Skip to main content
Version: 1.0

Test Realtime Features

Import libraries and select your workspace​

import tecton
import pandas
from datetime import datetime

ws = tecton.get_workspace("prod")

Realtime Feature Views with no Feature View Dependencies​

Load a Realtime Feature View​

fv = ws.get_feature_view("transaction_amount_is_high")
fv.summary()

Execute a Realtime Feature View Online​

Because this Realtime Feature View has no Batch or Stream Feature View dependencies it only needs request data to execute.

fv.get_online_features(request_data={"amt": 17000}).to_dict()
Out: {"transaction_amount_is_high": True}

Execute a Realtime Feature View Offline​

Create an events DataFrame with the request data to transform.

events = pandas.DataFrame({"amt": [200, 20]})
features_df = fv.get_features_for_events(events=events).to_pandas()
display(features_df)
amttransaction_amount_is_high__transaction_amount_is_high
0200True
120False

Realtime Feature Views with Feature View Dependencies​

Load a Realtime Feature View​

fv = ws.get_feature_view("transaction_amount_is_higher_than_average")

Run Feature View Transformation Pipeline with Mock Inputs​

To use run_transformation() on Realtime Feature View, all Feature View inputs must be mocked.

transaction_request = pandas.DataFrame([{"amt": 100.0}])

user_transaction_amount_metrics = pandas.DataFrame([{"amt_mean_1d_10m": 200.0}])

result = fv.run_transformation(
transaction_request=transaction_request,
user_transaction_amount_metrics=user_transaction_amount_metrics,
)
print(result)
{"transaction_amount_is_higher_than_average": 0}

Execute Realtime Feature View with Dependencies Offline​

Because this Realtime Feature View depends on another Feature View, we will first preview the dependent Feature View. We can use this to select keys that will be needed for historical lookups in order for the Realtime Feature View to run. The dependent column in the Feature View below is amt_mean_1d_10m.

dependent_fv = ws.get_feature_view("user_transaction_amount_metrics")
result_dataframe = dependent_fv.get_features_in_range(
start_time=datetime(2022, 5, 1), end_time=datetime(2022, 5, 4), from_source=True
).to_pandas()
display(result_dataframe)
user_idamt_sum_1h_10mamt_sum_1d_10mamt_sum_3d_10mamt_mean_1h_10mamt_mean_1d_10mamt_mean_3d_10m_valid_from_valid_to
0user_1120.66120.661404.69120.66120.66351.1732022-05-01 00:00:002022-05-02 00:00:00
1user_2134.73134.73144.07134.73134.7372.0352022-05-01 00:00:002022-05-02 00:00:00
2user_376.4576.45181.476.4576.4560.46672022-05-01 00:00:002022-05-02 00:00:00
3user_32.1378.58183.532.1339.2945.88252022-05-02 00:00:002022-05-03 00:00:00
4user_39.9688.54193.499.9629.513338.6982022-05-03 00:00:002022-05-04 00:00:00

Create an events DataFrame with events to look up. For more information on the events dataframe, check out Selecting Sample Keys and Timestamps.

events = pandas.DataFrame(
{
"user_id": [
"user_2",
"user_337750317412",
"user_222506789984",
"user_337750317412",
],
"timestamp": [
datetime(2022, 5, 1, 19, 51),
datetime(2022, 5, 2, 5, 0),
datetime(2022, 5, 1, 21, 11),
datetime(2022, 5, 1, 7, 21),
],
"amt": [71.82, 80.98, 400.55, 66.57],
}
)
display(events)
user_idtimestampamt
0user_22022-05-01 19:51:0071.82
1user_32022-05-02 05:00:0080.98
2user_12022-05-01 21:11:00400.55
3user_32022-05-01 07:21:0066.57
result_dataframe = fv.get_features_for_events(events=events, from_source=True).to_pandas()
display(result_dataframe)
user_idtimestampamttransaction_amount_is_higher_than_average__transaction_amount_is_higher_than_average
0user_22022-05-01 19:51:0071.820
1user_32022-05-02 05:00:0080.981
2user_12022-05-01 21:11:00400.551
3user_32022-05-03 07:21:0026.570

Execute Realtime Feature View with Dependencies Online​

# This will compare a transaction amount against the user's most recent average 24h transaction amount.
fv.get_online_features(request_data={"amt": 150}, join_keys={"user_id": "user_724235628997"}).to_dict()
Out: {"transaction_amount_is_higher_than_average": 1}

Was this page helpful?