Outline

1. Application
2. Dataset Trained on
3. Model
4. How to use this project?
5. What's next?

1. Application

Machine learning based web app that predicts the Math exam score for a student based on their input feature values. You may access the web app here (https://math-grade-predictor.onrender.com/). Snippet of the web interface:

• The API was built with Flask framework.
• Deployed with Render.

You may clone this repository and run the following commands on your terminal to ensure successful access to the app on your local host:

pip install -r requirements.txt

Then:

python3 app.py

Illustration Digram

The following Lucid chart illustrates how the web app works on the high level:

2. Dataset Trained on

The dataset stud.csv contains 1001 rows and 8 columns representing students records of their:

• Gender: female or male
• Race/ethnicity: denoted by groups
• Parental level of education: bachelor's degree, associate degree, some college, high school, master's degree
• Lunch: wether the student received standard or free/reduced lunch
• Test preparation course: wether the student completed a test prep course or not
• Math score: numeric value of the student's grade in Math ( the feature to be predicted)
• Reading score: numeric value of the student's grade in reading
• Writing score: numeric value of the student's grade in writing

3. Model

• The model trainer runs the following set of algorithms from the sklearn library on the preprocessed features to observe which performs the best:
  • Random forest
  • Decision Tree
  • Gradient Boosting
  • Linear Regression
  • K-Neighbors
  • XGBRegressor
  • AdaBoost Regressor

Then pickles the best model which in this case is acheived by the Ridge linear model.

4. How to use this project?

This project could be used a guide in creating a standard end-to-end ML web application with your choice of the API framework; You may fork this repository and add your own custom features or even retrain the models on your own dataset as long as you perform the necessary exploratory data analysis, feature engineering and selection, and data pre-processing. Please feel free to also open issues if you have encountered a bug or have a suggestion on how to improve the application or any parts of the pipeline.

5. What's next/in progress?

• In progress: Deployment with AmazonECS and Docker images.