ys-li

Thanks for your advice, @SingleZombie Yeah, adding `type=int` and setting a lower number will be better. Besides, Adding an annotation creator will be helpful. Now, shall we achieve the above...

Metrics are defined in `https://github.com/open-mmlab/mmediting/blob/master/mmedit/core/evaluation/metrics.py`. You can add your metrics in `https://github.com/open-mmlab/mmediting/blob/master/mmedit/core/evaluation/metrics.py` or create a new file in `https://github.com/open-mmlab/mmediting/blob/master/mmedit/core/evaluation/` to add your metrics. Besides, it will be better if you...

1. `val_set5` is another dataset called [`Set5`](https://paperswithcode.com/dataset/set5) 2. You can crop images by `https://github.com/open-mmlab/mmediting/blob/master/tools/data/super-resolution/div2k/preprocess_div2k_dataset.py` 3. Mean and std are provided in EDSR. Besides, we cannot give firm advice on whether...

I'm sorry for the previous code doesn't support inference of RefSR methods such as TTSR, and I have fixed it in #716 @ThorPham now you can download the PR branch...

#716 have been merged into master, now you can run inference of TTSR by ``` python demo/restoration_demo.py configs/restorers/ttsr/ttsr-gan_x4_c64b16_g1_500k_CUFED.py checkpoint/ttsr/ttsr-gan_x4_c64b16_g1_500k_CUFED_20210626-2ab28ca0.pth /home/thorpham/Documents/challenge/test/img_5.jpg outputs/demo_out_baboon.png --imshow --ref_path [REF_PATH] ``` where `REF_PATH` is the file...

@dzz416 If an image whose format is `png`, it may have a hidden channel: `alpha`. I think you can check the channel number of images by ``` img=cv2.imread('path', flag=cv2.IMREAD_UNCHANGED) print(img.shape)...