https://github.com/anothersamwilson/flaskpredictor

Simple Flask App that accepts json and returns prediction
https://github.com/anothersamwilson/flaskpredictor

Last synced: 3 months ago
JSON representation

Simple Flask App that accepts json and returns prediction

• Host: GitHub
• URL: https://github.com/anothersamwilson/flaskpredictor
• Owner: AnotherSamWilson
• Created: 2019-06-20T02:28:06.000Z (almost 6 years ago)
• Default Branch: master
• Last Pushed: 2019-06-20T23:37:21.000Z (almost 6 years ago)
• Last Synced: 2025-01-08T04:52:47.588Z (5 months ago)
• Language: Python
• Size: 137 KB
• Stars: 2
• Watchers: 2
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
Purpose

-------

This project is for personal improvement and satisfaction. The most

difficult step for (most) data scientists is model deployment. Flask

makes that easy enough with their slick Python library. In this project,

we train a model with lightgbm and create an app that will predict a

house selling price from the following columns:

-   LotArea

-   Neighborhood

-   Condition1

-   Condition2

-   OverallQual

-   OverallCond

-   YearBuilt

-   SaleCondition

There are hundreds (maybe thousands) of kernels on Kaggle that show you

how to train a model that can predict housing prices on this exact

dataset. However, none of them actually deploy the model so that one

could query it and recieve a prediction. That is what this project does.

It is a work in progress, and will be improved upon as I find new

features to add.

The requirements.txt file is pretty lightweight, and only includes the

packages needed for this specific project (plus base Python) so feel

free to copy the environment.

Usage

-----

The web service can receive 3 valid commands:

### POST with json body

The post to the web service should include a body with the following

json schema:

    {

      "LotArea": Integer,

      "Neighborhood": String,

      "Condition1": String,

      "Condition2": String,

      "OverallQual": Integer,

      "OverallCond": Integer,

      "YearBuilt": Integer,

      "SaleCondition": String

    }

This json represends ‘samples’ which we would like to predict. For

example, the following payload:

    {

      "LotArea": 9600,

      "Neighborhood": "Veenker",

      "Condition1": "Feedr",

      "Condition2": "Norm",

      "OverallQual": 6,

      "OverallCond": 8,

      "YearBuilt": 1976,

      "SaleCondition": "Normal"

    }

Will return:

    {

      "SalePrice": 186272.93

    }

### Shutdown

Simply send a GET request to the path /shutdown to shut the app down

gracefully.

### GET

Send a GET request to the root path to recieve a refreshable key. Used

for debugging.