Hi, should I use torchmetrics MulticlassF1Score or skicit-learn sklearn.metrics.f1_score for calculating F1 score of my classification model? I’m a beginner and I want to know what should I use to be professional. Both seem to be the same and I don’t know what role does one solve over another.
torchmetrics should be able to accept PyTorch tensors so might be more convenient as you would be able to pass e.g. CUDATensors directly to their methods.