Disaster Response Pipeline Project

Summary

This project does 3 things.

1. Process messages and categories into a SQLLite database.
2. Use ML to train a classifier for the messages. This will attempt to classify a message into up to 36 different categories.
3. Host a webapp that gives summary stats about the training data and allows interactive message classification.

Instructions

1. Run the following commands in the project's root directory to set up your database and model.

   • To run ETL pipeline that cleans data and stores in database python ./data/process_data.py ./data/disaster_messages.csv ./data/disaster_categories.csv ./data/DisasterResponse.db
   • To run ML pipeline that trains classifier and saves python ./models/train_classifier.py ./data/DisasterResponse.db ./models/classifier.pkl

2. Run the following command in the main directory to run your web app. python ./app/run.py

3. Go to http://localhost:3001/

Libaries

• python 3.6.6
• plotly 3.3
• pandas 0.23.4
• sqlalchemy 1.2.12
• nltk 3.3.0
• py-xgboost 0.72
• scikit-learn 0.20.0
• jsonschema 2.6.0
• flask 1.0.2

Files

• app
  • /templates HTML templates for flask
  • run.py Starts the flask webpage
• data
  • disaster_categories.csv raw category data
  • disaster_messages.csv raw message data
  • DisasterResponse.db sqllite: cleaned message/category data stored in messages table
  • process_data.py Run with two raw files to generate the database (cleaned data)
• models
  • classifier.pkl Pre-trained classifier, stored as a gzip pickle file
  • train_classifier.py Run with database file to generate a trained model for use in the webpage