li-chongyi / water-net_code Goto Github PK

when i run the main_test.py, it always be killed after 4 Reading checkpints.

Raw has 890 images, should I randomly select 800 from them for training 90 for the validation set？

[2019-04-25 13:50:06,274] DEBUG: b'tIME' 41 7 (unknown)
[2019-04-25 13:50:06,274] DEBUG: STREAM b'IDAT' 60 8192
[2019-04-25 13:50:06,493] DEBUG: STREAM b'IHDR' 16 13
[2019-04-25 13:50:06,493] DEBUG: STREAM b'IDAT' 41 65536
[2019-04-25 13:50:06,553] DEBUG: STREAM b'IHDR' 16 13
[2019-04-25 13:50:06,553] DEBUG: STREAM b'tIME' 41 7
[2019-04-25 13:50:06,553] DEBUG: b'tIME' 41 7 (unknown)
[2019-04-25 13:50:06,553] DEBUG: STREAM b'IDAT' 60 8192
[2019-04-25 13:50:06,615] DEBUG: STREAM b'IHDR' 16 13
[2019-04-25 13:50:06,615] DEBUG: STREAM b'tIME' 41 7
[2019-04-25 13:50:06,615] DEBUG: b'tIME' 41 7 (unknown)
[2019-04-25 13:50:06,615] DEBUG: STREAM b'IDAT' 60 8192
[2019-04-25 13:50:06,678] DEBUG: STREAM b'IHDR' 16 13
[2019-04-25 13:50:06,678] DEBUG: STREAM b'tIME' 41 7
[2019-04-25 13:50:06,678] DEBUG: b'tIME' 41 7 (unknown)
[2019-04-25 13:50:06,678] DEBUG: STREAM b'IDAT' 60 8192
[] Reading checkpoints...
INFO:tensorflow:Restoring parameters from checkpoint\coarse_112\coarse.model-1500
[2019-04-25 13:50:06,897] INFO: Restoring parameters from checkpoint\coarse_112\coarse.model-1500
[] Load SUCCESS
42.309471130371094
An exception has occurred, use %tb to see the full traceback.

SystemExit

The paper said,"Flipping and rotation are used to obtain 7 additional augmented versions of original training data.",but I not find the part in the code.

May I ask what is the file extension of coarse.model-50?

What are these images use for (WB Real, GC Real, CE Real)?
And how are the generated?

Can you propose the origin image ,because the picture is too small

原始代码给出的一组预处理后的图像用于训练，预处理的图像大小是112×112的，
我使用原始大小的尺寸的图片进行训练，结果报错：ValueError: setting an array element with a sequence.
我该怎样修改代码，使得可以对原始图像大小进行训练和测试？

we find the image 100x100, is that normal?

Hi All,

The model is not getting trained. Everytime an error is raised :Loading Failed.

Can you please help me to point out the error in the training file?

Is there any specific folder in which the training images needs to be kept. Please let me know its urgent

I try to run "Python main_test.py". It provides me this error. Could you upload your vgg.py file?
Thanks

Hi All,

I am getting error while training the data. The error says "Loading Failed".
In the current code of model_train, I am trying not to use VGG as background.

Can it be the reason for getting the error?

Please help me to resolve the issue.

when I run the train.py,the issue hapened.

UIQM and UCIQE
If convenient, can you share the code to how to calculate them

If convenient, can you share the metric code (UCIQE UICM) to calculate the figs.

For your convenience, we provide a set of training and testing data. You can find them by unziping "a set of training and testing data". However, the training data and testing data are diffrent from those used in our paper.

图片（RAW,ce,gc,wb）的名字与论文种的是对应的但是提供的是小尺寸的，为了方便跑起来代码

When I tried your work on my own computer, I got a value like 22.3476 as a PSNR value. I wonder if there is a problem with the metric code I have. Could you please share the metric code you are using?