Skip to content

A graduate-level introduction to reinforcement learning as a framework for modeling, optimization, and control, connecting dynamic models, data, and applications beyond standard benchmarks.

Notifications You must be signed in to change notification settings

pierrelux/rlbook

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

rlbook

Notes on executable content

This project is built with the new MyST “book-theme” site generator. Unlike the old Sphinx-based pipeline, the static HTML does not automatically embed the last outputs that were present in your .ipynb / MyST source. A few rules keep plots and widgets visible both on localhost:3000 and on GitHub Pages:

  1. Every code cell that produces a figure should follow this pattern:

    ```{code-cell} python
    :tags: [hide-input]
    
    #  label: fig-my-figure-id
    #  caption: Short description of the figure.
    
    %config InlineBackend.figure_format = 'retina'
    import matplotlib.pyplot as plt
    ...
    plt.tight_layout()
    ```
    

    Notes:

    • Use regular Python comments (# label:) for metadata, not MyST directives (#|)
    • Do not add a :::{figure} embed after the cell—the figure displays directly from the cell output
    • Do not call plt.show(); end with plt.tight_layout() instead
    • The %config InlineBackend.figure_format = 'retina' ensures high-DPI output
  2. Always build with notebook execution enabled. The helper script already enforces this:

    source publish.sh  # runs BASE_URL=/rlbook uv run jupyter-book build --html --execute

    Skipping --execute will re-use whatever cache happens to exist (or nothing, if you changed the cell), which can lead to empty output.

  3. When iterating locally, uv run jupyter-book start --execute --port 3000 reproduces exactly what GitHub Pages will host, including cached PNG outputs.

Following those three steps prevents regressions where the site appears correct only when a live kernel is attached.

About

A graduate-level introduction to reinforcement learning as a framework for modeling, optimization, and control, connecting dynamic models, data, and applications beyond standard benchmarks.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •