grad-cam-pytorch
grad-cam-pytorch copied to clipboard
PyTorch re-implementation of Grad-CAM (+ vanilla/guided backpropagation, deconvnet, and occlusion sensitivity maps)
### The difference between my model and resnet50 is only the last 3 layers,shown as below: (avgpool): AdaptiveAvgPool2d(output_size=(1, 1)) (fc): Linear(in_features=2048, out_features=1000, bias=True) ) **(bottleneck): Sequential( (0): AdaptiveAvgPool2d(output_size=(1, 1)) (1):...
Hello! Have you tested this code with EfficientNet (https://github.com/lukemelas/EfficientNet-PyTorch)? I tried to do that, but I have very unrealistic visualizations with huge attention in one place, nearly identical for every...