15. Manage feature life cycle

Manage Feature Life Cycle¶

In this section, you'll learn to adjust the readiness of features, ensuring that the system and its users know whether a feature is primed for production.

As data evolves, it’s essential to:

• Update the table's default settings to mirror these changes.
• Introduce a new version of the feature that syncs with the updated settings.
• Curate a fresh feature list version to harness these new default feature versions.

When there are modifications in the source table's availability or freshness, we can generate new feature versions with adjusted feature job settings. If data quality is compromised, a new feature version can be crafted with specific cleaning operations to mitigate the emerging quality challenges.

For undisturbed Machine Learning operations relying on these features, it's imperative to maintain the availability of old feature versions. This ensures that any ML tasks dependent on them continue smoothly.

In [1]:

import featurebyte as fb
import pandas as pd

# Set your profile to the tutorial environment
fb.use_profile("tutorial")

catalog_name = "Grocery Dataset Tutorial"
catalog = fb.Catalog.activate(catalog_name)

import featurebyte as fb import pandas as pd # Set your profile to the tutorial environment fb.use_profile("tutorial") catalog_name = "Grocery Dataset Tutorial" catalog = fb.Catalog.activate(catalog_name)

16:48:11 | INFO     | Using configuration file at: /Users/viktor/.featurebyte/config.yaml
16:48:11 | INFO     | Active profile: tutorial (https://tutorials.featurebyte.com/api/v1)
16:48:11 | INFO     | SDK version: 0.6.0.dev121
16:48:11 | INFO     | No catalog activated.
16:48:11 | INFO     | 10 feature lists, 59 features deployed
16:48:11 | INFO     | Using profile: tutorial
16:48:11 | INFO     | Using configuration file at: /Users/viktor/.featurebyte/config.yaml
16:48:11 | INFO     | Active profile: tutorial (https://tutorials.featurebyte.com/api/v1)
16:48:12 | INFO     | SDK version: 0.6.0.dev121
16:48:12 | INFO     | No catalog activated.
16:48:12 | INFO     | 10 feature lists, 59 features deployed
16:48:12 | INFO     | Catalog activated: Grocery Dataset Tutorial

Get feature from catalog¶

In [2]:

# Get CUSTOMER_Avg_of_invoice_Amount_28d
customer_avg_of_invoice_amount_28d = catalog.get_feature("CUSTOMER_Avg_of_invoice_Amount_28d")

# Get CUSTOMER_Avg_of_invoice_Amount_28d customer_avg_of_invoice_amount_28d = catalog.get_feature("CUSTOMER_Avg_of_invoice_Amount_28d")

Check feature definition file¶

In [3]:

customer_avg_of_invoice_amount_28d.definition

customer_avg_of_invoice_amount_28d.definition

Out[3]:

# Generated by SDK version: 0.6.0.dev121
from bson import ObjectId
from featurebyte import ColumnCleaningOperation
from featurebyte import DisguisedValueImputation
from featurebyte import EventTable
from featurebyte import FeatureJobSetting
from featurebyte import ValueBeyondEndpointImputation


# event_table name: "GROCERYINVOICE"
event_table = EventTable.get_by_id(ObjectId("6564b7ebbeba6c193e0fe3bc"))
event_view = event_table.get_view(
    view_mode="manual",
    drop_column_names=["record_available_at"],
    column_cleaning_operations=[
        ColumnCleaningOperation(
            column_name="Amount",
            cleaning_operations=[
                DisguisedValueImputation(
                    imputed_value=None, disguised_values=[-99.0, -98.0]
                ),
                ValueBeyondEndpointImputation(
                    type="less_than", end_point=0.0, imputed_value=0.0
                ),
                ValueBeyondEndpointImputation(
                    type="greater_than", end_point=2000.0, imputed_value=2000.0
                ),
            ],
        )
    ],
)
grouped = event_view.groupby(
    by_keys=["GroceryCustomerGuid"], category=None
).aggregate_over(
    value_column="Amount",
    method="avg",
    windows=["28d"],
    feature_names=["CUSTOMER_Avg_of_invoice_Amount_28d"],
    feature_job_setting=FeatureJobSetting(
        blind_spot="120s", frequency="3600s", time_modulo_frequency="120s"
    ),
    skip_fill_na=True,
)
feat = grouped["CUSTOMER_Avg_of_invoice_Amount_28d"]
output = feat
output.save(_id=ObjectId("6564b8f0a69f1a9a43bdfd83"))

Update Feature Readiness¶

In [4]:

