Can someone please tell me if I have a different number of images per class during training on how to tackle the class imbalance problem in PyTorch?
How to solve this “class imbalance problem” of a classification model in PyTorch?
Asked
Active
Viewed 70 times
0
-
Does this answer your question? [Multilabel classification with class imbalance in Pytorch](https://stackoverflow.com/questions/58206286/multilabel-classification-with-class-imbalance-in-pytorch) – m13op22 Apr 20 '20 at 02:53
-
@HS-nebula, thanks a lot for the reply. I already went through that but I am looking to solve the problem for a classification model. Will those solutions for object detectors work for image classification models? Any suggestions will be highly helpful. – MSD Paul Apr 20 '20 at 03:23
-
Wait, re-reading your question: how many images do you have per class? Can you share the distribution? Each instance in your dataset is an image, with the columns the various features of those images? Can you provide a [MCVE](https://stackoverflow.com/help/minimal-reproducible-example)? – m13op22 Apr 20 '20 at 04:22