Presentation slides in markdown, converted to LaTeX Beamer and then PDF with automatic annotations for equations: all the fun and none of the work! LLM-friendly.
is generated from this:
### Kernel Ridge Regression (KRR)
$$\mathbf{K}_{ij} = {k}(\mathbf{x}_i, \mathbf{x}_j)$$
[[ {k} ]] Kernel function
[[ \mathbf{x}_i, \mathbf{x}_j ]] Features of training points
[[ \mathbf{K}_{ij} ]] Kernel matrix element
$$\mathbf{w} = (\mathbf{K} + \lambda\, \mathbf{I}_N)^{-1}\mathbf{y}$$
[[ \mathbf{K} ]] Kernel matrix ($N\times N$)
[[ \mathbf{w} ]] Model weights
[[ \lambda ]] Regularization
[[ \mathbf{I}_N ]] Identity matrix ($N\times N$)
[[ \mathbf{y} ]] Training labels
$$\hat y(\mathbf{x}_q) = \sum_{i=1}^{ N } w_i {k}(\mathbf{x}_i, \mathbf{x}_q)$$
[[ \hat y ]] Prediction
[[ \mathbf{x}_q ]] Query
[[ w_i ]] Weight of $i$-th training point
[[ \mathbf{x}_i ]] Training point features
##### Title #####- Title page## Section- Section slide### Slide Title- Regular slide### !Hidden- Hidden slide### ?Summary- Summary slide with orange header
-|-- Column break---- Section break within columns
$$ equation $$with[[ term ]] explanation- Annotated equations:::image.pdf: Caption- Images- ````plot: Caption` - Matplotlib plots with axes
- ````schematic: Caption` - Matplotlib diagrams without tick marks
- ````language` - Syntax-highlighted code
| Header |- Tables[1] Text- Numbered footnote[*] Text- Unnumbered footnote// Comment- Ignored># file.md- Include file:icon:- Icons in headings, https://phosphoricons.com/
pip install click matplotlib numpy tqdm cairosvg pygmentsRequirements:
- Python 3.x
- XeLaTeX (via TeX Live or similar)
- latexmk
- Fira Sans font
$ python -m autoslide.cli lecture.md
Parsed 15 slides
Generating 8 figures...
Generating figures: 100%|██████████| 8/8 [00:12<00:00, 1.5s/figure]
Generated lecture-autoslide/lecture.tex
Compiling LaTeX to PDF...
LaTeX compilation successful
PDF copied to lecture.pdfAutoSlide creates an output directory, generates LaTeX, compiles to PDF, and copies the result back.