Midterm Project: Bank Marketing Use case

The data is related with direct marketing campaigns of a Portuguese banking institution. The marketing campaigns were based on phone calls. Often, more than one contact to the same client was required, in order to access if the product (bank term deposit) would be ('yes') or not ('no') subscribed.

Data Set

1. Download the dataset from the UCI Machine Learning Repository.
2. Unzip the file.
3. Use the bank-additional-full.csv file (which contains all records and attributes).
4. Place the file in the project's Data/ folder.

Note: The dataset contains 41,188 records and 20 attributes

Data description:

Variable	Type	Description
age	Numeric	Client's age.
job	Categorical	Type of job (e.g., "admin.", "blue-collar", "entrepreneur", "housemaid", "management", "retired", "self-employed", "services", "student", "technician", "unemployed", "unknown").
marital	Categorical	Marital status ("married", "divorced", "single"). Note: "divorced" includes divorced or widowed.
education	Categorical	Education level ("unknown", "secondary", "primary", "tertiary").
default	Binary	Has credit in default? ("yes", "no").
balance	Numeric	Average yearly balance, in euros.
housing	Binary	Has housing loan? ("yes", "no").
loan	Binary	Has personal loan? ("yes", "no").
contact	Categorical	Contact communication type ("unknown", "telephone", "cellular").
day	Numeric	Last contact day of the month.
month	Categorical	Last contact month of year ("jan", "feb", ..., "dec").
duration	Numeric	Last contact duration, in seconds.
campaign	Numeric	Number of contacts performed during this campaign and for this client (includes last contact).
pdays	Numeric	Number of days that passed by after the client was last contacted from a previous campaign (-1 means client was not previously contacted).
previous	Numeric	Number of contacts performed before this campaign and for this client.
poutcome	Categorical	Outcome of the previous marketing campaign ("unknown", "other", "failure", "success").
y	Binary	Target Variable: Has the client subscribed a term deposit? ("yes", "no").

Project Structure

This repository is organized into two main components: exploration and production source code.

dtc-ml-midterm-project/
│
├── conf/                       # Configuration files (Hydra)
│   └── config.yaml
│
├── data/                       # Raw and processed data
│   └── bank-additional-full.csv
│
├── notebooks/                  # Jupyter notebooks for experimentation
│   └── EDA.ipynb
│
├── src/                        # Source code for production
│   ├── main.py                 # FastAPI app entry point
│   ├── predict.py              # Client script for testing API
│   └── train.py                # Training pipeline & artifact generation
│
├── .env                        # Environment variables (Path configuration)
├── .gitignore                  # Files to ignore in Git
├── Dockerfile                  # Instructions to build the image
├── pyproject.toml              # Project dependencies (uv)
└── README.md                   # Project documentation

Environment Setup

Create the .env file by copying the template:
Copy .env.template and modify with your path configurations

cp .env.template .env

Deployment & Usage

Run the following commands to build and start the FastAPI infrastructure container:

docker build -t bank-marketing:latest .
docker run -p 8000:8000 bank-marketing:latest

The API will be available at http://localhost:8000/docs

Inference example:

{
  "job": "admin.",
  "marital": "married",
  "education": "university.degree",
  "default": "no",
  "housing": "yes",
  "loan": "no",
  "contact": "cellular",
  "month": "aug",
  "day_of_week": "thu",
  "age": 35,
  "duration": 210,
  "campaign": 1,
  "pdays": 999,
  "previous": 0,
  "poutcome": "nonexistent",
  "emp_var_rate": 1.4,
  "cons_price_idx": 93.444,
  "cons_conf_idx": -36.1,
  "euribor3m": 4.963,
  "nr_employed": 5228.1
}