# List features and readiness
display(catalog.list_features())

# List features and readiness display(catalog.list_features())

	id	name	dtype	readiness	online_enabled	tables	primary_tables	entities	primary_entities	created_at
0	6564b93614db530858940f50	CUSTOMER_vs_OVERALL_item_TotalCost_across_prod...	FLOAT	PRODUCTION_READY	False	[GROCERYINVOICE, INVOICEITEMS, GROCERYPRODUCT]	[INVOICEITEMS]	[customer]	[customer]	2023-11-27T15:44:00.351000
1	6564b91c69ee318ce739930c	CUSTOMER_Latest_invoice_Amount_Z_Score_to_invo...	FLOAT	PRODUCTION_READY	False	[GROCERYINVOICE]	[GROCERYINVOICE]	[customer]	[customer]	2023-11-27T15:43:32.311000
2	6564b8f0a69f1a9a43bdfd85	CUSTOMER_Std_of_invoice_Amount_28d	FLOAT	DRAFT	False	[GROCERYINVOICE]	[GROCERYINVOICE]	[customer]	[customer]	2023-11-27T15:42:56.115000
3	6564b8f0a69f1a9a43bdfd84	CUSTOMER_Std_of_invoice_Amount_14d	FLOAT	PRODUCTION_READY	False	[GROCERYINVOICE]	[GROCERYINVOICE]	[customer]	[customer]	2023-11-27T15:42:54.994000
4	6564b8f0a69f1a9a43bdfd83	CUSTOMER_Avg_of_invoice_Amount_28d	FLOAT	DRAFT	False	[GROCERYINVOICE]	[GROCERYINVOICE]	[customer]	[customer]	2023-11-27T15:42:53.425000
5	6564b8f0a69f1a9a43bdfd82	CUSTOMER_Avg_of_invoice_Amount_14d	FLOAT	PRODUCTION_READY	False	[GROCERYINVOICE]	[GROCERYINVOICE]	[customer]	[customer]	2023-11-27T15:42:52.201000
6	6564b8f0a69f1a9a43bdfd80	CUSTOMER_Count_of_invoice_28d	FLOAT	DRAFT	False	[GROCERYINVOICE]	[GROCERYINVOICE]	[customer]	[customer]	2023-11-27T15:42:51.056000
7	6564b8f0a69f1a9a43bdfd7e	CUSTOMER_Count_of_invoice_14d	FLOAT	PRODUCTION_READY	False	[GROCERYINVOICE]	[GROCERYINVOICE]	[customer]	[customer]	2023-11-27T15:42:50.243000
8	6564b8f0a69f1a9a43bdfd7d	CUSTOMER_Latest_invoice_Amount	FLOAT	PRODUCTION_READY	False	[GROCERYINVOICE]	[GROCERYINVOICE]	[customer]	[customer]	2023-11-27T15:42:49.486000
9	6564b8c8e991187ee84de74e	CUSTOMER_Age_band	VARCHAR	PRODUCTION_READY	False	[GROCERYCUSTOMER]	[GROCERYCUSTOMER]	[customer]	[customer]	2023-11-27T15:42:12.722000
10	6564b8c8e991187ee84de744	CUSTOMER_Age	INT	DRAFT	False	[GROCERYCUSTOMER]	[GROCERYCUSTOMER]	[customer]	[customer]	2023-11-27T15:42:05.100000

In [5]:

# Update the readiness of the feature you want to share with others to Public Draft 
customer_avg_of_invoice_amount_28d.update_readiness("PUBLIC_DRAFT")

# Update the readiness of the feature you want to share with others to Public Draft customer_avg_of_invoice_amount_28d.update_readiness("PUBLIC_DRAFT")

In [6]:

# Check readiness
print(
    f" {customer_avg_of_invoice_amount_28d.name} readiness:",
    customer_avg_of_invoice_amount_28d.readiness,
)

# Check readiness print( f" {customer_avg_of_invoice_amount_28d.name} readiness:", customer_avg_of_invoice_amount_28d.readiness, )

 CUSTOMER_Avg_of_invoice_Amount_28d readiness: PUBLIC_DRAFT

Collect additional information on a feature¶

In [7]:

# Get metadata on the feature
customer_avg_of_invoice_amount_28d.info()

# Get metadata on the feature customer_avg_of_invoice_amount_28d.info()

Out[7]:

