JuliaGPU / JuliaGPU/KernelAbstractions.jl

3d convolutions

Open
#259 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Julia
Stars
523
Forks
88
Avg merge
1d 11h
Merged PRs (30d)
25

Description

Hello I do not see anything that will make it impossible to use this package with 3d convolutions (for example for each voxel in CT scan I want to calculate the mean and standard deviation of its neighbours ... ) yet I can not find a way I looked into NNlib of FLux but I was unable to fuse it with this library I also have done Nvidia Cuda C course so I understand basics of indexing etc. ... still I suppose that there is already good performant and correct way to do it in julia, that I am just unaware of , I would be very gratefull for any help in pointing out how to start it.

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by comparing the 3D convolution needs described here with the relevant APIs in NNlib and Flux, which the report mentions. Determine whether KernelAbstractions already exposes a suitable path for voxel-neighbour calculations; done would require a documented, performant and correct way to perform the requested 3D operation.

Written by the indexing model from the issue text.

Assessment

Tech stack
julia
Domain
backend
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.