint if not os.path.exists(lora_path): File "C:\Program Files\Python310\lib\genericpath.py", line 19, in exists os.stat(path) TypeError: stat: path should be string, bytes, os.PathLike or integer, not list ```
Update API: fixing POST /dreambooth/model_config
I had enough with Gradio mischief
|
Need to fix some issues with sampling first. Will re-open when ready. |
|
Superseded by #880 |
|
@ExponentialML what do you suggest for for faces and for styles differently? also what learning rates do you suggest accordingly for 8bitadam optimizer and for lion? thank you so much |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Improves the LoRA ranking settings by having them be configurable separately.
What brought about this change was the text encoder having a low amount of weight at the highest setting (100). The max is now at the LoRA implementation's recommended maximum of 768. I'm leaving the UNET for LoRA at 128 as it's great for most cases.
Lora UNETnow has a maximum rank of 128.Lora Text Encodernow has a maximum rank of 768.