Skip to main content

02 · Python for machine learning

Turn understanding
into working code.

Start with small, understandable programs. Then connect arrays, data tables and careful preparation to a model you can fit and evaluate.

New to code? Python from zero →Python Essentials →Cheat sheet →Browse Python modules ↓

Recommended sequence

Build your Python toolkit.

Seven paths. New to code? Start with Python from zero, then Essentials, or open the skill you need. Module-specific practice, assessments and saved progress remain within each module.

  1. 01Start here · parts 1–3 free

    Python from zero

    Read and write your first Python, with no experience needed.

    6 short lessons · = versus ==, if and for, functions, dictionaries, classes and objects, and deploying a model. Start here if you have never coded.

    Start Python from zero
  2. 02Available to explore

    Python Essentials

    Write small programs and explain what each operation does.

    8 lessons · Values, types, conditions, loops, functions and meaningful model inputs. Recommended first module.

    Open Python Essentials
  3. 03Available to explore

    NumPy

    Follow arrays, shapes and vectorized predictions.

    6 lessons · Array operations, broadcasting, matrix products and reductions. Helpful foundation: Python Essentials.

    Open NumPy
  4. 04Available to explore

    pandas

    Keep data, labels and customer identities aligned.

    6 lessons · Tables, selection, missing values, grouping, joins and dataset audits.

    Open pandas
  5. 05Available to explore

    Visualization

    Use charts to inspect data and explain what they show.

    5 lessons · Distributions, relationships, box plots and misleading visual claims.

    Open Visualization
  6. 06Available to explore

    Data Preparation

    Prepare inputs without leaking evaluation information.

    6 lessons · Splits, missing values, encoding, scaling and fold-local pipelines.

    Open Data Preparation
  7. 07Available to explore

    scikit-learn

    Fit, predict and evaluate a complete model workflow.

    5 lessons · Estimators, baselines, metrics, validation, pipelines and comparison. Review arrays and preparation as needed.

    Open scikit-learn

Connect code to a model.

Use the same habits—clear inputs, meaningful operations and honest evaluation—when studying the algorithms themselves.

Explore Algorithms →
Earlier Python practice and learning records

The original P01 prototype and its records remain available. Any supported copy into Essentials still requires preview and explicit opt-in; this page does not import progress.

View supported learning records and resources

Existing lessons are available to explore. Manual accessibility and genuine notebook-host verification remain separate evidence requirements.

Notebooks

Download a notebook to run a module on your own computer or in Google Colab. They match the lessons and never award mastery by themselves.