Ray 1.12.1

1.12.1

Ray-1.12.1

Added 1
  • Resume building ray-ml Docker images for CPU after they were stopped in Ray 1.9
Fixed 5
  • Fix bug with memory limit fetching when running Ray in a container, allowing Ray to work on Google Colab again
  • Fix MLflow run to start under the correct experiment for Ray Train and Ray Tune integrations
  • Fix APPO in eager mode for RLlib
  • Fix Alphastar for TF2 with tracing enabled in RLlib
  • Fix replica leak in anonymous namespaces for Ray Serve

Patch release with the following fixes:

View original

Upgraded? How did it go?

Discussion