/NYC-SAT-Scores-Server

API for NYC SAT Scores

Primary LanguageJupyter Notebook

NYC SAT DATA

Model Training Notebook

Goal

I conducted an in-depth analysis of NYC SAT scores and racial demographics, training a predictive model to highlight disparities in the education system. To empower broader access to these insights, I developed and deployed a full-stack web application where users can input relevant data and receive predicted scores, thereby fostering awareness and informing discussions on educational equity.

How to Run Locally

*Prerequisites

*Server

  • from the NYC-SAT-Scores-Server directory
  • run

    pip install -r requirements.txt

  • run

    start_server.sh

Testing
  • run

    pytest

*Client

  • from the NYC-SAT-Scores-Client directory
  • run

    npm install

  • run

    start_server.sh host

*Database

  • Connection String -> Server=localhost;Database=master;Trusted_Connection=True;

Technologies Used

  • Python
  • Pandas
  • Flask
  • Pytest
  • Sqlite3

Future Updates

  • Create a regression model to predict SAT score from demographic makeup, and visualization
  • Write unit tests for scores route
  • Replace iterrows with df apply
  • Remove duplicate data in Filter function
  • Unit tests for formatScores utility
  • Use reduce instead for enumerate for formatScores
  • Allow user input of data

Screenshots