mlwpy_code

Code from the Pearson Addison-Wesley book Machine Learning with Python for Everyone

Versioning Note

The code here is the latest/greatest version. Updated releases are planned annually in August. The version here has been updated to work with the most recent versions of its dependencies (e.g., scikit-learn and pandas). If you want a quick link to the latest release (with convenient downloads), click here.

If you want the exact version of the code that came with the book, you'll want the 1.0 release which you can

• view the source of
• or download directly.

Python Environment Setup

See make_environments.md (here) for details.

Package Versions

Here are the versions of important packages as of the Feb 2024 code update. As a reminder, you can install specific versions of packages using anaconda/conda as documented here. For example, conda install scikit-learn=1.3.0.

My plan is to update the book software to the most recent versions of software around once a year (likely near August/September). Life's been crazy, so the timeline has slipped a bit. Sorry.

I'll update the software versions listed here when I do so. The full list of package versions is available in book_env_2024_feb.yml produced by conda env export --from-history > book_env_2024_feb.yml. Actually, it's not anymore because locking package versions limits platform independence. But, the important ones are listed here.

matplotlib                3.8.0
numpy                     1.22.3
opencv                    4.6.0
pandas                    2.0.3
patsy                     0.5.3
pymc3                     3.11.4
py-xgboost                1.7.3 
scikit-learn              1.3.0  
scipy                     1.7.3
seaborn                   0.12.2
statsmodels               0.14.0  
tensorflow                2.10.0