This implements a slightly modified LeNet-5 [LeCun et al., 1998a] and achieves an accuracy of ~99% on the MNIST dataset.
Install all dependencies using the following command
$ pip install -r requirements.txt
Start the visdom
server for visualization
$ python -m visdom.server
Start the training procedure
$ python run.py
See epoch train loss live graph at http://localhost:8097
.
[1] Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner. "Gradient-based learning applied to document recognition." Proceedings of the IEEE, 86(11):2278-2324, November 1998.