LeNet-5
LeNet-5 is a pioneering convolutional neural network (CNN) architecture developed by Yann LeCun and his colleagues in 1998. It was designed for handwritten digit recognition and played a significant role in the development of deep learning.
Architecture:
- Input: 32x32 grayscale image
- C1: Convolutional layer with 6 filters of size 5x5, followed by a subsampling layer (average pooling) with a 2x2 filter and stride of 2.
- C3: Convolutional layer with 16 filters of size 5x5, followed by a subsampling layer (average pooling) with a 2x2 filter and stride of 2.
- F5: Fully connected layer with 120 neurons.
- F6: Fully connected layer with 84 neurons.
- Output: 10 neurons corresponding to the 10 digit classes (0-9).