pyclesperanto-prototype 0.0.3

OpenCL based GPU-accelerated image processing (an early prototype)

pyclesperanto

pyclesperanto is a prototype for clEsperanto - a multi-platform multi-language framework for GPU-accelerated image procesing. It uses OpenCL kernels from CLIJ

Right now, this is very preliminary.

Installation

• Get a python environment, e.g. via mini-conda
• Install pyopencl and gputools.

If installation of gputools doesn't work because of issues with pyopencl for Windows, consider downloading a precompiled wheel (e.g. from here ) and installing it manually:

pip install pyopencl-2019.1.1+cl12-cp37-cp37m-win_amd64.whl
pip install gputools

Troubleshooting installation

If you receive an error like

DLL load failed: The specified procedure could not be found.

Try downloading and installing a pyopencl with a lower cl version, e.g. cl12 : pyopencl-2020.1+cl12-cp37-cp37m-win_amd64

Example code

An example is available in this script. Basically, you import the methods from clEsperanto you need:

import pyclesperanto_prototype as cle

You can then push an image to the GPU and create memory there:

import numpy as np

# push an array to the GPU
flip = cle.push(np.array([2, 4, 6, 8, 10, 12, 14, 16, 18, 20]))

# create memory for the output
flop = cle.create((10,))

And then you can call methods in the GPU without the need for learning OpenCL:

# add a constant to all pixels
cle.add_image_and_scalar(flip, flop, 100.0)

# print result
print(flop)