WasteLink is a FastAPI-based machine learning service designed to classify waste into five categories. This tool helps automate waste sorting to improve recycling efficiency. This API was built as the backend ML service for a mobile waste management app. The model endpoint accepts image uploads and returns waste classification results in real time.
- Custom ML Model: Powered by a MobileNetV2 architecture fine-tuned on custom waste datasets.
- FastAPI Backend: High-performance asynchronous API for real-time predictions.
- Cloud Deployment: Hosted on Render for global accessibility.
main.py: The core FastAPI application logic.wastelink_fixed.h5: The trained TensorFlow model (H5 format for stability).requirements.txt: List of Python dependencies.
The model can identify the following:
- General
- Metal
- Organic
- Paper
- Plastic
- Clone the repository:
git clone [https://github.com/NjeriCodeCraft/validation.git]