Feature

name	CUSTOMER_Avg_of_invoice_Amount_28d
created_at	2023-11-27 15:42:53
updated_at	2023-11-27 15:48:17
description	None
entities	name serving_names catalog_name 0 customer [GROCERYCUSTOMERGUID] Grocery Dataset Tutorial
primary_entity	name serving_names catalog_name 0 customer [GROCERYCUSTOMERGUID] Grocery Dataset Tutorial
tables	name status catalog_name 0 GROCERYINVOICE PUBLIC_DRAFT Grocery Dataset Tutorial
version_count	1
catalog_name	Grocery Dataset Tutorial
dtype	FLOAT
primary_table	name status catalog_name 0 GROCERYINVOICE PUBLIC_DRAFT Grocery Dataset Tutorial
default_version_mode	AUTO
default_feature_id	6564b8f0a69f1a9a43bdfd83
version	this V231127 default V231127
readiness	this PUBLIC_DRAFT default PUBLIC_DRAFT
table_feature_job_setting	this table_name feature_job_setting 0 GROCERYINVOICE {'blind_spot': '120s', 'frequency': '3600s', 'time_modulo_frequency': '120s'} default table_name feature_job_setting 0 GROCERYINVOICE {'blind_spot': '120s', 'frequency': '3600s', 'time_modulo_frequency': '120s'}
table_cleaning_operation	this table_name column_cleaning_operations 0 GROCERYINVOICE [{'column_name': 'Amount', 'cleaning_operations': [{'imputed_value': None, 'type': 'disguised', 'disguised_values': [-99.0, -98.0]}, {'imputed_value': 0.0, 'type': 'less_than', 'end_point': 0.0}, {'imputed_value': 2000.0, 'type': 'greater_than', 'end_point': 2000.0}]}] default table_name column_cleaning_operations 0 GROCERYINVOICE [{'column_name': 'Amount', 'cleaning_operations': [{'imputed_value': None, 'type': 'disguised', 'disguised_values': [-99.0, -98.0]}, {'imputed_value': 0.0, 'type': 'less_than', 'end_point': 0.0}, {'imputed_value': 2000.0, 'type': 'greater_than', 'end_point': 2000.0}]}]
versions_info	None
metadata	input_columns Input0 data GROCERYINVOICE column_name Amount semantic None derived_columns X0 name Amount inputs ['Input0'] transforms [] aggregations F0 name CUSTOMER_Avg_of_invoice_Amount_28d column X0 function avg keys ['GroceryCustomerGuid'] window 28d category None filter False aggregation_type groupby post_aggregation None
namespace_description	Avg of invoice Amount for the customer over a 28d period.

Update Default Feature Job Setting at the table level¶

In [8]:

# Get GROCERYINVOICE table
invoice_table = catalog.get_table("GROCERYINVOICE")

# Get GROCERYINVOICE table invoice_table = catalog.get_table("GROCERYINVOICE")

In [9]:

# Get current Default Feature Job Setting
invoice_table.default_feature_job_setting

# Get current Default Feature Job Setting invoice_table.default_feature_job_setting

Out[9]:

FeatureJobSetting(blind_spot='120s', frequency='3600s', time_modulo_frequency='120s')

In [10]:

# List past analysis
past_analysis = invoice_table.list_feature_job_setting_analysis()

# List past analysis past_analysis = invoice_table.list_feature_job_setting_analysis()

In [11]:

# Get past analysis
analysis_id = past_analysis.id.to_list()[0]
analysis = fb.FeatureJobSettingAnalysis.get_by_id(analysis_id)

# Get past analysis analysis_id = past_analysis.id.to_list()[0] analysis = fb.FeatureJobSettingAnalysis.get_by_id(analysis_id)

In [12]:

# Backtest new setting
new_feature_job_setting = fb.FeatureJobSetting(
    blind_spot='240s',
    frequency=invoice_table.default_feature_job_setting.frequency,
    time_modulo_frequency=invoice_table.default_feature_job_setting.time_modulo_frequency
)
backtest_result = analysis.backtest(feature_job_setting=new_feature_job_setting)

# Backtest new setting new_feature_job_setting = fb.FeatureJobSetting( blind_spot='240s', frequency=invoice_table.default_feature_job_setting.frequency, time_modulo_frequency=invoice_table.default_feature_job_setting.time_modulo_frequency ) backtest_result = analysis.backtest(feature_job_setting=new_feature_job_setting)

Done! |████████████████████████████████████████| 100% in 6.5s (0.16%/s)         

Feature Job Setting Analysis Report

Backtest Result

For the feature job setting:
- Frequency = 3600 s / Job time modulo frequency = 120 s / Blind spot = 240 s
The backtest found that all records would have been processed on time.

In [13]:

# Update Default Feature Job Setting
invoice_table.update_default_feature_job_setting(new_feature_job_setting)

# Update Default Feature Job Setting invoice_table.update_default_feature_job_setting(new_feature_job_setting)

Note:

• This new Default Feature Job Setting will be used by default by any new feature using the table.
• In the Enterprise platform, an approval process is associated with the change in the Default Feature Job Setting and a request to create new versions of existing features is automatically triggered.

