https://github.com/engichang1467/digit-recognition-nn

Build a basic neural network model with Tensorflow and Keras to train the program to recognize all the handwritten digit in the MNIST dataset
https://github.com/engichang1467/digit-recognition-nn

mnist mnist-handwriting-recognition neural-networks tensorflow

Last synced: 3 months ago
JSON representation

Build a basic neural network model with Tensorflow and Keras to train the program to recognize all the handwritten digit in the MNIST dataset

• Host: GitHub
• URL: https://github.com/engichang1467/digit-recognition-nn
• Owner: engichang1467
• Created: 2020-12-29T20:43:58.000Z (over 5 years ago)
• Default Branch: main
• Last Pushed: 2020-12-29T20:48:01.000Z (over 5 years ago)
• Last Synced: 2025-12-26T16:37:24.554Z (6 months ago)
• Topics: mnist, mnist-handwriting-recognition, neural-networks, tensorflow
• Language: Jupyter Notebook
• Homepage:
• Size: 61.5 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# Handwritten Digit Guesser Neural Network

- Build a basic neural network model with Tensorflow and Keras to train the program to recognize all the handwritten digit in the MNIST dataset

- Test if the prediction is correct on Jupyter Notebook

## Requirements

- Python 3.7

- Tensorflow

- Keras

- Numpy

- Matplotlib

- Jupyter

## Testing our Model

![](src/num7.png)

![](src/num5.png)

## Give It A Shot!!

- Make sure you have all the packages installed from ```requirements.txt```

```

pip install -m requirements.txt

```

- Build and train the model, and you will get a model directory ```numReader.model```, which will be use for testing later

```

python3 buildModel.py

```

- Use the Jupyter Notebook (```testModel.ipynb```) to test how accurate our model is