This repository contains ETL and NLP pipelines as well as a Flask app that I have created as a part of Udacity's Data Science Nanodegree.

Installation
Project Motivation
File Descriptions
Results
Licensing & Acknowledgements

The following libraries have been installed alongside a standard Anaconda distribution of Python:

pandas
numpy
nltk
sqlalchemy
sklearn
sys
re
flask
langdetect
iso639
plotly
json

The code should run with no issues using Python 3.6.7.

Make sure that all your packages are up-to-date if you want to use my code!

This code has been written to complete an assignment which forms a part of Udacity's Data Science Nanodegree.

The aim of the project was threefold:

• to build an ETL pipeline - combine text messages related to disasters with their labels/categories
• to create an NLP pipeline - clean the text, extract features, perform classification, evaluate the model and save it
• to prepare a Flask app - visualise the characteristics of the training data; allow any user to type a message and run the model to check how it would be classified

Please note that the repository DOES NOT contain the pickled model needed to make predictions in the Flask app! It has not been added due to its large size and GitHub's restrictions.

Apart from the README file, there is a number of files in this repository:

- app
| - template
| |- master.html  # main page of web app
| |- go.html  # classification result page of web app
|- run.py  # Flask file that runs app

- data
|- disaster_categories.csv  # data to process 
|- disaster_messages.csv  # data to process
|- process_data.py # ETL pipeline
|- discover_languages.py # script to detect languages 
|- InsertDatabaseName.db # database where tables are stored 

- models
|- train_classifier.py # NLP pipeline

- README.md

To prepare all files needed by the app, run the commands in the following order:

# In the "data" folder:
python process_data.py disaster_messages.csv disaster_categories.csv DisasterResponse.db
python discover_languages.py

# In the "models" folder:
python train_classifier.py ../data/DisasterResponse.db classifier.pkl

Your data is now ready and your pickled model has been created!

Finally, you can run the app!

Feel free to use the code here as you would like!