search

gladzhang / ART

PyTorch code for our ICLR 2023 paper "Accurate Image Restoration with Attention Retractable Transformer".
Apache License 2.0
148 stars 17 forks source link

CUDA error: no kernel image is available for execution on the device #4

Open KyriaAnnwyn opened 2 years ago

KyriaAnnwyn commented 2 years ago

Pytorch 1.12 + cuda 11.6 basicsr 1.4.2

Get this error:

2022-10-25 17:25:58,284 INFO: Loading ART model from experiments/pretrained_models/SR_ART_x2.pth, with param key: [params]. 2022-10-25 17:25:58,402 INFO: Model [ARTModel] is created. 2022-10-25 17:25:58,402 INFO: Testing example... Traceback (most recent call last): File "basicsr/test.py", line 51, in test_pipeline(root_path) File "basicsr/test.py", line 46, in test_pipeline model.validation(test_loader, current_iter=opt['name'], tb_logger=None, save_img=opt['val']['save_img']) File "/home/SuperResolution/ART/artenv/lib/python3.8/site-packages/basicsr/models/base_model.py", line 48, in validation self.nondist_validation(dataloader, current_iter, tb_logger, save_img) File "/home/SuperResolution/ART/artenv/lib/python3.8/site-packages/basicsr/models/sr_model.py", line 157, in nondist_validation self.test() File "/home/SuperResolution/ART/basicsr/models/art_model.py", line 23, in test img = F.pad(self.lq, (0, mod_pad_w, 0, mod_pad_h), 'reflect') RuntimeError: CUDA error: no kernel image is available for execution on the device

gladzhang commented 2 years ago

Please check if there exists a file in the testing folder datasets\example. If you still meet this error, check the corresponding yml file in the folder options\apply. In yml file, the dataroot_lq should be the same with the provided testing example files folder. Thanks.