This repository contains the code for a waste classification web application built with Streamlit. The application allows users to take a photo using a webcam connected to a Khadas VIM2 and classify the type of waste into 5 categories: cardboard, glass, metal, paper, and plastic.

To set up and run the web application, follow these steps:

• Python 3.7+
• Streamlit
• TensorFlow
• OpenCV
• PIL (Pillow)

1. Clone this repository:

git clone https://github.com/your_username/waste-classification-webapp.git

2. Navigate to the project directory:

cd waste-classification-webapp

3. Install the required packages using the requirements.txt file:

pip install -r requirements.txt

4. Place your trained models (model.h5, model.tflite, model_pruned.tflite, model_quantized.tflite, model_pruned_quantized.tflite) in the project directory.

1. Start the Streamlit web application:

streamlit run app.py

2. Open the web application in your browser using the URL provided in the terminal.

3. Choose the desired model from the dropdown menu.

4. Take a photo using the webcam connected to the Khadas VIM2.

5. The application will classify the waste in the image into one of the 5 categories: cardboard, glass, metal, paper, or plastic.

• OpenAI for the GPT model that helped in generating parts of this README.
• Streamlit for the awesome web app framework.
• TensorFlow for the machine learning library used in this project.
• OpenCV for the image processing library used in this project.