def __init__(
self, repo_path: Optional[str] = None, config: Optional[RepoConfig] = None,
):
""" Initializes a new FeatureStore object. Used to manage a feature store.
Args:
repo_path: Path to a `feature_store.yaml` used to configure the feature store
config (RepoConfig): Configuration object used to configure the feature store
"""
if repo_path is not None and config is not None:
raise ValueError("You cannot specify both repo_path and config")
if config is not None:
self.repo_path = Path(os.getcwd())
self.config = config
elif repo_path is not None:
self.repo_path = Path(repo_path)
self.config = load_repo_config(Path(repo_path))
else:
raise ValueError("Please specify one of repo_path or config")
registry_config = self.config.get_registry_config()
self._registry = Registry(
registry_path=registry_config.path,
repo_path=self.repo_path,
cache_ttl=timedelta(seconds=registry_config.cache_ttl_seconds),
)
self._tele = Telemetry()
def __init__(
self, repo_path: Optional[str] = None, config: Optional[RepoConfig] = None,
):
self.repo_path = repo_path
if repo_path is not None and config is not None:
raise ValueError("You cannot specify both repo_path and config")
if config is not None:
self.config = config
elif repo_path is not None:
self.config = load_repo_config(Path(repo_path))
else:
self.config = RepoConfig(
registry="./registry.db",
project="default",
provider="local",
online_store=OnlineStoreConfig(
local=LocalOnlineStoreConfig(path="online_store.db")
),
)
registry_config = self.config.get_registry_config()
self._registry = Registry(
registry_path=registry_config.path,
cache_ttl=timedelta(seconds=registry_config.cache_ttl_seconds),
)
self._tele = Telemetry()
def registry_dump_command():
"""
Print contents of the metadata registry
"""
cli_check_repo(Path.cwd())
repo_config = load_repo_config(Path.cwd())
tele = Telemetry()
tele.log("registry-dump")
registry_dump(repo_config, repo_path=Path.cwd())
def teardown_command():
"""
Tear down deployed feature store infrastructure
"""
cli_check_repo(Path.cwd())
repo_config = load_repo_config(Path.cwd())
tele = Telemetry()
tele.log("teardown")
teardown(repo_config, Path.cwd())
def init_command(project_directory, minimal: bool, template: str):
"""Create a new Feast repository"""
if not project_directory:
project_directory = generate_project_name()
if minimal:
template = "minimal"
tele = Telemetry()
tele.log("init")
init_repo(project_directory, template)
def registry_dump_command(ctx: click.Context):
"""
Print contents of the metadata registry
"""
repo = ctx.obj["CHDIR"]
cli_check_repo(repo)
repo_config = load_repo_config(repo)
tele = Telemetry()
tele.log("registry-dump")
registry_dump(repo_config, repo_path=repo)
def teardown_command(ctx: click.Context):
"""
Tear down deployed feature store infrastructure
"""
repo = ctx.obj["CHDIR"]
cli_check_repo(repo)
repo_config = load_repo_config(repo)
tele = Telemetry()
tele.log("teardown")
teardown(repo_config, repo)
def apply_total_command():
"""
Create or update a feature store deployment
"""
cli_check_repo(Path.cwd())
repo_config = load_repo_config(Path.cwd())
tele = Telemetry()
tele.log("apply")
try:
apply_total(repo_config, Path.cwd())
except FeastProviderLoginError as e:
print(str(e))
def apply_total_command(ctx: click.Context):
"""
Create or update a feature store deployment
"""
repo = ctx.obj["CHDIR"]
cli_check_repo(repo)
repo_config = load_repo_config(repo)
tele = Telemetry()
tele.log("apply")
try:
apply_total(repo_config, repo)
except FeastProviderLoginError as e:
print(str(e))
def __init__(
self,
repo_path: Optional[str] = None,
config: Optional[RepoConfig] = None,
):
if repo_path is not None and config is not None:
raise ValueError("You cannot specify both repo_path and config")
if config is not None:
self.repo_path = Path(os.getcwd())
self.config = config
elif repo_path is not None:
self.repo_path = Path(repo_path)
self.config = load_repo_config(Path(repo_path))
else:
raise ValueError("Please specify one of repo_path or config")
registry_config = self.config.get_registry_config()
self._registry = Registry(
registry_path=registry_config.path,
repo_path=self.repo_path,
cache_ttl=timedelta(seconds=registry_config.cache_ttl_seconds),
)
self._tele = Telemetry()
def __init__(self, options: Optional[Dict[str, str]] = None, **kwargs):
"""
The Feast Client should be initialized with at least one service url
Please see constants.py for configuration options. Commonly used options
or arguments include:
core_url: Feast Core URL. Used to manage features
serving_url: Feast Serving URL. Used to retrieve features
project: Sets the active project. This field is optional.
core_secure: Use client-side SSL/TLS for Core gRPC API
serving_secure: Use client-side SSL/TLS for Serving gRPC API
enable_auth: Enable authentication and authorization
auth_provider: Authentication provider – "google" or "oauth"
if auth_provider is "oauth", the following fields are mandatory –
oauth_grant_type, oauth_client_id, oauth_client_secret, oauth_audience, oauth_token_request_url
Args:
options: Configuration options to initialize client with
**kwargs: Additional keyword arguments that will be used as
configuration options along with "options"
"""
if options is None:
options = dict()
self._config = Config(options={**options, **kwargs})
self._core_service_stub: Optional[CoreServiceStub] = None
self._serving_service_stub: Optional[ServingServiceStub] = None
self._auth_metadata: Optional[grpc.AuthMetadataPlugin] = None
self._registry_impl: Optional[Registry] = None
# Configure Auth Metadata Plugin if auth is enabled
if self._config.getboolean(opt.ENABLE_AUTH):
self._auth_metadata = feast_auth.get_auth_metadata_plugin(
self._config)
self._tele = Telemetry()
class FeatureStore:
"""
A FeatureStore object is used to define, create, and retrieve features.
