Disaster Response ETL/ML Pipeline Project

ETL and Machine learning pipeline using python The ETL/ML project is part of Data Science Nanodegree Program by Udacity in collaboration with Figure Eight.

The Project is subdivided into:

1. Preprocessing of data and ETL Pipeline (including data storage into a sql database)
2. Building trained model through machine Learning Pipeline and GridSearchCV to tune the model hyperparameters
3. Model deployment(Web app) with flask app.

Dependencies

• Python 3+
• NumPy==1.12.1
• Pandas==2.0.15
• Sciki-Learn 0.23+
• NLTK (Natural Language Process Libraries)3.2.5+
• SQLAlchemy 2.3+
• Flask 0.12+
• Plotly==2.0.15
• gunicorn 19.9+

Instruction on how to run the program while in the main project directory (ETL_ML_Pipleine_project)

- To run the ETL pipeline (which extracts the data, transforms it and loads it into the database) type:

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

- To run the Machine learning (ML) pipeline(which perform feature extraction, trains, predict and saves the model, type:

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

- To run the app:

    Navigate to the app directory and type python run.py
    Then go to http://0.0.0.0:3001/