Class Central is learner-supported. When you buy through links on our site, we may earn an affiliate commission.

XGBoost and Data Leakage in Machine Learning - Day 14 of 30 Days of ML

1littlecoder via YouTube

Start learning Write review

Dive into Day 14 of Kaggle's 30 Days of ML Challenge, focusing on XGBoost and data leakage in machine learning. Explore the power of gradient boosting, a technique that dominates many Kaggle competitions and achieves state-of-the-art results on various datasets. Learn how to build and optimize models using XGBoost, and gain insights from a comprehensive tutorial. Understand the critical concept of data leakage, its potential to ruin models in subtle and dangerous ways, and discover effective prevention strategies. Complete hands-on exercises to reinforce your learning and wrap up Kaggle's Intermediate ML Course. Access additional resources, including related videos and a StatQuest tutorial on XGBoost, to deepen your understanding of these essential machine learning concepts.

Syllabus

Kaggle 30 Days of ML (Day 14) - XGBoost, Data Leakage - Learn Python ML in 30 Days

Taught by

1littlecoder

Reviews

5.0 rating, based on 1 Class Central review

Start your review of XGBoost and Data Leakage in Machine Learning - Day 14 of 30 Days of ML

Unza Khan

XGBoost Part: I understand how gradient boosting works, the purpose of key hyperparameters (learning rate, max depth, n_estimators, subsample, etc.), and how to control overfitting. I can interpret feature importance and apply XGBoost for classifica…

XGBoost Part: I understand how gradient boosting works, the purpose of key hyperparameters (learning rate, max depth, n_estimators, subsample, etc.), and how to control overfitting. I can interpret feature importance and apply XGBoost for classification, regression, and ranking problems.

Data Leakage Part: I can identify situations where future information accidentally leaks into training data, such as wrong use of target encoding, preprocessing before splitting, or leakage in time-series setups. I know how to prevent it using proper train-test splits, correct cross-validation, and by placing preprocessing inside a pipeline.

End Outcome: By the end, I can confidently build and tune XGBoost models while ensuring the model’s performance is real and not inflated due to leakage.

Start learning

Learn AI, Data Science & Business — Earn Certificates That Get You Hired

PowerBI Data Analyst - Create visualizations and dashboards from scratch

Taught by

Master Production-Ready Machine Learning, Step by Step

Kaggle 30 Days of ML Learning - Beginner-Friendly Machine Learning Challenge

Kaggle Competition Submission - Learn Python ML in 30 Days - Day 11

Python Libraries and Operator Overloading for Machine Learning - Day 7 of 30 Days of ML

Kaggle 30 Days of ML - Day 12 - Missing Values and Categorical Variables

Python Strings and Dictionaries for Machine Learning - Day 6 of 30 Days of ML

Build with Azure OpenAI, Copilot Studio & Agentic Frameworks — Microsoft Certified Ad

Harvard’s Intro to Python: Earn a Free Certificate

15 Best Python Courses for 2026: Learn the Most Popular Language

9 Best Django Courses for 2026 (Free & Paid): Build Real Python Web Apps

12 Best Data Analysis Courses for 2026: From Data to Insights

13 Best Data Science Courses for 2026

Never Stop Learning.