sensein / senselab

senselab is a Python package that simplifies building pipelines for biometric (e.g. speech, voice, video, etc) analysis.
http://sensein.group/senselab/
Apache License 2.0
8 stars 3 forks source link

Add support for generating confusion matrices #107

Closed 900miles closed 1 month ago

900miles commented 1 month ago

Description

Confusion matrices are a very useful visualization tool that senselab should support in its plotting functions

Example: output_44_0_huca103b841108efc6bc2697d38201aa3e_36627_3101c8cc78b6694d507fc029805ac757

Tasks

Freeform Notes

No response

900miles commented 1 month ago

Already implemented in sci-kit (https://scikit-learn.org/stable/modules/generated/sklearn.metrics.confusion_matrix.html), though should probably combine this with matplotlib to produce a nice picture. Perhaps a function parameter to return either a nice plot or the matrix itself, or both?

900miles commented 1 month ago

Actually nvm this is implemented exactly as intended here: https://scikit-learn.org/stable/modules/generated/sklearn.metrics.ConfusionMatrixDisplay.html#sklearn.metrics.ConfusionMatrixDisplay.from_predictions

900miles commented 1 month ago

From @satra: "things like metrics and confusion matrices should just be used from scikit-learn. no real need to incorporate them at this point."

I'll close this for now. If we decide this would be good to implement in the future we can reopen the issue or create a new one.