Quick Start Tutorial: Reusing Features¶

Learning Objectives¶

In this tutorial you will learn:

How to access catalogs of data, entities, features, and feature lists
How to search for features suitable for the unit of analysis
How to understand an existing feature
How to create new features from existing features
How to create a new feature list from existing features

Set up the prerequisites¶

Learning Objectives

In this section you will:

start your local featurebyte server
import libraries
learn the about catalogs
activate a pre-built catalog

Load the featurebyte library and connect to the local instance of featurebyte¶

In [1]:

            
                Copied!
                
# library imports
import pandas as pd
import numpy as np
import random

# load the featurebyte SDK
import featurebyte as fb

# start the local server, then wait for it to be healthy before proceeding
fb.playground()
# library imports
import pandas as pd
import numpy as np
import random

# load the featurebyte SDK
import featurebyte as fb

# start the local server, then wait for it to be healthy before proceeding
fb.playground()

02:29:43 | INFO     | Using configuration file at: /home/chester/.featurebyte/config.yaml
02:29:43 | INFO     | Active profile: local (http://127.0.0.1:8088)
02:29:43 | INFO     | SDK version: 0.2.2
02:29:43 | INFO     | Active catalog: default
02:29:43 | INFO     | 0 feature list, 0 feature deployed
02:29:43 | INFO     | (1/4) Starting featurebyte services
 Container mongo-rs  Running
 Container redis  Running
 Container featurebyte-server  Running
 Container spark-thrift  Running
 Container featurebyte-worker  Running
 Container mongo-rs  Waiting
 Container mongo-rs  Waiting
 Container redis  Waiting
 Container redis  Healthy
 Container mongo-rs  Healthy
 Container mongo-rs  Healthy
02:29:44 | INFO     | (2/4) Creating local spark feature store
02:29:44 | INFO     | (3/4) Import datasets
02:29:45 | INFO     | Dataset grocery already exists, skipping import
02:29:45 | INFO     | Dataset healthcare already exists, skipping import
02:29:45 | INFO     | Dataset creditcard already exists, skipping import
02:29:45 | INFO     | (4/4) Playground environment started successfully. Ready to go! 🚀

Create a pre-built catalog for this tutorial, with the data, metadata, and features already set up¶

Note that creating a pre-built catalog is not a step you will do in real-life. This is a function specific to this quick-start tutorial to quickly skip over many of the preparatory steps and get you to a point where you can materialize features.

In a real-life project you would do data modeling, declaring the tables, entities, and the associated metadata. This would not be a frequent task, but forms the basis for best-practice feature engineering.

In [2]:

            
                Copied!
                
# get the functions to create a pre-built catalog
from prebuilt_catalogs import *

# create a new catalog for this tutorial
catalog = create_tutorial_catalog(PrebuiltCatalog.QuickStartReusingFeatures)
# get the functions to create a pre-built catalog
from prebuilt_catalogs import *

# create a new catalog for this tutorial
catalog = create_tutorial_catalog(PrebuiltCatalog.QuickStartReusingFeatures)

Cleaning up existing tutorial catalogs

02:29:45 | INFO     | Catalog activated: quick start model training 20230511:0224

Cleaning catalog: quick start model training 20230511:0224
  1 historical feature tables
  1 observation tables
Done! |████████████████████████████████████████| 100% in 6.1s (0.17%/s)         
Done! |████████████████████████████████████████| 100% in 6.0s (0.17%/s)

02:29:59 | INFO     | Catalog activated: default
02:29:59 | INFO     | Catalog activated: quick start reusing features 20230511:0229

Building a quick start catalog for reusing features named [quick start reusing features 20230511:0229]
Creating new catalog
Catalog created
Registering the source tables
Registering the entities
Tagging the entities to columns in the data tables
Populating the feature store with example features
Saving Feature(s) |████████████████████████████████████████| 5/5 [100%] in 1.6s 
Loading Feature(s) |████████████████████████████████████████| 5/5 [100%] in 1.3s

Example: Load the tables and views¶

In [3]:

            
                Copied!
                
