You are viewing documentation about an older version (1.6.4). View latest version

snowflake.ml.feature_store.FeatureStore

class snowflake.ml.feature_store.FeatureStore(session: Session, database: str, name: str, default_warehouse: str, *, creation_mode: CreationOption = CreationOption.FAIL_IF_NOT_EXIST)

Bases: object

FeatureStore provides APIs to create, materialize, retrieve and manage feature pipelines.

Creates a FeatureStore instance.

Parameters:

  • session – Snowpark Session to interact with Snowflake backend.

  • database – Database to create the FeatureStore instance.

  • name – Target FeatureStore name, maps to a schema in the database.

  • default_warehouse – Default warehouse for feature store compute.

  • creation_mode – If FAIL_IF_NOT_EXIST, feature store throws when required resources not already exist; If CREATE_IF_NOT_EXIST, feature store will create required resources if they not already exist. Required resources include schema and tags. Note database must already exist in either mode.

Raises:

  • SnowflakeMLException – [ValueError] default_warehouse does not exist.

  • SnowflakeMLException – [ValueError] Required resources not exist when mode is FAIL_IF_NOT_EXIST.

  • SnowflakeMLException – [RuntimeError] Failed to find resources.

  • SnowflakeMLException – [RuntimeError] Failed to create feature store.

Example:

>>> from snowflake.ml.feature_store import (
...     FeatureStore,
...     CreationMode,
... )

>>> # Create a new Feature Store:
>>> fs = FeatureStore(
...     session=session,
...     database="MYDB",
...     name="MYSCHEMA",
...     default_warehouse="MYWH",
...     creation_mode=CreationMode.CREATE_IF_NOT_EXIST
... )

>>> # Connect to an existing Feature Store:
>>> fs = FeatureStore(
...     session=session,
...     database="MYDB",
...     name="MYSCHEMA",
...     default_warehouse="MYWH",
...     creation_mode=CreationMode.FAIL_IF_NOT_EXIST
... )

Methods

delete_entity(name: str) → None

Delete a previously registered Entity.

Parameters:

name – Name of entity to be deleted.

Raises:

  • SnowflakeMLException – [ValueError] Entity with given name not exists.

  • SnowflakeMLException – [RuntimeError] Failed to alter schema or drop tag.

  • SnowflakeMLException – [RuntimeError] Failed to find resources.

Example:

>>> fs = FeatureStore(...)
>>> e_1 = Entity("my_entity", ['col_1'], desc='My first entity.')
>>> fs.register_entity(e_1)
>>> fs.list_entities().show()
-----------------------------------------------------------
|"NAME"     |"JOIN_KEYS"  |"DESC"            |"OWNER"     |
-----------------------------------------------------------
|MY_ENTITY  |["COL_1"]    |My first entity.  |REGTEST_RL  |
-----------------------------------------------------------

>>> fs.delete_entity("my_entity")
>>> fs.list_entities().show()
-------------------------------------------
|"NAME"  |"JOIN_KEYS"  |"DESC"  |"OWNER"  |
-------------------------------------------
|        |             |        |         |
-------------------------------------------
delete_feature_view(feature_view: FeatureView) → None
delete_feature_view(feature_view: str, version: str) → None

Delete a FeatureView.

Parameters:

  • feature_view – FeatureView object or name to delete.

  • version – Optional version of feature view. Must set when argument feature_view is a str.

Raises:

SnowflakeMLException – [ValueError] FeatureView is not registered.

Example:

>>> fs = FeatureStore(...)
>>> fv = FeatureView('FV0', ...)
>>> fv1 = fs.register_feature_view(fv, 'FIRST')
>>> fv2 = fs.register_feature_view(fv, 'SECOND')
>>> fs.list_feature_views().select('NAME', 'VERSION').show()
----------------------
|"NAME"  |"VERSION"  |
----------------------
|FV0     |SECOND     |
|FV0     |FIRST      |
----------------------

>>> # delete with name and version
>>> fs.delete_feature_view('FV0', 'FIRST')
>>> fs.list_feature_views().select('NAME', 'VERSION').show()
----------------------
|"NAME"  |"VERSION"  |
----------------------
|FV0     |SECOND     |
----------------------

