Enhances the changelog generation script to categorize updates into Frontmatter, Chapters, Labs, and Appendix sections.
This improves readability and organization by grouping related changes together. It also adds logic to determine if the sections should be open or closed (details tag) based on if it's the latest entry. Finally, it correctly titles each of the items in the chapter lookup.
Massively reorganizes the textbook's file structure, improving maintainability, and adding references to datasets.
The updates involve sorting files into a new folder structure with subfolders within the images directory based on file type.
Distributes references so each chapter now has its own reference files.
Addresses minor textual inconsistencies across multiple QMD files to enhance readability and maintain consistency.
Updates the dash processing script to handle both em-dashes and triple hyphens, improving the accuracy of text transformations in QMD files. This ensures consistent formatting and enhances the overall quality of the generated content.
Updates the prompt template for better handling of Markdown formatting, including lists, bold/italic text, inline code, and citations. It also ensures that bullets are respected during paragraph splitting.
This allows the script to correctly process more complex documents and bulleted lists, improving the quality of the rewritten text.
Introduces a script that leverages OpenAI's GPT models to
reformat em-dash clauses in .qmd files, replacing them with
more formal structures like commas or parentheses.
The script processes files in either interactive or batch mode,
allowing users to review and accept changes, or automatically
apply them across a directory of files. It utilizes the gpt-4o model
to rewrite the clauses, ensuring the preservation of existing
Markdown formatting and citations.
This improves the formal tone of academic writing by addressing
the overuse of em-dashes.
Adds a script that uses OpenAI to rewrite clauses enclosed in em-dashes in .qmd files.
The script provides both interactive and batch modes, allowing users to review and accept changes, or automatically process files. It refactors em-dash clauses using commas or parentheses for improved formal writing style.
Adds a pre-commit hook that checks for quarantined images on macOS.
This prevents accidentally committing images that have quarantine flags set, which can cause issues for other users.
Updates regular expressions to correctly identify labels
defined with "%%| label:" and listings defined with
attributes. This ensures that the script accurately finds
all unreferenced labels in the documentation.
Switches the OpenAI model to "gpt-4o" for potentially improved performance and cost.
Adds a function to strip YAML fence blocks (```yaml ... ```) from the raw output of the OpenAI model. This improves the robustness of the YAML parsing by handling cases where the model includes these fences.
Refines the prompt used for generating revised section headers for the textbook.
The new prompt includes specific guidelines for conciseness, hierarchical awareness, consistent tone, and formatting.
It also emphasizes the importance of evaluating all headers and provides clear instructions for the expected output format.
Migrates to the new OpenAI client library for improved compatibility
and to align with the latest API standards.
This change replaces the deprecated `openai.ChatCompletion.create` method
with `client.chat.completions.create`, ensuring the script remains
functional with future OpenAI API updates.
Refactors the `fixtitle.py` script to accept file or directory input via command-line arguments, enhancing usability and flexibility. Includes argument validation and usage instructions for improved user experience.
The `header_update.py` script is updated to allow processing of single files in addition to directories and includes validation. Code is factored into functions for better readability and maintainability.
Addresses a typo in the privacy_security file.
Updates the regex for identifying labels in the
find_unreferenced_labels.py script to handle both
{#fig-xyz and #| label: fig-xyz formats.
Adds a check for unresolved references in the
find_unreferenced_labels.py script.
Updates the find_unreferenced_labels.py script to provide
more informative messages about label and reference status.
The file xiao_esp32s3.qmd has been edited.
The file setup.qmd has been edited.
The file image_classification.qmd has been edited.
The file object_detetion.qmd has been edited.
The file kws.qmd has been edited.
The file motion_classification.qmd has been edited.
Anything defined as @fig @tbl, etc. that is not referenced in the text will raise an error, so there are no figures or sections or tables, etc., that are just dangling.
