Skip to content

feast-cli-commands.md

Latest commit

 

History

History
556 lines (411 loc) · 20.4 KB

feast-cli-commands.md

File metadata and controls

556 lines (411 loc) · 20.4 KB
Raw

Feast CLI reference

Overview

The Feast CLI comes bundled with the Feast Python package. It is immediately available after installing Feast.

Usage: feast [OPTIONS] COMMAND [ARGS]...

  Feast CLI

  For more information, see our public docs at https://docs.feast.dev/

Options:
  -c, --chdir TEXT  Switch to a different feature repository directory before
                    executing the given subcommand.

  --help            Show this message and exit.

Commands:
  apply                    Create or update a feature store deployment
  configuration            Display Feast configuration
  delete                   Delete a Feast object from the registry
  demo-notebooks           Generate demo Jupyter notebooks for the project
  entities                 Access entities
  feature-views            Access feature views
  init                     Create a new Feast repository
  materialize              Run a (non-incremental) materialization job to...
  materialize-incremental  Run an incremental materialization job to ingest...
  permissions              Access permissions
  registry                 Manage the feature registry
  registry-dump            Print contents of the metadata registry
  teardown                 Tear down deployed feature store infrastructure
  version                  Display Feast SDK version

Global Options

The Feast CLI provides the following global options. Place them before the subcommand.

chdir (-c, --chdir)

This command allows users to run Feast CLI commands in a different folder from the current working directory.

feast -c path/to/my/feature/repo apply

The repository directory can also be set with the FEATURE_REPO_DIR_ENV_VAR environment variable. An explicit --chdir option takes precedence.

feature-store-yaml (-f, --feature-store-yaml)

Use a specific YAML configuration file instead of the default feature_store.yaml in the feature repository. Pass the path to the file, not its containing directory.

feast -c path/to/my/feature/repo -f path/to/config/production.yaml configuration

An explicit relative --feature-store-yaml path is resolved from the current working directory, independently of --chdir. The FEAST_FS_YAML_FILE_PATH environment variable provides a fallback configuration file path when --feature-store-yaml is omitted.

log-level (--log-level)

Set the logging level to DEBUG, INFO, WARNING, ERROR, or CRITICAL (case-insensitive). The default is warning.

feast --log-level info -c path/to/my/feature/repo apply

Apply

Creates or updates a feature store deployment

feast apply

Options:

  • --skip-source-validation: Skip validation of data sources (don't check if tables exist)
  • --skip-feature-view-validation: Skip validation of feature views. Use with caution as this skips important checks
# Skip only data source validation
feast apply --skip-source-validation

# Skip only feature view validation
feast apply --skip-feature-view-validation

# Skip both validations
feast apply --skip-source-validation --skip-feature-view-validation

What does Feast apply do?

  1. Feast will scan Python files in your feature repository and find all Feast object definitions, such as feature views, entities, and data sources.
  2. Feast will validate your feature definitions (e.g. for uniqueness of features). This validation can be skipped using the --skip-feature-view-validation flag if the type/validation system is being overly strict.
  3. Feast will sync the metadata about Feast objects to the registry. If a registry does not exist, then it will be instantiated. The standard registry is a simple protobuf binary file that is stored on disk (locally or in an object store).
  4. Feast CLI will create all necessary feature store infrastructure. The exact infrastructure that is deployed or configured depends on the provider configuration that you have set in feature_store.yaml. For example, setting local as your provider will result in a sqlite online store being created.

