ERROR with checkpoint

[13135675192]

Hi! I'm learning the RDN. Luckly,I find your works.thanks! but I meet some troubles,when I run it ,The problem is following :OSError: Unable to open file (unable to open file: name = './checkpoint\train.h5', errno = 2, error message = 'No such file or directory', flags = 0, o_flags = 0).and I have load the DIV2K ! some idea.Thank you!

[hengchuan]

@13135675192 Hi, did you run the code on windows? I have modified the code, but not tested it on windows yet. You could pull the latest commit and try it.

[13135675192]

Thanks ,I run it on windows,It is ok.now ,I can use the DIV2K/DIV2K_train_LR_bicubic/X3 to train it,and set the "matlab_bicubic"is False?

[13135675192]

Excuse me, I tested it directly. How did I find that the image in the result folder didn't turn into HD, but it was even worse?

[hengchuan]

@13135675192 Did you test the pretrained model? It needs to be tested with --matlab_bicubic=True.

[13135675192]

first putting matlab_bicubic=True, then training the data, and finally testing it?

[13135675192]

Thanks ,I know it.

[13135675192]

Sorry, I bothered you many times, I want to retrain it, I need “--matlab_bicubic=True”？？

[hengchuan]

Never mind, “--matlab_bicubic=True” is not necessary but should be same when training and testing.

[13135675192]

Excuse me, I don't quite understand what you mean, then how do I need to set it up when I train? Because I use the DIV2k_HR dataset

[hengchuan]

You could run python main.py or python main.py --matlab_bicubic=True. While from my observation, setting "--matlab_bicubic=True" leads to better performance.

[13135675192]

Thank you for your patience and thank you for your help.