hilab-git / ca-net Goto Github PK

Hi,
Thanks for sharing your code. Can I find your fetus model available online?

Should i train all 5 folders and compute average?

Traceback (most recent call last):
File "main.py", line 470, in
main(args)
File "main.py", line 305, in main
model = Test_Model[args.id](args, args.num_input, args.num_classes)
File "/home/SA/ljc/ca/CA-Net-master/Models/networks/network.py", line 28, in init
self.conv1 = conv_block(self.in_channels, filters[0])
File "/home/SA/ljc/ca/CA-Net-master/Models/layers/modules.py", line 21, in init
nn.Conv2d(ch_in, ch_out, kernel_size=3, stride=1, padding=1, bias=True),
File "/home/SA/anaconda3/envs/ljctorchpy3.6/lib/python3.6/site-packages/torch/nn/modules/conv.py", line 332, in init
False, _pair(0), groups, bias, padding_mode)
File "/home/SA/anaconda3/envs/ljctorchpy3.6/lib/python3.6/site-packages/torch/nn/modules/conv.py", line 22, in init
if in_channels % groups != 0:
TypeError: unsupported operand type(s) for %: 'Namespace' and 'int'
i have tried a lot versions pytorch,1.4.0/1.7.0/1.10.0, but don't work
do you konw how to fix it, thx!

Why does setting different batch_size affect the test results?

运行validation.py时，发生报错RuntimeError: The first supplied array does not contain any binary object，该怎么解决呢，谢谢！

In your code, drop_out=True

I think this line of code will affect the results when testing. Because you init the dropout layer in forward function, the model.eval() can not change the status of this layer.

You can test it with the following code

import torch
from torch import nn
import numpy as np

class Net(nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.out = nn.Dropout2d(0.5)

    def forward(self, x):
        out = nn.Dropout2d(0.5)(x)
        # out = self.out(x)
        return out

if __name__ == '__main__':
    model = Net()
    model.eval()
    input_npy = np.array([[1.0, 2.0], [3.0, 4.0]])
    input_tensor = torch.from_numpy(input_npy)
    output = model(input_tensor)
    print(output)

If I didn't understand your code correctly, sorry in advance

Should the bilinear interpolation（Image.BILINEAR） of label scaling be changed to nearest neighbor interpolation（Image.NEAREST）？

Dear author：
如何得到表1的结果呢，想学习一下

Run validation.Py file, cannot be generated the attention weight graph in the result folder.

Is it necessary to use these annotated codes and modify them?

In the paper, it's written that the images are randomly divided into three subsets for training, validating and testing, respectively.
While in this repo, authors perform 5-folds cross-validation.
So, which one is the true setting presented in the paper?

Hello, may I ask how to save the attention weight map in the middle step of the network. Thank you very much.