Terjemahan disediakan oleh mesin penerjemah. Jika konten terjemahan yang diberikan bertentangan dengan versi bahasa Inggris aslinya, utamakan versi bahasa Inggris.

import boto3
import sagemaker
from sagemaker.session import Session

sagemaker_session = sagemaker.Session()
region = sagemaker_session.boto_region_name
boto_session = boto3.Session(region_name=region)
role = sagemaker.get_execution_role()
default_bucket = sagemaker_session.default_bucket()
prefix = 'sagemaker-featurestore'
offline_feature_store_bucket = 's3://{}/{}'.format(default_bucket, prefix)

sagemaker_client = boto_session.client(service_name='sagemaker', region_name=region)
featurestore_runtime = boto_session.client(service_name='sagemaker-featurestore-runtime', region_name=region)

feature_store_session = Session(
    boto_session=boto_session,
    sagemaker_client=sagemaker_client,
    sagemaker_featurestore_runtime_client=featurestore_runtime
)

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import io

s3_client = boto3.client(service_name='s3', region_name=region)

fraud_detection_bucket_name = 'sagemaker-featurestore-fraud-detection'
identity_file_key = 'sampled_identity.csv'
transaction_file_key = 'sampled_transactions.csv'

identity_data_object = s3_client.get_object(Bucket=fraud_detection_bucket_name, Key=identity_file_key)
transaction_data_object = s3_client.get_object(Bucket=fraud_detection_bucket_name, Key=transaction_file_key)

identity_data = pd.read_csv(io.BytesIO(identity_data_object['Body'].read()))
transaction_data = pd.read_csv(io.BytesIO(transaction_data_object['Body'].read()))

identity_data = identity_data.round(5)
transaction_data = transaction_data.round(5)

identity_data = identity_data.fillna(0)
transaction_data = transaction_data.fillna(0)

# Feature transformations for this dataset are applied before ingestion into FeatureStore.
# One hot encode card4, card6
encoded_card_bank = pd.get_dummies(transaction_data['card4'], prefix = 'card_bank')
encoded_card_type = pd.get_dummies(transaction_data['card6'], prefix = 'card_type')

transformed_transaction_data = pd.concat([transaction_data, encoded_card_type, encoded_card_bank], axis=1)
transformed_transaction_data = transformed_transaction_data.rename(columns={"card_bank_american express": "card_bank_american_express"})

from sagemaker.feature_store.feature_group import FeatureGroup
feature_group_name = "some string for a name"
feature_group = FeatureGroup(name=feature_group_name, sagemaker_session=feature_store_session)

import time
from time import gmtime, strftime, sleep
from sagemaker.feature_store.feature_group import FeatureGroup

identity_feature_group_name = 'identity-feature-group-' + strftime('%d-%H-%M-%S', gmtime())
transaction_feature_group_name = 'transaction-feature-group-' + strftime('%d-%H-%M-%S', gmtime())

identity_feature_group = FeatureGroup(name=identity_feature_group_name, sagemaker_session=feature_store_session)
transaction_feature_group = FeatureGroup(name=transaction_feature_group_name, sagemaker_session=feature_store_session)

record_identifier_name = "TransactionID"
event_time_feature_name = "EventTime"
current_time_sec = int(round(time.time()))
identity_data[event_time_feature_name] = pd.Series([current_time_sec]*len(identity_data), dtype="float64")
transformed_transaction_data[event_time_feature_name] = pd.Series([current_time_sec]*len(transaction_data), dtype="float64")

identity_feature_group.load_feature_definitions(data_frame=identity_data); # output is suppressed
transaction_feature_group.load_feature_definitions(data_frame=transformed_transaction_data); # output is suppressed

# create a FeatureGroup
feature_group.create(
    description = "Some info about the feature group",
    feature_group_name = feature_group_name,
    record_identifier_name = record_identifier_name,
    event_time_feature_name = event_time_feature_name,
    feature_definitions = feature_definitions,
    role_arn = role,
    s3_uri = offline_feature_store_bucket,
    enable_online_store = True,
    online_store_kms_key_id = None,
    offline_store_kms_key_id = None,
    disable_glue_table_creation = False,
    data_catalog_config = None,
    tags = ["tag1","tag2"])

identity_feature_group.create(
    s3_uri=offline_feature_store_bucket,
    record_identifier_name=record_identifier_name,
    event_time_feature_name=event_time_feature_name,
    role_arn=role,
    enable_online_store=True
)

transaction_feature_group.create(
    s3_uri=offline_feature_store_bucket,
    record_identifier_name=record_identifier_name,
    event_time_feature_name=event_time_feature_name,
    role_arn=role,
    enable_online_store=True
)

status = feature_group.describe().get("FeatureGroupStatus")

feature_group.describe()

sagemaker_client.list_feature_groups()

feature_group.ingest(
    data_frame=feature_data, max_workers=3, wait=True
)

record_identifier_value = str(2990130)
featurestore_runtime.get_record(FeatureGroupName=transaction_feature_group_name, RecordIdentifierValueAsString=record_identifier_value)

...
'Record': [{'FeatureName': 'TransactionID', 'ValueAsString': '2990130'},
  {'FeatureName': 'isFraud', 'ValueAsString': '0'},
  {'FeatureName': 'TransactionDT', 'ValueAsString': '152647'},
  {'FeatureName': 'TransactionAmt', 'ValueAsString': '75.0'},
  {'FeatureName': 'ProductCD', 'ValueAsString': 'H'},
  {'FeatureName': 'card1', 'ValueAsString': '4577'},
...

print(feature_group.as_hive_ddl())

Contoh output:

CREATE EXTERNAL TABLE IF NOT EXISTS sagemaker_featurestore.identity-feature-group-27-19-33-00 (
  TransactionID INT
  id_01 FLOAT
  id_02 FLOAT
  id_03 FLOAT
  id_04 FLOAT
 ...

identity_query = identity_feature_group.athena_query()
transaction_query = transaction_feature_group.athena_query()

identity_table = identity_query.table_name
transaction_table = transaction_query.table_name

# Athena query
query_string = 'SELECT * FROM "'+transaction_table+'" LEFT JOIN "'+identity_table+'" ON "'+transaction_table+'".transactionid = "'+identity_table+'".transactionid'

# run Athena query. The output is loaded to a Pandas dataframe.
dataset = pd.DataFrame()
identity_query.run(query_string=query_string, output_location='s3://'+default_s3_bucket_name+'/query_results/')
identity_query.wait()
dataset = identity_query.as_dataframe()

feature_group.delete()

identity_feature_group.delete()
transaction_feature_group.delete()