Skip to content

Make created timestamp column optional #1108

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
feast-ci-bot merged 2 commits into from
Oct 29, 2020
Merged

Make created timestamp column optional #1108

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
5fb0eb0
Make created timestamp column optional
khorshuheng Oct 29, 2020
44ea663
Fix argument for FileSource under e2e tests
khorshuheng Oct 29, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 1 addition & 2 deletions docs/user-guide/feature-retrieval.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -36,8 +36,7 @@ feature_refs = [
# Define entity source
entity_source = FileSource(
"event_timestamp",
"created_timestamp",
"parquet",
ParquetFormat(),
"gs://some-bucket/customer"
)

Expand Down
4 changes: 1 addition & 3 deletions sdk/python/feast/cli.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -485,9 +485,7 @@ def get_historical_features(features: str, entity_df_path: str, destination: str

entity_df["event_timestamp"] = pandas.to_datetime(entity_df["event_timestamp"])

uploaded_df = client.stage_dataframe(
entity_df, "event_timestamp", "created_timestamp"
)
uploaded_df = client.stage_dataframe(entity_df, "event_timestamp")

job = client.get_historical_features(features.split(","), uploaded_df,)
print(job.get_output_file_uri())
Expand Down
20 changes: 7 additions & 13 deletions sdk/python/feast/client.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -881,10 +881,11 @@ def get_historical_features(

Examples:
>>> from feast import Client
>>> from feast.data_format import ParquetFormat
>>> from datetime import datetime
>>> feast_client = Client(core_url="localhost:6565")
>>> feature_refs = ["bookings:bookings_7d", "bookings:booking_14d"]
>>> entity_source = FileSource("event_timestamp", "parquet", "gs://some-bucket/customer")
>>> entity_source = FileSource("event_timestamp", ParquetFormat(), "gs://some-bucket/customer")
>>> feature_retrieval_job = feast_client.get_historical_features(
>>> feature_refs, entity_source, project="my_project")
>>> output_file_uri = feature_retrieval_job.get_output_file_uri()
Expand Down Expand Up @@ -916,10 +917,7 @@ def get_historical_features(
df_export_path.name, bucket, entity_staging_uri.path.lstrip("/")
)
entity_source = FileSource(
"event_timestamp",
"created_timestamp",
ParquetFormat(),
entity_staging_uri.geturl(),
"event_timestamp", ParquetFormat(), entity_staging_uri.geturl(),
)

return start_historical_feature_retrieval_job(
Expand Down Expand Up @@ -955,12 +953,13 @@ def get_historical_features_df(

Examples:
>>> from feast import Client
>>> from feast.data_format import ParquetFormat
>>> from datetime import datetime
>>> from pyspark.sql import SparkSession
>>> spark = SparkSession.builder.getOrCreate()
>>> feast_client = Client(core_url="localhost:6565")
>>> feature_refs = ["bookings:bookings_7d", "bookings:booking_14d"]
>>> entity_source = FileSource("event_timestamp", "parquet", "gs://some-bucket/customer")
>>> entity_source = FileSource("event_timestamp", ParquetFormat, "gs://some-bucket/customer")
>>> df = feast_client.get_historical_features(
>>> feature_refs, entity_source, project="my_project")
"""
Expand Down Expand Up @@ -1011,11 +1010,6 @@ def start_stream_to_online_ingestion(
return start_stream_to_online_ingestion(feature_table, extra_jars or [], self)

def stage_dataframe(
self,
df: pd.DataFrame,
event_timestamp_column: str,
created_timestamp_column: str,
self, df: pd.DataFrame, event_timestamp_column: str,
) -> FileSource:
return stage_dataframe(
df, event_timestamp_column, created_timestamp_column, self
)
return stage_dataframe(df, event_timestamp_column, self)
16 changes: 8 additions & 8 deletions sdk/python/feast/data_source.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -348,13 +348,13 @@ class DataSource:
def __init__(
self,
event_timestamp_column: str,
created_timestamp_column: str,
field_mapping: Optional[Dict[str, str]] = dict(),
created_timestamp_column: Optional[str] = "",
field_mapping: Optional[Dict[str, str]] = None,
date_partition_column: Optional[str] = "",
):
self._event_timestamp_column = event_timestamp_column
self._created_timestamp_column = created_timestamp_column
self._field_mapping = field_mapping
self._field_mapping = field_mapping if field_mapping else {}
self._date_partition_column = date_partition_column

def __eq__(self, other):
Expand Down Expand Up @@ -409,7 +409,7 @@ def created_timestamp_column(self):
@created_timestamp_column.setter
def created_timestamp_column(self, created_timestamp_column):
"""
Sets the event timestamp column of this data source
Sets the created timestamp column of this data source
"""
self._created_timestamp_column = created_timestamp_column

