[P] Data Processing Support in Ray 1.2 (Spark, Dask, Mars, Modin) : MachineLearning

Hey All,

Just wanted to mention that the new version of Ray has data processing and features like:

  • Support for spilling to external storage

  • New support for libraries in the Python data processing ecosystem (Spark, Dask)

You can now use data processing APIs like Spark/Dask DataFrames directly alongside distributed ML libraries on Ray like XGBoost and Horovod.

Let me know what you think!

Source link

We will be happy to hear your thoughts

      Leave a reply

      Enable registration in settings - general