opengvlab / cafo Goto Github PK

Great job! However, I believe that when using few-shot learning, real-world scenarios often involve dealing with samples that are difficult to obtain. In such cases, can pre-trained models such as GPT and DALL-E provide good results?

Hi, are there any plans to release the paper on arxiv? If not, please let me know what the best way to cite your work would be by providing a bibtex entry. Thanks!

请问，用GPT3生成类别文本的代码有嘛？ 可以公开一下嘛？ 谢谢

https://image-net.org/challenges/LSVRC/index.php
I saw a lot of versions: 2010,2012, ...

Is it convenient to provide a link that can be downloaded directly?

Hi，excellent work！I just want to know could you release the code or more detailed information of your Dall-e's generated image, I saw some well-packaged files but did not know to do reproduce the process. Thanks a lot~

Congratulation on your excellent work! It is truly fascinating, and I'm currently conducting experiments based on your code.

However, I'm encountering some difficulties in reproducing your ImageNet 16 shots results. Despite following your config file without making any modifications, the highest test accuracy I've been able to achieve is 68.59, which is slightly lower compared to the reported number of 68.79. I would greatly appreciate any insights or comments you might have regarding this outcome.

I use the imagenet 2012 and downlaoded the val dataset like your dataset.md, but got an error like the title. So what folder in ../val? The val dataset only have images!

I hope this message finds you well! I recently stumbled upon your amazing open source repository, and I must say, I'm truly impressed with your work. I wanted to reach out and kindly request something important: adding a license to your repository. Your project has great potential, and by including a license, you can provide clarity and legal protection for users and contributors while fostering collaboration.

Would you consider adding a commonly used open source license, such as the MIT License or Apache License 2.0?

Thank you!

Traceback (most recent call last):
File "C:\Users\fengxiaohu\Downloads\Compressed\FSL\Prompt\CaFo-main\main.py", line 276, in
main()
File "C:\Users\fengxiaohu\Downloads\Compressed\FSL\Prompt\CaFo-main\main.py", line 175, in main
assert (os.path.exists(args.config))
File "C:\Users\fengxiaohu\Anaconda3\envs\Anaconda3\lib\genericpath.py", line 19, in exists
os.stat(path)
TypeError: stat: path should be string, bytes, os.PathLike or integer, not NoneType
如何解决呢？感谢回复

Thank you for your excellent work. I found you provide the few-shot yaml file, but I couldn't find the zero-shot files. Can you give me some modified suggestions?

Traceback (most recent call last):
File "D:\few-shot-code\2023CaFo-main\CaFo-main\main.py", line 276, in
main()
File "D:\few-shot-code\2023CaFo-main\CaFo-main\main.py", line 234, in main
dino_cache_keys, dino_cache_values = build_dino_cache_model(cfg, dino_model, train_loader_cache)
File "D:\few-shot-code\2023CaFo-main\CaFo-main\utils.py", line 96, in build_dino_cache_model
cache_keys = cache_keys.permute(1, 0)
RuntimeError: permute(sparse_coo): The number of dimensions in the tensor input does not match the length of the desired ordering of dimensions i.e. input.dim() = 4 is not equal to len(dims) = 2
Has the author encountered this problem and how to solve it?

The link of StanfordCars does not exist. Could you provide another download link?