{% hint style="info" %} The --skip-feature-view-validation flag is particularly useful for On-Demand Feature Views (ODFVs) with complex transformations that may fail validation. However, use it with caution and please report any validation issues to the Feast team on GitHub. {% endhint %}

{% hint style="warning" %} feast apply (when configured to use cloud provider like gcp or aws) will create cloud infrastructure. This may incur costs. {% endhint %}

{% hint style="info" %} Important: feast apply only registers or updates objects found in your Python files. It does not delete objects that you've removed from your code. To delete objects from the registry, you must use the feast delete command or explicit delete methods in the Python SDK. See the Delete command below and the Registry documentation for details. {% endhint %}

Configuration

Display the repository configuration as YAML, including defaults populated when the configuration is loaded. The output is not an exhaustive list of effective defaults: fields that have not been set are omitted. For example, entity_key_serialization_version can be absent from the output when it is omitted from feature_store.yaml, even though Feast uses its default value of 3.

feast configuration

For a repository that explicitly configures the following values, an excerpt of the output looks like:

project: foo
registry: data/registry.db
provider: local
online_store:
    type: sqlite
    path: data/online_store.db
offline_store: 
    type: dask 
entity_key_serialization_version: 3
auth:
    type: no_auth

Delete

Delete a Feast object from the registry by its name.

feast delete <OBJECT_NAME>

What does feast delete do?

The feast delete command removes a Feast object (such as a feature view, entity, data source, feature service, etc.) from the registry. The command will:

  1. Search for the object by name across all object types (entities, feature views, feature services, data sources, saved datasets, validation references, etc.)
  2. Delete the first matching object found
  3. Remove any associated infrastructure

Example:

# Delete a feature view named "driver_hourly_stats"
feast delete driver_hourly_stats

# Delete an entity named "driver"
feast delete driver

{% hint style="warning" %} The delete operation is permanent and will remove the object from the registry. Make sure you want to delete the object before running this command. {% endhint %}

{% hint style="info" %} If multiple objects have the same name across different types, feast delete will delete the first one it finds. For programmatic deletion with more control, use the Python SDK methods like store.delete_feature_view(), store.delete_feature_service(), etc. {% endhint %}

Demo Notebooks

Generate tailored demo Jupyter notebooks for each Feast project found in the current directory.

feast demo-notebooks

The command searches for feature_store.yaml in the current directory and every file inside the feast-config/ directory. Each file is treated as a separate project config, and notebooks are created under ./feast-demo-notebooks/<project>/.

The generated notebooks adapt to your project configuration (online/offline store types, authentication, vector search) and cover:

  • Feature store overview — explore registered entities, feature views, and services.
  • Historical feature retrieval — build training datasets with point-in-time correct joins.
  • Online feature serving — materialize features and retrieve them at low latency.

Options:

  • -o, --output-dir — Directory where the notebooks are written. Default: ./feast-demo-notebooks.
  • --overwrite — Overwrite existing notebooks if the output directory already exists.
feast demo-notebooks -o ./my-notebooks --overwrite

You can also use the --chdir global option to point at a different feature repository:

feast -c /path/to/feature_repo demo-notebooks

The same functionality is available via the Python SDK:

from feast import copy_demo_notebooks

copy_demo_notebooks(output_dir="./feast-demo-notebooks", repo_path=".")

For more details see the Demo Notebooks tutorial.

Entities

List all registered entities

feast entities list

Options:
  --tags TEXT  Filter by tags (e.g. --tags 'key:value' --tags 'key:value,
               key:value, ...'). Items return when ALL tags match.

NAME       DESCRIPTION    TYPE
driver_id  driver id      ValueType.INT64

Feature views

List all registered feature views

feast feature-views list

Options:
  --tags TEXT  Filter by tags (e.g. --tags 'key:value' --tags 'key:value,
               key:value, ...'). Items return when ALL tags match.

NAME                 ENTITIES    TYPE
driver_hourly_stats  {'driver'}  FeatureView

List version history for a feature view

feast feature-views list-versions FEATURE_VIEW_NAME

VERSION  TYPE          CREATED              VERSION_ID
v0       feature_view  2024-01-15 10:30:00  a1b2c3d4-...
v1       feature_view  2024-01-16 14:22:00  e5f6g7h8-...

Init

Creates a new feature repository

feast init my_repo_name

Creating a new Feast repository in /projects/my_repo_name.

