Reproduction of DVM and BladeDISC - two ML compiler papers on dynamic tensor shape compilation. Includes benchmark scripts, 81 CSV measurements, 8 publication-quality figures, and a full report.
-
Updated
Jun 12, 2026 - Python
Reproduction of DVM and BladeDISC - two ML compiler papers on dynamic tensor shape compilation. Includes benchmark scripts, 81 CSV measurements, 8 publication-quality figures, and a full report.
A creative project using HTML5 Canvas to generate a complex digital landscape or portrait with interactive elements. The artwork features dynamic shapes, curves, and gradients, ensuring a 3D effect rather than a flat design.
A simple experiment applying compressible flow principles to soften distributed gradient communication stalls.
Add a description, image, and links to the dynamic-shapes topic page so that developers can more easily learn about it.
To associate your repository with the dynamic-shapes topic, visit your repo's landing page and select "manage topics."