>>> # delete with feature view object
>>> fs.delete_feature_view(fv2)
>>> fs.list_feature_views().select('NAME', 'VERSION').show()
----------------------
|"NAME"  |"VERSION"  |
----------------------
|        |           |
----------------------
generate_dataset(name: str, spine_df: DataFrame, features: List[Union[FeatureView, FeatureViewSlice]], *, version: Optional[str] = None, spine_timestamp_col: Optional[str] = None, spine_label_cols: Optional[List[str]] = None, exclude_columns: Optional[List[str]] = None, include_feature_view_timestamp_col: bool = False, desc: str = '', output_type: Literal['dataset'] = 'dataset') → Dataset
generate_dataset(name: str, spine_df: DataFrame, features: List[Union[FeatureView, FeatureViewSlice]], *, output_type: Literal['table'], version: Optional[str] = None, spine_timestamp_col: Optional[str] = None, spine_label_cols: Optional[List[str]] = None, exclude_columns: Optional[List[str]] = None, include_feature_view_timestamp_col: bool = False, desc: str = '') → DataFrame

Generate dataset by given source table and feature views.

Parameters:

  • name – The name of the Dataset to be generated. Datasets are uniquely identified within a schema by their name and version.

  • spine_df – Snowpark DataFrame to join features into.

  • features – A list of FeatureView or FeatureViewSlice which contains features to be joined.

  • version – The version of the Dataset to be generated. If none specified, the current timestamp will be used instead.

  • spine_timestamp_col – Name of timestamp column in spine_df that will be used to join time-series features. If spine_timestamp_col is not none, the input features also must have timestamp_col.

  • spine_label_cols – Name of column(s) in spine_df that contains labels.

  • exclude_columns – Name of column(s) to exclude from the resulting training set.

  • include_feature_view_timestamp_col – Generated dataset will include timestamp column of feature view (if feature view has timestamp column) if set true. Default to false.

  • desc – A description about this dataset.

  • output_type – (Deprecated) The type of Snowflake storage to use for the generated training data.

Returns:

If output_type is “dataset” (default), returns a Dataset object. If output_type is “table”, returns a Snowpark DataFrame representing the table.

Raises:

  • SnowflakeMLException – [ValueError] Invalid output_type specified.

  • SnowflakeMLException – [RuntimeError] Dataset name/version already exists.

  • SnowflakeMLException – [RuntimeError] Failed to find resources.

Example:

>>> fs = FeatureStore(session, ...)
>>> # Assume you already have feature view registered.
>>> fv = fs.get_feature_view("MY_FV", "1")
>>> # Spine dataframe has same join keys as the entity of fv.
>>> spine_df = session.create_dataframe(["1", "2"], schema=["id"])
>>> my_dataset = fs.generate_dataset(
...     "my_dataset"
...     spine_df,
...     [fv],
... )
>>> # Current timestamp will be used as default version name.
>>> # You can explicitly overwrite by setting a version.
>>> my_dataset.list_versions()
['2024_07_12_11_26_22']

>>> my_dataset.read.to_snowpark_dataframe().show(n=3)
-------------------------------------------------------
|"QUALITY"  |"FIXED_ACIDITY"     |"VOLATILE_ACIDITY"  |
-------------------------------------------------------
|3          |11.600000381469727  |0.5799999833106995  |
|3          |8.300000190734863   |1.0199999809265137  |
|3          |7.400000095367432   |1.184999942779541   |
-------------------------------------------------------
generate_training_set(spine_df: DataFrame, features: List[Union[FeatureView, FeatureViewSlice]], *, save_as: Optional[str] = None, spine_timestamp_col: Optional[str] = None, spine_label_cols: Optional[List[str]] = None, exclude_columns: Optional[List[str]] = None, include_feature_view_timestamp_col: bool = False) → DataFrame

Generate a training set from the specified Spine DataFrame and Feature Views. Result is materialized to a Snowflake Table if save_as is specified.

Parameters:

  • spine_df – Snowpark DataFrame to join features into.

  • features – A list of FeatureView or FeatureViewSlice which contains features to be joined.

  • save_as – If specified, a new table containing the produced result will be created. Name can be a fully qualified name or an unqualified name. If unqualified, defaults to the Feature Store database and schema

  • spine_timestamp_col – Name of timestamp column in spine_df that will be used to join time-series features. If spine_timestamp_col is not none, the input features also must have timestamp_col.

  • spine_label_cols – Name of column(s) in spine_df that contains labels.

