<Yun>

May i know how you estimate the ZCA matrix?

1

Hello, I found your ZCA matrix and thanks for your upload. I know how to calculate ZCA matrix, but for CIFAR10 or 100 the matrix is huge and my hardware...

hi, I love your work! just one point: I think the original paper and code applies BN layer, so you can also include BN layer in your code. thanks

hi, i read through the model and I don't understand which part of the model represents the name "Lap"-SRN it is just a feature extraction with CNN and reconstruction with...

https://github.com/JingyunLiang/SwinIR/blob/6545850fbf8df298df73d81f3e8cba638787c8bd/models/network_swinir.py#L451 i think in your paper you have multiple layers?

SelfReformer is now on TMM and code is released

1

Hi, Thanks for your table. Here is the update for "SelfReformer" It was accepted by TMM this year, and the article's name changed to "Towards Complete and Detail-Preserved Salient Object...

dataset呢？

2

这个代码怎么用啊 dataset呢？ 能提供一下吗 谢谢

scaling factor is not working at all

3

hi, when we try to train on our own dataset, your scaling factor is not working at all. the code is so poor written, takes me so long to fix...

Missing key(s) in state_dict: "prior"

9

when i load your model, it gives error: RuntimeError: Error(s) in loading state_dict for MLNet: Missing key(s) in state_dict: "prior". code i am using: PATH = "./mlnet/pretrain.model" model.load_state_dict(torch.load(PATH,map_location=torch.device('cpu'))) thanks in...

Where is the ZCA matrix?

2

Hi, I read your paper and some of your model uses ZCA. I am not familiar with caffe at all and did not see how you apply the ZCA in...

为啥用matlab跑出来的结果和python的不一样

1

你好， 请问这个python的实现有和matlab的结果做比较吗？ 我拿了几个模型的预测结果分别用matlab和python跑出来结果不一样呢。s measure误差较大。 用你提供的单张图片跑出来的结果： matlab：max-F:0.9608. avg-f:0.9024 mae:0.0199 avg-E:0.9749 S-score:0.9315 python：max-F:0.9608 avg-f:0.9102 mae: 0.0199 avg-E:0.9740 S-score:0.9304 而且在matlab版本中，e-measure是用的均值，你的代码返回最大值， 我上面列的是我改为均值后的。 f-measure的误差我可以理解，因为matlab用的不一样公式。其余的误差差不多都在0.009，还算可以接受的水平。