RuiyangJu / mnist-canvas

A web canvas that you can draw and see the MNIST classification result distribution.

Home Page:https://ruiyangju.github.io/mnist-canvas/

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Introduction

A web canvas that you can draw and see the MNIST classification result distribution.

Requirements

  • Python 3.6+
  • Pytorch 0.4.0+
  • Pandas 0.23.4+
  • NumPy 1.14.3+

Usage

Train the model:

  python3 main.py

Convert pth to onnx:

  python3 convert.py

optional arguments:

  --lr                default=1e-3    learning rate
  --epoch             default=30      number of epochs tp train for
  --trainBatchSize    default=128     training batch size
  --testBatchSize     default=128     test batch size
  --optimizer         default=adam    optimizer

About

A web canvas that you can draw and see the MNIST classification result distribution.

https://ruiyangju.github.io/mnist-canvas/

License:MIT License


Languages

Language:Python 41.8%Language:JavaScript 25.0%Language:HTML 17.9%Language:CSS 15.3%