UCL

UCL copied to clipboard

Update README.md

1

typo: configs/barlowm_c100.yaml -> configs/barlow_c100.yaml and 1 more

jundev1l2l

Hey authors, I am not sure that the learning is updated in the code. While I saw something to define the learning rate update strategy, there is nothing to be...

JonnesLin

Reproduce issues

6

Hi authors, Ideally, we can get the same result as the paper reported if I set the cl_model to "mixup", right? But I only got 70.95% acc on this way...

JonnesLin

Hi, I found a difference between your code and the way the `Average Forgetting` is calculated in the paper. - paper `Average Forgetting`:  - code `line 113` in `main.py`:...

zuael

Hi, I fixed the code for calculating the `average forgetting` in `main.py` and `linear_eval_alltasks.py`, and updated the configs.

zuael

Dear Author, Thank you for your excellent work! I have a few questions regarding the evaluation process at UCL: After the representation learning phase, the paper mentions measuring performance using...

pgh2874