/web-prediction

A RESTful web service that allows you to upload a training data set, train a Logistic Regression classifier, and use it to make predictions.

Primary LanguagePythonCreative Commons Zero v1.0 UniversalCC0-1.0

webprediction

This code sample is a bare bones demo of the following workflow:

  1. load a training dataset from a local file
  2. train a classifier with your favorite machine learning algorithm
  3. launch a RESTful web service to apply the trained model to future test data

This is a simple way of integrating a prediction module into a larger web-based analytics platform

USAGE: python demo.py <ip>:<port>

Then open <ip>:<port> in your web browser

Tested on: Python 2.7, sklearn 0.13