Data Science Deployed

An extensible end-to-end data science solution created using Flask, Celery, SQL ALchemy and Rabbit MQ , deployed on AMAZON EC2 and Microsoft server instances

Applications currently served:

1. Salary Prediction System

Configuration on local development environment

1. Check out the project and install dependencies from the requirements.txt

2. Register the app on google drive API (to upload the results) and save teh credentials in the app folder.

3. Edit the template files and save them without .template extension.

4. You can use any message broker and backend data base, edit the URI's accordingly.

5. Run authenticate.py to check for Google Drive connectivity.

6. Open a terminal and run upload_results.py

7. Open another terminal and start celery broker using teh command celery worker -A app.celery --pool=eventlet --loglevel=info.

8. Finally run the flask app.

Data set can be downloaded from https://drive.google.com/drive/folders/1mJcUpmtPWAe7NlM6UUqe8lkpmFGw_V08?usp=sharing.
This application can be extended to any number of use cases, just add the URL in application stack page and route it like salary app

Demo
https://github.com/sourabhparime/AWSFLASK/blob/master/DATA%20SCIENCE%20DEPLOYED.pdf

Improved analysis script to be uploaded soon

Datasets, problem statement and analysis script inspired from Kyle