"""
config: RepoConfig
repo_path: Optional[str]
_registry: Registry
def __init__(
self, repo_path: Optional[str] = None, config: Optional[RepoConfig] = None,
):
self.repo_path = repo_path
if repo_path is not None and config is not None:
raise ValueError("You cannot specify both repo_path and config")
if config is not None:
self.config = config
elif repo_path is not None:
self.config = load_repo_config(Path(repo_path))
else:
self.config = RepoConfig(
registry="./registry.db",
project="default",
provider="local",
online_store=OnlineStoreConfig(
local=LocalOnlineStoreConfig(path="online_store.db")
),
)
registry_config = self.config.get_registry_config()
self._registry = Registry(
registry_path=registry_config.path,
cache_ttl=timedelta(seconds=registry_config.cache_ttl_seconds),
)
self._tele = Telemetry()
def version(self) -> str:
"""Returns the version of the current Feast SDK/CLI"""
return get_version()
@property
def project(self) -> str:
return self.config.project
def _get_provider(self) -> Provider:
return get_provider(self.config)
def refresh_registry(self):
"""Fetches and caches a copy of the feature registry in memory.
Explicitly calling this method allows for direct control of the state of the registry cache. Every time this
method is called the complete registry state will be retrieved from the remote registry store backend
(e.g., GCS, S3), and the cache timer will be reset. If refresh_registry() is run before get_online_features()
is called, then get_online_feature() will use the cached registry instead of retrieving (and caching) the
registry itself.
Additionally, the TTL for the registry cache can be set to infinity (by setting it to 0), which means that
refresh_registry() will become the only way to update the cached registry. If the TTL is set to a value
greater than 0, then once the cache becomes stale (more time than the TTL has passed), a new cache will be
downloaded synchronously, which may increase latencies if the triggering method is get_online_features()
"""
self._tele.log("refresh_registry")
registry_config = self.config.get_registry_config()
self._registry = Registry(
registry_path=registry_config.path,
cache_ttl=timedelta(seconds=registry_config.cache_ttl_seconds),
)
self._registry.refresh()
def list_entities(self, allow_cache: bool = False) -> List[Entity]:
"""
Retrieve a list of entities from the registry
Args:
allow_cache (bool): Whether to allow returning entities from a cached registry
Returns:
List of entities
"""
self._tele.log("list_entities")
return self._registry.list_entities(self.project, allow_cache=allow_cache)
def list_feature_views(self) -> List[FeatureView]:
"""
Retrieve a list of feature views from the registry
Returns:
List of feature views
"""
self._tele.log("list_feature_views")
return self._registry.list_feature_views(self.project)
def get_entity(self, name: str) -> Entity:
"""
Retrieves an entity.
Args:
name: Name of entity
Returns:
Returns either the specified entity, or raises an exception if
none is found
"""
self._tele.log("get_entity")
return self._registry.get_entity(name, self.project)
def get_feature_view(self, name: str) -> FeatureView:
"""
Retrieves a feature view.
Args:
name: Name of feature view
Returns:
Returns either the specified feature view, or raises an exception if
none is found
"""
self._tele.log("get_feature_view")
return self._registry.get_feature_view(name, self.project)
def delete_feature_view(self, name: str):
"""
Deletes a feature view or raises an exception if not found.
Args:
name: Name of feature view
"""
self._tele.log("delete_feature_view")
return self._registry.delete_feature_view(name, self.project)
def apply(self, objects: List[Union[FeatureView, Entity]]):
"""Register objects to metadata store and update related infrastructure.
The apply method registers one or more definitions (e.g., Entity, FeatureView) and registers or updates these
objects in the Feast registry. Once the registry has been updated, the apply method will update related
infrastructure (e.g., create tables in an online store) in order to reflect these new definitions. All
operations are idempotent, meaning they can safely be rerun.
Args: objects (List[Union[FeatureView, Entity]]): A list of FeatureView or Entity objects that should be
registered
Examples:
Register a single Entity and FeatureView.
>>> from feast.feature_store import FeatureStore
>>> from feast import Entity, FeatureView, Feature, ValueType, FileSource
>>> from datetime import timedelta
>>>
>>> fs = FeatureStore()
>>> customer_entity = Entity(name="customer", value_type=ValueType.INT64, description="customer entity")
>>> customer_feature_view = FeatureView(
>>> name="customer_fv",
>>> entities=["customer"],
>>> features=[Feature(name="age", dtype=ValueType.INT64)],
>>> input=FileSource(path="file.parquet", event_timestamp_column="timestamp"),
>>> ttl=timedelta(days=1)
>>> )
>>> fs.apply([customer_entity, customer_feature_view])
"""
self._tele.log("apply")
# TODO: Add locking
# TODO: Optimize by only making a single call (read/write)
views_to_update = []
for ob in objects:
if isinstance(ob, FeatureView):
self._registry.apply_feature_view(ob, project=self.config.project)
views_to_update.append(ob)
elif isinstance(ob, Entity):
self._registry.apply_entity(ob, project=self.config.project)
else:
raise ValueError(
f"Unknown object type ({type(ob)}) provided as part of apply() call"
)
self._get_provider().update_infra(
project=self.config.project,
tables_to_delete=[],
tables_to_keep=views_to_update,
partial=True,
)
def get_historical_features(
self, entity_df: Union[pd.DataFrame, str], feature_refs: List[str],
) -> RetrievalJob:
"""Enrich an entity dataframe with historical feature values for either training or batch scoring.