  • exclude_columns – Name of column(s) to exclude from the resulting training set.

  • include_feature_view_timestamp_col – Generated dataset will include timestamp column of feature view (if feature view has timestamp column) if set true. Default to false.

Returns:

Returns a Snowpark DataFrame representing the training set.

Raises:

  • SnowflakeMLException – [RuntimeError] Materialized table name already exists

  • SnowflakeMLException – [RuntimeError] Failed to create materialized table.

Example:

>>> fs = FeatureStore(session, ...)
>>> # Assume you already have feature view registered.
>>> fv = fs.get_feature_view("MY_FV", "1")
>>> # Spine dataframe has same join keys as the entity of fv.
>>> spine_df = session.create_dataframe(["1", "2"], schema=["id"])
>>> training_set = fs.generate_training_set(
...     spine_df,
...     [fv],
...     save_as="my_training_set",
... )
>>> print(type(training_set))
<class 'snowflake.snowpark.table.Table'>

>>> print(training_set.queries)
{'queries': ['SELECT  *  FROM (my_training_set)'], 'post_actions': []}
get_entity(name: str) → Entity

Retrieve previously registered Entity object.

Parameters:

name – Entity name.

Returns:

Entity object.

Raises:

  • SnowflakeMLException – [ValueError] Entity is not found.

  • SnowflakeMLException – [RuntimeError] Failed to retrieve tag reference information.

  • SnowflakeMLException – [RuntimeError] Failed to find resources.

Example:

>>> fs = FeatureStore(...)
>>> # e_1 is a local object that hasn't registered to Snowflake backend yet.
>>> e_1 = Entity("my_entity", ['col_1'], desc='My first entity.')
>>> fs.register_entity(e_1)

>>> # e_2 is a local object that points a backend object in Snowflake.
>>> e_2 = fs.get_entity("my_entity")
>>> print(e_2)
Entity(name=MY_ENTITY, join_keys=['COL_1'], owner=REGTEST_RL, desc=My first entity.)
get_feature_view(name: str, version: str) → FeatureView

Retrieve previously registered FeatureView.

Parameters:

  • name – FeatureView name.

  • version – FeatureView version.

Returns:

FeatureView object.

Raises:

SnowflakeMLException – [ValueError] FeatureView with name and version is not found, or incurred exception when reconstructing the FeatureView object.

Example:

>>> fs = FeatureStore(...)
>>> # draft_fv is a local object that hasn't materiaized to Snowflake backend yet.
>>> draft_fv = FeatureView(
...     name='foo',
...     entities=[e1],
...     feature_df=session.sql('...'),
...     desc='this is description',
... )
>>> fs.register_feature_view(feature_view=draft_fv, version='v1')

>>> # fv is a local object that maps to a Snowflake backend object.
>>> fv = fs.get_feature_view('foo', 'v1')
>>> print(f"name: {fv.name}")
>>> print(f"version:{fv.version}")
>>> print(f"desc:{fv.desc}")
name: FOO
version:v1
desc:this is description
get_refresh_history(feature_view: FeatureView, version: Optional[str] = None, *, verbose: bool = False) → DataFrame
get_refresh_history(feature_view: str, version: str, *, verbose: bool = False) → DataFrame

Get refresh hisotry statistics about a feature view.

Parameters:

  • feature_view – A registered feature view object, or the name of feature view.

  • version – Optional version of feature view. Must set when argument feature_view is a str.

  • verbose – Return more detailed history when set true.

Returns:

A dataframe contains the refresh history information.

Example:

>>> fs = FeatureStore(...)
>>> fv = fs.get_feature_view(name='MY_FV', version='v1')
>>> # refresh with name and version
>>> fs.refresh_feature_view('MY_FV', 'v1')
>>> fs.get_refresh_history('MY_FV', 'v1').show()
-----------------------------------------------------------------------------------------------------
|"NAME"    |"STATE"    |"REFRESH_START_TIME"        |"REFRESH_END_TIME"          |"REFRESH_ACTION"  |
-----------------------------------------------------------------------------------------------------
|MY_FV$v1  |SUCCEEDED  |2024-07-10 14:53:58.504000  |2024-07-10 14:53:59.088000  |INCREMENTAL       |
-----------------------------------------------------------------------------------------------------

