Skip to content

04 A Train Baseline Model

View code on GitHub Get the Video Course

Train a baseline Logistic Regression model for credit card fraud detection using sklearn Pipeline with business metrics.

Students build the training script incrementally across 7 versions - each version adds exactly ONE concept.

Pre-requisite: Python Environment Setup

Terminal
# Create conda environment
conda create -n mlops-env1 python=3.14 -c conda-forge -y

# Activate environment
conda activate mlops-env1

# Install dependencies (locked versions)
cd ccfd-project
pip install -r requirements.txt

Note: All sections in this course use the same mlops-env1 environment. You only need to create it once. After that, just activate it with conda activate mlops-env1 before running any scripts.



Incremental Build: v1 → v7

TRAINING PIPELINE - BUILD IT STEP BY STEP
==========================================

  v1          v2          v3           v4            v5            v6           v7
  Skeleton    Load        Split        Pipeline      Evaluate      Save         Save
  (stubs)     Data        Train/Test   & Train       & Metrics     Pipeline     Metrics
                                                                                JSON
  ┌─────┐   ┌─────┐    ┌─────┐     ┌─────┐       ┌─────┐      ┌─────┐      ┌─────┐
  │     │   │STEP1│    │STEP1│     │STEP1│       │STEP1│      │STEP1│      │STEP1│
  │     │   │Load │───>│Load │───> │Load │───>   │Load │───>  │Load │───>  │Load │
  │     │   │     │    │STEP2│     │STEP2│       │STEP2│      │STEP2│      │STEP2│
  │ ALL │   │     │    │Split│───> │Split│───>   │Split│───>  │Split│───>  │Split│
  │STUBS│   │     │    │     │     │STEP3│       │STEP3│      │STEP3│      │STEP3│
  │     │   │     │    │     │     │Pipe │───>   │Pipe │───>  │Pipe │───>  │Pipe │
  │     │   │     │    │     │     │STEP4│       │STEP4│      │STEP4│      │STEP4│
  │     │   │     │    │     │     │Train│───>   │Train│───>  │Train│───>  │Train│
  │     │   │     │    │     │     │     │       │STEP5│      │STEP5│      │STEP5│
  │     │   │     │    │     │     │     │       │Eval │───>  │Eval │───>  │Eval │
  │     │   │     │    │     │     │     │       │     │      │STEP6│      │STEP6│
  │     │   │     │    │     │     │     │       │     │      │Save │───>  │Save │
  │     │   │     │    │     │     │     │       │     │      │Model│      │Model│
  │     │   │     │    │     │     │     │       │     │      │     │      │STEP7│
  │     │   │     │    │     │     │     │       │     │      │     │      │Save │
  │     │   │     │    │     │     │     │       │     │      │     │      │JSON │
  └─────┘   └─────┘    └─────┘     └─────┘       └─────┘      └─────┘      └─────┘

Version Summary

Version Folder What's New New Functions
v1 v1_base_ccfd-project/ Project skeleton - all files are stubs -
v2 v2_load_data_ccfd-project/ STEP 1: Load CSV data load_data()
v3 v3_split_train_test_ccfd-project/ STEP 2: Stratified train/test split split_data(), get_feature_columns()
v4 v4_create_and_train_pipeline_ccfd-project/ STEP 3-4: Create & train sklearn Pipeline create_preprocessor(), create_pipeline()
v5 v5_evaluate_and_metrics_ccfd-project/ STEP 5: Evaluate with technical + business metrics print_metrics(), calculate_business_metrics()
v6 v6_save_pipeline_ccfd-project/ STEP 6: Save trained pipeline to disk save_pipeline()
v7 v7_save_metrics_ccfd-project/ STEP 7: Save metrics JSON for tracking save_metrics_json()
Final ccfd-project/ Complete reference with full docstrings All functions

Project Structure (Final - ccfd-project)

ccfd-project/
├── p1_01_generate_initial_dataset.py    # Generate synthetic dataset
├── p3_01_train_model_baseline.py        # Training script (7 steps)
├── requirements.txt                     # Python dependencies
└── utils/
    ├── __init__.py                      # Package exports
    ├── data_utils.py                    # load_data, split_data, create_preprocessor
    ├── model_utils.py                   # create_pipeline, save_pipeline
    └── metrics_utils.py                 # print_metrics, save_metrics_json

