https://github.com/activatedgeek/lenet-5

PyTorch implementation of LeNet-5 with live visualization
https://github.com/activatedgeek/lenet-5

cnn convolutional-neural-networks deep-learning deep-neural-networks lenet5 machine-learning pytorch

Last synced: 1 day ago
JSON representation

PyTorch implementation of LeNet-5 with live visualization

• Host: GitHub
• URL: https://github.com/activatedgeek/lenet-5
• Owner: activatedgeek
• Created: 2017-11-12T04:25:33.000Z (about 7 years ago)
• Default Branch: master
• Last Pushed: 2023-01-27T01:11:49.000Z (almost 2 years ago)
• Last Synced: 2024-12-17T10:05:30.520Z (9 days ago)
• Topics: cnn, convolutional-neural-networks, deep-learning, deep-neural-networks, lenet5, machine-learning, pytorch
• Language: Python
• Homepage:
• Size: 10.7 KB
• Stars: 224
• Watchers: 4
• Forks: 94
• Open Issues: 7
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# LeNet-5

This implements a slightly modified LeNet-5 [LeCun et al., 1998a] and achieves an accuracy of ~99% on the [MNIST dataset](http://yann.lecun.com/exdb/mnist/).

![Epoch Train Loss visualization](https://i.imgur.com/h4h7CrF.gif)

## Setup

Install all dependencies using the following command

```

$ pip install -r requirements.txt

```

## Usage

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`](http://localhost:8097).

The trained model will be exported as ONNX to `lenet.onnx`. The `lenet.onnx` file can be viewed with [Neutron](https://www.electronjs.org/apps/netron)

## References

[[1](http://yann.lecun.com/exdb/publis/pdf/lecun-98.pdf)] 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.