xl-tang3 / UAUDeblur

(CVPR 23) Uncertainty-Aware Unsupervised Image Deblurring with Deep Residual Prior
57 stars 4 forks source link

RuntimeError: The size of tensor a (938) must match the size of tensor b (470) at non-singleton dimension 3 #2

Closed lvxiaoqian closed 1 year ago

lvxiaoqian commented 1 year ago

when running the UaUDeblur_demo.py, I meet an error as follows: line 150, in closure self.img_blurred_torch) + self.Lambda1 self.sploss(dct_2d(self.s)) + self.Lambda2 self.sploss(self.w)+0.05*self.tv(self.img_out_torch) RuntimeError: The size of tensor a (938) must match the size of tensor b (470) at non-singleton dimension 3

How to solve this problem?

lvxiaoqian commented 1 year ago

and what the version of the pytorch?

xl-tang3 commented 1 year ago

It's PyTorch 1.10.1 and personally this error does not occur in my environment. ---- Replied Message ---- | From | @.> | | Date | 07/14/2023 11:05 | | To | xl-tang01/UAUDeblur @.> | | Cc | Subscribed @.***> | | Subject | Re: [xl-tang01/UAUDeblur] TypeError: can't convert np.ndarray of type numpy.complex128. (Issue #2) |

and what the version of the pytorch?

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you are subscribed to this thread.Message ID: @.***>

lvxiaoqian commented 1 year ago

Thanks! I've solved the problem. How to show experimental results for color images?