ceccode / income-prediction

Prediction model to determine whether a person makes over 50K a year

Income prediction

Machine Learning Zoomcamp midterm project: https://github.com/alexeygrigorev/mlbookcamp-code/tree/master/course-zoomcamp/07-midterm-project.

Income classification

Prediction task is to determine whether a person makes over 50K a year.

Dataset

• https://www.kaggle.com/lodetomasi1995/income-classification
• Copy of the dataset: income_evaluation_dataset.csv

Notebook

• Colab:https://colab.research.google.com/drive/14O27isuk8rJFNOo_srXqnHl_nuBfa6GE
• copy of the notebook here: notebook.ipynb

TLDR; Use the API

Predict if a certain person income is up to 50K using /predict API.

Endpoint:

• https://ml-zoomcamp-midterm-project.herokuapp.com

API:

• /heartbeat
• /predict

Predict API

curl --location --request POST 'https://ml-zoomcamp-midterm-project.herokuapp.com/predict' \
--header 'Content-Type: application/json' \
--data-raw '{
    "age": 45,
    "workclass": "private",
    "fnlwgt": 209642,
    "education": "masters",
    "education-num": 14,
    "marital-status": "married-civ-spouse",
    "occupation": "exec-managerial",
    "relationship": "husband",
    "race": "white",
    "sex": "male",
    "capital-gain": 0,
    "capital-loss": 0,
    "hours-per-week": 45,
    "native-country": "italy"
}'

Or download Postman collection: Income Prediction.postman_collection.json

Run locally with docker

Docker

docker build -t income-prediction ./src
docker run -it -p 9696:9696 income-prediction:latest

Predict

curl --location --request POST 'http://localhost:9696/predict' \
--header 'Content-Type: application/json' \
--data-raw '{
    "age": 45,
    "workclass": "private",
    "fnlwgt": 209642,
    "education": "masters",
    "education-num": 14,
    "marital-status": "married-civ-spouse",
    "occupation": "exec-managerial",
    "relationship": "husband",
    "race": "white",
    "sex": "male",
    "capital-gain": 0,
    "capital-loss": 0,
    "hours-per-week": 45,
    "native-country": "italy"
}'

Run locally without docker

Python required. For more see: https://packaging.python.org/tutorials/installing-packages/.

in /src dir:

pip install pipenv
pipenv install
pipenv shell
python predict.py

Install libomp required by xgboot:

MacOS:

brew install libomp

Ubuntu

RUN apt-get update && apt-get install -y libgomp1

Deploy to heroku and Run

Heroku account needed. Install heroku cli.

Docker file

Update docker file:

#ENTRYPOINT ["gunicorn", "--bind=0.0.0.0:9696", "predict:app"] 
ENTRYPOINT ["gunicorn", "predict:app"]

login

heroku login
heroku container:login

create app in heroku

heroku create your-app-name

Push docker image to Heroku

heroku container:push web -a your-app-name

Deploy container on Heroku

heroku container:release web -a your-app-name

Launch app

Test is up: https://your-app-name-xyz.herokuapp.com/heartbeat

Predict

Consume /predict API:

curl --location --request POST 'http://ml-zoomcamp-midterm-project.herokuapp.com/predict' \
--header 'Content-Type: application/json' \
--data-raw '{
    "age": 45,
    "workclass": "private",
    "fnlwgt": 209642,
    "education": "masters",
    "education-num": 14,
    "marital-status": "married-civ-spouse",
    "occupation": "exec-managerial",
    "relationship": "husband",
    "race": "white",
    "sex": "male",
    "capital-gain": 0,
    "capital-loss": 0,
    "hours-per-week": 45,
    "native-country": "italy"
}'

Postman collection

Download postman collection

• Income Prediction.postman_collection.json

and import it in your postman account to run it.

LICENSE

LICENSE