Update Default Cleaning Operations at the table level¶

In [14]:

# Get GROCERYINVOICE table
invoice_table = catalog.get_table("GROCERYINVOICE")

# Get GROCERYINVOICE table invoice_table = catalog.get_table("GROCERYINVOICE")

In [15]:

# Get Info on columns
columns_info = pd.DataFrame(invoice_table.info(verbose=True)['columns_info'])
display(columns_info)

# Get Info on columns columns_info = pd.DataFrame(invoice_table.info(verbose=True)['columns_info']) display(columns_info)

	name	dtype	entity	semantic	critical_data_info	description
0	GroceryInvoiceGuid	VARCHAR	invoice	event_id	None	Unique identifier of each row in the table, in...
1	GroceryCustomerGuid	VARCHAR	customer	None	None	Unique identifier for each customer, in GUID f...
2	Timestamp	TIMESTAMP	None	event_timestamp	None	The GMT timestamp of when this invoice transac...
3	tz_offset	VARCHAR	None	time_zone	None	The local timezone offset of the invoice event.
4	record_available_at	TIMESTAMP	None	record_creation_timestamp	None	A timestamp for when this row was added to the...
5	Amount	FLOAT	None	None	{'cleaning_operations': [{'imputed_value': Non...	The total amount of the invoice, including all...

In [16]:

# Get Current Cleaning Operation for Amount column
for info in columns_info.loc[columns_info.name=="Amount"]["critical_data_info"]:
    print(info)

# Get Current Cleaning Operation for Amount column for info in columns_info.loc[columns_info.name=="Amount"]["critical_data_info"]: print(info)

{'cleaning_operations': [{'imputed_value': None, 'type': 'disguised', 'disguised_values': [-99.0, -98.0]}, {'imputed_value': 0.0, 'type': 'less_than', 'end_point': 0.0}, {'imputed_value': 2000.0, 'type': 'greater_than', 'end_point': 2000.0}]}

In [17]:

# Update Cleaning Operations by adding -96 as a new disguised missing value
new_cleaning_operations = [
    fb.DisguisedValueImputation(disguised_values=[-99, -98, -96], imputed_value=None),
    fb.ValueBeyondEndpointImputation(
        type="less_than", end_point=0, imputed_value=0
    ),
    fb.ValueBeyondEndpointImputation(
        type="greater_than", end_point=2000, imputed_value=2000
    ),
]
invoice_table["Amount"].update_critical_data_info(
    cleaning_operations=new_cleaning_operations
)

# Update Cleaning Operations by adding -96 as a new disguised missing value new_cleaning_operations = [ fb.DisguisedValueImputation(disguised_values=[-99, -98, -96], imputed_value=None), fb.ValueBeyondEndpointImputation( type="less_than", end_point=0, imputed_value=0 ), fb.ValueBeyondEndpointImputation( type="greater_than", end_point=2000, imputed_value=2000 ), ] invoice_table["Amount"].update_critical_data_info( cleaning_operations=new_cleaning_operations )

Note:

• This new Default Cleaning Operations will be used by default by any new feature using the table column.
• In the Enterprise platform, an approval process is associated with the change in the Default Cleaning Operations and a request to create new versions of existing features using the table column is automatically triggered.

Change feature job setting and cleaning operations of a feature¶

In [18]:

# Get feature CUSTOMER_Avg_of_invoice_Amount_14d
customer_avg_of_invoice_amount_14d = catalog.get_feature("CUSTOMER_Avg_of_invoice_Amount_14d")

# Get feature CUSTOMER_Avg_of_invoice_Amount_14d customer_avg_of_invoice_amount_14d = catalog.get_feature("CUSTOMER_Avg_of_invoice_Amount_14d")

In [19]:

# Get current feature job setting
customer_avg_of_invoice_amount_14d.info()["table_feature_job_setting"]

# Get current feature job setting customer_avg_of_invoice_amount_14d.info()["table_feature_job_setting"]

Out[19]:

{'this': [{'table_name': 'GROCERYINVOICE',
   'feature_job_setting': {'blind_spot': '120s',
    'frequency': '3600s',
    'time_modulo_frequency': '120s'}}],
 'default': [{'table_name': 'GROCERYINVOICE',
   'feature_job_setting': {'blind_spot': '120s',
    'frequency': '3600s',
    'time_modulo_frequency': '120s'}}]}

In [20]:

# Get current cleaning operations
customer_avg_of_invoice_amount_14d.info()["table_cleaning_operation"]

# Get current cleaning operations customer_avg_of_invoice_amount_14d.info()["table_cleaning_operation"]

Out[20]:

