Data Science Full Course for Beginners (3 Hours) — Zero to a Real Project – Source Code
An end-to-end introduction to data science in Python using real public datasets. It covers NumPy and pandas, cleaning messy data, grouping and charts, then basic statistics and machine learning (regression, classification, k-means). A two-part final project loads, explores, models and evaluates a dataset from start to finish.
- Duration
- 3:11:05
- Tools
- Python 3.14.7, numpy 2.5.3, pandas 3.0.6, matplotlib 3.11.2, seaborn 0.13.2, scipy 1.18.1, scikit-learn 1.9.1
Watch the course
Bundled datasets keep their own licences
The MIT licence covers the code only. The datasets in data/ keep their own licences, listed in DATA_LICENSES.txt:
penguins.csv– palmerpenguins, CC0 1.0.customers_messy.csv– invented practice data.
The California Housing data is not included. The project loads it with scikit-learn's fetch_california_housing() on first run, so you need an internet connection once.
Chapters
- 00:00Welcome & the finished project
- 04:59Setup: Python, venv, packages, data
- 12:55Python refresher for data
- 24:19NumPy arrays & vectorised maths
- 36:41pandas: Series, DataFrames, inspecting data
- 46:51Selecting, filtering & sorting
- 57:27Cleaning messy data
- 1:08:37groupby, merge & pivot tables
- 1:19:27Charts with matplotlib & seaborn
- 1:37:33Statistics: distributions, correlation, t-test
- 1:53:08Machine learning: linear regression
- 2:10:04Classification, confusion matrix & overfitting
- 2:26:04Unsupervised learning: k-means clustering
- 2:34:30Final project 1: load, clean, explore
- 2:46:17Final project 2: model, evaluate, conclude
- 2:58:05Expert map: what to learn next
File: data-science-full-course-code.zip
Size: 20.8 KB (21,281 bytes)
SHA-256:b3536cda102875b54d25fe9722e067a3a9f565387e177f9427566ba05f70bd46
Licence: code is released under the MIT licence, copyright PyInns. See LICENSE and README inside the zip. Free learning code, not production code.