correctly swapping precision and recall

samleoqh / MSCG-Net

Multi-view Self-Constructing Graph Convolutional Networks with Adaptive Class Weighting Loss for Semantic Segmentation

MIT License

68 stars 28 forks source link

Open luarzou opened 1 year ago

luarzou commented 1 year ago

I guess the variables for fp and fn have to be swapped as indicated by the indices of the confusion matrix.