Data Science and Machine Learning with Python
The data science and introductory AI lessons are available. Applied pipelines, the final project, assessment, and certificate are still being developed.
Complete both Scientific Computing with Python and Statistics with Python. You may take those two courses in either order.
Milestone sequence
Each milestone combines its units in an ungraded challenge. Complete the unit challenges as you go, then use the milestone challenge to connect those skills. Progress is self-reported in this browser on this device.
- Answer a Question with Tabular Data — 1 units; milestone challenge.
- Evaluate a Model Responsibly — 1 units; milestone challenge.
Before you start
- Designed for: learners who want to turn data questions into careful analyses and responsibly evaluated first models.
- You are ready when: you can work with arrays, make and explain plots, reason about small numerical experiments, describe distributions, and explain sampling variability and uncertainty.
- Practice setup: use Google Colab with pandas, Seaborn, and the introduced machine-learning libraries. A local environment is recommended for larger portfolio projects and reproducible dependency management.
What the current lessons teach
- load, inspect, clean, summarize, and visualize tabular data;
- communicate evidence, uncertainty, and limitations;
- identify features, labels, datasets, and prediction tasks;
- train and evaluate a first model and recognize overfitting;
- explain where deep-learning libraries and transformers fit;
- identify privacy, bias, and accountability concerns.
Current units
1. Data Science
Load, inspect, clean, summarize, visualize, and communicate tabular data.
2. Machine Learning and AI
Train and evaluate a first model, understand the broader library landscape, and consider responsible AI.
Planned expansion
- preprocessing, feature engineering, pipelines, baselines, and leakage;
- model selection, cross-validation, metrics, and error analysis;
- interpretation and communication for model results;
- an end-to-end project with a model card;
- a clear optional boundary for deep learning and transformers.
Practice, assessment, and next step
Current lessons provide runnable data workflows, Colab notebooks, and formative quizzes. They count toward local progress, but the applied pipeline, final project rubric, and certificate assessment are not complete. After the current material, use the Project Toolkit to plan a small analysis while the full capstone is developed.