>>> # refresh with feature view object
>>> fs.refresh_feature_view(fv)
>>> fs.get_refresh_history(fv).show()
-----------------------------------------------------------------------------------------------------
|"NAME"    |"STATE"    |"REFRESH_START_TIME"        |"REFRESH_END_TIME"          |"REFRESH_ACTION"  |
-----------------------------------------------------------------------------------------------------
|MY_FV$v1  |SUCCEEDED  |2024-07-10 14:54:06.680000  |2024-07-10 14:54:07.226000  |INCREMENTAL       |
|MY_FV$v1  |SUCCEEDED  |2024-07-10 14:53:58.504000  |2024-07-10 14:53:59.088000  |INCREMENTAL       |
-----------------------------------------------------------------------------------------------------
list_entities() → DataFrame

List all Entities in the FeatureStore.

Returns:

Snowpark DataFrame containing the results.

Example:

>>> fs = FeatureStore(...)
>>> e_1 = Entity("my_entity", ['col_1'], desc='My first entity.')
>>> fs.register_entity(e_1)
>>> fs.list_entities().show()
-----------------------------------------------------------
|"NAME"     |"JOIN_KEYS"  |"DESC"            |"OWNER"     |
-----------------------------------------------------------
|MY_ENTITY  |["COL_1"]    |My first entity.  |REGTEST_RL  |
-----------------------------------------------------------
list_feature_views(*, entity_name: Optional[str] = None, feature_view_name: Optional[str] = None) → DataFrame

List FeatureViews in the FeatureStore. If entity_name is specified, FeatureViews associated with that Entity will be listed. If feature_view_name is specified, further reducing the results to only match the specified name.

Parameters:

  • entity_name – Entity name.

  • feature_view_name – FeatureView name.

Returns:

FeatureViews information as a Snowpark DataFrame.

Example:

>>> fs = FeatureStore(...)
>>> draft_fv = FeatureView(
...     name='foo',
...     entities=[e1, e2],
...     feature_df=session.sql('...'),
...     desc='this is description',
... )
>>> fs.register_feature_view(feature_view=draft_fv, version='v1')
>>> fs.list_feature_views().select("name", "version", "desc").show()
--------------------------------------------
|"NAME"  |"VERSION"  |"DESC"               |
--------------------------------------------
|FOO     |v1         |this is description  |
--------------------------------------------
load_feature_views_from_dataset(ds: Dataset) → List[Union[FeatureView, FeatureViewSlice]]

Retrieve FeatureViews used during Dataset construction.

Parameters:

ds – Dataset object created from feature store.

Returns:

List of FeatureViews used during Dataset construction.

Raises:

ValueError – if dataset object is not generated from feature store.

Example:

>>> fs = FeatureStore(session, ...)
>>> # Assume you already have feature view registered.
>>> fv = fs.get_feature_view("MY_FV", "1.0")
>>> # Spine dataframe has same join keys as the entity of fv.
>>> spine_df = session.create_dataframe(["1", "2"], schema=["id"])
>>> my_dataset = fs.generate_dataset(
...     "my_dataset"
...     spine_df,
...     [fv],
... )
>>> fvs = fs.load_feature_views_from_dataset(my_dataset)
>>> print(len(fvs))
1

>>> print(type(fvs[0]))
<class 'snowflake.ml.feature_store.feature_view.FeatureView'>

>>> print(fvs[0].name)
MY_FV

>>> print(fvs[0].version)
1.0
read_feature_view(feature_view: str, version: str) → DataFrame
read_feature_view(feature_view: FeatureView) → DataFrame

Read values from a FeatureView.

Parameters:

  • feature_view – A FeatureView object to read from, or the name of feature view. If name is provided then version also must be provided.

  • version – Optional version of feature view. Must set when argument feature_view is a str.

Returns:

Snowpark DataFrame(lazy mode) containing the FeatureView data.

Raises:

  • SnowflakeMLException – [ValueError] version argument is missing when argument feature_view is a str.

  • SnowflakeMLException – [ValueError] FeatureView is not registered.

