Zeema
New design
Catalogue / Compute

ML platform

computeml-platform

A training workspace with compute targets, datasets and a model registry.

Category
Compute
Properties
6
Components
62
Rules
1
Properties · 6 defined by this kind

A node of this kind resolves each property from its own props, then the component's defaults, then the default below.

propertytypevaluesrequireddefault
computeInstancesinteger—no—
experimentTrackingboolean—notrue
gpuEnabledboolean—no—
managedboolean—nofalse
modelRegistryboolean—notrue
publicAccessboolean—no—
Rules for this kind · 1
Stateful node without zone redundancyerror
stateful-node-without-zone-redundancy

A stateful compute node pinned to exactly one availability zone disappears with that zone, and its state goes with it. Nodes inside a cell are exempt, because there the cell is deliberately single-zone and redundancy comes from the other cells.

  • — all of these hold:
  • — the node's `stateless` is false
  • — the node sits inside exactly 1 availability-zone group
  • — this does not hold:
  • — the node sits inside at least one cell group
Components · 62
vendor-neutral
Generic ML platformml-platformgenericthe kind itself
alibabacloud
Platform for AIalibaba-pai1 defaults
aws
AWS Clean Rooms MLaws-clean-rooms-ml4 defaultsAmazon SageMaker AIaws-sagemaker4 defaultsAmazon SageMaker Canvasaws-sagemaker-canvas3 defaultsAmazon SageMaker Feature Storeaws-sagemaker-feature-store3 defaultsManaged MLflow on Amazon SageMaker AIaws-sagemaker-mlflow3 defaultsAmazon SageMaker Unified Studioaws-sagemaker-unified-studio3 defaults
azure
Microsoft Foundryazure-ai-foundry4 defaultsAzure Machine Learningazure-machine-learning3 defaults
default
Abacus.AIabacus-ai3 defaultsAlteryxalteryx3 defaultsArgillaargilla3 defaultsC3 AI Platformc3-ai3 defaultsClearMLclearml3 defaultsCometcomet-ml3 defaultsCVATcvat3 defaultsDataikudataiku3 defaultsDataRobotdatarobot3 defaultsDatasaurdatasaur3 defaultsDetermined AIdetermined4 defaultsDomino Data Labdomino-data-lab3 defaultsDVCdvc3 defaultsEncordencord3 defaultsFeastfeast3 defaultsFlyteflyte3 defaultsH2O.aih2o3 defaultsKilikili3 defaultsKNIMEknime3 defaultsKubeflowkubeflow3 defaultsLabel Studiolabel-studio3 defaultsLabelboxlabelbox3 defaultsMetaflowmetaflow2 defaultsMLflowmlflow3 defaultsneptune.aineptune-ai3 defaultsNVIDIA DGX Cloudnvidia-dgx-cloud2 defaultsPachydermpachyderm3 defaultsPolyaxonpolyaxon3 defaultsProdigyprodigy4 defaultsRapidMiner AI Hubrapidminer3 defaultsRoboflowroboflow4 defaultsScale AIscale-ai3 defaultsSnorkel Flowsnorkel3 defaultsSuperAnnotatesuperannotate3 defaultsSurge AIsurge-ai3 defaultsTolokatoloka3 defaultsUnion.aiunion-ai3 defaultsV7 Darwinv73 defaultsValohaivalohai3 defaultsWeights & Biasesweights-and-biases4 defaultsZenMLzenml3 defaults
digitalocean
Paperspacedigitalocean-paperspace2 defaults
gcp
Colab Enterprisegoogle-colab-enterprise4 defaultsVertex AI Neural Architecture Searchgoogle-neural-architecture-search3 defaultsVertex AIgoogle-vertex-ai3 defaultsVertex AI Feature Storegoogle-vertex-ai-feature-store3 defaultsVertex AI Workbenchgoogle-vertex-ai-workbench5 defaults
ibm
IBM Watson Knowledge Studioibm-knowledge-studio2 defaultsIBM watsonx.aiibm-watsonx-ai1 defaultsIBM watsonx.governanceibm-watsonx-governance3 defaults
oci
OCI Data Labelingoci-data-labeling4 defaultsOCI Data Scienceoci-data-science5 defaults
Source
# yaml-language-server: $schema=../schema/kind.schema.json
# The workspace that produces models, where ml-inference is only the endpoint that serves one. publicAccess and gpuEnabled have no default so exposure and capacity rules can catch a workspace that never stated them.
schemaVersion: "1"
type: "kind"
id: "ml-platform"
category: "compute"
properties:
  computeInstances:
    type: "integer"
  gpuEnabled:
    type: "boolean"
  modelRegistry:
    type: "boolean"
    default: true
  experimentTracking:
    type: "boolean"
    default: true
  publicAccess:
    type: "boolean"
  managed:
    type: "boolean"
    default: false

The same page as markdown, for an agent: /catalogue/kinds/ml-platform.md