'NoneType' object is not subscriptable about grad-cam.pytorch

[yizt]

@Iron4dam 感谢您的反馈;在main.py中增加了如下代码;测试resnet34没有错误；不清楚您是怎么改动channel数的, 可否详细描述下，谢谢！

    elif net_name in ['resnet34']:
        net = models.resnet34(pretrained=pretrain)

[adamxyang]

@yizt 感谢回复！我的model是：

model = models.resnet34(pretrained=True)

kernel = model.conv1.weight
new_conv = nn.Conv2d(4, 64, kernel_size=7, stride=2, padding=3, bias=False)
with torch.no_grad():
    new_conv.weight[:,:] = torch.stack([torch.mean(kernel, 1)]*4, dim=1)
model.conv1 = new_conv

model.fc = torch.nn.Linear(512, 2)

input也是四个channel（类似RGBY），dimension是（1，4，2048，2048）

[yizt]

@Iron4dam 您，我测试报了如下错：

Traceback (most recent call last):
  File "main.py", line 211, in <module>
    main(arguments)
  File "main.py", line 172, in main
    mask = grad_cam(inputs, args.class_id)  # cam mask
  File "E:\pyspace\Grad-CAM.pytorch\interpretability\grad_cam.py", line 60, in __call__
    output = self.net(inputs)  # [1,num_classes]
  File "F:\Anaconda3-5.0.1\envs\pytorch\lib\site-packages\torch\nn\modules\module.py", line 547, in __call__
    result = self.forward(*input, **kwargs)
  File "F:\Anaconda3-5.0.1\envs\pytorch\lib\site-packages\torchvision\models\resnet.py", line 196, in forward
    x = self.conv1(x)
  File "F:\Anaconda3-5.0.1\envs\pytorch\lib\site-packages\torch\nn\modules\module.py", line 547, in __call__
    result = self.forward(*input, **kwargs)
  File "F:\Anaconda3-5.0.1\envs\pytorch\lib\site-packages\torch\nn\modules\conv.py", line 343, in forward
    return self.conv2d_forward(input, self.weight)
  File "F:\Anaconda3-5.0.1\envs\pytorch\lib\site-packages\torch\nn\modules\conv.py", line 340, in conv2d_forward
    self.padding, self.dilation, self.groups)
RuntimeError: Expected object of scalar type Double but got scalar type Float for argument #2 'weight'

[adamxyang]

@yizt 这个我好像没有遇到，不过我加了一个inputs.requires_grad=True就解决了之前的问题，谢谢！