Example:

>>> fs = FeatureStore(...)
>>> # Read from feature view name and version.
>>> fs.read_feature_view('foo', 'v1').show()
------------------------------------------
|"NAME"  |"ID"  |"TITLE"  |"AGE"  |"TS"  |
------------------------------------------
|jonh    |1     |boss     |20     |100   |
|porter  |2     |manager  |30     |200   |
------------------------------------------

>>> # Read from feature view object.
>>> fv = fs.get_feature_view('foo', 'v1')
>>> fs.read_feature_view(fv).show()
------------------------------------------
|"NAME"  |"ID"  |"TITLE"  |"AGE"  |"TS"  |
------------------------------------------
|jonh    |1     |boss     |20     |100   |
|porter  |2     |manager  |30     |200   |
------------------------------------------
refresh_feature_view(feature_view: FeatureView) → None
refresh_feature_view(feature_view: str, version: str) → None

Manually refresh a feature view.

Parameters:

  • feature_view – A registered feature view object, or the name of feature view.

  • version – Optional version of feature view. Must set when argument feature_view is a str.

Example:

>>> fs = FeatureStore(...)
>>> fv = fs.get_feature_view(name='MY_FV', version='v1')

>>> # refresh with name and version
>>> fs.refresh_feature_view('MY_FV', 'v1')
>>> fs.get_refresh_history('MY_FV', 'v1').show()
-----------------------------------------------------------------------------------------------------
|"NAME"    |"STATE"    |"REFRESH_START_TIME"        |"REFRESH_END_TIME"          |"REFRESH_ACTION"  |
-----------------------------------------------------------------------------------------------------
|MY_FV$v1  |SUCCEEDED  |2024-07-10 14:53:58.504000  |2024-07-10 14:53:59.088000  |INCREMENTAL       |
-----------------------------------------------------------------------------------------------------

>>> # refresh with feature view object
>>> fs.refresh_feature_view(fv)
>>> fs.get_refresh_history(fv).show()
-----------------------------------------------------------------------------------------------------
|"NAME"    |"STATE"    |"REFRESH_START_TIME"        |"REFRESH_END_TIME"          |"REFRESH_ACTION"  |
-----------------------------------------------------------------------------------------------------
|MY_FV$v1  |SUCCEEDED  |2024-07-10 14:54:06.680000  |2024-07-10 14:54:07.226000  |INCREMENTAL       |
|MY_FV$v1  |SUCCEEDED  |2024-07-10 14:53:58.504000  |2024-07-10 14:53:59.088000  |INCREMENTAL       |
-----------------------------------------------------------------------------------------------------
register_entity(entity: Entity) → Entity

Register Entity in the FeatureStore.

Parameters:

entity – Entity object to be registered.

Returns:

A registered entity object.

Raises:

SnowflakeMLException – [RuntimeError] Failed to find resources.

Example:

>>> fs = FeatureStore(...)
>>> e = Entity('BAR', ['A'], desc='entity bar')
>>> fs.register_entity(e)
>>> fs.list_entities().show()
--------------------------------------------------
|"NAME"  |"JOIN_KEYS"  |"DESC"      |"OWNER"     |
--------------------------------------------------
|BAR     |["A"]        |entity bar  |REGTEST_RL  |
--------------------------------------------------
register_feature_view(feature_view: FeatureView, version: str, *, block: bool = True, overwrite: bool = False) → FeatureView

Materialize a FeatureView to Snowflake backend. Incremental maintenance for updates on the source data will be automated if refresh_freq is set. NOTE: Each new materialization will trigger a full FeatureView history refresh for the data included in the

FeatureView.

Parameters:

  • feature_view – FeatureView instance to materialize.

  • version – version of the registered FeatureView. NOTE: Version only accepts letters, numbers and underscore. Also version will be capitalized.

  • block – Specify whether the FeatureView backend materialization should be blocking or not. If blocking then the API will wait until the initial FeatureView data is generated. Default to true.

  • overwrite – Overwrite the existing FeatureView with same version. This is the same as dropping the FeatureView first then recreate. NOTE: there will be backfill cost associated if the FeatureView is being continuously maintained.

Returns:

A materialized FeatureView object.

Raises:

  • SnowflakeMLException – [ValueError] FeatureView entity has not been registered.

  • SnowflakeMLException – [ValueError] Warehouse or default warehouse is not specified.

  • SnowflakeMLException – [RuntimeError] Failed to create dynamic table, task, or view.

  • SnowflakeMLException – [RuntimeError] Failed to find resources.

