Bruno Ma

@weiyx16 Greetings. Have you ever tried CLIP on the StanfordCars dataset? I can only get a ~47%/~48% acc without/with the "prompt ensembling" trick, far away from ~55% as reported in...

I solved my problem. It is data augmentation that cheats me.

> > > I use the code in the `README.md (Zero-Shot Prediction)` to test the acc of ViT/B-32 on the CIFAR100 dataset and get the result of about 62% by...

> Hey, I could do it successfully using `clean` and `legacy_pytorch` modes. I just needed to add the extension 'JPEG' to https://github.com/GaParmar/clean-fid/blob/main/cleanfid/utils.py#L50 because ImageNet-1k images have that extension. I think...

Thanks for your reply! Following the link you sent, I've confirmed that Numba is installed correctly. Besides, my GPU is a 12 GB Titan V, and my CPU has 125...

@jayaram-r Besides, the test set is normalized by `transforms.Normalize(*NORMALIZE_IMAGES['cifar10'])]` in [generate_samples.py](https://github.com/jayaram-r/adversarial-detection/blob/master/expts/generate_samples.py), the bound of image pixels are rescaled from [0, 1] into [-2.429, 2.754]. Correspondingly, I think the epsilon (1/255)...

I am a huge fan of adversarial training, and I have also checked your previous work [CV_A-FAN](https://github.com/VITA-Group/CV_A-FAN). In the classification task of CV_A-FAN, there seems no additional "adversarial" branch within...

Yeah yeah, I think `nerf_adv` should be replaced with `nerf_fine`/`nerf` as well.

Hello @reynoldscem . It takes me about 10 days.

I'm sorry. After I download the [pretrained models](https://github.com/dunbar12138/DSNeRF/blob/main/download_models.sh) that you provided, I find only a **50000.tar** file and a **test** directory under every folder. I think the **config.txt** and **args.txt**...