Add additional WD Tagger generation options #312
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
Added support for a minimum character-tag probability. If a character tag's probability is less than this value, it will be discarded. Set to `1.01` to disable character tags. Set to `0.01` to accept any matched character tags. Note: the existing 'Minimum tag probability' must _also_ be met for a tag to be included in results. Added additional generated-tag sorting option. Generated tags may now be sorted by: - `Highest probability` -- The existing behavior. - `Alphabetical` -- the (lowercased) alphabetical order of the tag names. - `Unsorted (Model Default)` -- tags are left in the order the model generated them. The order of this is model-dependant. Tag categories indexes (general, character, rating) are now stored as sets instead of lists for faster lookup performance. This change slightly increases the RAM footprint, but does not affect VRAM. Instead of rating-category tags being removed from the tag list for each generation, rating-category tags are no longer added to the overall tag list. Tag probability is now rounded to `2` decimal points when determining if a tag passes the minimum probability values. This fixes an issue where a tag's displayed probability could be `0.01` lower than the minimum probability.
|
I think it would be reasonable to move the new sort feature to the Settings dialog as well. I'm not sure how often users would actually swap to different values? |
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Outdated suggestions cannot be applied.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.
New
Added support for a minimum character-tag probability. If a character tag's probability is less than this value, it will be discarded. Set to
1.01to disable character tags. Set to0.01to accept any matched character tags. Note: the existing 'Minimum tag probability' must also be met for a tag to be included in results.Added additional generated-tag sorting option. Generated tags may now be sorted by:
Highest probability-- The existing behavior.Alphabetical-- the (lowercased) alphabetical order of the tag names.Unsorted (Model Default)-- tags are left in the order the model generated them. The order of this is model-dependant.Changes to existing behavior
Tag categories indexes (general, character, rating) are now stored as sets instead of lists for faster lookup performance. This change slightly increases the RAM footprint, but does not affect VRAM.
Instead of rating-category tags being removed from the tag list for each generation, rating-category tags are no longer added to the overall tag list.
Bug fix
Tag probability is now rounded to
2decimal points when determining if a tag passes the minimum probability values. This fixes an issue where a tag's displayed probability could be0.01lower than the minimum probability.