Air Quality Classification Web App

This web app is a demonstration project that classifies the air quality of a given location based on various air quality metrics like PM2.5, PM10, NO2, SO2, CO, and others. The app is built using Flask and aims to showcase an end-to-end machine learning pipeline, including data ingestion, transformation, model training, and prediction.

Table of Contents

• Project Overview
• Technologies Used
• Setup and Instalaltion
• Usage
• File Descriptions
• Model Training and Evaluation
• License

Project Overview

This web application allows users to input air quality data (such as temperature, humidity, PM2.5, PM10, NO2, SO2, and CO) and receive a classification of the air quality. The project involves the following key steps:

1. Data Ingestion: Raw air quality data is ingested, cleaned, and split into training and test datasets.
2. Data Transformation: The data is preprocessed, including missing value imputation and scaling.
3. Model Training: Various machine learning models are trained to predict air quality based on the features.
4. Prediction: The trained model predicts air quality based on the user-provided input.

Technologies Used

• Flask: A lightweight Python web framework for building web applications.
• Scikit-learn: For machine learning tasks, including data preprocessing, model training, and evaluation.
• Pandas: For data manipulation and analysis.
• NumPy: For numerical computations.
• Matplotlib and Seaborn: For data visualization.

Setup and Installation

1. Clone the repository

git clone https://github.com/rahul5021/air-quality-classification.git
cd air-quality-classification

2. Install dependencies

pip install -r requirements.txt

Running the Application

1. Run the Flask application

python app.py

The app will be available at http://127.0.0.1:5000/.

Usage

• Open the app in your browser.
• Select the location.
• Click on the predict button to get the air quality classification.

File Descriptions

• app.py: The main application file that runs the Flask server.
• data_ingestion.py: Responsible for loading and splitting the dataset into training and test sets.
• data_transfomation.py: Performs data preprocessing, including scaling and encoding.
• model_training.py: Contains the logic for training different machine learning models and selecting the best one based on accuracy.
• requirements.txt: Contains all the Python libraries required to run the project.
• setup.py: Script to install the project as a package.
• logger.py: Custom logger for logging events and errors.
• utils.py: Utility functions for data manipulation and evaluating model.
• exception.py: Custom exceptions for handling errors.

Model Training and Evaluation

The project uses the following models for classification:

• Random Forest Classifier
• Decision Tree

The models are evaluated based on accuracy, and the best-performing model is selected for deployment. The performance of the model is tested using a holdout test set, and accuracy metrics are logged.

License

This project is licensed under the MIT License - see the LICENSE file for details.