Rainfall-Prediction-Using-Machine-Learning

This project predicts whether it will rain or not based on weather features like pressure, humidity, dew point, cloud cover, sunshine, wind direction, and wind speed. We use a Random Forest Classifier, a popular ML algorithm, trained on historical weather data. The model learns patterns and helps us forecast rain chances. This project is a web-based application that predicts whether it will rain today based on real-world weather input features like pressure, humidity, cloud cover, sunshine, etc. It uses a Random Forest Classifier trained on historical weather data and is deployed as an interactive Streamlit app.

---

Live : https://rainfall-prediction-using-machine-learning.streamlit.app/

📌 Problem Statement

Predicting rainfall is crucial in weather forecasting. It helps:

• Farmers plan irrigation and harvesting.
• People manage daily travel plans.
• Cities prepare for flood alerts or water conservation.

Traditional methods require large infrastructure, but machine learning can make accurate predictions from past weather patterns.

---

🎯 Objective

To build a machine learning model that:

• Takes weather features as input.
• Predicts whether it will rain or not.
• Provides results through a simple and user-friendly web app.

---

📊 Features of the App

✅ Predicts Rain or No Rain
✅ Easy-to-use UI built with Streamlit
✅ Takes 7 weather parameters as input
✅ Displays prediction with icons: ☀️ / 🌧️
✅ Trained using Random Forest Classifier
✅ Model saved and loaded using joblib

---

🧪 Input Features Used for Prediction

Feature	Description
pressure	Atmospheric pressure (hPa)
dewpoint	Dew point temperature (°C)
humidity	Relative humidity (%)
cloud	Cloud cover (%)
sunshine	Sunshine duration (hours)
winddirection	Wind direction in degrees (°)
windspeed	Wind speed in km/h

---

🧠 Machine Learning Model

• Algorithm Used: RandomForestClassifier (from Scikit-learn)
• Model Training Steps:
  • Data Preprocessing
  • Train-Test Split
  • Feature Selection
  • Model Tuning with GridSearchCV
• Evaluation Metrics:
  • Accuracy
  • Confusion Matrix
  • Classification Report
• Model Deployment:
  • Saved as .pkl file using joblib
  • Loaded into the Streamlit app for prediction

---

🛠️ Tech Stack

Tool/Library	Purpose
Python	Programming Language
Pandas & NumPy	Data handling and processing
Scikit-learn	Machine Learning model
Matplotlib & Seaborn	Data Visualization
Joblib	Model saving/loading
Streamlit	Web App frontend

---