LeNet-5
IMAGE
This demonstrator presents one of the first deep convolutional networks proposed by Yann Lecun in the article Y. Lecun, L. Bottou, Y. Bengio, P. Haffner, "Gradient-based learning applied to document recognition". Proceedings of the IEEE. 86 (11): 2278–2324, 1998.
LeNet is a serie of convolutional neural network architectures created by a research group of AT&T Bell Laboratories between 1988 and 1998, around Yann Lecun. They have been designed to read very small pictures in gray levels of handwritten numbers and letters and were used in automated teller machines to read cheques.
These networks needed, at the time they were conceived, many days to be trained. Nowadays, this takes a few minutes.
This demonstrator uses the TensorFlowJS technology.
Contributor
Olivier Lezoray