-
Notifications
You must be signed in to change notification settings - Fork 11
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.
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
Boosting not working #5
Comments
It is worth adding that I modified the original code as described in XLearning-SCU/2022-IJCV-TCL#3 |
Hi, it seems that in that batch there are no confident predictions selected as pseudo labels. In line 367 in loss.py, the cluster loss would be set to zero in that case. Thus the cluster projector did not receive grad, which leads to the above error. You could manually check the loss value. |
Thank you for quick response! On CIFAR it might be caused by the low number of epochs (50 vs the paper's 1000), but on my custom dataset I trained over 1000 epochs and it had the same issue.
|
What is the target cluster number of your custom dataset? As pointed out in the paper, when the cluster number is large, a sharper temperature in cluster-level loss is recommended. |
Hello, I have encountered the same problem. Have you solved it? |
Hi, sorry but I haven't worked on it much anymore. After learning the problem was the low confidence and high number of clusters I just reduced the number of clusters, each having more samples. It worked pretty well. |
Thank you for your reply. I encountered this problem when replacing my dataset, and my dataset only has 10 categories. I tried to modify the confidence parameter in InstanceLossBoost, but there was an error |
Training for 50 epochs on CIFAR-10 with
and then boosting with
Throws the following error:
This occurs in the middle of the 2nd boosting epoch.
Full log: tcl_cifar_exception.txt
This first occurred while I was boosting on a custom dataset, so I tried on CIFAR to see if it was caused by the model itself or the dataset. I think #3 (comment) was referring to this too.
The text was updated successfully, but these errors were encountered: