search

Caleydo / confusionflow-ui

Visualize learning dynamics in multi-class image classifiers (e.g., convolutional neural networks) 💬 @gfrogat @thinkh
https://confusionflow.org
BSD 3-Clause "New" or "Revised" License
1 stars 0 forks source link

Update precision/recall after subsetting of confusion matrix #239

Open thinkh opened 5 years ago

thinkh commented 5 years ago

Currently the metrics curves (precision, recall, F1 score) are always computed for the whole confusion matrix. When filtering classes (i.e., subsetting the confusion matrix) these curves must be updated based on the selected set of classes.

Performance metrics for 10 classes:

image

Same performance metrics for 4 classes: image

@gfrogat Check how the subsetting/computation is handled.

thinkh commented 5 years ago

It seems that it is already correct, please check the computed performance values.