how to test？
cheun726 opened this issue · 11 comments
I have got the pre-trained models， and i followed the test commend in demo.sh. But i got the following answer。
D:\softwarezijianzhuangde\anaconda\envs\pytorch-1.9\python.exe "D:/data/experiments code/code/1/Non-Local-Sparse-Attention/src/main.py" --dir_data ../benchmarkdata/benchmark/benchmark --model NLSN --chunk_size 144 --data_test Set5+Set14+B100+Urban100 --n_hashes 4 --chop --save_results --rgb_range 1 --data_range 801-900 --scale 2 --n_feats 256 --n_resblocks 32 --res_scale 0.1 --pre_train ../experiment/test/model/model_x2.pt --test_only
Making model...
Loading model from ../experiment/test/model/model_x2.pt
Total params: 41.80M
Evaluation:
[Set5 x2] PSNR: nan (Best: nan @epoch 1)
[Set14 x2] PSNR: nan (Best: nan @epoch 1)
[B100 x2] PSNR: nan (Best: nan @epoch 1)
[Urban100 x2] PSNR: nan (Best: nan @epoch 1)
Forward: 0.00s
Saving...
Total: 0.00s
0it [00:00, ?it/s]
0it [00:00, ?it/s]
0it [00:00, ?it/s]
0it [00:00, ?it/s]
Process finished with exit code 0
please help me.
Hi, please set the dir_data to the directory that contains the benchmark folder.
i have done so.i set it up like this
parser.add_argument('--dir_data', type=str, default='D:\data\experiments code\code\1\Non-Local-Sparse-Attention\benchmarkdata\benchmark\benchmark',
help='dataset directory')
Hi, the directory should contain the folder benchmark not the folder like Set5. In your case, it may be 'D:\data\experiments code\code\1\Non-Local-Sparse-Attention\benchmarkdata\benchmark\’.
My operating system is Windows 10.Do I need to change demo.sh?I changed it to look like this.Is that right
--dir_data ../benchmark --model NLSN --chunk_size 144 --data_test Set5+Set14+B100+Urban100 --n_hashes 4 --chop --save_results --rgb_range 1 --data_range 801-900 --scale 2 --n_feats 256 --n_resblocks 32 --res_scale 0.1 --pre_train ../experiment/test/model/model_x2.pt --test_only
I changed it as you said.And that's what happened
D:\softwarezijianzhuangde\anaconda\envs\pytorch-1.9\python.exe "D:/data/experiments code/code/1/Non-Local-Sparse-Attention/src/main.py" --dir_data ../benchmark --model NLSN --chunk_size 144 --data_test Set5+Set14+B100+Urban100 --n_hashes 4 --chop --save_results --rgb_range 1 --data_range 801-900 --scale 2 --n_feats 256 --n_resblocks 32 --res_scale 0.1 --pre_train ../experiment/test/model/model_x2.pt --test_only
Making model...
Loading model from ../experiment/test/model/model_x2.pt
0it [00:00, ?it/s]
0it [00:00, ?it/s]
0it [00:00, ?it/s]
0it [00:00, ?it/s]
Total params: 41.80M
Evaluation:
[Set5 x2] PSNR: nan (Best: nan @epoch 1)
[Set14 x2] PSNR: nan (Best: nan @epoch 1)
[B100 x2] PSNR: nan (Best: nan @epoch 1)
[Urban100 x2] PSNR: nan (Best: nan @epoch 1)
Forward: 0.00s
Saving...
Total: 0.00s
Process finished with exit code 0
The problem is the datasets are not correctly loaded. After you extracting the datasets, the structure should like this,
—/somepath/somefolder
——benchmark
———Set5
———Set14
———B100
———Urban100
And set the dir_data as /somepath/somefolder
It does not work.
In option.py,i set the dir_data like this
parser.add_argument('--dir_data', type=str, default='D:\data\experiments code\code\1\Non-Local-Sparse-Attention\benchmarkdata\benchmark',
help='dataset directory')
In demo.sh,i set it like this
--dir_data ../D:/data/experiments code/code/1/Non-Local-Sparse-Attention/benchmarkdata --model NLSN --chunk_size 144 --data_test Set5+Set14+B100+Urban100 --n_hashes 4 --chop --save_results --rgb_range 1 --data_range 801-900 --scale 2 --n_feats 256 --n_resblocks 32 --res_scale 0.1 --pre_train ../experiment/test/model/model_x2.pt --test_only
I don't know the exact reasons.please help me.Thanks
Hi, after extracting the zip file, you should get a folder with name “benchmark”. What is the parent folder that contains the benchmark folder? Please set the dir_data to a path pointing to the parent folder in the demo.sh.
In addition, please make sure the grammar for windows is correct.
A new problem has arisen。
D:\softwarezijianzhuangde\anaconda\envs\pytorch-1.9\python.exe "D:/data/experiments code/code/1/Non-Local-Sparse-Attention/src/main.py" --dir_data ../benchmarkdata --model NLSN --chunk_size 144 --data_test Set5+Set14+B100+Urban100 --n_hashes 4 --chop --save_results --rgb_range 1 --data_range 801-900 --scale 2 --n_feats 256 --n_resblocks 32 --res_scale 0.1 --pre_train ../experiment/test/model/model_x2.pt --test_only
Making model...
Loading model from ../experiment/test/model/model_x2.pt
Total params: 41.80M
Evaluation:
0%| | 0/5 [00:03<?, ?it/s]
Traceback (most recent call last):
File "D:/data/experiments code/code/1/Non-Local-Sparse-Attention/src/main.py", line 35, in
main()
File "D:/data/experiments code/code/1/Non-Local-Sparse-Attention/src/main.py", line 28, in main
while not t.terminate():
File "D:\data\experiments code\code\1\Non-Local-Sparse-Attention\src\trainer.py", line 141, in terminate
self.test()
File "D:\data\experiments code\code\1\Non-Local-Sparse-Attention\src\trainer.py", line 102, in test
self.ckp.save_results(d, filename[0], save_list, scale)
File "D:\data\experiments code\code\1\Non-Local-Sparse-Attention\src\utility.py", line 159, in save_results
self.queue.put(('{}{}.png'.format(filename, p), tensor_cpu))
AttributeError: 'checkpoint' object has no attribute 'queue'
Process finished with exit code 1
Hi, I am not sure about this problem. It may be because torch version and windows platform. The code is tested on Linux.
This code is based on EDSR and you may want to look this issue for reference sanghyun-son/EDSR-PyTorch#91.
thank you very much