🕘 Meeting notes: Machine Learning WG

Dec 2, 2019


@Noah B 
@Ryan A 
@Joe H 
@Tom 7
@Sebastian A 

Quick update:
  • Noah: AGU next week, presentations on ML for climate modeling on Thursday
  • Ryan: 
  • Changes to data catalogs, turning off free egress (requester pays)
  • Still working on ML environment in the cloud
  • Tom: gcsfs now supports requester pays and fallback project auth. 
  • Sebastian: new data scientist at UW, a bit of ML background, going to work on GPUs on the cloud
  • Joe
  • ML Cloud Environment
  • ML for met disaggregation: Looking into best practices with packaging trained ML models in a python api.
  • Tom pointed us to ONNX which looks promising.


  • AGU ”ML for climate modeling” session at 2:20PM on Thursday Dec 12, 13:40-15:40
  • Workflows and pipelines in k8s:
  • Would be cool to try dask-gateway on kubeflow

Action Item

  • Thank Martin Durant? about gcsfs

Nov 4, 2019


@mention yourself and add others
@Joe H Joe Hamman
@Noah B  Noah Brenowitz
@Jbednar Jim Bednar
@David G  David Gagne
@Stephan R 


  • Benchmark datasets (Stephan)
  • (maybe) Hydrologic model replacement efforts at USGS, Alabama, NCAR, etc. (Joe and/or Jeff)


  • Quick update
  • Stephan, weather benchmarks
  • benchmark of state to state simulation for weather forecasting