This method joins historical feature data from one or more feature views to an entity dataframe by using a time
travel join.
Each feature view is joined to the entity dataframe using all entities configured for the respective feature
view. All configured entities must be available in the entity dataframe. Therefore, the entity dataframe must
contain all entities found in all feature views, but the individual feature views can have different entities.
Time travel is based on the configured TTL for each feature view. A shorter TTL will limit the
amount of scanning that will be done in order to find feature data for a specific entity key. Setting a short
TTL may result in null values being returned.
Args:
entity_df (Union[pd.DataFrame, str]): An entity dataframe is a collection of rows containing all entity
columns (e.g., customer_id, driver_id) on which features need to be joined, as well as a event_timestamp
column used to ensure point-in-time correctness. Either a Pandas DataFrame can be provided or a string
SQL query. The query must be of a format supported by the configured offline store (e.g., BigQuery)
feature_refs: A list of features that should be retrieved from the offline store. Feature references are of
the format "feature_view:feature", e.g., "customer_fv:daily_transactions".
Returns:
RetrievalJob which can be used to materialize the results.
Examples:
Retrieve historical features using a BigQuery SQL entity dataframe
>>> from feast.feature_store import FeatureStore
>>>
>>> fs = FeatureStore(config=RepoConfig(provider="gcp"))
>>> retrieval_job = fs.get_historical_features(
>>> entity_df="SELECT event_timestamp, order_id, customer_id from gcp_project.my_ds.customer_orders",
>>> feature_refs=["customer:age", "customer:avg_orders_1d", "customer:avg_orders_7d"]
>>> )
>>> feature_data = job.to_df()
>>> model.fit(feature_data) # insert your modeling framework here.
"""
self._tele.log("get_historical_features")
all_feature_views = self._registry.list_feature_views(
project=self.config.project
)
feature_views = _get_requested_feature_views(feature_refs, all_feature_views)
provider = self._get_provider()
job = provider.get_historical_features(
self.config,
feature_views,
feature_refs,
entity_df,
self._registry,
self.project,
)
return job
def materialize_incremental(
self, end_date: datetime, feature_views: Optional[List[str]] = None,
) -> None:
"""
Materialize incremental new data from the offline store into the online store.
This method loads incremental new feature data up to the specified end time from either
the specified feature views, or all feature views if none are specified,
into the online store where it is available for online serving. The start time of
the interval materialized is either the most recent end time of a prior materialization or
(now - ttl) if no such prior materialization exists.
Args:
end_date (datetime): End date for time range of data to materialize into the online store
feature_views (List[str]): Optional list of feature view names. If selected, will only run
materialization for the specified feature views.
Examples:
Materialize all features into the online store up to 5 minutes ago.
>>> from datetime import datetime, timedelta
>>> from feast.feature_store import FeatureStore
>>>
>>> fs = FeatureStore(config=RepoConfig(provider="gcp", registry="gs://my-fs/", project="my_fs_proj"))
>>> fs.materialize_incremental(end_date=datetime.utcnow() - timedelta(minutes=5))
"""
self._tele.log("materialize_incremental")
feature_views_to_materialize = []
if feature_views is None:
feature_views_to_materialize = self._registry.list_feature_views(
self.config.project
)
else:
for name in feature_views:
feature_view = self._registry.get_feature_view(
name, self.config.project
)
feature_views_to_materialize.append(feature_view)
# TODO paging large loads
for feature_view in feature_views_to_materialize:
start_date = feature_view.most_recent_end_time
if start_date is None:
if feature_view.ttl is None:
raise Exception(
f"No start time found for feature view {feature_view.name}. materialize_incremental() requires either a ttl to be set or for materialize() to have been run at least once."
)
start_date = datetime.utcnow() - feature_view.ttl
provider = self._get_provider()
provider.materialize_single_feature_view(
feature_view, start_date, end_date, self._registry, self.project
)
def materialize(
self,
start_date: datetime,
end_date: datetime,
feature_views: Optional[List[str]] = None,
) -> None:
"""
Materialize data from the offline store into the online store.
This method loads feature data in the specified interval from either
the specified feature views, or all feature views if none are specified,
into the online store where it is available for online serving.
Args:
start_date (datetime): Start date for time range of data to materialize into the online store
end_date (datetime): End date for time range of data to materialize into the online store
feature_views (List[str]): Optional list of feature view names. If selected, will only run
materialization for the specified feature views.
Examples:
Materialize all features into the online store over the interval
from 3 hours ago to 10 minutes ago.
>>> from datetime import datetime, timedelta
>>> from feast.feature_store import FeatureStore
>>>
>>> fs = FeatureStore(config=RepoConfig(provider="gcp"))
>>> fs.materialize(
>>> start_date=datetime.utcnow() - timedelta(hours=3), end_date=datetime.utcnow() - timedelta(minutes=10)
>>> )
"""
self._tele.log("materialize")
feature_views_to_materialize = []
if feature_views is None:
feature_views_to_materialize = self._registry.list_feature_views(
self.config.project
)
else:
for name in feature_views:
feature_view = self._registry.get_feature_view(
name, self.config.project
)
feature_views_to_materialize.append(feature_view)
# TODO paging large loads
for feature_view in feature_views_to_materialize:
provider = self._get_provider()
provider.materialize_single_feature_view(
feature_view, start_date, end_date, self._registry, self.project
)
def get_online_features(
self, feature_refs: List[str], entity_rows: List[Dict[str, Any]],
) -> OnlineResponse:
"""
Retrieves the latest online feature data.
Note: This method will download the full feature registry the first time it is run. If you are using a
remote registry like GCS or S3 then that may take a few seconds. The registry remains cached up to a TTL
duration (which can be set to infinitey). If the cached registry is stale (more time than the TTL has
passed), then a new registry will be downloaded synchronously by this method. This download may
introduce latency to online feature retrieval. In order to avoid synchronous downloads, please call
refresh_registry() prior to the TTL being reached. Remember it is possible to set the cache TTL to
infinity (cache forever).
