I have a multi-class classifier and I'm trying to visualize how the confusion matrix values (ie. tn, fp, fn, tp) of each class change over time.
For example, let's say I have three classes of flowers A, B, C.
Flower A:
- Confusion matrix values of (10, 3, 2, 5) in interval 1
- Confusion matrix values of (8, 5, 5, 2) in interval 2
- Confusion matrix values of (10, 3, 6, 1) in interval 3
Flower B:
- Confusion matrix values of (9, 6, 3, 2) in interval 1
- Confusion matrix values of (8, 3, 4, 5) in interval 2
- Confusion matrix values of (5, 0, 7, 8) in interval 3
Flower C:
- Confusion matrix values of (1, 0, 9, 10) in interval 1
- Confusion matrix values of (5, 4, 9, 2) in interval 2
- Confusion matrix values of (6, 3, 8, 3) in interval 3
What would be some ideal ways to visualize the result?
question from:
https://stackoverflow.com/questions/65916396/how-to-visualize-confusion-matrix-values-of-sub-classes-over-time 与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…