Welcome to the README file for the Pokedex Android app built with Kotlin. This app integrates an Image Classification Machine Learning model as a backend, hosted on AWS using a Flask API. The app serves as a comprehensive guide to Pokemon, featuring a delightful user interface with support for both dark and light modes.
The Pokedex Android app is designed to provide users with an intuitive and captivating experience while exploring the world of Pokemon. It combines a state-of-the-art Machine Learning model with a user-friendly interface to offer features like Pokemon information browsing, image classification, and customization options.
- Browse and explore information about all Pokemon generations.
- Switch between dark and light themes for a comfortable experience.
- Integrated Search feature by name, type or number for ease of access
- Use the camera or upload images to identify Generation 1 Pokemon.
- Collect all Generation 1 Pokemon by scanning them with the integrated feature.
- Enjoy smooth transitions and animations for an appealing interface.
Here are some screenshots showcasing the app's user interface and features:
Light Mode
Dark Mode
- Machine Learning Model: Utilizing TensorFlow's Transfer Learning, incorporating EfficientNet and ResNet models for accurate image classification.
- AWS Services: Hosting the Machine Learning model on Amazon Web Services for scalability and reliability.
- Flask API: Serving as a bridge between the app and the Machine Learning model, processing image data and returning predicted Pokemon details.
- Language: Kotlin
- Architecture: MVVM (Model-View-ViewModel) pattern for clean separation of concerns.
- Network Calls: Retrofit library for efficient API calls and data loading.
- UI Components:
- RecyclerView: Smooth lists for displaying Pokemon information.
- Glide: Seamless image loading and caching.
- Material Components: Consistent UI elements and themes for an attractive look and feel.
- Asynchronous Operations: Coroutines for handling asynchronous tasks efficiently.
The Pokedex Android app follows the MVVM (Model-View-ViewModel) architectural pattern, facilitating a clean and modular structure for efficient development and maintenance. This pattern divides the app into three distinct components:
-
Model: Responsible for managing the data and business logic. This includes the Machine Learning model hosted on AWS, data structures, and repositories crucial for app functionality.
-
View: Focuses on presenting the user interface elements and handling user interactions. The app's screens, themes, and animations are managed within the View component.
-
ViewModel: Acts as a bridge between the Model and the View. It holds UI-related data and logic, facilitating communication between components while maintaining data integrity.
By adopting the MVVM architecture, the Pokedex app gains several benefits:
-
Separation of Concerns: Each component has a well-defined responsibility, leading to a structured and easily maintainable codebase.
-
Testability: Business logic within the ViewModel can be independently tested, enhancing overall reliability.
-
Scalability: Adding features or making changes becomes more straightforward, as components remain isolated.
-
Data Binding: Coupled with ViewModel, data binding streamlines UI-data interaction, resulting in a smoother user experience.
In essence, the MVVM architecture empowers the Pokedex app with a foundation that promotes code quality, maintenance efficiency, and a delightful user experience. It seamlessly integrates various technologies and libraries while laying the groundwork for future enhancements.
- Explore Pokemon information across generations.
- Identify Generation 1 Pokemon using the integrated ML model.
- Collect all Generation 1 Pokemon by scanning them.
- Customize the app's theme through the profile page.
This project is licensed under the Apache-2.0 license. See the Apache-2.0 license file for details.
Have questions or feedback? Feel free to reach out:
- LinkedIn: Niraj