Skip to content

feat: allow to preload ML models when running inference (#224) #487

feat: allow to preload ML models when running inference (#224)

feat: allow to preload ML models when running inference (#224) #487

Triggered via push September 28, 2023 09:54
Status Success
Total duration 47m 21s
Artifacts
This run and associated checks have been archived and are scheduled for deletion. Learn more about checks retention

build.yml

on: push
Matrix: Build and test
Fit to window
Zoom out
Zoom in