♻️ Smart Waste Classification – Week 3 Deep Learning Model + Web App (Streamlit)

A Sustainability-Themed AI Project

📌 Overview This project is part of the Skill4Future AI/ML Internship (Sustainability Theme). In Week 3, the goal is to:

->Train a CNN model to classify Organic vs Recyclable waste ->Preprocess and load dataset from Kaggle ->Generate training accuracy graph ->Save the trained model ->Build a simple Streamlit Web App for prediction ->Upload everything to GitHub

📁 Dataset Dataset (Kaggle): 🔗 https://www.kaggle.com/datasets/techsash/waste-classification-data

The dataset contains 2 classes: O → Organic Waste R → Recyclable Waste Folder structure: DATASET/ │── TRAIN/ │ ├── O/ │ └── R/ │ └── TEST/ ├── O/ └── R/ 🧠 Model (CNN) A simple CNN model was trained: Input size: 128×128×3 Optimizer: Adam Loss: Categorical Crossentropy Epochs: 10 Classes: 2 Files saved as: waste_cnn_model.h5 accuracy_plot.png

📙 Notebook Training code is inside: 📄 Week3-train.ipynb

This notebook includes: -Dataset loading -Preprocessing -Model training -Accuracy plot -Model saving

🌐 Streamlit Web App File: app2.py This app allows users to upload an image and predicts: -Organic Waste -Recyclable Waste

pip install -r requirements.txt streamlit run app2.py

📂 Repository Structure WEEK3/ │── README.md │── Week3-train.ipynb │── app2.py │── accuracy_plot.png │── waste_cnn_model.h5 (LFS) │── requirements.txt

✍️ Author

S. Prathik GitHub: https://github.com/prathik-05