This repository contains an Android application that demonstrates the implementation of a Convolutional Neural Network (CNN) for handwritten digit classification. The project serves educational purposes, showcasing how to train a neural network and integrate it into an Android application. PPP
The application is built using Jetpack Compose and implements a CNN model trained on the MNIST dataset. The model was trained using Google Colab, and the resulting TensorFlow Lite model is integrated into the Android application. The CNN model was trained using the MNIST dataset in Google Colab. You can find the training notebook here: Training Notebook
- Modern UI built with Jetpack Compose
- Real-time digit classification
- Integration of TensorFlow Lite model
- Clean architecture implementation
- Built using Jetpack Compose for modern UI development
- Uses TensorFlow Lite for model inference
The project follows a clean architecture approach with the following main components:
- UI Layer (Compose)
- Model Integration
- Clone the repository
- Open the project in Android Studio
- Build and run the application
The project uses the following main dependencies:
- Jetpack Compose
- TensorFlow Lite
- AndroidX Core
- Material Design Components
This project was inspired by various sources and learning materials:
- A Proposal for the Dartmouth Summer Research Project on Artificial Intelligence
- Levels of AGI for Operationalizing Progress on the Path to AGI
- Build a handwritten digit classifier app with TensorFlow Lite
- Google Machine Learning Crash Course
- APRENDE ¿Qué son las Redes Neuronales? de Dot CSV
- Introduction to Generative AI
- The Impact of Generative AI on Critical Thinking
- Ask a Techspert: What is generative AI?
- Con Jetpack Compose , lleva el desarrollo de interfaces al siguiente nivel
This project is open source and available under the MIT License.
Contributions are welcome! Please feel free to submit a Pull Request.