Expand Down Expand Up @@ -498,10 +498,10 @@ class FileSource(DataSource):
def __init__(
self,
event_timestamp_column: str,
created_timestamp_column: str,
file_format: FileFormat,
file_url: str,
field_mapping: Optional[Dict[str, str]] = dict(),
created_timestamp_column: Optional[str] = "",
field_mapping: Optional[Dict[str, str]] = None,
date_partition_column: Optional[str] = "",
):
super().__init__(
Expand Down Expand Up @@ -555,9 +555,9 @@ class BigQuerySource(DataSource):
def __init__(
self,
event_timestamp_column: str,
created_timestamp_column: str,
table_ref: str,
field_mapping: Optional[Dict[str, str]] = dict(),
created_timestamp_column: Optional[str] = "",
field_mapping: Optional[Dict[str, str]] = None,
date_partition_column: Optional[str] = "",
):
super().__init__(
Expand Down
5 changes: 1 addition & 4 deletions sdk/python/feast/pyspark/abc.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -458,10 +458,7 @@ def start_stream_to_online_ingestion(

@abc.abstractmethod
def stage_dataframe(
self,
df: pandas.DataFrame,
event_timestamp_column: str,
created_timestamp_column: str,
self, df: pandas.DataFrame, event_timestamp_column: str,
) -> FileSource:
"""
Upload a pandas dataframe so it is available to the Spark cluster.
Expand Down
2 changes: 1 addition & 1 deletion sdk/python/feast/pyspark/historical_feature_retrieval_job.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -86,7 +86,7 @@ def __init__(
format: str,
path: str,
event_timestamp_column: str,
created_timestamp_column: Optional[str] = None,
created_timestamp_column: Optional[str] = "",
field_mapping: Optional[Dict[str, str]] = None,
options: Optional[Dict[str, str]] = None,
):
Expand Down
8 changes: 2 additions & 6 deletions sdk/python/feast/pyspark/launcher.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -243,10 +243,6 @@ def start_stream_to_online_ingestion(
)


def stage_dataframe(
df, event_timestamp_column: str, created_timestamp_column: str, client: "Client"
) -> FileSource:
def stage_dataframe(df, event_timestamp_column: str, client: "Client") -> FileSource:
launcher = resolve_launcher(client._config)
return launcher.stage_dataframe(
df, event_timestamp_column, created_timestamp_column,
)
return launcher.stage_dataframe(df, event_timestamp_column)
5 changes: 1 addition & 4 deletions sdk/python/feast/pyspark/launchers/aws/emr.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -291,9 +291,7 @@ def start_stream_to_online_ingestion(

return EmrStreamIngestionJob(self._emr_client(), job_ref)

def stage_dataframe(
self, df: pandas.DataFrame, event_timestamp: str, created_timestamp_column: str
) -> FileSource:
def stage_dataframe(self, df: pandas.DataFrame, event_timestamp: str) -> FileSource:
with tempfile.NamedTemporaryFile() as f:
df.to_parquet(f)
file_url = _s3_upload(
Expand All @@ -304,7 +302,6 @@ def stage_dataframe(
)
return FileSource(
event_timestamp_column=event_timestamp,
created_timestamp_column=created_timestamp_column,
file_format=ParquetFormat(),
file_url=file_url,
)
Expand Down
4 changes: 1 addition & 3 deletions sdk/python/feast/pyspark/launchers/gcloud/dataproc.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -205,9 +205,7 @@ def start_stream_to_online_ingestion(
cancel_fn = partial(self.dataproc_cancel, operation.metadata.job_id)
return DataprocStreamingIngestionJob(operation, cancel_fn)

def stage_dataframe(
self, df, event_timestamp_column: str, created_timestamp_column: str,
):
def stage_dataframe(self, df, event_timestamp_column: str):
raise NotImplementedError

def get_job_by_id(self, job_id: str) -> SparkJob:
Expand Down
4 changes: 1 addition & 3 deletions sdk/python/feast/pyspark/launchers/standalone/local.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -224,9 +224,7 @@ def start_stream_to_online_ingestion(
ui_port,
)

def stage_dataframe(
self, df, event_timestamp_column: str, created_timestamp_column: str,
):
def stage_dataframe(self, df, event_timestamp_column: str):
raise NotImplementedError