                    
                    
                
                

        
# get the tables for this workspace
grocery_customer_table = catalog.get_table("GROCERYCUSTOMER")
grocery_items_table = catalog.get_table("INVOICEITEMS")
grocery_invoice_table = catalog.get_table("GROCERYINVOICE")
grocery_product_table = catalog.get_table("GROCERYPRODUCT")

# create the views
grocery_customer_view = grocery_customer_table.get_view()
grocery_invoice_view = grocery_invoice_table.get_view()
grocery_items_view = grocery_items_table.get_view()
grocery_product_view = grocery_product_table.get_view()
# get the tables for this workspace
grocery_customer_table = catalog.get_table("GROCERYCUSTOMER")
grocery_items_table = catalog.get_table("INVOICEITEMS")
grocery_invoice_table = catalog.get_table("GROCERYINVOICE")
grocery_product_table = catalog.get_table("GROCERYPRODUCT")

# create the views
grocery_customer_view = grocery_customer_table.get_view()
grocery_invoice_view = grocery_invoice_table.get_view()
grocery_items_view = grocery_items_table.get_view()
grocery_product_view = grocery_product_table.get_view()

Accessing Catalogs¶

Learning Objectives:

In this section you will learn how to display catalogs of:

tables
entities
features
feature lists

Example: A catalog of tables¶

In [4]:

            
                Copied!
                
# list the tables in the catalog
catalog.list_tables()
# list the tables in the catalog
catalog.list_tables()

Out[4]:

	id	name	type	status	entities	created_at
0	645c5330b37fce40c0e3c89d	GROCERYPRODUCT	dimension_table	PUBLIC_DRAFT	[groceryproduct]	2023-05-11 02:30:09.678
1	645c532db37fce40c0e3c89c	INVOICEITEMS	item_table	PUBLIC_DRAFT	[groceryinvoice, groceryproduct]	2023-05-11 02:30:06.967
2	645c532ab37fce40c0e3c89b	GROCERYINVOICE	event_table	PUBLIC_DRAFT	[groceryinvoice, grocerycustomer]	2023-05-11 02:30:03.446
3	645c5328b37fce40c0e3c89a	GROCERYCUSTOMER	scd_table	PUBLIC_DRAFT	[grocerycustomer, frenchstate]	2023-05-11 02:30:01.588

In [5]:

            
                Copied!
                
# load a table
grocery_customer_table = catalog.get_table("GROCERYCUSTOMER")

# show the metadata
grocery_customer_table.info()
# load a table
grocery_customer_table = catalog.get_table("GROCERYCUSTOMER")

# show the metadata
grocery_customer_table.info()

Out[5]:

{
  'name': 'GROCERYCUSTOMER',
  'created_at': '2023-05-11T02:30:01.588000',
  'updated_at': '2023-05-11T02:30:10.171000',
  'status': 'PUBLIC_DRAFT',
  'catalog_name': 'quick start reusing features 20230511:0229',
  'record_creation_timestamp_column': 'record_available_at',
  'table_details': {
    'database_name': 'spark_catalog',
    'schema_name': 'GROCERY',
    'table_name': 'GROCERYCUSTOMER'
  },
  'entities': [
    {
      'name': 'frenchstate',
      'serving_names': [
        'FRENCHSTATE'
      ],
      'catalog_name': 'quick start reusing features 20230511:0229'
    },
    {
      'name': 'grocerycustomer',
      'serving_names': [
        'GROCERYCUSTOMERGUID'
      ],
      'catalog_name': 'quick start reusing features 20230511:0229'
    }
  ],
  'semantics': [
    'scd_surrogate_key_id',
    'scd_natural_key_id'
  ],
  'column_count': 18,
  'columns_info': None,
  'natural_key_column': 'GroceryCustomerGuid',
  'effective_timestamp_column': 'ValidFrom',
  'surrogate_key_column': 'RowID',
  'end_timestamp_column': None,
  'current_flag_column': 'CurrentRecord'
}

Example: A catalog of entities¶

In [6]:

            
                Copied!
                
