#223 · Primary category: Education & Research

NN-SVG

d3 deep-learning diagrams drawing machine-learning neural-network svg threejs

Publication-ready NN-architecture schematics.

Project last updated:06/02/26

GitHub Stars

5.7K

Forks

751

Contributors

12

License

MIT

Why we included this project

Putting together a clean neural network diagram for a paper or talk usually means fighting a vector editor for an hour, and NN-SVG removes that step by generating the figure from parameters in the browser. You choose one of its three drawing styles, the fully connected, LeNet-style convolutional, or AlexNet-style deep network layouts, then tune size, color, and layout until the schematic matches the visual conventions of your field. The result exports as SVG, so lines stay crisp whether the figure ends up in a PDF, a journal page, or a slide. Researchers who have hand-placed boxes and arrows will recognize the relief immediately, and the same controls make the tool a reasonable way to show students how layers connect. Since the output is standard vector graphics, it drops into most publishing workflows without extra cleanup.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category