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
Hi, first of all congrats on your great work and thank you for sharing.
I'm trying to reproduce your results on tiered-ImageNet. I already trained the model. When running the testing code, which "iters" value should I load the trained model from to obtain your results displayed in your paper? That parameter is indicated for the testing phase on mini-Imagenet, but it is not for tiered-ImageNet.
Thank you for your answer.
The text was updated successfully, but these errors were encountered:
oveilleux
changed the title
Testing - Model iteration to load tieredImageNet
Testing - Model iteration to load for tieredImageNet
Mar 1, 2021
oveilleux
changed the title
Testing - Model iteration to load for tieredImageNet
Testing - Model checkpoint to load for tieredImageNet
Mar 3, 2021
In our paper, we monitor the best validation performance and choose the corresponding iteration to test the model. This iteration may differ according to different training processes with random seeds.
Hi, first of all congrats on your great work and thank you for sharing.
I'm trying to reproduce your results on tiered-ImageNet. I already trained the model. When running the testing code, which "iters" value should I load the trained model from to obtain your results displayed in your paper? That parameter is indicated for the testing phase on mini-Imagenet, but it is not for tiered-ImageNet.
Thank you for your answer.
The text was updated successfully, but these errors were encountered: