This is a toy example of robust regreesion with huber loss, which is suitable when training samples have outliers. The ground truth is a simple line. The results below show that huber loss could get rid of the outliers.
p3i0t/robust_regression_with_huber_loss
Folders and files
| Name | Name | Last commit date | ||
|---|---|---|---|---|
