search

dvlab-research / Parametric-Contrastive-Learning

Parametric Contrastive Learning (ICCV2021) & GPaCo (TPAMI 2023)
https://arxiv.org/abs/2107.12028
MIT License
241 stars 32 forks source link

Question about CIFAR-LT experiment #5

Closed caisarl76 closed 3 years ago

caisarl76 commented 3 years ago

Hi! thank you for your interesting work!

I have a question on hyperparameters in CIFAR-LT dataset experiments

The paper explains alpha and moco temperature, but there is no explanation for the rest of the hyper-parameters(beta, gamma, augmentation stradegy). Can you please provide it?

jiequancui commented 3 years ago

Hi, thank you for your questions.

We have updated our code. Training code and configurations for CIFAR-LT are provided now.

jiequancui commented 3 years ago

Hi,

There seems something wrong with the current cifar configuration. We are working on it now. I will update it later.

jiequancui commented 3 years ago

Hi,

We have updated the training code and configurations for CIFAR-LT.

Yancy10-1 commented 2 years ago

Hi,

We have updated the training code and configurations for CIFAR-LT.

Hi, CIFAR-LT experiment is so interesting. Is the experiment CIFAR-100 LT with imbalance factor 100 mentioned in the paper? Does the project provide full CIFAR-100 experiment?

jiequancui commented 2 years ago

Hi,

The code on full CIFAR-100 for GPaCo will be released soon

jiequancui commented 2 years ago

Hi,

The training scripts on full CIFAR-100 for GPaCo are released. Scripts on CIFAR-LT will be released later.