FluxML/Flux.jl

Does Flux support N-dimensional convolutions, where N>3?

Open

#451 opened on Oct 22, 2018

View on GitHub
 (20 comments) (0 reactions) (0 assignees)Julia (4,725 stars) (619 forks)batch import
enhancementhelp wanted

Description

While this works:

Conv((1,1,1), 1=>1)(rand(1,1,1,1,1)) #3D convolution

This throws an error:

Conv((1,1,1,1), 1=>1)(rand(1,1,1,1,1,1)) #4D convolution
#ERROR: MethodError: no method matching conv!(::Array{Float64,6}, ::Array{Float64,6}, ::Array{Float64,6}; pad=(0, 0, 0, 0), stride=(1, 1, 1, 1), dilation=(1, 1, 1, 1))

I'm running Julia 0.7 and:

[587475ba] Flux v0.6.7+ #master (https://github.com/FluxML/Flux.jl.git)
[872c559c] NNlib v0.4.2+ #master (https://github.com/FluxML/NNlib.jl.git)

Contributor guide

Tech stack
None
Domain
machine learning
Issue type
feature
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
4
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
3-5 days
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
stale
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
clear
Prerequisites
JuliaFluxconvolution operations
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
30
Research direction
Investigate the conv! function in Flux and NNlib to understand why it only supports up to 3D convolutions. The error indicates no method matching conv! for 6D arrays. Extending the convolution to arbitrary N dimensions would require modifying the underlying implementation to handle dynamic kernel sizes and padding. Look at the existing code for 3D convolutions and generalize it, possibly leveraging broadcast or loop structures. Check if there are any open PRs or discussions in the issue comments for guidance.