search

csjliang / LPTN

Official implementation of the paper 'High-Resolution Photorealistic Image Translation in Real-Time: A Laplacian Pyramid Translation Network' in CVPR 2021
Apache License 2.0
413 stars 47 forks source link

recompute_scale_factor is not meaningful with an explicit size #30

Closed lkp520 closed 3 years ago

lkp520 commented 3 years ago

When I use dataroot_gt: datasets/FiveK/FiveK_480p/test/B dataroot_lq: datasets/FiveK/FiveK_480p/test/A to evaluation, I meet the problem:

Traceback (most recent call last): File "/data/likengpeng/LPTN/codes/test.py", line 56, in main() File "/data/likengpeng/LPTN/codes/test.py", line 48, in main model.validation( File "/data/likengpeng/LPTN/codes/models/base_model.py", line 47, in validation self.nondist_validation(dataloader, current_iter, tb_logger, File "/data/likengpeng/LPTN/codes/models/lptn_test_model.py", line 75, in nondist_validation self.test() File "/data/likengpeng/LPTN/codes/models/lptn_test_model.py", line 41, in test self.output = self.net_g(self.lq) File "/home/amax/anaconda3/envs/deoldify/lib/python3.9/site-packages/torch/nn/modules/module.py", line 727, in _call_impl result = self.forward(*input, **kwargs) File "/data/likengpeng/LPTN/codes/models/archs/LPTN_arch.py", line 186, in forward real_A_up = nn.functional.interpolate(pyr_A[-1], size=(pyr_A[-2].shape[2], pyr_A[-2].shape[3])) File "/home/amax/anaconda3/envs/deoldify/lib/python3.9/site-packages/torch/nn/functional.py", line 3110, in interpolate raise ValueError("recompute_scale_factor is not meaningful with an explicit size.") ValueError: recompute_scale_factor is not meaningful with an explicit size.

How can I solve it? Thank you!