# list the entities in the catalog
catalog.list_entities()
# list the entities in the catalog
catalog.list_entities()

Out[6]:

	id	name	serving_names	created_at
0	645c5331b37fce40c0e3c8a1	frenchstate	[FRENCHSTATE]	2023-05-11 02:30:09.982
1	645c5331b37fce40c0e3c8a0	groceryproduct	[GROCERYPRODUCTGUID]	2023-05-11 02:30:09.910
2	645c5331b37fce40c0e3c89f	groceryinvoice	[GROCERYINVOICEGUID]	2023-05-11 02:30:09.840
3	645c5331b37fce40c0e3c89e	grocerycustomer	[GROCERYCUSTOMERGUID]	2023-05-11 02:30:09.776

In [7]:

            
                Copied!
                
# list the entity relationships in the catalog
catalog.list_relationships()
# list the entity relationships in the catalog
catalog.list_relationships()

Out[7]:

	id	relationship_type	entity	related_entity	relation_table	relation_table_type	enabled	created_at	updated_at
0	645c53329c28d6ed179a1a73	child_parent	groceryinvoice	grocerycustomer	GROCERYINVOICE	event_table	True	2023-05-11 02:30:10.464	None
1	645c53329c28d6ed179a1a6c	child_parent	grocerycustomer	frenchstate	GROCERYCUSTOMER	scd_table	True	2023-05-11 02:30:10.220	None

In [8]:

            
                Copied!
                
# load an entity
customer_entity = catalog.get_entity("grocerycustomer")

# show the metadata
customer_entity.info()
# load an entity
customer_entity = catalog.get_entity("grocerycustomer")

# show the metadata
customer_entity.info()

Out[8]:

{
  'name': 'grocerycustomer',
  'created_at': '2023-05-11T02:30:09.776000',
  'updated_at': '2023-05-11T02:30:10.426000',
  'serving_names': [
    'GROCERYCUSTOMERGUID'
  ],
  'catalog_name': 'quick start reusing features 20230511:0229'
}

Example: A catalog of features¶

In [9]:

            
                Copied!
                
# list the features in the catalog
catalog.list_features()
# list the features in the catalog
catalog.list_features()

Out[9]:

