/Digit-Recognition

Classifying handwritten single digits using Neural Network.

Primary LanguageJupyter NotebookMIT LicenseMIT

Digit-Recognition

Classifying handwritten single digits using Neural Network.

About Dataset

The MNIST Dataset contains 60,000 28x28 grayscale images of the 10 digits in the training set, along with a test set of 10,000 images.

Custom Input

To test the model against custom input, add (.png) images in the Images folder and replace the path in the variable 'input_path'. Three sample images are provided in the Images folder

Colab Notebook link

https://colab.research.google.com/drive/1yUjhD670vi1kyHGj2cYLjjaQ71TKLTUv?usp=sharing