Visualize Convolutional layer of Alexnet

vision
1

Hello All,
I am newbie into this field. I was trying to visualize the feature map/activation map of Alexnet.
I am working on cifar-10 dataset and was trying to visualize the kernels and feature map.

AlexNet(
(features): Sequential(
(0): Conv2d(3, 64, kernel_size=(11, 11), stride=(4, 4), padding=(2, 2))
(1): ReLU(inplace)
(2): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)
(3): Conv2d(64, 192, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))
(4): ReLU(inplace)
(5): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)
(6): Conv2d(192, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(7): ReLU(inplace)
(8): Conv2d(384, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(9): ReLU(inplace)
(10): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(11): ReLU(inplace)
(12): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)
)
(avgpool): AdaptiveAvgPool2d(output_size=(6, 6))
(classifier): Sequential(
(0): Dropout(p=0.5)
(1): Linear(in_features=9216, out_features=4096, bias=True)
(2): ReLU(inplace)
(3): Dropout(p=0.5)
(4): Linear(in_features=4096, out_features=2048, bias=True)
(5): ReLU(inplace)
(6): Linear(in_features=2048, out_features=1024, bias=True)
(7): ReLU(inplace)
(8): Linear(in_features=1024, out_features=10, bias=True)
)
)

I have few questions on visualization.

  1. How do I visualize feature map of Alexnet?
  2. Is there any supporting code for visualizing the convolutional layers of Alexnet?

Thanks in advance…

2

You could use forward hooks to get the activation maps and visualize them as described here. The topic also includes another post to visualize kernels.