You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Sep 12, 2023. It is now read-only.
* Add cuda/cpu arguments everywhere necessary
* Fix argument in RemovebgServer init function
* Rename device as compute_device
* Update Dockerfile and docker-compose.yml to handle the extra env COMPUTE_DEVICE
* Add an extra argument "compute_device" to force computation on either cuda (like before) or cpu (this is new)
* Update dependencies
* Update dependencies
* Update dependencies
* Fix im_for_trimap_tr being always copied to gpu
* Fix core tests foreground_anchors
* Update README
Sign up for freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Comes from kubeflow/training-operator#1111.
Seems
kubeflow/common
doesn't have this patch yet. We need to update the reconcile logic.The text was updated successfully, but these errors were encountered: