1 Fundamentals of convolution
1.1 Movement and step size
- How can I add zero if the input and output dimensions remain the same? (F: convolution kernel dimension, P:padding layer)
- Parameter calculation method
- conclusion
1.2 Functions of 1*1 convolution kernel (dimensionality reduction and parameter calculation)
1.3 pooling
1.4 Stanford Experiment website
1.5 Keras.js experimental website
2 AlexNet Network (2012 ImageNet champion, the first application of deep convolutional network to large-scale image classification
)
2.1 an overview of the
2.2 Deep network architecture