search

arthurdouillard / incremental_learning.pytorch

A collection of incremental learning paper implementations including PODNet (ECCV20) and Ghost (CVPR-W21).
MIT License
383 stars 60 forks source link

About UCIR configuration #42

Closed BinahHu closed 3 years ago

BinahHu commented 3 years ago

Thank you so much for this excellent work!

I am a little confusing about the epoch number for ucir in the latest version (ucir_cifar100.yaml, line 42). It seems that you change it from 160 to 1. However, in the original implementation of UCIR, such number is 160. Moreover, I tried to reproduce UCIR using the current configuration, the results are awful (15.17% average acc).

I am wondering if this is a typo, or you have a higher level of consideration? Thank you so much!

arthurdouillard commented 3 years ago

Thanks for the kind words :)

I was just testing the model when I used 1 epoch, the correct number is obviously 160 as you well remarked. Unfortunatly, even with 160 epochs my results were a little under those of the original paper if I recall.

BinahHu commented 3 years ago

Thank you for your reply!

It's fine if the reproduced performance is a bit lower. That's acceptable!