CineMatch 🎬

A modern Flutter-based movie recommendation application that leverages on-device machine learning (ONNX) to provide personalized, lightning-fast movie suggestions.

✨ Features

• On-Device Machine Learning: Utilizes onnxruntime with a quantized model and ml_linalg to generate recommendations directly on your device, ensuring privacy and instant results.
• Fast Local Storage: Uses Hive for caching movie data, managing the user's watchlist, and storing preferences completely offline.
• Robust State Management: Powered by Riverpod for a scalable, reactive, and testable architecture.
• Declarative Routing: Implemented using GoRouter for smooth, deep-linkable, and predictable navigation.
• Beautiful UI: Features an elegant modern dark theme built with Material Design and beautiful typography via Google Fonts.
• Seamless Networking: API integrations handled efficiently via Dio.

🛠️ Tech Stack

• Framework: Flutter (Dart ^3.11.5)
• State Management: Riverpod (flutter_riverpod, riverpod_annotation)
• Local Database: Hive
• Networking: Dio
• Machine Learning: ONNX Runtime & ml_linalg
• Routing: GoRouter

🚀 Getting Started

Prerequisites

• Flutter SDK ^3.11.5 or higher
• Dart SDK
• An active IDE (VS Code, Android Studio, or IntelliJ)

Installation

1. Clone the repository:

   git clone <repository_url>
   cd cinematch

2. Install dependencies:

   flutter pub get

3. Environment Setup: Create a .env file in the root directory to store your API keys and configuration variables:

   touch .env
   # Add your required environment variables to the .env file

4. Code Generation: Generate the required Hive adapters and Riverpod providers:

   dart run build_runner build -d

5. Run the app:

   flutter run

📁 Project Structure

• lib/core/ - Core configurations, theming, and constants.
• lib/data/ - Data layer including models, DTOs, data sources, and repositories.
• lib/features/ - Application features partitioned by domain, containing specific UI and controllers.
• lib/shared/ - Shared widgets, providers, and utilities used across multiple features.
• assets/ml/ - Contains the ONNX quantized model (model_quantized.onnx), tokenizer configuration, and movie embeddings required for the local ML engine.
• assets/images/ - Application image assets and launcher icons.

🤝 Contributing

Contributions, issues, and feature requests are always welcome!

📝 License

This project is licensed under the MIT License.