	id	name	dtype	readiness	online_enabled	tables	primary_tables	entities	primary_entities	created_at
0	645c5347b37fce40c0e3c8c0	InvoiceUniqueProductGroupCount	FLOAT	DRAFT	False	[GROCERYINVOICE, INVOICEITEMS, GROCERYPRODUCT]	[INVOICEITEMS]	[groceryinvoice]	[groceryinvoice]	2023-05-11 02:30:33.185
1	645c5346b37fce40c0e3c8bc	InvoiceDiscountAmount	FLOAT	DRAFT	False	[GROCERYINVOICE, INVOICEITEMS]	[INVOICEITEMS]	[groceryinvoice]	[groceryinvoice]	2023-05-11 02:30:31.101
2	645c5344b37fce40c0e3c8ba	InvoiceItemCount	FLOAT	DRAFT	False	[GROCERYINVOICE, INVOICEITEMS]	[INVOICEITEMS]	[groceryinvoice]	[groceryinvoice]	2023-05-11 02:30:29.423
3	645c5343b37fce40c0e3c8b8	CustomerYearOfBirth	INT	DRAFT	False	[GROCERYCUSTOMER]	[GROCERYCUSTOMER]	[grocerycustomer]	[grocerycustomer]	2023-05-11 02:30:27.690
4	645c5341b37fce40c0e3c8b4	CustomerSpend_14d	FLOAT	DRAFT	False	[GROCERYINVOICE]	[GROCERYINVOICE]	[grocerycustomer]	[grocerycustomer]	2023-05-11 02:30:26.874
5	645c5340b37fce40c0e3c8b2	CustomerInventory_24w	OBJECT	DRAFT	False	[GROCERYINVOICE, INVOICEITEMS, GROCERYPRODUCT]	[INVOICEITEMS]	[grocerycustomer]	[grocerycustomer]	2023-05-11 02:30:25.298
6	645c533eb37fce40c0e3c8b0	CustomerInventory_28d	OBJECT	DRAFT	False	[GROCERYINVOICE, INVOICEITEMS, GROCERYPRODUCT]	[INVOICEITEMS]	[grocerycustomer]	[grocerycustomer]	2023-05-11 02:30:23.367
7	645c533cb37fce40c0e3c8ae	StateMeanLongitude	FLOAT	DRAFT	False	[GROCERYCUSTOMER]	[GROCERYCUSTOMER]	[frenchstate]	[frenchstate]	2023-05-11 02:30:20.604
8	645c533bb37fce40c0e3c8ac	StateMeanLatitude	FLOAT	DRAFT	False	[GROCERYCUSTOMER]	[GROCERYCUSTOMER]	[frenchstate]	[frenchstate]	2023-05-11 02:30:19.879
9	645c533ab37fce40c0e3c8aa	StateAvgInvoiceAmount_28d	FLOAT	DRAFT	False	[GROCERYCUSTOMER, GROCERYINVOICE]	[GROCERYINVOICE]	[frenchstate]	[frenchstate]	2023-05-11 02:30:19.007
10	645c5337b37fce40c0e3c8a8	StateInventory_28d	OBJECT	DRAFT	False	[GROCERYCUSTOMER, GROCERYINVOICE, INVOICEITEMS...	[INVOICEITEMS]	[frenchstate]	[frenchstate]	2023-05-11 02:30:17.076
11	645c5336b37fce40c0e3c8a4	StatePopulation	FLOAT	DRAFT	False	[GROCERYCUSTOMER]	[GROCERYCUSTOMER]	[frenchstate]	[frenchstate]	2023-05-11 02:30:14.988
12	645c5335b37fce40c0e3c8a2	StateName	VARCHAR	DRAFT	False	[GROCERYCUSTOMER]	[GROCERYCUSTOMER]	[grocerycustomer]	[grocerycustomer]	2023-05-11 02:30:13.971

In [10]:

            
                Copied!
                
# load a feature
state_population = catalog.get_feature("StatePopulation")

# show the metadata
state_population.info()
# load a feature
state_population = catalog.get_feature("StatePopulation")

# show the metadata
state_population.info()

Out[10]:

