-
I can't find the test code
Please help me
Can you provide test code
-
Could you provide some instructions on how to test the pre-trained models?
-
The provided C++ api for running VOE is not enough and the sample examples use the Onnx_task.hpp which is not generic for all models so can you please provide a working example including the needed he…
-
# Error
When trying to run the `TestRCAN_scripts.sh`, here comes the error
```
$ sh TestRCAN_scripts.sh
Traceback (most recent call last):
File "main.py", line 4, in
import data
File "…
-
Please take a look at https://github.com/ChaofWang/Awesome-Super-Resolution and https://github.com/YapengTian/Single-Image-Super-Resolution for other algorithms that you can implement in order to be c…
-
hi authors,
i‘ve only tested the performance of cutblur once by using `python main.py --model CARN --augs cutblur --alpha 0.7 --dataset RealSR --scale 4 --camera all --dataset_root ./input/RealSR/ -…
-
In the paper:
To further reduce parameter burdens in attention connections, we re-formulate the direct fully-connected layer by two lightweight fully-connected layers; the output sizes are C/r and …
-
Hi Yulun,
Thank you for sharing your excellent work.
While answering this [issue](https://github.com/yulunzhang/RCAN/issues/4#issuecomment-406669422), You mentioned that your model takes 70s on …
swz30 updated
5 years ago
-
您好 我在论文复现的时候出现了一些问题，首先，是学习率的设置 您论文中说的是每200次衰减一半，然后代码设置里是说的每50次衰减为0.6，按照代码直接跑的结果是37.7多，改完学习率以后跑的结果是37.9多。
随后模型中LSRAG中soca后面是没有conv层的 然后您加了conv层 我去掉跑的话第一代结果为8.多，然后就直接关掉了 请问您加这一层的作用是什么?
最后 我又添加了SSRG中最后…
-
Thank you for sharing codes to us. While reproducing the code, I came across some minor issues that I would like to ask you for advice on, as follows:
![KE0_L@XDKXGJW@(6P4X1AP1](https://user-images.g…