metalcompute 0.1.0

A python library to run metal compute kernels on MacOS

metalcompute for Python

A python library to run metal compute kernels on MacOS 12.0 (Monterey)

Installations

Install latest stable release from PyPI:

> python3 -m pip install metalcompute

Install latest unstable version from Github:

> python3 -m pip install git+https://github.com/baldand/py-metal-compute.git

Install locally from source:

> python3 -m pip install -e .

Basic test

Example execution from M1-based Mac running MacOS 12.0:

> python3 tests/basic.py
Calculating sin of 1234567 values
Expected value: 0.9805107116699219 Received value: 0.9807852506637573
Metal compute took: 0.0040209293365478516 s
Reference compute took: 0.1068720817565918 s

Interface

import metalcompute as mc

mc.init() 
# Call before use

mc.compile(program, function_name)
# Will raise exception with details if metal kernel has errors

mc.run(input_f32_or_u8_array, output_f32_or_u8_array, kernel_call_count)
# Run the kernel once with supplied input data, 
# filling supplied output data
# Specify number of kernel calls

mc.release()
# Call after use

Examples

Measure TFLOPS of GPU

> metalcompute-measure-flops
Running compute intensive Metal kernel to measure TFLOPS...
Estimated GPU TFLOPS: 2.50825

Render a 3D image with raymarching

# Usage: metalcompute-raymarch [<width> <height> [<output image file: PNG, JPG>]]

> metalcompute-raymarch.py 1024 1024 raymarch.jpg
Render took 0.0119569s

Mandelbrot set

# Usage: metalcompute-mandelbrot [<width> <height> [<output image file: PNG, JPG>]]

> metalcompute-mandelbrot
Rendering mandelbrot set using Metal compute, res:4096x4096, iters:8192
Render took 0.401446s
Writing image to mandelbrot.png
Image encoding took 1.35182s

Status

This is an early preview version.