Propensity To Buy Model

Demonstrates how to predict propensity to buy using a well known bank marketing dataset. The data is related with direct marketing campaigns (phone calls) of a Portuguese banking institution. The classification goal is to predict if the client will subscribe a term deposit (variable y).

Prerequisites

1. Python 3.6+
2. Source-to-Image
3. Make
4. (Optional) Httpie (only if you want to run make test-endpoint)

Quick Start

1. Install Python dependencies:

pip install -r requirements.txt

2. Train the model(s):

make train

3. Run unit tests:

make test

4. Build the Docker image:

make build

References

Bank Marketing Dataset from UCI