Skip to main content
Data Science

Return to Question

edited body
Source Link
Ethan
Ethan
  • 1.7k
  • 9
  • 24
  • 39

I have a set of datadataset with a few strongly imbalanced classes, eg. the smallest class is about 54 times smaller than the largest. Therefore, data augmentation in order to equalize the size of classes seems to melike a bad idea to me (in the example above each image would have to be augmented 54 times on average). So I thought that I could do less augmentation of minority classes, and then use class weights in the loss function. Is this approach better than the mere augmentation or just the use of class weights  ?

I have a set of data with a few strongly imbalanced classes, eg the smallest class is about 54 times smaller than the largest. Therefore, data augmentation in order to equalize the size of classes seems to me a bad idea (in the example above each image would have to be augmented 54 times on average). So I thought that I could do less augmentation of minority classes, and then use class weights in the loss function. Is this approach better than the mere augmentation or just the use of class weights  ?

I have a dataset with a few strongly imbalanced classes, eg. the smallest class is about 54 times smaller than the largest. Therefore, data augmentation in order to equalize the size of classes seems like a bad idea to me (in the example above each image would have to be augmented 54 times on average). So I thought that I could do less augmentation of minority classes and then use class weights in the loss function. Is this approach better than the mere augmentation or just the use of class weights?

Source Link
I.D.M
I.D.M
  • 175
  • 1
  • 10

CNN - imbalanced classes, class weights vs data augmentation

I have a set of data with a few strongly imbalanced classes, eg the smallest class is about 54 times smaller than the largest. Therefore, data augmentation in order to equalize the size of classes seems to me a bad idea (in the example above each image would have to be augmented 54 times on average). So I thought that I could do less augmentation of minority classes, and then use class weights in the loss function. Is this approach better than the mere augmentation or just the use of class weights ?