Example:

>>> fs = FeatureStore(...)
>>> # draft_fv is a local object that hasn't materiaized to Snowflake backend yet.
>>> feature_df = session.sql("select f_1, f_2 from source_table")
>>> draft_fv = FeatureView("my_fv", [entities], feature_df)
>>> print(draft_fv.status)
FeatureViewStatus.DRAFT

>>> fs.list_feature_views().select("NAME", "VERSION", "SCHEDULING_STATE").show()
-------------------------------------------
|"NAME"  |"VERSION"  |"SCHEDULING_STATE"  |
-------------------------------------------
|        |           |                    |
-------------------------------------------

>>> # registered_fv is a local object that maps to a Snowflake backend object.
>>> registered_fv = fs.register_feature_view(draft_fv, "v1")
>>> print(registered_fv.status)
FeatureViewStatus.ACTIVE

>>> fs.list_feature_views().select("NAME", "VERSION", "SCHEDULING_STATE").show()
-------------------------------------------
|"NAME"  |"VERSION"  |"SCHEDULING_STATE"  |
-------------------------------------------
|MY_FV   |v1         |ACTIVE              |
-------------------------------------------
resume_feature_view(feature_view: FeatureView) → FeatureView
resume_feature_view(feature_view: str, version: str) → FeatureView

Resume a previously suspended FeatureView.

Parameters:

  • feature_view – FeatureView object or name to resume.

  • version – Optional version of feature view. Must set when argument feature_view is a str.

Returns:

A new feature view with updated status.

Example:

>>> fs = FeatureStore(...)
>>> # you must already have feature views registered
>>> fv = fs.get_feature_view(name='MY_FV', version='v1')
>>> fs.suspend_feature_view('MY_FV', 'v1')
>>> fs.list_feature_views().select("NAME", "VERSION", "SCHEDULING_STATE").show()
-------------------------------------------
|"NAME"  |"VERSION"  |"SCHEDULING_STATE"  |
-------------------------------------------
|MY_FV   |v1         |SUSPENDED           |
-------------------------------------------

>>> fs.resume_feature_view('MY_FV', 'v1')
>>> fs.list_feature_views().select("NAME", "VERSION", "SCHEDULING_STATE").show()
-------------------------------------------
|"NAME"  |"VERSION"  |"SCHEDULING_STATE"  |
-------------------------------------------
|MY_FV   |v1         |ACTIVE              |
-------------------------------------------
retrieve_feature_values(spine_df: DataFrame, features: Union[List[Union[FeatureView, FeatureViewSlice]], List[str]], *, spine_timestamp_col: Optional[str] = None, exclude_columns: Optional[List[str]] = None, include_feature_view_timestamp_col: bool = False) → DataFrame

Enrich spine dataframe with feature values. Mainly used to generate inference data input. If spine_timestamp_col is specified, point-in-time feature values will be fetched.

Parameters:

  • spine_df – Snowpark DataFrame to join features into.

  • features – List of features to join into the spine_df. Can be a list of FeatureView or FeatureViewSlice, or a list of serialized feature objects from Dataset.

  • spine_timestamp_col – Timestamp column in spine_df for point-in-time feature value lookup.

  • exclude_columns – Column names to exclude from the result dataframe.

  • include_feature_view_timestamp_col – Generated dataset will include timestamp column of feature view (if feature view has timestamp column) if set true. Default to false.

Returns:

Snowpark DataFrame containing the joined results.

Raises:

ValueError – if features is empty.

Example:

>>> fs = FeatureStore(...)
>>> # Assume you already have feature view registered.
>>> fv = fs.get_feature_view('my_fv', 'v1')
>>> # Spine dataframe has same join keys as the entity of fv.
>>> spine_df = session.create_dataframe(["1", "2"], schema=["id"])
>>> fs.retrieve_feature_values(spine_df, [fv]).show()
--------------------
|"END_STATION_ID"  |
--------------------
|505               |
|347               |
|466               |
--------------------
suspend_feature_view(feature_view: FeatureView) → FeatureView
suspend_feature_view(feature_view: str, version: str) → FeatureView

Suspend an active FeatureView.

Parameters:

  • feature_view – FeatureView object or name to suspend.

  • version – Optional version of feature view. Must set when argument feature_view is a str.

Returns:

A new feature view with updated status.