.
├── data
│   └── driver_stats.parquet
├── example.py
└── feature_store.yaml

It's also possible to use other templates

feast init -t gcp my_feature_repo

or to set the name of the new project

feast init -t gcp my_feature_repo

Materialize

Load data from feature views into the online store.

With timestamps:

feast materialize 2020-01-01T00:00:00 2022-01-01T00:00:00

Without timestamps (materializes all available data):

feast materialize --disable-event-timestamp

Load data for specific feature views:

feast materialize -v driver_hourly_stats 2020-01-01T00:00:00 2022-01-01T00:00:00

feast materialize --disable-event-timestamp -v driver_hourly_stats

The --disable-event-timestamp flag lets you materialize all available data (from 1970-01-01 up to the current UTC time) without specifying start and end timestamps. Rows keep the event timestamps from the source.

Materializing 1 feature views from 2020-01-01 to 2022-01-01

driver_hourly_stats:
100%|██████████████████████████| 5/5 [00:00<00:00, 5949.37it/s]

Materialize incremental

For regular (non-on-demand) feature views, load data into the online store beginning from each feature view's most recent materialization end date.

For a regular feature view with no previous materialization, the start date is the current time minus its ttl. A zero TTL uses a lookback of 52 weeks; a feature view with no TTL must first be materialized with feast materialize to establish a start date. The initial lookback is relative to the current time, rather than the supplied end timestamp.

feast materialize-incremental 2022-01-01T00:00:00

Permissions

List permissions

List all registered permission

feast permissions list

Options:
  --tags TEXT    Filter by tags (e.g. --tags 'key:value' --tags 'key:value,
                 key:value, ...'). Items return when ALL tags match.
  -v, --verbose  Print the resources matching each configured permission

+-----------------------+-------------+-----------------------+-----------+----------------+-------------------------+
| NAME                  | TYPES       | NAME_PATTERNS         | ACTIONS   | ROLES          | REQUIRED_TAGS           |
+=======================+=============+=======================+===========+================+================+========+
| reader_permission1234 | FeatureView | transformed_conv_rate | DESCRIBE  | reader         | -                       |
|                       |             | driver_hourly_stats   | DESCRIBE  | reader         | -                       |
+-----------------------+-------------+-----------------------+-----------+----------------+-------------------------+
| writer_permission1234 | FeatureView | transformed_conv_rate | CREATE    | writer         | -                       |
+-----------------------+-------------+-----------------------+-----------+----------------+-------------------------+
| special               | FeatureView | special.*             | DESCRIBE  | admin          | test-key2 : test-value2 |
|                       |             |                       | UPDATE    | special-reader | test-key : test-value   |
+-----------------------+-------------+-----------------------+-----------+----------------+-------------------------+

verbose option describes the resources matching each configured permission:

feast permissions list -v

Permissions:

permissions
├── reader_permission1234 ['reader']
│   └── FeatureView: none
└── writer_permission1234 ['writer']
    ├── FeatureView: none
    │── OnDemandFeatureView: ['transformed_conv_rate_fresh', 'transformed_conv_rate']
    └── BatchFeatureView: ['driver_hourly_stats', 'driver_hourly_stats_fresh']

Describe a permission

Describes the provided permission

feast permissions describe permission-name
name: permission-name
types:
- FEATURE_VIEW
namePattern: transformed_conv_rate
requiredTags:
  required1: required-value1
  required2: required-value2
actions:
- DESCRIBE
policy:
  roleBasedPolicy:
    roles:
    - reader
tags:
  key1: value1
  key2: value2

Permission check

The permissions check command is used to identify resources that lack the appropriate permissions based on their type, name, or tags.

This command is particularly useful for administrators when roles, actions, or permissions have been modified or newly configured. By running this command, administrators can easily verify which resources and actions are not protected by any permission configuration, ensuring that proper security measures are in place.

