Real Time Weather Classification using Machine Learning

This repository contains a multi-page Streamlit web application that demonstrates a complete end-to-end machine learning workflow. The application handles everything from rate-limited API data collection and CSV storage to interactive data exploration, feature selection, model training, and live prediction.

The primary goal of this project is to build a classification model capable of predicting weather conditions (e.g., 'Rain', 'Clouds', 'Clear') based on various meteorological features collected from the OpenWeatherMap API.

---

✨ Key Features

• Rate-Limited Data Collection: Safely collects weather data for thousands of cities using a robust, rate-limited, and asynchronous approach to respect API limits.
• Interactive Data Exploration: A dedicated page to view the raw dataset, analyze column statistics, and visualize feature distributions with interactive Plotly charts.
• Dynamic Feature & Target Selection: Interactively select which columns to use as input features and which to set as the prediction target for the model.
• ML Model Training: Train classification models (like Random Forest and Logistic Regression) with a single click.
• Clear Model Evaluation: Instantly view model performance with clear, table-based metrics, including accuracy, precision, recall, a confusion matrix, and a full classification report.
• Manual & Live Prediction: Make predictions by manually entering feature values or by using real-time weather data fetched directly from the API for any city.

---

🛠️ Technology Stack

• Application Framework: Streamlit
• Data Manipulation: Pandas, NumPy
• Machine Learning: Scikit-learn
• Data Visualization: Plotly
• API Requests: Requests

---

📂 Project Structure

.
├── pages/
│   ├── 02_Data_Exploration.py
|   ├── 02_Data_Collection.py
│   ├── 03_Feature_Selection.py
│   ├── 04_Model_Training.py
│   ├── 05__Manual_Prediction.py
│   └── 06_Live_API_Prediction.py
├── Home.py                             # Main entry point of the app
├── README.md                           # You are here <──
├── current_city_list.json              # Default city list for randomize
├── data_utils.py                       # Helper functions for data loading & API calls
├── derived_cities_for_collection.csv   # Randomized city list
├── model_utils.py                      # Helper functions for preprocessing & model training
├── pexels-jplenio-1118873.jpg          # Photo assets
├── rate_limited_weather_data.csv       # Weather dataset
└── requirements.txt                    # Python package dependencies

---

▶️ Deployed Streamlit Link

(weather-classification-ml.streamlit.app)

---

🚀 Setup and Installation

Follow these steps to get the application running on your local machine.

1. Prerequisites

• Python 3.8 - 3.11
• An API Key from OpenWeatherMap (the free tier is sufficient).

2. Clone the Repository

git clone [https://github.com/your-username/your-repo-name.git](https://github.com/your-username/your-repo-name.git)
cd your-repo-name

3. Create a Virtual Environment

It is highly recommended to use a virtual environment to manage dependencies.

# For Windows
python -m venv venv
venv\Scripts\activate

# For macOS/Linux
python3 -m venv venv
source venv/bin/activate

4. Install Dependencies

Run this command to install all required libraries from the requirements.txt file.

pip install -r requirements.txt

▶️ How to Run the Application

With your virtual environment activated, run the following command from the project's root directory:

streamlit run Home.py

Your web browser should automatically open a new tab with the running application.

---

📋 How to Use the App

1. Data Collection:

   • Navigate to the Rate Limited Data Collection page (the home page).
   • You can load an existing weather data CSV file or start a new collection process.
   • For a new collection, enter your API key in the configuration section, set the number of cities to sample from the source file (derived_cities_for_collection.csv), and click "Prepare Randomized City List".
   • Once the city list is ready, click "Collect Weather Data" to begin.

2. Dataset Exploration:

   • Go to the Dataset Exploration page from the sidebar.
   • View the complete dataset, see detailed info on each column, and use the interactive charts to analyze feature distributions.

3. Feature Selection:

   • On the Feature Selection page, choose your prediction target (e.g., General Weather Category).
   • Select the features you believe will be useful for the prediction. Irrelevant features are automatically filtered out.

4. Model Training:

   • On the Model Training page, select a model from the dropdown menu.
   • Click the "Train Model" button to start the training and evaluation process.
   • The results, including performance metrics and a confusion matrix, will be displayed and will persist even if you navigate to other pages.

5. Prediction:

   • Use the Manual Prediction page to manually input feature values and get a prediction.
   • Use the Live API Prediction page to enter a city name and get a prediction based on real-time weather data.

---