PiCANet-Implementation
PiCANet-Implementation copied to clipboard
Ugness
Pytorch Implementation of PiCANet: Learning Pixel-wise Contextual Attention for Saliency Detection
Why do I have this problem during training? Epoch : 0 0%|1 | 11/10554 [00:07
数据集问题
你好， 感谢你所做的贡献。 请问我在用DUTS-TR数据集训练网络的时候，我把数据集按照说明需要的方式：images/a.jpg masks/a.jpg，这样放置了，但为什么还是报错‘size of input tensor and input format are different. tensor shape: (1, 1, 28, 28), input_format: CHW’，请问能否解答一下，谢谢
hi, I found an error in your code. in the training part, you wrote 'if i + epo * len(dataloader) % decay_step == 0 and i != 0:' by doing...
I didn't change the network, but the loss didn't fall. I thought the lr was too big or small ,but even I set it bigger or smaller, the loss curve...
Please follow the steps in the Execution Guideline in README.md. https://github.com/Ugness/PiCANet-Implementation#execution-guide You can put the files anywhere, but you should give the directory as command-line options. If you meet any...
Hello，can you tell me how can I use the Pretrained Model and dataset to make the code work please? Where should I put the downloaded files to make it work?
I find it very time consuming to use F.unfold() function. Do you have the same problem？
A ERROR
new ERROR，WHY？？？ Downloading: "https://download.pytorch.org/models/vgg16-397923af.pth" to C:\Users\Administrator/.torch\models\vgg16-397923af.pth 553433881it [20:34, 448334.95it/s] Epoch : 0 0%| | 0/24 [00:00
There is still have the memory leak problem when running train.py downloaded from https://github.com/Ugness/PiCANet-Implementation.
