search

fperazzi / proSR

GNU General Public License v3.0
27 stars 10 forks source link

RuntimeError: CuDNN error: CUDNN_STATUS_SUCCESS #1

Open Goru1890 opened 3 years ago

Goru1890 commented 3 years ago

I'm trying to test your project just for fun, every time I try to run it I get a runtime error. I'm not sure if it uses my GPU.

python test.py -i C:\Users\Usuario\Pictures\marcos\64039.jpg --checkpoint C:\Users\Usuario\Documents\IAPOR\proSRGAN\proSRGAN\proSRGAN_x4.pth --scale 4 -o C:\Users\Usuario\Pictures\marcos\new

←[0;97m2021-08-29 19:18:03: phase: phase.TEST←[0;97m←[0m ←[0;97m2021-08-29 19:18:03: checkpoint: proSRGAN_x4.pth←[0;97m←[0m {'D': {'act_params': {'negative_slope': 0.2}, 'act_type': 'LRELU', 'input_residual': True, 'ndf': 64, 'scale_overhead': True, 'update_freq': 2, 'use_lsgan': True, 'warmup_epochs': 0, 'which_epoch': 'latest', 'which_model_netD': 'srgan'}, 'G': {'bn_size': 4, 'growth_rate': 40, 'level_compression': -1, 'level_config': [[8, 8, 8, 8, 8, 8, 8, 8, 8], [8, 8, 8], [8]], 'max_num_feature': 312, 'max_scale': 8, 'num_init_features': 160, 'ps_woReLU': False, 'res_factor': 0.2, 'residual_denseblock': True, 'vgg': [2, 4], 'vgg_mean_pool': True}, 'data': {'input_size': [48, 36, 24], 'scale': [2, 4, 8]}, 'test': {'dataset': {'downscale': False, 'path': {'source': '', 'target': 'data/datasets/Set14'}, 'stddev': [0.0039215, 0.0039215, 0.0039215]}, 'fast_validation': -1}, 'train': {'D_lr': 0.0001, 'batch_size': 16, 'dataset': {'downscale': False, 'mean': [0.4488, 0.4371, 0.404], 'path': {'source': '', 'target': 'data/datasets/DIV2K/DIV2K_train_HR'}, 'stddev': [0.0039215, 0.0039215, 0.0039215]}, 'epochs': 500, 'gan_loss_weight': 1, 'growing_steps': [0.12, 0.25, 0.45, 0.6, 1.0], 'io': {'eval_epoch_freq': 10, 'print_errors_freq': 100, 'save_model_freq': 20}, 'l1_loss_weight': 0, 'lr': 0.0001, 'lr_decay': 0.5, 'lr_schedule_patience': 30, 'smallest_lr': 1e-05, 'vgg_loss_weight': [0.5, 2]}} ←[0;97m2021-08-29 19:24:35: Saving images in: C:\Users\Usuario\Pictures\marcos\new←[0;97m←[0m Traceback (most recent call last): File "test.py", line 156, in output = model(input,args.scale).cpu() + data['bicubic'] File "C:\Users\Usuario\anaconda3\envs\proSR\lib\site-packages\torch\nn\modules\module.py", line 477, in call result = self.forward(*input, kwargs) File "C:\Users\Usuario\Documents\IAPOR\proSR\lib\prosr\models\generators.py", line 121, in forward feats = self.get_init_conv(log2(upscale_factor))(x) File "C:\Users\Usuario\anaconda3\envs\proSR\lib\site-packages\torch\nn\modules\module.py", line 477, in call result = self.forward(*input, *kwargs) File "C:\Users\Usuario\Documents\IAPOR\proSR\lib\prosr\models\layers.py", line 64, in forward return self.conv(x) File "C:\Users\Usuario\anaconda3\envs\proSR\lib\site-packages\torch\nn\modules\module.py", line 477, in call result = self.forward(input, kwargs) File "C:\Users\Usuario\anaconda3\envs\proSR\lib\site-packages\torch\nn\modules\container.py", line 91, in forward input = module(input) File "C:\Users\Usuario\anaconda3\envs\proSR\lib\site-packages\torch\nn\modules\module.py", line 477, in call result = self.forward(*input, **kwargs) File "C:\Users\Usuario\anaconda3\envs\proSR\lib\site-packages\torch\nn\modules\conv.py", line 301, in forward self.padding, self.dilation, self.groups) RuntimeError: CuDNN error: CUDNN_STATUS_SUCCESS

SamHSlva commented 2 years ago

Were you able to fix it?

ayub567 commented 1 year ago

Hello dear

i am facing this issue

Traceback (most recent call last): File "test.py", line 3, in from prosr import Phase File "/content/proSR/prosr/init.py", line 2, in from . import metrics File "/content/proSR/prosr/metrics.py", line 4, in from skimage.measure import compare_psnr, compare_ssim ImportError: cannot import name 'compare_psnr' from 'skimage.measure' (/usr/local/lib/python3.8/dist-packages/skimage/measure/init.py)

please replay