bibinmjose / dsgramner

An analysis based on National Achievement Survey

Insights from this project

Github repo for the project.

Setup

To reproduce the project clone repository: git clone https://github.com/bibinmjose/dsgramner.git

unzip gramener-usecase-nas.zip to root folder /dsgramner if doesn't exist

Original ipython notebook is in root folder while an html redering can be accessed from here.

jupyter-notebook National\ Achievement\ Survey.ipynb

Execute all the cells in jupyter notebook

raw .html of ipython notebook can be found at root folder as well.

Dependencies

The packages used include :

• pandas
• numpy
• seaborn
• sklearn
• matplotlib

Data and Documentation

All the major documentation of this project can be accessed from the markdown file or the html in folder docs.

Datasets used for analysis is provided by "Gramener" and can be found in gramener-usecase-nas. The dataset is based on national achievement survey conducted by NCERT for 8th grade students.

Model

Since inference is the goal of the questions asked, a linear model is a good choice than a complex one.

To run the model, cd into /model and run python LR_model.py

To Do

• Make a simple linear model
• Split each categorical variable into its own category in the linear model.

By Bibin Jose