Args:
feature_refs: List of feature references that will be returned for each entity.
Each feature reference should have the following format:
"feature_table:feature" where "feature_table" & "feature" refer to
the feature and feature table names respectively.
Only the feature name is required.
entity_rows: A list of dictionaries where each key-value is an entity-name, entity-value pair.
Returns:
OnlineResponse containing the feature data in records.
Examples:
>>> from feast import FeatureStore
>>>
>>> store = FeatureStore(repo_path="...")
>>> feature_refs = ["sales:daily_transactions"]
>>> entity_rows = [{"customer_id": 0},{"customer_id": 1}]
>>>
>>> online_response = store.get_online_features(
>>> feature_refs, entity_rows, project="my_project")
>>> online_response_dict = online_response.to_dict()
>>> print(online_response_dict)
{'sales:daily_transactions': [1.1,1.2], 'sales:customer_id': [0,1]}
"""
self._tele.log("get_online_features")
provider = self._get_provider()
entities = self.list_entities(allow_cache=True)
entity_name_to_join_key_map = {}
for entity in entities:
entity_name_to_join_key_map[entity.name] = entity.join_key
join_key_rows = []
for row in entity_rows:
join_key_row = {}
for entity_name, entity_value in row.items():
try:
join_key = entity_name_to_join_key_map[entity_name]
except KeyError:
raise Exception(
f"Entity {entity_name} does not exist in project {self.project}"
)
join_key_row[join_key] = entity_value
join_key_rows.append(join_key_row)
entity_row_proto_list = _infer_online_entity_rows(join_key_rows)
union_of_entity_keys = []
result_rows: List[GetOnlineFeaturesResponse.FieldValues] = []
for entity_row_proto in entity_row_proto_list:
union_of_entity_keys.append(_entity_row_to_key(entity_row_proto))
result_rows.append(_entity_row_to_field_values(entity_row_proto))
all_feature_views = self._registry.list_feature_views(
project=self.config.project, allow_cache=True
)
grouped_refs = _group_refs(feature_refs, all_feature_views)
for table, requested_features in grouped_refs:
entity_keys = _get_table_entity_keys(
table, union_of_entity_keys, entity_name_to_join_key_map
)
read_rows = provider.online_read(
project=self.project, table=table, entity_keys=entity_keys,
)
for row_idx, read_row in enumerate(read_rows):
row_ts, feature_data = read_row
result_row = result_rows[row_idx]
if feature_data is None:
for feature_name in requested_features:
feature_ref = f"{table.name}__{feature_name}"
result_row.statuses[
feature_ref
] = GetOnlineFeaturesResponse.FieldStatus.NOT_FOUND
else:
for feature_name in feature_data:
feature_ref = f"{table.name}__{feature_name}"
if feature_name in requested_features:
result_row.fields[feature_ref].CopyFrom(
feature_data[feature_name]
)
result_row.statuses[
feature_ref
] = GetOnlineFeaturesResponse.FieldStatus.PRESENT
return OnlineResponse(GetOnlineFeaturesResponse(field_values=result_rows))
class Client:
"""
Feast Client: Used for creating, managing, and retrieving features.
"""
def __init__(self, options: Optional[Dict[str, str]] = None, **kwargs):
"""
The Feast Client should be initialized with at least one service url
Please see constants.py for configuration options. Commonly used options
or arguments include:
core_url: Feast Core URL. Used to manage features
serving_url: Feast Serving URL. Used to retrieve features
project: Sets the active project. This field is optional.
core_secure: Use client-side SSL/TLS for Core gRPC API
serving_secure: Use client-side SSL/TLS for Serving gRPC API
enable_auth: Enable authentication and authorization
auth_provider: Authentication provider – "google" or "oauth"
if auth_provider is "oauth", the following fields are mandatory –
oauth_grant_type, oauth_client_id, oauth_client_secret, oauth_audience, oauth_token_request_url
Args:
options: Configuration options to initialize client with
**kwargs: Additional keyword arguments that will be used as
configuration options along with "options"
"""
if options is None:
options = dict()
self._config = Config(options={**options, **kwargs})
self._core_service_stub: Optional[CoreServiceStub] = None
self._serving_service_stub: Optional[ServingServiceStub] = None
self._auth_metadata: Optional[grpc.AuthMetadataPlugin] = None
self._registry_impl: Optional[Registry] = None
# Configure Auth Metadata Plugin if auth is enabled
if self._config.getboolean(opt.ENABLE_AUTH):
self._auth_metadata = feast_auth.get_auth_metadata_plugin(
self._config)
self._tele = Telemetry()
@property
def config(self) -> Config:
return self._config
@property
def _core_service(self):
"""
Creates or returns the gRPC Feast Core Service Stub
Returns: CoreServiceStub
"""
if not self._core_service_stub:
channel = create_grpc_channel(
url=self._config.get(opt.CORE_URL),
enable_ssl=self._config.getboolean(opt.CORE_ENABLE_SSL),
enable_auth=self._config.getboolean(opt.ENABLE_AUTH),
ssl_server_cert_path=self._config.get(
opt.CORE_SERVER_SSL_CERT),
auth_metadata_plugin=self._auth_metadata,
timeout=self._config.getint(opt.GRPC_CONNECTION_TIMEOUT),
)
self._core_service_stub = CoreServiceStub(channel)
return self._core_service_stub
@property
def _use_object_store_registry(self) -> bool:
return self._config.exists(opt.REGISTRY_PATH)
@property
def _registry(self):
if self._registry_impl is None:
self._registry_impl = Registry(self._config.get(opt.REGISTRY_PATH))
return self._registry_impl
@property
def _serving_service(self):
"""
Creates or returns the gRPC Feast Serving Service Stub. If both `opentracing`
and `grpcio-opentracing` are installed, an opentracing interceptor will be
instantiated based on the global tracer.
