This repository contains machine learning models of Single Digit Recognition, designed to be deployed using ONNX and utilized in a Streamlit-based web application. The app provides an interactive interface for performing this task using neural network architectures. Check here to see other ML tasks.
For more information about the training process, please check the sn-dgt-recognition.ipynb file in the training folder.
If you encounter message This app has gone to sleep due to inactivity, click Yes, get this app back up! button to wake the app back up.
If the demo page is not working, you can fork or clone this repository and run the application locally by following these steps:
-
Clone the repository:
git clone https://github.com/verneylmavt/st-sn-dgt-recognition.git cd st-sn-dgt-recognition -
Install the required dependencies:
pip install -r requirements.txt
-
Run the Streamlit app:
streamlit run app.py
I acknowledge the use of the MNIST dataset provided by Yann LeCun, Corinna Cortes, and Christopher J.C. Burges. This dataset has been instrumental in conducting the research and developing this project.
- Dataset Name: MNIST (Modified National Institute of Standards and Technology) database
- Source: http://yann.lecun.com/exdb/mnist/
- Description: This dataset contains 70,000 grayscale images of handwritten digits (0-9), each 28x28 pixels in size. It is commonly used for training and testing in the field of machine learning and has become a benchmark for evaluating image processing systems.
I deeply appreciate the efforts of Yann LeCun, Corinna Cortes, and Christopher J.C. Burges in making this dataset available.