indyfree / tailor

Clustering Algorithm for clustering retail products according to custom requirements.

tailor - Tailored Data Solution

Clustering Algorithm for clustering retail products according to custom requirements.

Requirements

• python 3.6
• GNU make

Installation

This project is intended to run on Mac or Linux. On Windows it should also be runnable via the Linux Subsystem.

Clone the repository

> git clone https://github.com/indyfree/tailor

Install required packages

Installs dependencies with pip:

> make requirements

Provide server information

Create a file .env in the project root:

> cd tailorit
> touch .env

Edit the file with an editor of your choice to provide credentials to the tailorit server. The file should look like this:

export TAILORIT_SERVER_ADDRESS=[address]
export TAILORIT_USER=[user]
export TAILORIT_PW=[password]

Where [address], [user] and [password] have to be substituted with the respective values.

Get the data

Download the provided data to data/raw.

> make data

Run the project

This project is set up twofold:

1. Custom functions and algorithms live in the tailor python package in src/tailor.
2. A walkthrough through the data science process and visualizations are made with jupyter notebooks in notebooks. Required functions from tailor are imported.

Run the jupyter notebooks with:

> make jupyter

You can access the jupyter notebooks via your webbrowser at http://localhost:8888/.