def get_job_by_id(self, job_id: str) -> SparkJob:
Expand Down
24 changes: 19 additions & 5 deletions sdk/python/tests/test_historical_feature_retrieval.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -179,7 +179,10 @@ def transactions_feature_table(spark, client):
spark, "transactions", schema, df_data
)
file_source = FileSource(
"event_timestamp", "created_timestamp", ParquetFormat(), file_uri
event_timestamp_column="event_timestamp",
created_timestamp_column="created_timestamp",
file_format=ParquetFormat(),
file_url=file_uri,
)
features = [
Feature("total_transactions", ValueType.DOUBLE),
Expand Down Expand Up @@ -225,7 +228,10 @@ def bookings_feature_table(spark, client):
temp_dir, file_uri = create_temp_parquet_file(spark, "bookings", schema, df_data)

file_source = FileSource(
"event_timestamp", "created_timestamp", ParquetFormat(), file_uri
event_timestamp_column="event_timestamp",
created_timestamp_column="created_timestamp",
file_format=ParquetFormat(),
file_url=file_uri,
)
features = [Feature("total_completed_bookings", ValueType.INT32)]
max_age = Duration()
Expand Down Expand Up @@ -270,7 +276,11 @@ def bookings_feature_table_with_mapping(spark, client):
temp_dir, file_uri = create_temp_parquet_file(spark, "bookings", schema, df_data)

file_source = FileSource(
"datetime", "created_datetime", ParquetFormat(), file_uri, {"id": "driver_id"}
event_timestamp_column="datetime",
created_timestamp_column="created_datetime",
file_format=ParquetFormat(),
file_url=file_uri,
field_mapping={"id": "driver_id"},
)
features = [Feature("total_completed_bookings", ValueType.INT32)]
max_age = Duration()
Expand Down Expand Up @@ -309,7 +319,9 @@ def test_historical_feature_retrieval_from_local_spark_session(
spark, "customer_driver_pair", schema, df_data
)
customer_driver_pairs_source = FileSource(
"event_timestamp", "created_timestamp", ParquetFormat(), file_uri
event_timestamp_column="event_timestamp",
file_format=ParquetFormat(),
file_url=file_uri,
)
joined_df = client.get_historical_features_df(
["transactions:total_transactions", "bookings:total_completed_bookings"],
Expand Down Expand Up @@ -356,7 +368,9 @@ def test_historical_feature_retrieval_with_field_mappings_from_local_spark_sessi
]
temp_dir, file_uri = create_temp_parquet_file(spark, "drivers", schema, df_data)
entity_source = FileSource(
"event_timestamp", "created_timestamp", ParquetFormat(), file_uri
event_timestamp_column="event_timestamp",
file_format=ParquetFormat(),
file_url=file_uri,
)
joined_df = client.get_historical_features_df(
["bookings:total_completed_bookings"], entity_source,
Expand Down
8 changes: 4 additions & 4 deletions tests/e2e/test_historical_features.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -45,10 +45,10 @@ def test_historical_features(feast_client: Client, local_staging_path: str):
Feature("total_transactions", ValueType.DOUBLE),
],
batch_source=FileSource(
"event_timestamp",
"created_timestamp",
ParquetFormat(),
os.path.join(local_staging_path, "transactions"),
event_timestamp_column="event_timestamp",
created_timestamp_column="created_timestamp",
file_format=ParquetFormat(),
file_url=os.path.join(local_staging_path, "transactions"),
),
max_age=max_age,
)
Expand Down
16 changes: 8 additions & 8 deletions tests/e2e/test_online_features.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -47,10 +47,10 @@ def test_offline_ingestion(feast_client: Client, local_staging_path: str):
entities=["s2id"],
features=[Feature("unique_drivers", ValueType.INT64)],
batch_source=FileSource(
"event_timestamp",
"event_timestamp",
ParquetFormat(),
os.path.join(local_staging_path, "batch-storage"),
event_timestamp_column="event_timestamp",
created_timestamp_column="event_timestamp",
file_format=ParquetFormat(),
file_url=os.path.join(local_staging_path, "batch-storage"),
),
)

Expand Down Expand Up @@ -92,10 +92,10 @@ def test_streaming_ingestion(
entities=["s2id"],
features=[Feature("unique_drivers", ValueType.INT64)],
batch_source=FileSource(
"event_timestamp",
"event_timestamp",
ParquetFormat(),
os.path.join(local_staging_path, "batch-storage"),
event_timestamp_column="event_timestamp",
created_timestamp_column="event_timestamp",
file_format=ParquetFormat(),
file_url=os.path.join(local_staging_path, "batch-storage"),
),
stream_source=KafkaSource(
"event_timestamp",
Expand Down