FluxML / FluxML/Torch.jl

Feature request: 3D conv

Open
#31 1 comment 2 reactions 0 assignees View on GitHub
Dominant language
Julia
Stars
240
Forks
18
PR merge metrics
No merged PRs in 30d

Description

Right now does not seem to support 3D convolutions. I tried to convert my model with a bunch of conv3D layers and I get

`
ERROR: LoadError: "expected output_padding to be a single integer value or a list of 3 values to match the convolution dimensions, but got output_padding=[0, 0]
`

Contributor guide

No contributing guide indexed for this repository

Research direction

Start by tracing how convolution layers are converted and how output_padding is built for different convolution dimensions. Reproduce the reported conversion with a model containing Conv3D layers. Done means 3D convolutions convert without the output_padding dimension error.

Written by the indexing model from the issue text.

Assessment

Tech stack
julia, pytorch
Domain
machine-learning
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.