Will training code be released?

[ozkilim]

Will training code be released?

[NoahVl]

Hey all,

For the "ML Reproducibility Challenge 2021 Fall Edition" we tried to reproduce this paper by training the models from scratch. I have included the repository we made, which includes our attempt at recreating the training code.

As can be seen from our results in the paper, this code is a starting point and might not reproduce the original paper's results out of the box. It should be mentioned that we were fairly bottlenecked by the GPUs available to us, this could be part of the reason why our images don't look stellar. However, it might help you in your efforts to reproduce the StylEx paper and improve on the findings of the paper.

Reproducibility study paper: https://rescience.github.io/bibliography/Vleuten_2022.html
Repo: https://github.com/NoahVl/Explaining-In-Style-Reproducibility-Study

[Johnson-yue]

+1

[kevinbro96]

+2

[hxngiee]

+3

[iDTer]

+4

[Xjmengnieer]

+5

[karray]

+6

[tarunkul]

+7

[snassimr]

+8

[evolu8]

+9

[Kwanseok-K]

+10

[qiminchen]

+11

[QWERDF007]

+12

[agenoves]

+13

[shang106121]

+14

[huuquan1994]

+15

[yenjia]

+16

[MichaelDoron]

+17

[jpainam]

+18

[ThibaultLSDC]

+19

[sgadgil6]

+20

[Anis-bourou]

+21

[szy96]

+22