Lightlines
By nvidia

tilegym-converting-cutile-to-julia

nvidia1.5k installs

Turn Python GPU kernels into tested Julia code.

What it does

Converts Python cuTile GPU kernels into standalone Julia cuTile.jl implementations with Julia-native tests.

When to use it

Use when converting cuTile Python kernels to Julia cuTile.jl or debugging and optimizing existing Julia translations.

How to use it

Give it a project containing a Python cuTile kernel; it writes the Julia implementation and tests, validates the translation, and runs the tests.

What you provide

  • An existing project

Uses

Latest review

What you need · 4

Requires a Julia version compatible with the [compat] julia declaration in the project's julia/Project.toml.

Requires Python to run the bundled static validator.


Access · 3

This skill

https://julialang.org/install/

Read

Reads Julia installation instructions

Julia

Execute

Installs a compatible Julia version