Returns: ServingServiceStub
"""
if not self._serving_service_stub:
channel = create_grpc_channel(
url=self._config.get(opt.SERVING_URL),
enable_ssl=self._config.getboolean(opt.SERVING_ENABLE_SSL),
enable_auth=self._config.getboolean(opt.ENABLE_AUTH),
ssl_server_cert_path=self._config.get(
opt.SERVING_SERVER_SSL_CERT),
auth_metadata_plugin=self._auth_metadata,
timeout=self._config.getint(opt.GRPC_CONNECTION_TIMEOUT),
)
try:
import opentracing
from grpc_opentracing import open_tracing_client_interceptor
from grpc_opentracing.grpcext import intercept_channel
interceptor = open_tracing_client_interceptor(
opentracing.global_tracer())
channel = intercept_channel(channel, interceptor)
except ImportError:
pass
self._serving_service_stub = ServingServiceStub(channel)
return self._serving_service_stub
def _extra_grpc_params(self) -> Dict[str, Any]:
return dict(
timeout=self._config.getint(opt.GRPC_CONNECTION_TIMEOUT),
metadata=self._get_grpc_metadata(),
)
@property
def core_url(self) -> str:
"""
Retrieve Feast Core URL
Returns:
Feast Core URL string
"""
return self._config.get(opt.CORE_URL)
@core_url.setter
def core_url(self, value: str):
"""
Set the Feast Core URL
Args:
value: Feast Core URL
"""
self._config.set(opt.CORE_URL, value)
@property
def serving_url(self) -> str:
"""
Retrieve Feast Serving URL
Returns:
Feast Serving URL string
"""
return self._config.get(opt.SERVING_URL)
@serving_url.setter
def serving_url(self, value: str):
"""
Set the Feast Serving URL
Args:
value: Feast Serving URL
"""
self._config.set(opt.SERVING_URL, value)
@property
def job_service_url(self) -> str:
"""
Retrieve Feast Job Service URL
Returns:
Feast Job Service URL string
"""
return self._config.get(opt.JOB_SERVICE_URL)
@job_service_url.setter
def job_service_url(self, value: str):
"""
Set the Feast Job Service URL
Args:
value: Feast Job Service URL
"""
self._config.set(opt.JOB_SERVICE_URL, value)
@property
def core_secure(self) -> bool:
"""
Retrieve Feast Core client-side SSL/TLS setting
Returns:
Whether client-side SSL/TLS is enabled
"""
return self._config.getboolean(opt.CORE_ENABLE_SSL)
@core_secure.setter
def core_secure(self, value: bool):
"""
Set the Feast Core client-side SSL/TLS setting
Args:
value: True to enable client-side SSL/TLS
"""
self._config.set(opt.CORE_ENABLE_SSL, value)
@property
def serving_secure(self) -> bool:
"""
Retrieve Feast Serving client-side SSL/TLS setting
Returns:
Whether client-side SSL/TLS is enabled
"""
return self._config.getboolean(opt.SERVING_ENABLE_SSL)
@serving_secure.setter
def serving_secure(self, value: bool):
"""
Set the Feast Serving client-side SSL/TLS setting
Args:
value: True to enable client-side SSL/TLS
"""
self._config.set(opt.SERVING_ENABLE_SSL, value)
@property
def job_service_secure(self) -> bool:
"""
Retrieve Feast Job Service client-side SSL/TLS setting
Returns:
Whether client-side SSL/TLS is enabled
"""
return self._config.getboolean(opt.JOB_SERVICE_ENABLE_SSL)
@job_service_secure.setter
def job_service_secure(self, value: bool):
"""
Set the Feast Job Service client-side SSL/TLS setting
Args:
value: True to enable client-side SSL/TLS
"""
self._config.set(opt.JOB_SERVICE_ENABLE_SSL, value)
def version(self, sdk_only=False):
"""
Returns version information from Feast Core and Feast Serving
"""
import pkg_resources
try:
sdk_version = pkg_resources.get_distribution("feast").version
except pkg_resources.DistributionNotFound:
sdk_version = "local build"
if sdk_only:
return sdk_version
result = {
"sdk": {
"version": sdk_version
},
"serving": "not configured",
"core": "not configured",
}
if self.serving_url:
serving_version = self._serving_service.GetFeastServingInfo(
GetFeastServingInfoRequest(),
timeout=self._config.getint(opt.GRPC_CONNECTION_TIMEOUT),
metadata=self._get_grpc_metadata(),
).version
result["serving"] = {
"url": self.serving_url,
"version": serving_version
}
if not self._use_object_store_registry and self.core_url:
core_version = self._core_service.GetFeastCoreVersion(
GetFeastCoreVersionRequest(),
timeout=self._config.getint(opt.GRPC_CONNECTION_TIMEOUT),
metadata=self._get_grpc_metadata(),
).version
result["core"] = {"url": self.core_url, "version": core_version}
return result
@property
def project(self) -> str:
"""
Retrieve currently active project
Returns:
Project name
"""
if not self._config.get(opt.PROJECT):
raise ValueError("No project has been configured.")
return self._config.get(opt.PROJECT)
def set_project(self, project: Optional[str] = None):
"""
Set currently active Feast project
Args:
project: Project to set as active. If unset, will reset to the default project.
