fix: compare the feature vector against medians, not poisoned means - #41
Merged
Conversation
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
El panel comparaba la magnitud de cada bloque contra la media sobre el split de test. El bloque de manos tiene media 3.37 y mediana 0.73: tres de sus 69 floats son la muñeca relativa al torso, y en el 0.01% de fotogramas donde MediaPipe colapsa el ancho de hombros a casi cero esa división llega a seis cifras. Resultado: el panel marcaba en rojo un vector perfectamente sano, por un factor de 3.
Ahora compara contra medianas (0.73 / 0.73 / 0.51 / 0.15).
Además separa el contador de salida en "Palabras del vocabulario" y "Letras deletreadas": estaban sumados, así que el panel decía "2 palabras" mientras el vocabulario había sido vetado en todas las ventanas — justo el caso que había que poder distinguir.