TigerStream

This Web-App is developed for the submission of Build a Web-App with TigerGraph using Streamlit & Graphistry Hackathon. ## Features - Integration of TigerGraph database for Movie Recommendation starter-kit. - Interactive Web application developed with Streamlit. - Graph visualizations using Graphistry and Pyviz. ## Installed GSQL queries - UserStatistics - Input : user id - Output : average rating, no. of movies rated, average rating, timestamp of rated movie - SimilarPeople - Input : user id, no. of similar user - Output : similar movie list and similar movie count of N user's with respect to given user id - RecommendMovies - Input : user id, no. of similar user, no. of movies to recommend - Output : list of movies along with their genre and other calculated matrices ## Streamlit features in action - Multiple Page navigation - Components for displaying saved html files and iframe - Beta columns for side-by-side elements - Plotly integration for interactive visualization - DataFrame Pagination - Tooltip on Hover ## System Configurations - This Project is developed and Tested with below mentioned system configurations. ``` - Operating System : Ubuntu 18.04 64 bit - RAM : 4 GB - Python version : 3.8.7 ``` ## Project setup - Follow below steps to obtain necessary credentials. You can also watch videos of TigerGraph on [YouTube](https://www.youtube.com/playlist?list=PLq4l3NnrSRp7om_qw4ciNbxslMONszkoX) for these steps. ``` - Login / Register on https://tgcloud.io/ - Create Solution using Movie Recommendation starter-kit, follow all default options. - Once the solution is ready launch GraphStudio and sign in. - Generate secret token by heading to AdminPortal -> User Management - Update config.json with tigergraph credentials. - Login / Register on https://hub.graphistry.com/ - Update Graphistry credentials in config.json ``` - Visit all the pages on the left panel of GraphStudio and execute necessary steps. - Copy GSQL queries present under asset/queries folder from this repo then go to Write Queries section in GraphStudio, create new query and paste there. - Install all the queries in order to use them as API by clicking on Install all queries button. ## Environment setup - Create virtual environment and install all the dependencies mentioned in requirements.txt ## Run project ``` - streamlit run TigerStream.py ``` ## Linter - [Flake8](https://realpython.com/python-pep8/#linters) extension available in VS Code is used to analyze code and flag errors. ## Demo - Watch demo of this application on [YouTube](https://www.youtube.com/watch?v=lBdxM13H16Y).