> feast permissions check


The following resources are not secured by any permission configuration:
NAME                         TYPE
driver                       Entity
driver_hourly_stats_fresh    FeatureView
The following actions are not secured by any permission configuration (Note: this might not be a security concern, depending on the used APIs):
NAME                         TYPE                 UNSECURED ACTIONS
driver                       Entity               CREATE
                                                  DESCRIBE
                                                  UPDATE
                                                  DELETE
                                                  READ_ONLINE
                                                  READ_OFFLINE
                                                  WRITE_ONLINE
                                                  WRITE_OFFLINE
driver_hourly_stats_fresh    FeatureView          CREATE
                                                  DESCRIBE
                                                  UPDATE
                                                  DELETE
                                                  READ_ONLINE
                                                  READ_OFFLINE
                                                  WRITE_ONLINE
                                                  WRITE_OFFLINE

Based on the above results, the administrator can reassess the permissions configuration and make any necessary adjustments to meet their security requirements.

If no resources are accessible publicly, the permissions check command will return the following response:
> feast permissions check
The following resources are not secured by any permission configuration:
NAME    TYPE
The following actions are not secured by any permission configuration (Note: this might not be a security concern, depending on the used APIs):
NAME    TYPE    UNSECURED ACTIONS

List of the configured roles

List all the configured roles

feast permissions list-roles

Options:
  --verbose Print the resources and actions permitted to each configured
            role

ROLE NAME
admin
reader
writer

verbose option describes the resources and actions permitted to each managed role:

feast permissions list-roles -v

ROLE NAME          RESOURCE NAME               RESOURCE TYPE    PERMITTED ACTIONS
admin              driver_hourly_stats_source  FileSource       CREATE
                                                                DELETE
                                                                QUERY_OFFLINE
                                                                QUERY_ONLINE
                                                                DESCRIBE
                                                                UPDATE
admin              vals_to_add                 RequestSource    CREATE
                                                                DELETE
                                                                QUERY_OFFLINE
                                                                QUERY_ONLINE
                                                                DESCRIBE
                                                                UPDATE
admin              driver_stats_push_source    PushSource       CREATE
                                                                DELETE
                                                                QUERY_OFFLINE
                                                                QUERY_ONLINE
                                                                DESCRIBE
                                                                UPDATE
admin              driver_hourly_stats_source  FileSource       CREATE
                                                                DELETE
                                                                QUERY_OFFLINE
                                                                QUERY_ONLINE
                                                                DESCRIBE
                                                                UPDATE
admin              vals_to_add                 RequestSource    CREATE
                                                                DELETE
                                                                QUERY_OFFLINE
                                                                QUERY_ONLINE
                                                                DESCRIBE
                                                                UPDATE
admin              driver_stats_push_source    PushSource       CREATE
                                                                DELETE
                                                                QUERY_OFFLINE
                                                                QUERY_ONLINE
                                                                DESCRIBE
                                                                UPDATE
reader             driver_hourly_stats         FeatureView      DESCRIBE
reader             driver_hourly_stats_fresh   FeatureView      DESCRIBE
...

Registry

create-schema

Pre-create the SQL registry schema so the application does not need DDL privileges at runtime. Use this with schema_mode: verify or schema_mode: skip in your feature_store.yaml.

feast registry create-schema

This command only applies to SQL-based registries (registry_type: sql). It is safe to run multiple times — existing tables are not modified.

MLflow

MLflow integration utilities for GenAI dataset sync and DataSource validation. Requires pip install 'feast[mlflow]'. See MLflow Integration.

# Sync MLflow GenAI EvaluationDataset → FeatureView (via MlflowDatasetSource)
feast mlflow sync-dataset --feature-view mlflow_labels

# Preview flattened dataset records
feast mlflow preview-dataset --source agent-feedback-v3 --limit 10

# Validate an MlflowDatasetSource-backed FeatureView
feast mlflow validate-source eval_records

# List all FeatureViews backed by MlflowDatasetSource
feast mlflow list-sources

Teardown

Tear down deployed feature store infrastructure

feast teardown

Version

Print the current Feast version

feast version