{'this': [{'table_name': 'GROCERYINVOICE',
   'column_cleaning_operations': [{'column_name': 'Amount',
     'cleaning_operations': [{'imputed_value': None,
       'type': 'disguised',
       'disguised_values': [-99.0, -98.0]},
      {'imputed_value': 0.0, 'type': 'less_than', 'end_point': 0.0},
      {'imputed_value': 2000.0,
       'type': 'greater_than',
       'end_point': 2000.0}]}]}],
 'default': [{'table_name': 'GROCERYINVOICE',
   'column_cleaning_operations': [{'column_name': 'Amount',
     'cleaning_operations': [{'imputed_value': None,
       'type': 'disguised',
       'disguised_values': [-99.0, -98.0]},
      {'imputed_value': 0.0, 'type': 'less_than', 'end_point': 0.0},
      {'imputed_value': 2000.0,
       'type': 'greater_than',
       'end_point': 2000.0}]}]}]}

In [21]:

# Deprecate current default version
customer_avg_of_invoice_amount_14d.update_readiness("DEPRECATED")

# Deprecate current default version customer_avg_of_invoice_amount_14d.update_readiness("DEPRECATED")

In [22]:

# Create new version
new_version = customer_avg_of_invoice_amount_14d.create_new_version(
    table_feature_job_settings=[
        fb.TableFeatureJobSetting(
              table_name="GROCERYINVOICE",
              feature_job_setting=new_feature_job_setting
        )
    ],
    table_cleaning_operations=[
        fb.TableCleaningOperation(
              table_name="GROCERYINVOICE",
              column_cleaning_operations=[
                    fb.ColumnCleaningOperation(
                        column_name="Amount",
                        cleaning_operations=new_cleaning_operations
                    )
              ]
        )
    ]
)

# Create new version new_version = customer_avg_of_invoice_amount_14d.create_new_version( table_feature_job_settings=[ fb.TableFeatureJobSetting( table_name="GROCERYINVOICE", feature_job_setting=new_feature_job_setting ) ], table_cleaning_operations=[ fb.TableCleaningOperation( table_name="GROCERYINVOICE", column_cleaning_operations=[ fb.ColumnCleaningOperation( column_name="Amount", cleaning_operations=new_cleaning_operations ) ] ) ] )

In [23]:

# Check new version is the default
print(
    f"version_name: {new_version.version} \n",
    f"is the new version the new default? {new_version.is_default}"
)

# Check new version is the default print( f"version_name: {new_version.version} \n", f"is the new version the new default? {new_version.is_default}" )

version_name: V231127_1 
 is the new version the new default? True

In [24]:

# Check you get the new version from the catalog by default
customer_avg_of_invoice_amount_14d = catalog.get_feature("CUSTOMER_Avg_of_invoice_Amount_14d")
print(customer_avg_of_invoice_amount_14d.version)

# Check you get the new version from the catalog by default customer_avg_of_invoice_amount_14d = catalog.get_feature("CUSTOMER_Avg_of_invoice_Amount_14d") print(customer_avg_of_invoice_amount_14d.version)

V231127_1

In [25]:

# List versions
customer_avg_of_invoice_amount_14d.list_versions()

# List versions customer_avg_of_invoice_amount_14d.list_versions()

Out[25]:

	id	name	version	dtype	readiness	online_enabled	tables	primary_tables	entities	primary_entities	created_at	is_default
0	6564ba50aee16c542a9c58d1	CUSTOMER_Avg_of_invoice_Amount_14d	V231127_1	FLOAT	DRAFT	False	[GROCERYINVOICE]	[GROCERYINVOICE]	[customer]	[customer]	2023-11-27T15:48:33.139000	True
1	6564b8f0a69f1a9a43bdfd82	CUSTOMER_Avg_of_invoice_Amount_14d	V231127	FLOAT	DEPRECATED	False	[GROCERYINVOICE]	[GROCERYINVOICE]	[customer]	[customer]	2023-11-27T15:42:52.185000	False

In [26]:

# Upgrade readiness of the new version
customer_avg_of_invoice_amount_14d.update_readiness("PRODUCTION_READY")

# Upgrade readiness of the new version customer_avg_of_invoice_amount_14d.update_readiness("PRODUCTION_READY")

Upgrade readiness of a feature with discrepancies with table defaults¶

In [27]:

# Get the feature CUSTOMER_Std_of_invoice_Amount_28d
customer_std_of_invoice_amount_28d = catalog.get_feature("CUSTOMER_Std_of_invoice_Amount_28d")

# Get the feature CUSTOMER_Std_of_invoice_Amount_28d customer_std_of_invoice_amount_28d = catalog.get_feature("CUSTOMER_Std_of_invoice_Amount_28d")

In [28]:

# Check that guardrails prevent you from upgrading the feature to production ready
# because of discrepancies between the feature settings and the default settings.
try:
    customer_std_of_invoice_amount_28d.update_readiness("PRODUCTION_READY")
except Exception as e:
    print(e)