"""
if project is None:
project = opt().PROJECT
self._config.set(opt.PROJECT, project)
def list_projects(self) -> List[str]:
"""
List all active Feast projects
Returns:
List of project names
"""
if self._use_object_store_registry:
raise NotImplementedError(
"Projects are not implemented for object store registry.")
else:
response = self._core_service.ListProjects(
ListProjectsRequest(),
timeout=self._config.getint(opt.GRPC_CONNECTION_TIMEOUT),
metadata=self._get_grpc_metadata(),
) # type: ListProjectsResponse
return list(response.projects)
def create_project(self, project: str):
"""
Creates a Feast project
Args:
project: Name of project
"""
if self._use_object_store_registry:
raise NotImplementedError(
"Projects are not implemented for object store registry.")
else:
self._core_service.CreateProject(
CreateProjectRequest(name=project),
timeout=self._config.getint(opt.GRPC_CONNECTION_TIMEOUT),
metadata=self._get_grpc_metadata(),
) # type: CreateProjectResponse
def archive_project(self, project):
"""
Archives a project. Project will still continue to function for
ingestion and retrieval, but will be in a read-only state. It will
also not be visible from the Core API for management purposes.
Args:
project: Name of project to archive
"""
if self._use_object_store_registry:
raise NotImplementedError(
"Projects are not implemented for object store registry.")
else:
try:
self._core_service.ArchiveProject(
ArchiveProjectRequest(name=project),
timeout=self._config.getint(opt.GRPC_CONNECTION_TIMEOUT),
metadata=self._get_grpc_metadata(),
) # type: ArchiveProjectResponse
except grpc.RpcError as e:
raise grpc.RpcError(e.details())
# revert to the default project
if self._project == project:
self._project = opt().PROJECT
def apply(
self,
objects: Union[List[Union[Entity, FeatureTable]], Entity,
FeatureTable],
project: str = None,
):
"""
Idempotently registers entities and feature tables with Feast Core. Either a single
entity or feature table or a list can be provided.
Args:
objects: List of entities and/or feature tables that will be registered
Examples:
>>> from feast import Client
>>> from feast.entity import Entity
>>> from feast.value_type import ValueType
>>>
>>> feast_client = Client(core_url="localhost:6565")
>>> entity = Entity(
>>> name="driver_entity",
>>> description="Driver entity for car rides",
>>> value_type=ValueType.STRING,
>>> labels={
>>> "key": "val"
>>> }
>>> )
>>> feast_client.apply(entity)
"""
self._tele.log("apply")
if project is None:
project = self.project
if not isinstance(objects, list):
objects = [objects]
for obj in objects:
if isinstance(obj, Entity):
self._apply_entity(project, obj) # type: ignore
elif isinstance(obj, FeatureTable):
self._apply_feature_table(project, obj) # type: ignore
else:
raise ValueError(
f"Could not determine object type to apply {obj} with type {type(obj)}. Type must be Entity or FeatureTable."
)
def apply_entity(self,
entities: Union[List[Entity], Entity],
project: str = None):
"""
Deprecated. Please see apply().
"""
warnings.warn(
"The method apply_entity() is being deprecated. Please use apply() instead. Feast 0.10 and onwards will not support apply_entity().",
DeprecationWarning,
)
if project is None:
project = self.project
if not isinstance(entities, list):
entities = [entities]
for entity in entities:
if isinstance(entity, Entity):
self._apply_entity(project, entity) # type: ignore
continue
raise ValueError(
f"Could not determine entity type to apply {entity}")
def _apply_entity(self, project: str, entity: Entity):
"""
Registers a single entity with Feast
Args:
entity: Entity that will be registered
"""
if self._use_object_store_registry:
return self._registry.apply_entity(entity, project)
else:
entity.is_valid()
entity_proto = entity.to_spec_proto()
# Convert the entity to a request and send to Feast Core
try:
apply_entity_response = self._core_service.ApplyEntity(
ApplyEntityRequest(project=project,
spec=entity_proto), # type: ignore
timeout=self._config.getint(opt.GRPC_CONNECTION_TIMEOUT),
metadata=self._get_grpc_metadata(),
) # type: ApplyEntityResponse
except grpc.RpcError as e:
raise grpc.RpcError(e.details())
# Extract the returned entity
applied_entity = Entity.from_proto(apply_entity_response.entity)
# Deep copy from the returned entity to the local entity
entity._update_from_entity(applied_entity)
def list_entities(self,
project: str = None,
labels: Dict[str, str] = dict()) -> List[Entity]:
"""
Retrieve a list of entities from Feast Core
Args:
project: Filter entities based on project name
labels: User-defined labels that these entities are associated with
Returns:
List of entities
"""
if project is None:
project = self.project
if self._use_object_store_registry:
return self._registry.list_entities(project)
else:
filter = ListEntitiesRequest.Filter(project=project, labels=labels)
# Get latest entities from Feast Core
entity_protos = self._core_service.ListEntities(
ListEntitiesRequest(filter=filter),
metadata=self._get_grpc_metadata(),
) # type: ListEntitiesResponse
# Extract entities and return
entities = []
for entity_proto in entity_protos.entities:
entity = Entity.from_proto(entity_proto)
entity._client = self
entities.append(entity)
return entities
def get_entity(self, name: str, project: str = None) -> Entity:
"""
Retrieves an entity.
Args:
project: Feast project that this entity belongs to
name: Name of entity
Returns:
Returns either the specified entity, or raises an exception if
none is found
"""
self._tele.log("get_entity")
if project is None:
project = self.project
if self._use_object_store_registry:
return self._registry.get_entity(name, project)
else:
try:
get_entity_response = self._core_service.GetEntity(
GetEntityRequest(project=project, name=name.strip()),
metadata=self._get_grpc_metadata(),
) # type: GetEntityResponse
except grpc.RpcError as e:
raise grpc.RpcError(e.details())
entity = Entity.from_proto(get_entity_response.entity)
return entity
def apply_feature_table(
self,
feature_tables: Union[List[FeatureTable], FeatureTable],
project: str = None,
):
"""
Deprecated. Please see apply().