Example:

>>> fs = FeatureStore(...)
>>> # assume you already have feature views registered
>>> fv = fs.get_feature_view(name='MY_FV', version='v1')
>>> fs.suspend_feature_view('MY_FV', 'v1')
>>> fs.list_feature_views().select("NAME", "VERSION", "SCHEDULING_STATE").show()
-------------------------------------------
|"NAME"  |"VERSION"  |"SCHEDULING_STATE"  |
-------------------------------------------
|MY_FV   |v1         |SUSPENDED           |
-------------------------------------------

>>> fs.resume_feature_view('MY_FV', 'v1')
>>> fs.list_feature_views().select("NAME", "VERSION", "SCHEDULING_STATE").show()
-------------------------------------------
|"NAME"  |"VERSION"  |"SCHEDULING_STATE"  |
-------------------------------------------
|MY_FV   |v1         |ACTIVE              |
-------------------------------------------
update_default_warehouse(warehouse_name: str) → None

Update default warehouse for feature store.

Parameters:

warehouse_name – Name of warehouse.

Raises:

SnowflakeMLException – If warehouse does not exists.

Example:

>>> fs = FeatureStore(...)
>>> fs.update_default_warehouse("MYWH_2")
>>> draft_fv = FeatureView("my_fv", ...)
>>> registered_fv = fs.register_feature_view(draft_fv, '2.0')
>>> print(registered_fv.warehouse)
MYWH_2
update_entity(name: str, *, desc: Optional[str] = None) → Optional[Entity]

Update a registered entity with provided information.

Parameters:

  • name – Name of entity to update.

  • desc – Optional new description to apply. Default to None.

Raises:

SnowflakeMLException – Error happen when updating.

Returns:

A new entity with updated information or None if the entity doesn’t exist.

Example:

>>> fs = FeatureStore(...)

>>> e = Entity(name='foo', join_keys=['COL_1'], desc='old desc')
>>> fs.list_entities().show()
------------------------------------------------
|"NAME"  |"JOIN_KEYS"  |"DESC"    |"OWNER"     |
------------------------------------------------
|FOO     |["COL_1"]    |old desc  |REGTEST_RL  |
------------------------------------------------

>>> fs.update_entity('foo', desc='NEW DESC')
>>> fs.list_entities().show()
------------------------------------------------
|"NAME"  |"JOIN_KEYS"  |"DESC"    |"OWNER"     |
------------------------------------------------
|FOO     |["COL_1"]    |NEW DESC  |REGTEST_RL  |
------------------------------------------------
update_feature_view(name: str, version: str, *, refresh_freq: Optional[str] = None, warehouse: Optional[str] = None, desc: Optional[str] = None) → FeatureView
update_feature_view(name: FeatureView, version: Optional[str] = None, *, refresh_freq: Optional[str] = None, warehouse: Optional[str] = None, desc: Optional[str] = None) → FeatureView
Update a registered feature view.

Check feature_view.py for which fields are allowed to be updated after registration.

Parameters:

  • name – FeatureView object or name to suspend.

  • version – Optional version of feature view. Must set when argument feature_view is a str.

  • refresh_freq – updated refresh frequency.

  • warehouse – updated warehouse.

  • desc – description of feature view.

Returns:

Updated FeatureView.

Example:

>>> fs = FeatureStore(...)
>>> fv = FeatureView(
...     name='foo',
...     entities=[e1, e2],
...     feature_df=session.sql('...'),
...     desc='this is old description',
... )
>>> fv = fs.register_feature_view(feature_view=fv, version='v1')
>>> fs.list_feature_views().select("name", "version", "desc").show()
------------------------------------------------
|"NAME"  |"VERSION"  |"DESC"                   |
------------------------------------------------
|FOO     |v1         |this is old description  |
------------------------------------------------

>>> # update_feature_view will apply new arguments to the registered feature view.
>>> new_fv = fs.update_feature_view(
...     name='foo',
...     version='v1',
...     desc='that is new descption',
... )
>>> fs.list_feature_views().select("name", "version", "desc").show()
------------------------------------------------
|"NAME"  |"VERSION"  |"DESC"                   |
------------------------------------------------
|FOO     |v1         |THAT IS NEW DESCRIPTION  |
------------------------------------------------
Raises:

  • SnowflakeMLException – [RuntimeError] If FeatureView is not managed and refresh_freq is defined.

  • SnowflakeMLException – [RuntimeError] Failed to update feature view.