# Check that guardrails prevent you from upgrading the feature to production ready # because of discrepancies between the feature settings and the default settings. try: customer_std_of_invoice_amount_28d.update_readiness("PRODUCTION_READY") except Exception as e: print(e)

Discrepancies found between the promoted feature version you are trying to promote to PRODUCTION_READY, and the input table.
{'feature_job_setting': {'data_source': FeatureJobSetting(blind_spot='240s', frequency='3600s', time_modulo_frequency='120s'), 'promoted_feature': FeatureJobSetting(blind_spot='120s', frequency='3600s', time_modulo_frequency='120s')}, 'cleaning_operations': {'data_source': [ColumnCleaningOperation(column_name='Amount', cleaning_operations=[DisguisedValueImputation(imputed_value=None, disguised_values=[-99.0, -98.0, -96.0]), ValueBeyondEndpointImputation(imputed_value=0.0, type=less_than, end_point=0.0), ValueBeyondEndpointImputation(imputed_value=2000.0, type=greater_than, end_point=2000.0)])], 'promoted_feature': [ColumnCleaningOperation(column_name='Amount', cleaning_operations=[DisguisedValueImputation(imputed_value=None, disguised_values=[-99.0, -98.0]), ValueBeyondEndpointImputation(imputed_value=0.0, type=less_than, end_point=0.0), ValueBeyondEndpointImputation(imputed_value=2000.0, type=greater_than, end_point=2000.0)])]}}
Please fix these issues first before trying to promote your feature to PRODUCTION_READY.

In [29]:

# Ignore the guardrails if you are ok with the feature setting
customer_std_of_invoice_amount_28d.update_readiness("PRODUCTION_READY", ignore_guardrails=True)

# Ignore the guardrails if you are ok with the feature setting customer_std_of_invoice_amount_28d.update_readiness("PRODUCTION_READY", ignore_guardrails=True)

Create new version of a feature list¶

In [30]:

# Get feature list from catalog
simple_feature_list = catalog.get_feature_list("Customer Simple FeatureList")

# Get feature list from catalog simple_feature_list = catalog.get_feature_list("Customer Simple FeatureList")

Loading Feature(s) |████████████████████████████████████████| 7/7 [100%] in 0.8s

In [31]:

# Check Fraction of default features
simple_feature_list.default_feature_fraction

# Check Fraction of default features simple_feature_list.default_feature_fraction

Out[31]:

0.8571428571428571

In [32]:

# Create new version
new_version = simple_feature_list.create_new_version()

# Create new version new_version = simple_feature_list.create_new_version()

Loading Feature(s) |████████████████████████████████████████| 7/7 [100%] in 0.8s

In [33]:

# Check Fraction of default features of new version
new_version.default_feature_fraction

# Check Fraction of default features of new version new_version.default_feature_fraction

Out[33]:

1.0

In [34]:

# Check new version is the default
print(
    f"version_name: {new_version.version} \n",
    f"is the new version the new default? {new_version.is_default}"
)

# Check new version is the default print( f"version_name: {new_version.version} \n", f"is the new version the new default? {new_version.is_default}" )

version_name: V231127_1 
 is the new version the new default? True

In [35]:

# Get Default version from Catalog
simple_feature_list = catalog.get_feature_list("Customer Simple FeatureList")
print(simple_feature_list.version)

# Get Default version from Catalog simple_feature_list = catalog.get_feature_list("Customer Simple FeatureList") print(simple_feature_list.version)

Loading Feature(s) |████████████████████████████████████████| 7/7 [100%] in 0.7s
V231127_1

In [36]:

# List versions
simple_feature_list.list_versions()

# List versions simple_feature_list.list_versions()

Out[36]:

	id	name	version	online_frac	deployed	created_at	is_default
0	6564ba59aee16c542a9c58dd	Customer Simple FeatureList	V231127_1	0.0	False	2023-11-27T15:48:41.420000	True
1	6564b959d3f7244b7fc926a0	Customer Simple FeatureList	V231127	0.0	False	2023-11-27T15:44:36.950000	False

In [37]:

# Check Production Readiness
simple_feature_list.production_ready_fraction

# Check Production Readiness simple_feature_list.production_ready_fraction

Out[37]:

1.0

Delete Draft features and feature lists¶

In [38]:

