-
I would like to ask you whether you evaluate the direct effect of different layers on zero-shot classification accuracy on the test set of ImageNet or on the same validation set used to calculate the …
-
I installed the conda environment using the provided environment.yml. I am using RedHat OS.
```
Environment:
Python: 3.9.7
PyTorch: 1.13.0+cu117
Torchvision: 0.14.0+cu117
CUDA:…
-
Hi @YuYang0901, thank you for sharing your great work.
I am trying to run your code, could you please update the training command to train the ViT-L/14@336px model on waterbird dataset and how you …
-
Do you plan to release the best models for all the algorithms and datasets? Based on the paper, it is not clear what were the best hyperparameter values since it provides mean and std over top 5. It w…
-
May I know what the encoder used to extract features on WaterBirds is? I have used the ViT-L/14 embedding in the repo for linear probing. However, the performance is much higher (>80) than the number …
-
Hi, I build a factor template using random words and add it to "imagenet_factor_templates":
```
"helloworld": [
"",
", hello",
", world",
", !",
],
…
-
Hello, first of all, thank you for your wonderful summary of baselines on 'spurious correlation'.
I have a question about training code of 'DFR'.
In the original paper, D_{val} is splitted to D^1_…
-
Summary of request: Add a new organization to ROR
Name of organization: The Waterbird Society
Website: https://waterbirds.org
Link to publications:
Organization type: Nonprofit
Wikipedia page:…
-
Hi @izmailovpavel ! Thanks for your great work and I really enjoy it. In section 6 and figure 3, it was reported that 'random init' base model achieved 0.9 mean group accuracy, however in my experimen…
-
Hi,
when running `python main.py --config configs/Waterbirds/mlp.yaml`, I encountered: `NameError: name 'load_embeddings' is not defined` in main.py [line 91](https://github.com/lisadunlap/LADS/blob/…