Machine Learning Zoomcamp Capstone

Introduction

The organization datatalks.club has provided an online course entitled Machine Learning Zoomcamp to provide a practical introduction to machine learning. The course is based on the book Machine Learning Bookcamp by Alexey Grigorev.

This project was undertaken as a capstone project for the Machine Learning Zoomcamp course.

Project description

The service implemented here produces predictions for house prices based on data collected on residential property sales in Ames, Iowa from 2006 to 2010. The data was shared by Dean De Cock in his paper "Ames, Iowa: Alternative to the Boston Housing Data as an End of Semester Regression Project" . De Cock intended for the data to be used in projects by students taking coursework in statistics and, as the title of the paper suggests, is meant to replace the commonly used Boston housing dataset.

The project uses this data to produce a model and implements a service to produce predictions of the sale price of a house based on the features provided in the request itself.

Some files of interest:

notebook.ipynb: The Jupyter notebook used to perform EDA. Also, captures work to select and tune the estimator used to create the model.
train.py: Produces the model binary used by the service to perform predictions
predict.py: Implementation of the prediction service using Flask
predict_client.py: An example client used to request predictions
Pipfile: Defines the project dependencies
Procfile: Used by Heroku to run the prediction service

Getting started

This project requires Docker and optionally make. make is available typically readily available in Linux and can be installed on other OSs using their respective toolsets:

Mac: Homebrew or xcode
Windows: Windows Subsystem for Linux or Chocolatey

The following assumes that you have make installed. However, make is simply a wrapper to ease execution of the commands using Python or Docker, so please consult the Makefile to see the underlying commands used to work with the project.

To build the Docker image for the product, run make build.

To run the prediction service, run make run-service.

To run a Python client that makes example requests to the service, run make run-client-local to make a request to a local version of the prediction service or make run-client-remote if you'd like to make requests to the service running in Heroku.

The data used to train the model is available in data. However, it can be downloaded using make data.

Binary versions of the model are available in bin. However, they can be regenerated using make bin.

Deployment

The prediction service is available at https://zoomcamp-capstone.herokuapp.com. It can be deployed using make heroku-deploy and requires installation of the Heroku CLI. Authentication is required to run make heroku-deploy. To authenticate, use make heroku-login and you will be prompted to authenticate using a web browser. However, deployment is not necessary for testing as the service is available publicly (see the notes regarding make run-client-remote above for how to make a request). The file Procfile is included for Heroku's sake so it knows how to start the service (see the Heroku docs on Procfile for further detail).

Usage example

The service is available at https://zoomcamp-capstone.herokuapp.com/predict. A request should contain the non-target variables for a single property as defined in the data documentation. The response contains a predicted sales price value.

The example client sends a request using a random entry from the test data set. The following is an example request and response made by the client.

Request

{
  "ms_subclass": 20,
  "ms_zoning": "RL",
  "lot_area": 9612,
  "street": "Pave",
  "alley": "NA",
  "lot_shape": "Reg",
  "land_contour": "Lvl",
  "utilities": "AllPub",
  "lot_config": "Inside",
  "land_slope": "Gtl",
  "neighborhood": "Somerst",
  "condition_1": "Feedr",
  "condition_2": "Norm",
  "bldg_type": "1Fam",
  "house_style": "1Story",
  "overall_qual": 8,
  "overall_cond": 5,
  "year_built": 2008,
  "year_remod/add": 2009,
  "roof_style": "Gable",
  "roof_matl": "CompShg",
  "exterior_1st": "VinylSd",
  "exterior_2nd": "VinylSd",
  "mas_vnr_type": "Stone",
  "mas_vnr_area": 72.0,
  "exter_qual": "Gd",
  "exter_cond": "TA",
  "foundation": "PConc",
  "bsmt_qual": "Gd",
  "bsmt_cond": "TA",
  "bsmt_exposure": "No",
  "bsmtfin_type_1": "Unf",
  "bsmtfin_sf_1": 0.0,
  "bsmtfin_type_2": "Unf",
  "bsmtfin_sf_2": 0.0,
  "bsmt_unf_sf": 1468.0,
  "total_bsmt_sf": 1468.0,
  "heating": "GasA",
  "heating_qc": "Ex",
  "central_air": true,
  "electrical": "SBrkr",
  "1st_flr_sf": 1468,
  "2nd_flr_sf": 0,
  "low_qual_fin_sf": 0,
  "gr_liv_area": 1468,
  "bsmt_full_bath": 0.0,
  "bsmt_half_bath": 0.0,
  "full_bath": 2,
  "half_bath": 0,
  "bedroom_abvgr": 3,
  "kitchen_abvgr": 1,
  "kitchen_qual": "Gd",
  "totrms_abvgrd": 6,
  "functional": "Typ",
  "fireplaces": 1,
  "fireplace_qu": "Gd",
  "garage_type": "Attchd",
  "garage_finish": "Fin",
  "garage_cars": 3.0,
  "garage_area": 898.0,
  "garage_qual": "TA",
  "garage_cond": "TA",
  "paved_drive": true,
  "wood_deck_sf": 210,
  "open_porch_sf": 150,
  "enclosed_porch": 0,
  "3ssn_porch": 0,
  "screen_porch": 0,
  "pool_area": 0,
  "pool_qc": "NA",
  "fence": "NA",
  "misc_feature": "NA",
  "misc_val": 0,
  "mo_sold": 12,
  "yr_sold": 2009,
  "sale_type": "New",
  "sale_condition": "Partial"
}

Response

{
  "price": 239011.84
}

Dependencies

The project uses Pipenv to manage its dependencies. When used outside of Docker, the dependencies can be installed via pipenv install and the environment can be used via pipenv shell. To view the dependencies, see Pipfile.

robmoore / zoomcamp-capstone-project