# Get list of features together with their readiness
catalog.list_features()

# Get list of features together with their readiness catalog.list_features()

Out[38]:

	id	name	dtype	readiness	online_enabled	tables	primary_tables	entities	primary_entities	created_at
0	6564b93614db530858940f50	CUSTOMER_vs_OVERALL_item_TotalCost_across_prod...	FLOAT	PRODUCTION_READY	False	[GROCERYINVOICE, INVOICEITEMS, GROCERYPRODUCT]	[INVOICEITEMS]	[customer]	[customer]	2023-11-27T15:44:00.351000
1	6564b91c69ee318ce739930c	CUSTOMER_Latest_invoice_Amount_Z_Score_to_invo...	FLOAT	PRODUCTION_READY	False	[GROCERYINVOICE]	[GROCERYINVOICE]	[customer]	[customer]	2023-11-27T15:43:32.311000
2	6564b8f0a69f1a9a43bdfd85	CUSTOMER_Std_of_invoice_Amount_28d	FLOAT	PRODUCTION_READY	False	[GROCERYINVOICE]	[GROCERYINVOICE]	[customer]	[customer]	2023-11-27T15:42:56.115000
3	6564b8f0a69f1a9a43bdfd84	CUSTOMER_Std_of_invoice_Amount_14d	FLOAT	PRODUCTION_READY	False	[GROCERYINVOICE]	[GROCERYINVOICE]	[customer]	[customer]	2023-11-27T15:42:54.994000
4	6564b8f0a69f1a9a43bdfd83	CUSTOMER_Avg_of_invoice_Amount_28d	FLOAT	PUBLIC_DRAFT	False	[GROCERYINVOICE]	[GROCERYINVOICE]	[customer]	[customer]	2023-11-27T15:42:53.425000
5	6564ba50aee16c542a9c58d1	CUSTOMER_Avg_of_invoice_Amount_14d	FLOAT	PRODUCTION_READY	False	[GROCERYINVOICE]	[GROCERYINVOICE]	[customer]	[customer]	2023-11-27T15:42:52.201000
6	6564b8f0a69f1a9a43bdfd80	CUSTOMER_Count_of_invoice_28d	FLOAT	DRAFT	False	[GROCERYINVOICE]	[GROCERYINVOICE]	[customer]	[customer]	2023-11-27T15:42:51.056000
7	6564b8f0a69f1a9a43bdfd7e	CUSTOMER_Count_of_invoice_14d	FLOAT	PRODUCTION_READY	False	[GROCERYINVOICE]	[GROCERYINVOICE]	[customer]	[customer]	2023-11-27T15:42:50.243000
8	6564b8f0a69f1a9a43bdfd7d	CUSTOMER_Latest_invoice_Amount	FLOAT	PRODUCTION_READY	False	[GROCERYINVOICE]	[GROCERYINVOICE]	[customer]	[customer]	2023-11-27T15:42:49.486000
9	6564b8c8e991187ee84de74e	CUSTOMER_Age_band	VARCHAR	PRODUCTION_READY	False	[GROCERYCUSTOMER]	[GROCERYCUSTOMER]	[customer]	[customer]	2023-11-27T15:42:12.722000
10	6564b8c8e991187ee84de744	CUSTOMER_Age	INT	DRAFT	False	[GROCERYCUSTOMER]	[GROCERYCUSTOMER]	[customer]	[customer]	2023-11-27T15:42:05.100000

In [39]:

# Get draft features to delete
customer_count_of_invoice_28d = catalog.get_feature("CUSTOMER_Count_of_invoice_28d")
customer_age = catalog.get_feature("CUSTOMER_Age")

# Get draft features to delete customer_count_of_invoice_28d = catalog.get_feature("CUSTOMER_Count_of_invoice_28d") customer_age = catalog.get_feature("CUSTOMER_Age")

In [40]:

# Create a feature list that uses the 2 features we want to delete
new_feature_list = fb.FeatureList(
    [
        customer_count_of_invoice_28d,
        customer_age
    ],
    name="New List"
)
new_feature_list.save()

# Create a feature list that uses the 2 features we want to delete new_feature_list = fb.FeatureList( [ customer_count_of_invoice_28d, customer_age ], name="New List" ) new_feature_list.save()

Done! |████████████████████████████████████████| 100% in 3.3s (0.30%/s)         
Loading Feature(s) |████████████████████████████████████████| 2/2 [100%] in 0.5s

In [41]:

# Check that guardrails prevent you from deleting a feature that is still in use by a feature list
try:
    customer_age.delete()
except Exception as e:
    print(e)

# Check that guardrails prevent you from deleting a feature that is still in use by a feature list try: customer_age.delete() except Exception as e: print(e)

("Feature is still in use by feature list(s). Please remove the following feature list(s) first:\n[{'id': '6564ba60f5b3b570559367e7', 'name': 'New List', 'version': 'V231127'}]", 'Failed to delete specified object.')

