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

Finetune in PODNet #48

Closed KHao123 closed 2 years ago

KHao123 commented 2 years ago

The code conduct finetune stage in training phase. It seem not be mentioned in the paper. Is the result reported in paper obtained with finetune?

arthurdouillard commented 2 years ago

Hello,

Yes we do the same finetuning as UCIR. In the paper, it is only done for the CNN model but not for the NME model. You can check the provided options files in this github that were used to produce the paper.