Add Min-P sampling for the GPT model in the web UI #1118
+15
−3
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
modified: GPT_SoVITS/AR/models/t2s_model.py modified: GPT_SoVITS/AR/models/utils.py modified: GPT_SoVITS/inference_webui.py
modified: GPT_SoVITS/inference_webui.py
modified: GPT_SoVITS/inference_webui.py
modified: sweight.txt
deleted: sweight.txt
|
Very nice! 💯 |
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Outdated suggestions cannot be applied.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.
The intent of this PR is to add Min-P sampling for the GPT model in the web UI.
Min-P sampling selects tokens with a probability above min_p * max_prob, where max_prob is the maximum probability in the distribution and min_p is a float between 0.0 and 1.0. This allows the model to adapt to situations with multiple viable logit choices, without sacrificing reliability when only the top logit is suitable.
In my testing a value of 0.95 is a good balance between reliability and sample variance. This almost certainly depends on the specific fine tune and personal preference.