search

OatmealLiu / class-iNCD

PyTorch implementation for the paper Class-incremental Novel Class Discovery (ECCV 2022)
85 stars 10 forks source link

No such file or directory：tinyimagenet_200.txt #1

Closed zstarN70 closed 2 years ago

zstarN70 commented 2 years ago

Please tell me how to obtain the file? Or what format the file was generated from?

roysubhankar commented 2 years ago

Hi,

Did you follow the instructions for preparing the tinyimagenet dataset as described in our repo?

For TinyImagenet, to download and generate image folders to ./data/datasets/. Please follow https://github.com/tjmoon0104/pytorch-tiny-imagenet

zhunzhong07 commented 2 years ago

@zstarN70 Hi, thanks for your interest in our work.

If you extract the tiny-imagenet file, you can obtain the file wnids.txt. Please rename it to tinyimagenet_200.txt.

Sorry for this confusion.

We will update a more detailed instruction of preparing datasets. @OatmealLiu

OatmealLiu commented 2 years ago

@zstarN70 Hi! Thanks for running our code.

I uploaded the prepared datasets (CIFAR-10, CIFAR-100 and TinyImagenet) to drive. You can now download the dataset from: https://drive.google.com/file/d/1o73ehAii6AJpJQ5tMwfywE2JBiFC3Bv7/view?usp=sharing

After you downloaded it, you can move the downloaded datasets.zip file to ./data/ folder. Then you can unzip datasets.zip to ./data/ folder.

I also updated the dataset preparation instruction in our README.md file. You can also check it there.

Thanks!

Best regards, Miu

zstarN70 commented 2 years ago

@OatmealLiu @zhunzhong07 Thank you for your reply！Excellent work！