Disaster-Response-Pipelines

📌Demo

📌Objective

The objective of the project is to create a web application which will classify disaster messages.

📁 Data/File Description

• disaster_categories.csv - Contains the id, message that was sent and genre.
• disaster_messages.csv - Contains the id and the categories (related, offer, medical assistance..) the message belonged to.
• DisasterResponse.db - Database contain cleaned data
• process_data.py - Used for data cleaning and pre-processing
• train_classifier.py - Used to train the model
• classifier.pkl - Trained model
• run.py - To run Web App

🔑Prerequisites

All the required libraries are included in the file requirements.txt

⚠️TechStack/framework used

• Machine Learning
• Natural Language Processing
• Random Forest
• SQLite
• ETL

🚀Installation

1. Go to Project's root directory
2. Install required libraries.

pip install -r requirements.txt

3. To run ETL Pipeline that clean and store data

python data/process_data.py data/disaster_messages.csv data/disaster_categories.csv data/DisasterResponse.db

4. To run ML pipeline that trains the classifier model

python model/train_classifier.py data/DisasterResponse.db model/classifier.pkl

5. You can download the pre-trained model here

❤️Owner

Made With ❤️ by Sidharth kumar mohanty

😖Troubleshoot

Any issues??? Feel free to ask.Linkedin

If you find this repo useful,don't forget to give a ⭐