Cross-Scale-Non-Local-Attention icon indicating copy to clipboard operation
Cross-Scale-Non-Local-Attention copied to clipboard

About the Dataloader can't find __DataLoaderIter

Open ChunmengLiu1 opened this issue 4 years ago • 3 comments

Hi , thanks for the great work! when I run the code of the paper, I met the problem that "from torch.utils.data.dataloader import _DataLoaderIter". It's can't find _DataloaderIter. The environment is pytorch 1.1.0,torchvision 0.3.0,python 3.6,cuda 10.2,ubuntu 18.04,which I think the same as mentioned in the paper. I thought there was a problem with the pytorch version,so I changed it to 1.5.0, but I still coudn't fix it. Have you ever encountered such a problem and how to solve it ? thank you for your advice!

ChunmengLiu1 avatar Sep 21 '20 01:09 ChunmengLiu1

I had the same problem. Did you address it? Or can the author give the answer? Thank you for your reading.

Tomchenshi avatar May 10 '21 15:05 Tomchenshi

你好,我想问一下你的dir-data中的div2k的训练集是放在哪里的,我运行了程序会在dir-data下产生一个div2k/bin/div2k-train-hr文件同时报错File "main.py", line 23, in main loader = data.Data(args) File "/home/l88/usr/csnln/src/data/init.py", line 31, in init num_workers=args.n_threads, File "/home/l88/anaconda3/envs/csnln/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 176, in init sampler = RandomSampler(dataset) File "/home/l88/anaconda3/envs/csnln/lib/python3.6/site-packages/torch/utils/data/sampler.py", line 66, in init "value, but got num_samples={}".format(self.num_samples)) ValueError: num_samples should be a positive integer value, but got num_samples=0

Ralapfranco avatar Mar 28 '22 02:03 Ralapfranco

hello! i have a same quession.first,i set the dir_data to‘../src/model/DIV2K’.then run the main.py .it create DIV2K/bin/DIV2K_train_hr. At the same time,it tell me :ValueError: num_samples should be a positive integer value, but got num_samples=0 so what i will do? should i copy 800 images to ../../DIV2K/bin/DIV2K_train/?

Ralapfranco avatar Mar 28 '22 02:03 Ralapfranco