RuntimeError: CuDNN error: CUDNN_STATUS_SUCCESS #1
Description
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