Feast
Important Capabilities
Capability | Status | Notes |
---|---|---|
Descriptions | ✅ | Enabled by default. |
Detect Deleted Entities | ✅ | Enabled by default via stateful ingestion. |
Schema Metadata | ✅ | Enabled by default. |
Table-Level Lineage | ✅ | Enabled by default. |
This plugin extracts:
- Entities as
MLPrimaryKey
- Fields as
MLFeature
- Feature views and on-demand feature views as
MLFeatureTable
- Batch and stream source details as
Dataset
- Column types associated with each entity and feature
CLI based Ingestion
Starter Recipe
Check out the following recipe to get started with ingestion! See below for full configuration options.
For general pointers on writing and running a recipe, see our main recipe guide.
source:
type: feast
config:
# Coordinates
path: "/path/to/repository/"
# Options
environment: "PROD"
sink:
# sink configs
Config Details
- Options
- Schema
Note that a .
is used to denote nested fields in the YAML recipe.
Field | Description |
---|---|
path ✅ string | Path to Feast repository |
enable_owner_extraction boolean | If this is disabled, then we NEVER try to map owners. If this is enabled, then owner_mappings is REQUIRED to extract ownership. Default: False |
enable_tag_extraction boolean | If this is disabled, then we NEVER try to extract tags. Default: False |
environment string | Environment to use when constructing URNs Default: PROD |
fs_yaml_file One of string(path), null | Path to the feature_store.yaml file used to configure the feature store Default: None |
owner_mappings One of array, null | Mapping of owner names to owner types Default: None |
owner_mappings.map map(str,string) | |
stateful_ingestion One of StatefulIngestionConfig, null | Stateful Ingestion Config Default: None |
stateful_ingestion.enabled boolean | Whether or not to enable stateful ingest. Default: True if a pipeline_name is set and either a datahub-rest sink or datahub_api is specified, otherwise False Default: False |
The JSONSchema for this configuration is inlined below.
{
"$defs": {
"StatefulIngestionConfig": {
"additionalProperties": false,
"description": "Basic Stateful Ingestion Specific Configuration for any source.",
"properties": {
"enabled": {
"default": false,
"description": "Whether or not to enable stateful ingest. Default: True if a pipeline_name is set and either a datahub-rest sink or `datahub_api` is specified, otherwise False",
"title": "Enabled",
"type": "boolean"
}
},
"title": "StatefulIngestionConfig",
"type": "object"
}
},
"properties": {
"stateful_ingestion": {
"anyOf": [
{
"$ref": "#/$defs/StatefulIngestionConfig"
},
{
"type": "null"
}
],
"default": null,
"description": "Stateful Ingestion Config"
},
"path": {
"description": "Path to Feast repository",
"title": "Path",
"type": "string"
},
"fs_yaml_file": {
"anyOf": [
{
"format": "path",
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"description": "Path to the `feature_store.yaml` file used to configure the feature store",
"title": "Fs Yaml File"
},
"environment": {
"default": "PROD",
"description": "Environment to use when constructing URNs",
"title": "Environment",
"type": "string"
},
"owner_mappings": {
"anyOf": [
{
"items": {
"additionalProperties": {
"type": "string"
},
"type": "object"
},
"type": "array"
},
{
"type": "null"
}
],
"default": null,
"description": "Mapping of owner names to owner types",
"title": "Owner Mappings"
},
"enable_owner_extraction": {
"default": false,
"description": "If this is disabled, then we NEVER try to map owners. If this is enabled, then owner_mappings is REQUIRED to extract ownership.",
"title": "Enable Owner Extraction",
"type": "boolean"
},
"enable_tag_extraction": {
"default": false,
"description": "If this is disabled, then we NEVER try to extract tags.",
"title": "Enable Tag Extraction",
"type": "boolean"
}
},
"required": [
"path"
],
"title": "FeastRepositorySourceConfig",
"type": "object"
}
Code Coordinates
- Class Name:
datahub.ingestion.source.feast.FeastRepositorySource
- Browse on GitHub
Questions
If you've got any questions on configuring ingestion for Feast, feel free to ping us on our Slack.
Is this page helpful?