In [42]:

# Delete feature list first
new_feature_list.delete()

# Delete feature list first new_feature_list.delete()

In [43]:

# Then delete the 2 draft features
customer_age.delete()
customer_count_of_invoice_28d.delete()

# Then delete the 2 draft features customer_age.delete() customer_count_of_invoice_28d.delete()

In [44]:

# Check that guardrails prevent you from deleting a feature that is not a draft
customer_avg_of_invoice_amount_28d = catalog.get_feature("CUSTOMER_Avg_of_invoice_Amount_28d")
try:
    customer_avg_of_invoice_amount_28d.delete()
except Exception as e:
    print(e)

# Check that guardrails prevent you from deleting a feature that is not a draft customer_avg_of_invoice_amount_28d = catalog.get_feature("CUSTOMER_Avg_of_invoice_Amount_28d") try: customer_avg_of_invoice_amount_28d.delete() except Exception as e: print(e)

('Only feature with draft readiness can be deleted.', 'Failed to delete specified object.')

In [45]:

# Instead deprecate it
customer_avg_of_invoice_amount_28d.update_readiness("DEPRECATED")

# Instead deprecate it customer_avg_of_invoice_amount_28d.update_readiness("DEPRECATED")

In [46]:

# Get updated list of features together with their readiness
catalog.list_features()

# Get updated list of features together with their readiness catalog.list_features()

Out[46]:

	id	name	dtype	readiness	online_enabled	tables	primary_tables	entities	primary_entities	created_at
0	6564b93614db530858940f50	CUSTOMER_vs_OVERALL_item_TotalCost_across_prod...	FLOAT	PRODUCTION_READY	False	[GROCERYINVOICE, INVOICEITEMS, GROCERYPRODUCT]	[INVOICEITEMS]	[customer]	[customer]	2023-11-27T15:44:00.351000
1	6564b91c69ee318ce739930c	CUSTOMER_Latest_invoice_Amount_Z_Score_to_invo...	FLOAT	PRODUCTION_READY	False	[GROCERYINVOICE]	[GROCERYINVOICE]	[customer]	[customer]	2023-11-27T15:43:32.311000
2	6564b8f0a69f1a9a43bdfd85	CUSTOMER_Std_of_invoice_Amount_28d	FLOAT	PRODUCTION_READY	False	[GROCERYINVOICE]	[GROCERYINVOICE]	[customer]	[customer]	2023-11-27T15:42:56.115000
3	6564b8f0a69f1a9a43bdfd84	CUSTOMER_Std_of_invoice_Amount_14d	FLOAT	PRODUCTION_READY	False	[GROCERYINVOICE]	[GROCERYINVOICE]	[customer]	[customer]	2023-11-27T15:42:54.994000
4	6564b8f0a69f1a9a43bdfd83	CUSTOMER_Avg_of_invoice_Amount_28d	FLOAT	DEPRECATED	False	[GROCERYINVOICE]	[GROCERYINVOICE]	[customer]	[customer]	2023-11-27T15:42:53.425000
5	6564ba50aee16c542a9c58d1	CUSTOMER_Avg_of_invoice_Amount_14d	FLOAT	PRODUCTION_READY	False	[GROCERYINVOICE]	[GROCERYINVOICE]	[customer]	[customer]	2023-11-27T15:42:52.201000
6	6564b8f0a69f1a9a43bdfd7e	CUSTOMER_Count_of_invoice_14d	FLOAT	PRODUCTION_READY	False	[GROCERYINVOICE]	[GROCERYINVOICE]	[customer]	[customer]	2023-11-27T15:42:50.243000
7	6564b8f0a69f1a9a43bdfd7d	CUSTOMER_Latest_invoice_Amount	FLOAT	PRODUCTION_READY	False	[GROCERYINVOICE]	[GROCERYINVOICE]	[customer]	[customer]	2023-11-27T15:42:49.486000
8	6564b8c8e991187ee84de74e	CUSTOMER_Age_band	VARCHAR	PRODUCTION_READY	False	[GROCERYCUSTOMER]	[GROCERYCUSTOMER]	[customer]	[customer]	2023-11-27T15:42:12.722000

Concepts in this tutorial¶

• Feature versioning
• Default feature version
• Feature list versioning
• Default Feature Job Setting
• Feature Job Setting Recommendation
• Default Cleaning Operations

SDK reference for¶

• Feature
• FeatureList
• Feature.readiness
• Feature.update_readiness()
• Feature.info()
• EventTable.list_feature_job_setting_analysis()
• EventTable.create_new_feature_job_setting_analysis()
• EventTable.update_default_feature_job_setting()
• FeatureJobSettingAnalysis.get_by_id()
• FeatureJobSettingAnalysis.backtest()
• TableColumn.update_critical_data_info()
• Feature.create_new_version()
• Feature.list_versions()
• Feature.delete()
• FeatureList.create_new_version()
• FeatureList.list_versions()
• FeatureList.delete()
• FeatureList.production_ready_fraction
• FeatureList.default_feature_fraction