Qingyun

@ShreyaaChauhan You can set a trace before the coco.py, line 271 like this: ```Python3 if not label < len(self.cat_ids): import ipdb; ipdb.set_trace() ``` When the program runs to the breakpoint,...

@engin-work Your `self.cat_ids` may be an empty list? try `print(self.cat_ids)` and more information of your dataset and your modification to your code is required to help you.

Hi, Yiming. The target branch of your PR may be `refactor-detr`? You can check and click Edit to change it.

## Short-term Plan We plan to implement a base detector class for Transformer-based detector, TransformerDetector, move the transformer-related part in the original head module to the new detector module, and...

## Short-term Plan We decided to refactor the forward process of Transformer in the TransformerDetector class. Specifically, the whole forward process is wrapped with forward_Transformer, which is divided into about...

> The design LGTM. Only need to fix minor comments +1 Thank you very much for the review and the suggestions! The comments have been fixed! @Tai-Wang

> The whole design LGTM. Thanks very much for your review and the suggestions! The comments have been replied and modifications have been committed. @RangiLyu

We decide to complete the development of DINO on MMDetection 2.x in this pr. The code remains using the old style (the code has been refactored in MMDetection 3.x for...

@betterze @ElliotQi @liangbingzhao Why did your experiments not work? I tried ddim two, the result are as follows: diffusion method | SSIM(+) | PSNR(+) -- | -- | -- DDPM...

I'm a newcomer, my codebase refer to ddim_sample() in denoising_diffusion_pytorch. The exps both use `64×64 -> 512×512 on FFHQ-CelebaHQ` ckpt and `sr_sr3_16_128.json` config Are the ddim results better than ddpm...