You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
@nashory hi, I noticed that, when training the keypoint stage, you set use_l2_normalized_feature = True, what's the reason of setting this parameters? And what's more, I noticed that you set target_layer = layer3 by default, have you tried target_layer = layer4? If yes, which one is better?
Another question that confused me is, when I train finetune stage directly on google-landmark-dataset-top1k, I got acc1 over 97.5, how about your result on this stage?
Thank you and wait for your answer.
The text was updated successfully, but these errors were encountered:
@nashory hi, I noticed that, when training the keypoint stage, you set use_l2_normalized_feature = True, what's the reason of setting this parameters? And what's more, I noticed that you set target_layer = layer3 by default, have you tried target_layer = layer4? If yes, which one is better?
Another question that confused me is, when I train finetune stage directly on google-landmark-dataset-top1k, I got acc1 over 97.5, how about your result on this stage?
Thank you and wait for your answer.
The text was updated successfully, but these errors were encountered: