Data Science
From statistics and pandas to end-to-end projects — the complete foundation for turning raw data into decisions.
What Is Data Science?
Define data science, understand its core disciplines, and map the roles and workflow of a modern data team.
The Data Science Pipeline
Walk through the six stages of a data science project and learn where real projects fail — and why iteration matters.
Python for Data Science
The Python fundamentals every data scientist leans on: collections, comprehensions, file I/O, and reproducible notebooks.
NumPy: Arrays & Vectorization
Master the n-dimensional array, vectorized operations, and the broadcasting rules that make Python fast.
Pandas: DataFrames & Series
Load, inspect, filter, group and merge tabular data with the most important library in data science.
Data Cleaning & Wrangling
Fix missing values, wrong types, duplicates and inconsistent text — the 60–80% of real data work.
Exploratory Data Analysis
Use summaries and plots to interrogate a dataset before any modeling — find patterns, outliers and bugs.
Data Visualization
Build clear, honest charts with Matplotlib and Seaborn — and learn which chart answers which question.
Statistics Fundamentals
Descriptive statistics, samples vs populations, and the distributions that data science runs on.
Probability Distributions
The distributions behind data science: Bernoulli, binomial, Poisson, normal and more — and when to use each.
Hypothesis Testing
Null hypotheses, p-values, t-tests and the statistical decisions that separate real effects from noise.
Correlation & Causation
Measure correlation, then learn why correlation is not causation — and how to actually establish causes.
Feature Engineering
Turn raw data into features models can learn from: encodings, scaling, transforms and derived variables.
Train/Test Splits & Validation
Split data honestly, avoid leakage, and use holdout sets and cross-validation to measure real performance.
Regression Models
Predict continuous values with linear and regularized regression, and read the results honestly.
Classification Models
Predict categories with logistic regression, decision trees and k-NN — and understand what a classifier actually outputs.
Clustering
Discover structure in unlabeled data with k-means and hierarchical clustering — and evaluate the result.
Model Evaluation Metrics
Accuracy, precision, recall, F1, ROC-AUC and confusion matrices — the honest measurement of model quality.
Communicating Results
Turn analysis into decisions: structure a finding, choose the right chart, and write for stakeholders.
An End-to-End Data Science Project
Combine everything into a real project: from a raw dataset to a working, honestly-evaluated model.
Beyond the Basics: Your Data Science Roadmap
Where to go after this course: specializations, projects, and the learning paths that lead to a career.