Feast
ML platformfeast
Open source feature store serving the same features to training and online inference.
- Provider
- default
- Kind
- ML platform
- Category
- Compute
- Official docs
- https://docs.feast.dev/getting-started/quickstart
| property | type | values | required | kind default | this component |
|---|---|---|---|---|---|
| computeInstances | integer | — | no | — | inherited |
| experimentTracking | boolean | — | no | true | ▸ false |
| gpuEnabled | boolean | — | no | — | inherited |
| managed | boolean | — | no | false | ▸ false |
| modelRegistry | boolean | — | no | true | ▸ false |
| offlineStore | enum | file | bigquery | redshift | snowflake | spark | no | file | inherited |
| onlineStore | enum | sqlite | redis | dynamodb | datastore | postgres | no | sqlite | inherited |
| provider | enum | local | aws | gcp | azure | no | local | inherited |
| publicAccess | boolean | — | no | — | inherited |
| registryBackend | enum | file | sql | no | file | inherited |
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-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
nodes:
- { id: "feast", component: "feast" }Platform for AIalibaba-paiAWS Clean Rooms MLaws-clean-rooms-mlAmazon SageMaker AIaws-sagemakerAmazon SageMaker Canvasaws-sagemaker-canvasAmazon SageMaker Feature Storeaws-sagemaker-feature-storeManaged MLflow on Amazon SageMaker AIaws-sagemaker-mlflowAmazon SageMaker Unified Studioaws-sagemaker-unified-studioMicrosoft Foundryazure-ai-foundryAzure Machine Learningazure-machine-learningAbacus.AIabacus-aiAlteryxalteryxArgillaargillaC3 AI Platformc3-aiClearMLclearmlCometcomet-mlCVATcvatDataikudataikuDataRobotdatarobotDatasaurdatasaurDetermined AIdeterminedDomino Data Labdomino-data-labDVCdvcEncordencordFlyteflyteH2O.aih2oKilikiliKNIMEknimeKubeflowkubeflowLabel Studiolabel-studioLabelboxlabelboxMetaflowmetaflowGeneric ML platformml-platformMLflowmlflowneptune.aineptune-aiNVIDIA DGX Cloudnvidia-dgx-cloudPachydermpachydermPolyaxonpolyaxonProdigyprodigyRapidMiner AI HubrapidminerRoboflowroboflow
and 21 more → ML platform page# yaml-language-server: $schema=../../schema/component.schema.json # Feast: open source feature store serving the same features to training and to online inference. # Defaults from "Feast quickstart": `feast init` writes provider local, a SQLite online store and a # file (parquet) offline store, with the registry in a local file. schemaVersion: "1" type: "component" id: "feast" kind: "ml-platform" provider: "default" docs: "https://docs.feast.dev/getting-started/quickstart" defaults: experimentTracking: false modelRegistry: false managed: false properties: provider: type: "enum" values: ["local", "aws", "gcp", "azure"] default: "local" onlineStore: type: "enum" values: ["sqlite", "redis", "dynamodb", "datastore", "postgres"] default: "sqlite" offlineStore: type: "enum" values: ["file", "bigquery", "redshift", "snowflake", "spark"] default: "file" registryBackend: type: "enum" values: ["file", "sql"] default: "file"