https://github.com/daochenzha/cnn-for-mnist

A simple CNN that achieves 99.9% testing accuracy on MNIST
https://github.com/daochenzha/cnn-for-mnist

Last synced: over 1 year ago
JSON representation

A simple CNN that achieves 99.9% testing accuracy on MNIST

• Host: GitHub
• URL: https://github.com/daochenzha/cnn-for-mnist
• Owner: daochenzha
• Created: 2020-10-21T23:11:26.000Z (almost 6 years ago)
• Default Branch: main
• Last Pushed: 2020-10-21T23:33:43.000Z (almost 6 years ago)
• Last Synced: 2025-02-09T06:44:05.968Z (over 1 year ago)
• Language: Python
• Homepage:
• Size: 2.93 KB
• Stars: 0
• Watchers: 3
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# CNN-for-MNIST

A very simple CNN that achieves 99.9% testing accuracy on MNIST

## Installation

With `python3` and `pip3` installed, run

```

pip3 install -r requirements.txt

```

## Run the Code

```

python3 cnn_mnist.py

```

## Results

The means and stardard deviations across 5 experiments on validaiton and testing sets are as below.

| Validation Set      | Testing Set         |

| --------------------| --------------------|

| 0.9923 +- 0.0006    | 0.9990 +- 0.0002    |