Students Performance Analysis

This data set consists of the marks secured by the students in various subjects. Here we will try to understand the influence of the parents background, test preparation etc on students performance. Also we will try to analyze is there are any socio-economic pattern that could affect students to perform better on the exam.

This is my very first public notebook data exploration. Please leave comments or suggestions to improve this notebook and give me another insight. Hope you like it 😊

Kaggle Dataset Link: https://www.kaggle.com/spscientist/students-performance-in-exams

Notebook Link: https://www.kaggle.com/affand20/students-performance-deep-analysis

CLI-based Model Inference

I'm also made a simple CLI-based model inference where you can try to run this model by yourself locally.

Installation

git clone https://github.com/affand20/students-performance
cd student-performance/deployment
pip install -r requirements.txt

Run Model Inference

python3 cli.py

If you're curious of how I created the model, you can take a look at my prediction notebook at notebook folders or just click here

RESTful API Based Model Inference

Make sure you have installed the following packages listed on requirements.txt. Run this command to install the packages.

cd student-performance/deployment
pip install -r requirements.txt

Or simply just follow the Installation steps above.

How to run:

cd deployment
uvicorn app:app --reload

Go to browser and visit http://127.0.0.1:8000

Available API

GET /predict?gender=male&race=group E&parental_edu=master's degree&lunch=standard&test_prep=completed&math_score=88&read_score=65&write_score=92

POST /predict

Parameter/Body	Type	Available Value	Description
gender	string	Male, Female	Student's gender.
race	string	group A, group B, ... group E	Student's race/ethnicity.
parental_edu	string	Master's Degree, Bachelor's Degree, Associate's Degree, Some College, High School, Some High School	Student's parent education level.
lunch	string	Standard, Free/reduced	Student's lunch type.
test_prep	string	Completed, None	Student's taking test preparation course status.
math_score	int	0-100	Student's Math subject score.
read_score	int	0-100	Student's Reading subject score.
write_score	int	0-100	Student's Writing subject score.

The API available in 2 methods POST and GET. Both using the same path /predict and using the same list parameters. For the GET method, you should insert parameter as Query Parameter, while for the POST method, you have to insert parameter as Request Body.

Responses

{
    "predicted_score": 81.66666666666667
}

Example

• curl

curl -X POST "http://0.0.0.0:8000/predict" -H  "accept: application/json" -H  "Content-Type: application/json" -d "{\"gender\":\"male\",\"race\":\"group E\",\"parental_edu\":\"master's degree\",\"lunch\":\"standard\",\"test_prep\":\"completed\",\"math_score\":88,\"read_score\":65,\"write_score\":92}"

• Postman

Get in Touch

If you are interested to give me any suggestions, comments, or discuss projects, works, anything please contact me at:

• Mail: fahri.lafa@gmail.com
• Linkedin: https://www.linkedin.com/in/affandyfahrizain/

Thank you!