"""
warnings.warn(
"The method apply_feature_table() is being deprecated. Please use apply() instead. Feast 0.10 and onwards will not support apply_feature_table().",
DeprecationWarning,
)
if project is None:
project = self.project
if not isinstance(feature_tables, list):
feature_tables = [feature_tables]
for feature_table in feature_tables:
if isinstance(feature_table, FeatureTable):
self._apply_feature_table(project,
feature_table) # type: ignore
continue
raise ValueError(
f"Could not determine feature table type to apply {feature_table}"
)
def _apply_feature_table(self, project: str, feature_table: FeatureTable):
"""
Registers a single feature table with Feast
Args:
feature_table: Feature table that will be registered
"""
if self._use_object_store_registry:
return self._registry.apply_feature_table(feature_table, project)
else:
feature_table.is_valid()
feature_table_proto = feature_table.to_spec_proto()
# Convert the feature table to a request and send to Feast Core
try:
apply_feature_table_response = self._core_service.ApplyFeatureTable(
ApplyFeatureTableRequest(
project=project,
table_spec=feature_table_proto), # type: ignore
timeout=self._config.getint(opt.GRPC_CONNECTION_TIMEOUT),
metadata=self._get_grpc_metadata(),
) # type: ApplyFeatureTableResponse
except grpc.RpcError as e:
raise grpc.RpcError(e.details())
# Extract the returned feature table
applied_feature_table = FeatureTable.from_proto(
apply_feature_table_response.table)
# Deep copy from the returned feature table to the local entity
feature_table._update_from_feature_table(applied_feature_table)
def list_feature_tables(
self,
project: str = None,
labels: Dict[str, str] = dict()
) -> List[FeatureTable]:
"""
Retrieve a list of feature tables from Feast Core
Args:
project: Filter feature tables based on project name
Returns:
List of feature tables
"""
if project is None:
project = self.project
if self._use_object_store_registry:
return self._registry.list_feature_tables(project)
else:
filter = ListFeatureTablesRequest.Filter(project=project,
labels=labels)
# Get latest feature tables from Feast Core
feature_table_protos = self._core_service.ListFeatureTables(
ListFeatureTablesRequest(filter=filter),
metadata=self._get_grpc_metadata(),
) # type: ListFeatureTablesResponse
# Extract feature tables and return
feature_tables = []
for feature_table_proto in feature_table_protos.tables:
feature_table = FeatureTable.from_proto(feature_table_proto)
feature_table._client = self
feature_tables.append(feature_table)
return feature_tables
def get_feature_table(self,
name: str,
project: str = None) -> FeatureTable:
"""
Retrieves a feature table.
Args:
project: Feast project that this feature table belongs to
name: Name of feature table
Returns:
Returns either the specified feature table, or raises an exception if
none is found
"""
self._tele.log("get_feature_table")
if project is None:
project = self.project
if self._use_object_store_registry:
return self._registry.get_feature_table(name, project)
else:
try:
get_feature_table_response = self._core_service.GetFeatureTable(
GetFeatureTableRequest(project=project, name=name.strip()),
metadata=self._get_grpc_metadata(),
) # type: GetFeatureTableResponse
except grpc.RpcError as e:
raise grpc.RpcError(e.details())
return FeatureTable.from_proto(get_feature_table_response.table)
def delete_feature_table(self, name: str, project: str = None) -> None:
"""
Deletes a feature table.
Args:
project: Feast project that this feature table belongs to
name: Name of feature table
"""
if project is None:
project = self.project
if self._use_object_store_registry:
return self._registry.delete_feature_table(name, project)
else:
try:
self._core_service.DeleteFeatureTable(
DeleteFeatureTableRequest(project=project,
name=name.strip()),
metadata=self._get_grpc_metadata(),
)
except grpc.RpcError as e:
raise grpc.RpcError(e.details())
def list_features_by_ref(
self,
project: str = None,
entities: List[str] = list(),
labels: Dict[str, str] = dict(),
) -> Dict[FeatureRef, Feature]:
"""
Retrieve a dictionary of feature reference to feature from Feast Core based on filters provided.
Args:
project: Feast project that these features belongs to
entities: Feast entity that these features are associated with
labels: Feast labels that these features are associated with
Returns:
Dictionary of <feature references: features>
Examples:
>>> from feast import Client
>>>
>>> feast_client = Client(core_url="localhost:6565")
>>> features = feast_client.list_features(project="test_project", entities=["driver_id"], labels={"key1":"val1","key2":"val2"})
>>> print(features)
"""
if self._use_object_store_registry:
raise NotImplementedError(
"This function is not implemented for object store registry.")
else:
if project is None:
project = self.project
filter = ListFeaturesRequest.Filter(project=project,
entities=entities,
labels=labels)
feature_protos = self._core_service.ListFeatures(
ListFeaturesRequest(filter=filter),
metadata=self._get_grpc_metadata(),
) # type: ListFeaturesResponse
# Extract features and return
features_dict = {}
for ref_str, feature_proto in feature_protos.features.items():
feature_ref = FeatureRef.from_str(ref_str)
feature = Feature.from_proto(feature_proto)
features_dict[feature_ref] = feature
return features_dict
def ingest(
self,
feature_table: Union[str, FeatureTable],
source: Union[pd.DataFrame, str],
project: str = None,
chunk_size: int = 10000,
max_workers: int = max(CPU_COUNT - 1, 1),
timeout: int = int(opt().BATCH_INGESTION_PRODUCTION_TIMEOUT),
) -> None:
"""
Batch load feature data into a FeatureTable.
