New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Bug in Sub-graph loss #50

Open

fjchange opened this issue Jun 4, 2021 · 1 comment

fjchange commented Jun 4, 2021 •

edited

Loading

gtad/gtad_lib/loss_function.py

Line 23 in 6deb5b1

def subgraph_loss_func(pred_bm, gt_iou_map, bm_mask, lambda1=10):

def subgraph_loss_func(pred_bm, gt_iou_map, bm_mask, lambda1=10):
    pred_bm_wcs = pred_bm[:, 0].contiguous()
    pred_bm_mse = pred_bm[:, 1].contiguous()
->  wce_loss = pem_reg_loss_func(pred_bm_wcs, gt_iou_map, bm_mask)
->  mse_loss = pem_cls_loss_func(pred_bm_mse, gt_iou_map, bm_mask)
    subgraph_loss = wce_loss + lambda1* mse_loss
    return subgraph_loss

Here, it is obvious that the wce_loss and mse_loss are wrongly named and lead to the wrong weighted sum.

The text was updated successfully, but these errors were encountered:

Owner

frostinassiky commented Jun 4, 2021

Very good catch @fjchange . This is a bug.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment