avast / wanna-ml

Complete MLOps framework for Vertex-AI
MIT License
17 stars 3 forks source link

[FeatureRequest]: extend the docs #111

Open racinmat opened 3 months ago

racinmat commented 3 months ago

Contact Details

No response

Is your feature request related to a problem? Please describe

it's non-trivial to figure out the correct version and os for the vm_image add more examples e.g.

    environment:
     vm_image:
       framework: common
       version: cpu
       os: v20240306-debian-11-py310

leads to google.api_core.exceptions.NotFound: 404 The resource 'projects/deeplearning-platform-release/global/images/family/common-cpu-notebooks-v20240306-debian-11-py310' was not found and I don't know why it's wrong, if I should use the first or third column from the list

Describe the solution you'd like

more examples like

    environment:
     vm_image:
       framework: pytorch
       version: 1-9-xla
       os: debian-10

and explaining how this maps to list of images listed by gcloud compute images list --project deeplearning-platform-release | grep notebooks, e.g.

common-cpu-notebooks-20200229                                  deeplearning-platform-release  common-cpu-notebooks-scheduler                                  READY
common-cpu-notebooks-v20201230-debian-9                        deeplearning-platform-release  common-cpu-notebooks-debian-9                                   READY
common-cpu-notebooks-v20211105-ubuntu-1804                     deeplearning-platform-release  common-cpu-notebooks-ubuntu-1804                                READY
common-cpu-notebooks-v20240306                                 deeplearning-platform-release  common-cpu-notebooks                                            READY
common-cpu-notebooks-v20240306-debian-11                       deeplearning-platform-release  common-cpu-notebooks-debian-11                                  READY
common-cpu-notebooks-v20240306-debian-11-py310                 deeplearning-platform-release  common-cpu-notebooks-debian-11-py310                            READY
common-cpu-notebooks-v20240306-py310                           deeplearning-platform-release  common-cpu-notebooks-py310                                      READY
common-cpu-notebooks-v20240306-ubuntu-2004                     deeplearning-platform-release  common-cpu-notebooks-ubuntu-2004                                READY
common-cpu-notebooks-v20240306-ubuntu-2004-py310               deeplearning-platform-release  common-cpu-notebooks-ubuntu-2004-py310                          READY
common-cu100-notebooks-20200302                                deeplearning-platform-release  common-cu100-notebooks-scheduler                                READY
common-cu100-notebooks-v20201231-debian-9                      deeplearning-platform-release  common-cu100-notebooks-debian-9                                 READY

Describe alternatives you've considered

using

    environment:
     vm_image:
       framework: common
       version: cpu
       os: debian-11-py310

worked for me, but it was trial and error and I don't know why this one works and the other one not

Additional context

No response