UserWarning:Detected call of `lr_scheduler.step()` before `optimizer.step()`.

[navy63]

19-10-08 11:06:23.167 - INFO: Start training from epoch: 0, iter: 0 C:\Users\mayn\Anaconda3\envs\pytorch\lib\site-packages\torch\optim\lr_scheduler.py:82: UserWarning: Detected call of lr_scheduler.step() before optimizer.step(). In PyTorch 1.1.0 and later, you should call them in the opposite order: optimizer.step() before lr_scheduler.step(). Failure to do this will result in PyTorch skipping the first value of the learning rate schedule.See more details at https://pytorch.org/docs/stable/optim.html#how-to-adjust-learning-rate "https://pytorch.org/docs/stable/optim.html#how-to-adjust-learning-rate", UserWarning) 19-10-08 11:12:06.405 - INFO: <epoch: 0, iter: 100, lr:1.000e-04> l_pix: 1.0065e-01

[yuanjunchai]

Seemingly you are training SFTMD, have you changed train_SFTMD.yaml?

[navy63]

Yes, I am now training, single card 2080TI training, it seems that this takes 5 days, the training process can understand, do not understand is the test process, when testing, why we need to test the two process, and how the result is obtained. example: gpu_ids: [0] datasets: train: dataroot_GT: E:/Dataset/Flickr2K/Flickr2K_train/HR/x4 val: dataroot_GT: E:/Dataset/Set14_test/HR/x4

19-10-09 08:41:18.331 - INFO: <epoch:575, iter: 95,000> psnr: 26.535406 19-10-09 08:42:41.467 - INFO: <epoch:576, iter: 95,100, lr:1.355e-05> l_pix: 2.9932e-02 19-10-09 08:43:54.576 - INFO: <epoch:576, iter: 95,200, lr:1.347e-05> l_pix: 2.7815e-02 19-10-09 08:45:23.507 - INFO: <epoch:577, iter: 95,300, lr:1.338e-05> l_pix: 2.9500e-02 19-10-09 08:46:46.556 - INFO: <epoch:578, iter: 95,400, lr:1.330e-05> l_pix: 3.7441e-02

[yuanjunchai]

well, you could only test IKC or SFTMD(non-blind SR net), for more details please refer to the paper. Thank you!

[navy63]

OK，Thank you!

[yuanjunchai]

my pleasure.