Changing Class Weights In YOLOv10

[Lokesh-1015]

Question:
I trained my model on a dataset using YOLOv10. But my model did not perform well in some classes. When I examined my data set, I saw that there was very little data in those classes. Instead of methods such as data augmentation, I want my model to give more weight to classes with less data. How can I make this happen? What changes are supposed to be made while training model?

Dataset:
I am using vindr cxr dataset. Images are resized to 224x224.
Imabalance -
Class Name - Count:
Aortic enlargement - 7193
Cardiomegaly - 5443
Pleural thickening - 4884
Pulmonary fibrosis - 4659
Nodule/Mass - 2611
Lung Opacity - 2493
Pleural effusion - 2483
Other lesion - 2228
Infiltration - 1247
ILD - 1015
Calcification - 967
Consolidation - 556
Rib fracture - 300
Atelectasis - 280
Mediastinal shift - 270
Pneumothorax - 226
Enlarged PA - 188
Emphysema - 141
Lung cavity - 91
Lung cyst - 43
Clavicle fracture - 30
Edema - 19

Note:
I also used weighted dataloader, it did improve performance by some extent but still performance is not good at all.