Zeema
New design
Catalogue / ml-platform

Vertex AI Neural Architecture Search

gcpML platformgoogle-neural-architecture-search

Managed search over model architectures that runs trials as Vertex AI training jobs.

Provider
gcp
Kind
ML platform
Category
Compute
Official docs
https://cloud.google.com/vertex-ai/docs/training/neural-architecture-search/overview
Also known as
NAS
Properties · 3 set by this component
propertytypevaluesrequiredkind defaultthis component
computeInstancesinteger—no—inherited
experimentTrackingboolean—notrue▸ false
gpuEnabledboolean—no—inherited
managedboolean—nofalse▸ true
modelRegistryboolean—notrue▸ false
parallelTrials · component propertyinteger—no—inherited
proxyTask · component propertyboolean—nofalseinherited
publicAccessboolean—no—inherited
stage · component propertyenumsearch | trainnosearchinherited
Rules that apply · 1

Every rule whose selector covers this component's kind or its category. Whether a rule fires depends on the design: its properties, its edges and its groups.

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
Use it in a design
nodes:
  - { id: "neural-architecture-search", component: "google-neural-architecture-search" }
Other ML platform products · 61
Platform for AIalibaba-paialibabacloudml-platformAWS Clean Rooms MLaws-clean-rooms-mlawsml-platformAmazon SageMaker AIaws-sagemakerawsml-platformAmazon SageMaker Canvasaws-sagemaker-canvasawsml-platformAmazon SageMaker Feature Storeaws-sagemaker-feature-storeawsml-platformManaged MLflow on Amazon SageMaker AIaws-sagemaker-mlflowawsml-platformAmazon SageMaker Unified Studioaws-sagemaker-unified-studioawsml-platformMicrosoft Foundryazure-ai-foundryazureml-platformAzure Machine Learningazure-machine-learningazureml-platformAbacus.AIabacus-aiml-platformAlteryxalteryxml-platformArgillaargillaml-platformC3 AI Platformc3-aiml-platformClearMLclearmlml-platformCometcomet-mlml-platformCVATcvatml-platformDataikudataikuml-platformDataRobotdatarobotml-platformDatasaurdatasaurml-platformDetermined AIdeterminedml-platformDomino Data Labdomino-data-labml-platformDVCdvcml-platformEncordencordml-platformFeastfeastml-platformFlyteflyteml-platformH2O.aih2oml-platformKilikiliml-platformKNIMEknimeml-platformKubeflowkubeflowml-platformLabel Studiolabel-studioml-platformLabelboxlabelboxml-platformMetaflowmetaflowml-platformGeneric ML platformml-platformml-platformMLflowmlflowml-platformneptune.aineptune-aiml-platformNVIDIA DGX Cloudnvidia-dgx-cloudml-platformPachydermpachydermml-platformPolyaxonpolyaxonml-platformProdigyprodigyml-platformRapidMiner AI Hubrapidminerml-platform
and 21 more → ML platform page
Source
# yaml-language-server: $schema=../../schema/component.schema.json
# Vertex AI Neural Architecture Search: a managed search over model architectures that runs trials as
# Vertex AI training jobs ("Neural architecture search overview"). It produces architectures rather
# than registering models or tracking experiments itself.
schemaVersion: "1"
type: "component"
id: "google-neural-architecture-search"
kind: "ml-platform"
provider: "gcp"
docs: "https://cloud.google.com/vertex-ai/docs/training/neural-architecture-search/overview"
aliases: ["NAS"]
defaults:
  managed: true
  modelRegistry: false
  experimentTracking: false
properties:
  stage:
    type: "enum"
    values: ["search", "train"]
    default: "search"
  parallelTrials:
    type: "integer"
  proxyTask:
    type: "boolean"
    default: false

The same page as markdown, for an agent: /catalogue/components/google-neural-architecture-search.md