Padding and stride stride in convolutional layers hands-on deep learning v2 pytorch

Hyperparameters for convolutional layers:

• Kernel convolution kernel size
• Padding
• Stride

1. Padding in the convolutional layer

2 Stride in the convolutional layer

3. Code implementation

3. Q&A

• 1. Hyperparameter kernel size, the filling is generally kernel size -1, and the stride generally depends on whether the speed of calculation needs to be optimized.
• 2. The variable length of the convolution kernel is generally chosen to be an odd number, which is convenient for calculation.
• 3. Generally, everyone directly uses the classic network structure, such as ResNet.
• 4. The essence of machine learning is a compression algorithm. Machine learning will always lose information, just like information theory.
• 5. The bottom layer uses a large Kernel, and the upper layer uses a small Kernel
• 6. Different convolutional layers can be understood as identifying multiple different textures of an image.
• 7. The Industrial Revolution is about replacing human beings with machines. Science is also a process from expensive to cheap.

reference

https://www.bilibili.com/video/BV1Th411U7UN?p=1