Args:
feature_table (typing.Union[str, feast.feature_table.FeatureTable]):
FeatureTable object or the string name of the feature table
source (typing.Union[pd.DataFrame, str]):
Either a file path or Pandas Dataframe to ingest into Feast
Files that are currently supported:
* parquet
* csv
* json
project: Feast project to locate FeatureTable
chunk_size (int):
Amount of rows to load and ingest at a time.
max_workers (int):
Number of worker processes to use to encode values.
timeout (int):
Timeout in seconds to wait for completion.
Examples:
>>> from feast import Client
>>>
>>> client = Client(core_url="localhost:6565")
>>> ft_df = pd.DataFrame(
>>> {
>>> "datetime": [pd.datetime.now()],
>>> "driver": [1001],
>>> "rating": [4.3],
>>> }
>>> )
>>> client.set_project("project1")
>>>
>>> driver_ft = client.get_feature_table("driver")
>>> client.ingest(driver_ft, ft_df)
"""
self._tele.log("ingest")
if project is None:
project = self.project
if isinstance(feature_table, str):
name = feature_table
if isinstance(feature_table, FeatureTable):
name = feature_table.name
fetched_feature_table: Optional[FeatureTable] = self.get_feature_table(
name, project)
if fetched_feature_table is not None:
feature_table = fetched_feature_table
else:
raise Exception(f"FeatureTable, {name} cannot be found.")
# Check 1) Only parquet file format for FeatureTable batch source is supported
if (feature_table.batch_source
and issubclass(type(feature_table.batch_source), FileSource)
and isinstance(
type(feature_table.batch_source.file_options.file_format),
ParquetFormat)):
raise Exception(
f"No suitable batch source found for FeatureTable, {name}."
f"Only BATCH_FILE source with parquet format is supported for batch ingestion."
)
pyarrow_table, column_names = _read_table_from_source(source)
# Check 2) Check if FeatureTable batch source field mappings can be found in provided source table
_check_field_mappings(
column_names,
name,
feature_table.batch_source.event_timestamp_column,
feature_table.batch_source.field_mapping,
)
dir_path = None
with_partitions = False
if (issubclass(type(feature_table.batch_source), FileSource)
and feature_table.batch_source.date_partition_column):
with_partitions = True
dest_path = _write_partitioned_table_from_source(
column_names,
pyarrow_table,
feature_table.batch_source.date_partition_column,
feature_table.batch_source.event_timestamp_column,
)
else:
dir_path, dest_path = _write_non_partitioned_table_from_source(
column_names,
pyarrow_table,
chunk_size,
max_workers,
)
try:
if issubclass(type(feature_table.batch_source), FileSource):
file_url = feature_table.batch_source.file_options.file_url.rstrip(
"*")
_upload_to_file_source(file_url, with_partitions, dest_path,
self._config)
if issubclass(type(feature_table.batch_source), BigQuerySource):
bq_table_ref = feature_table.batch_source.bigquery_options.table_ref
feature_table_timestamp_column = (
feature_table.batch_source.event_timestamp_column)
_upload_to_bq_source(bq_table_ref,
feature_table_timestamp_column, dest_path)
finally:
# Remove parquet file(s) that were created earlier
print("Removing temporary file(s)...")
if dir_path:
shutil.rmtree(dir_path)
print(
"Data has been successfully ingested into FeatureTable batch source."
)
def _get_grpc_metadata(self):
"""
Returns a metadata tuple to attach to gRPC requests. This is primarily
used when authentication is enabled but SSL/TLS is disabled.
Returns: Tuple of metadata to attach to each gRPC call
"""
if self._config.getboolean(opt.ENABLE_AUTH) and self._auth_metadata:
return self._auth_metadata.get_signed_meta()
return ()
def get_online_features(
self,
feature_refs: List[str],
entity_rows: List[Dict[str, Any]],
project: Optional[str] = None,
) -> OnlineResponse:
"""
Retrieves the latest online feature data from Feast Serving.
Args:
feature_refs: List of feature references that will be returned for each entity.
Each feature reference should have the following format:
"feature_table:feature" where "feature_table" & "feature" refer to
the feature and feature table names respectively.
Only the feature name is required.
entity_rows: A list of dictionaries where each key-value is an entity-name, entity-value pair.
project: Optionally specify the the project override. If specified, uses given project for retrieval.
Overrides the projects specified in Feature References if also are specified.
Returns:
GetOnlineFeaturesResponse containing the feature data in records.
Each EntityRow provided will yield one record, which contains
data fields with data value and field status metadata (if included).
Examples:
>>> from feast import Client
>>>
>>> feast_client = Client(core_url="localhost:6565", serving_url="localhost:6566")
>>> feature_refs = ["sales:daily_transactions"]
>>> entity_rows = [{"customer_id": 0},{"customer_id": 1}]
>>>
>>> online_response = feast_client.get_online_features(
>>> feature_refs, entity_rows, project="my_project")
>>> online_response_dict = online_response.to_dict()
>>> print(online_response_dict)
{'sales:daily_transactions': [1.1,1.2], 'sales:customer_id': [0,1]}
"""
self._tele.log("get_online_features")
try:
response = self._serving_service.GetOnlineFeaturesV2(
GetOnlineFeaturesRequestV2(
features=_build_feature_references(
feature_ref_strs=feature_refs),
entity_rows=_infer_online_entity_rows(entity_rows),
project=project if project is not None else self.project,
),
timeout=self._config.getint(opt.GRPC_CONNECTION_TIMEOUT),
metadata=self._get_grpc_metadata(),
)
except grpc.RpcError as e:
raise grpc.RpcError(e.details())
response = OnlineResponse(response)
return response