Closed cys4 closed 6 years ago
Hello.
x4 upsampler consists of two consecutive x2 upsampler.
In our original approach, we randomly initialize x4 upsampler even when we load the weights from scale 2 model.
If you use strict=False, however, the first part of x4 upsampler is initialized with pre-trained x2 upsampler, and the second part is initialized randomly.
Although this situation is not intended, I think your solution will not make problems.
I will check the code and make it executable without any error message.
Thank you for reporting this error.
Sanghyun.
Okay. I see. Thank you for clarifying!
First of all, thank you for sharing this great work.
I have a problem when training a x4 model using a pre-trained x2 model (../experiment/EDSR_baseline_x2/model/model_best.pt) as follows.
Although all above errors could be removed by adding 'strict=False' to the call statement of 'load_statedict' (line 15) in './code/model/__init_\.py' as follows, I'm not sure this is a right way to handle this situation.
Please let me know if I'm missing something important.