How to Run (Any Version)

Terminal
# 1. Navigate to any version folder
cd v3_split_train_test_ccfd-project/       # or any v1-v7, or ccfd-project/

# 2. Generate data (required first - creates data/ folder)
python p1_01_generate_initial_dataset.py

# 3. Run the training script
python p3_01_train_model_baseline.py

Note: v1 is stubs only - nothing runs. v2+ will execute the steps implemented so far.


What the Pipeline Does

Input (8 raw features)
    │
    ▼
ColumnTransformer (Preprocessor)
├── OneHotEncoder  → 3 categorical (merchant_category, card_present, international)
└── StandardScaler → 5 numerical   (transaction_amount, transaction_hour, etc.)
    │
    ▼
LogisticRegression (C=1.0, class_weight='balanced')
    │
    ▼
Output: FRAUD / LEGITIMATE (with probability)

Utils Module - Function Inventory

data_utils.py

Function Added In Purpose
load_data() v2 Load CSV → DataFrame
get_feature_columns() v3 Return categorical + numerical column lists
split_data() v3 Stratified train/test split with feature info
create_preprocessor() v4 ColumnTransformer (OneHotEncoder + StandardScaler)

model_utils.py

Function Added In Purpose
create_pipeline() v4 Build sklearn Pipeline (preprocessor + model)
save_pipeline() v6 Save pipeline pickle + metadata JSON

metrics_utils.py

Function Added In Purpose
print_metrics() v5 Print technical + business metrics
calculate_business_metrics() v5 Compute net benefit, fraud caught/missed
save_metrics_json() v7 Save metrics to results/ as JSON

Expected Output (v7 / ccfd-project)

STEP 1: Loading Data
Loaded 10,000 transactions from data/credit_card_transactions_latest.csv

STEP 2: Splitting Data
Split: 8,000 train | 2,000 test

STEP 3: Creating Pipeline
Pipeline created:
   1. Preprocessor (OneHotEncoder + StandardScaler)
   2. Model (LogisticRegression)

STEP 4: Training Pipeline
Training complete!

STEP 5: Evaluating Performance
   F1 Score:     0.2329
   Precision:    0.1373
   Recall:       0.7667
   ROC AUC:      0.8837
   Net Benefit:  $39,625

STEP 6: Saving Pipeline
   Model saved to: models/p3_01_baseline/

STEP 7: Saving Metrics
   Metrics saved to: results/

======================================================================
BASELINE TRAINING COMPLETE!
======================================================================

Files Created (v7):

File Description
data/credit_card_transactions_latest.csv Generated by p1_01
models/p3_01_baseline/model_latest.pkl Trained sklearn Pipeline
models/p3_01_baseline/metadata_latest.json Model metadata + metrics
results/logistic_regression_*.json Detailed metrics JSON

Understanding the Metrics

Why Recall Matters Most in Fraud Detection

Cost of missing real fraud:   $1,500  (average fraud amount)
Cost of a false alarm:        $25     (investigation cost)

Missing fraud is 60x more expensive than a false alarm!

That's why we use class_weight='balanced' - prioritize catching fraud even at the cost of more false alarms.

Business Value (Net Benefit)

Net Benefit = (Fraud Caught × $1,475) - (False Alarms × $25) - (Fraud Missed × $1,500)

Even with low precision (13.7%), the model generates positive net benefit because catching fraud far outweighs false alarm costs.


Next Steps

Next Topic What You'll Do
04_02_Inference_API Serve Predictions Build FastAPI server, test via browser + scripts

Prefer to learn by watching?

The video course builds this whole project with you on screen, step by step.

Get the Video Course


04 Model Training and Inference Start the build: v1 Project Skeleton Next: 04 B Inference API

🎉 New Course
Machine Learning End-to-End: Credit Card Fraud Detection
$15.99 $84.99 81% OFF
Get the Video Course