{
  'name': 'StatePopulation',
  'created_at': '2023-05-11T02:30:14.988000',
  'updated_at': None,
  'entities': [
    {
      'name': 'frenchstate',
      'serving_names': [
        'FRENCHSTATE'
      ],
      'catalog_name': 'quick start reusing features 20230511:0229'
    }
  ],
  'primary_entity': [
    {
      'name': 'frenchstate',
      'serving_names': [
        'FRENCHSTATE'
      ],
      'catalog_name': 'quick start reusing features 20230511:0229'
    }
  ],
  'tables': [
    {
      'name': 'GROCERYCUSTOMER',
      'status': 'PUBLIC_DRAFT',
      'catalog_name': 'quick start reusing features 20230511:0229'
    }
  ],
  'default_version_mode': 'AUTO',
  'version_count': 1,
  'catalog_name': 'quick start reusing features 20230511:0229',
  'dtype': 'FLOAT',
  'primary_table': [
    {
      'name': 'GROCERYCUSTOMER',
      'status': 'PUBLIC_DRAFT',
      'catalog_name': 'quick start reusing features 20230511:0229'
    }
  ],
  'default_feature_id': '645c5336b37fce40c0e3c8a4',
  'version': {
    'this': 'V230511',
    'default': 'V230511'
  },
  'readiness': {
    'this': 'DRAFT',
    'default': 'DRAFT'
  },
  'table_feature_job_setting': {
    'this': [],
    'default': []
  },
  'table_cleaning_operation': {
    'this': [],
    'default': []
  },
  'versions_info': None,
  'metadata': {
    'input_columns': {
      'Input0': {
        'data': 'GROCERYCUSTOMER',
        'column_name': 'GroceryCustomerGuid',
        'semantic': 'scd_natural_key_id'
      },
      'Input1': {
        'data': 'GROCERYCUSTOMER',
        'column_name': 'ValidFrom',
        'semantic': None
      }
    },
    'derived_columns': {},
    'aggregations': {
      'F0': {
        'name': 'StatePopulation',
        'column': None,
        'function': 'count',
        'keys': [
          'State'
        ],
        'window': None,
        'category': None,
        'filter': False
      }
    },
    'post_aggregation': {
      'name': 'StatePopulation',
      'inputs': [
        'F0'
      ],
      'transforms': [
        'is_null',
        'conditional'
      ]
    }
  }
}

In [11]:

            
                Copied!
                
# show the feature lineage for the state population feature
display(state_population.definition)
# show the feature lineage for the state population feature
display(state_population.definition)

# Generated by SDK version: 0.2.2
from bson import ObjectId
from featurebyte import SCDTable


# scd_table name: "GROCERYCUSTOMER"
scd_table = SCDTable.get_by_id(ObjectId("645c5328b37fce40c0e3c89a"))
scd_view = scd_table.get_view(
    view_mode="manual",
    drop_column_names=["record_available_at", "CurrentRecord"],
    column_cleaning_operations=[],
)
feat = scd_view.groupby(by_keys=["State"], category=None).aggregate_asat(
    value_column=None,
    method="count",
    feature_name="StatePopulation",
    offset=None,
    backward=True,
    skip_fill_na=True,
)
feat[feat.isnull()] = 0
feat_1 = feat
feat_1.name = "StatePopulation"
output = feat_1

Example: A catalog of feature lists¶

In [12]:

            
                Copied!
                
# list the feature lists in the catalog
catalog.list_feature_lists()
# list the feature lists in the catalog
catalog.list_feature_lists()

Out[12]:

	id	name	num_feature	status	deployed	readiness_frac	online_frac	tables	entities	created_at
0	645c5349b37fce40c0e3c8c2	StateFeatureList	5	DRAFT	False	0.0	0.0	[GROCERYCUSTOMER, GROCERYINVOICE, INVOICEITEMS...	[frenchstate]	2023-05-11 02:30:36.305

In [13]:

            
                Copied!
                
# load the feature list
state_features = catalog.get_feature_list("StateFeatureList")

# show the metadata
state_features.info()
# load the feature list
state_features = catalog.get_feature_list("StateFeatureList")

# show the metadata
state_features.info()

Loading Feature(s) |████████████████████████████████████████| 5/5 [100%] in 1.3s

Out[13]:

{
  'name': 'StateFeatureList',
  'created_at': '2023-05-11T02:30:36.305000',
  'updated_at': None,
  'entities': [
    {
      'name': 'frenchstate',
      'serving_names': [
        'FRENCHSTATE'
      ],
      'catalog_name': 'quick start reusing features 20230511:0229'
    }
  ],
  'primary_entity': [
    {
      'name': 'frenchstate',
      'serving_names': [
        'FRENCHSTATE'
      ],
      'catalog_name': 'quick start reusing features 20230511:0229'
    }
  ],
  'tables': [
    {
      'name': 'GROCERYPRODUCT',
      'status': 'PUBLIC_DRAFT',
      'catalog_name': 'quick start reusing features 20230511:0229'
    },
    {
      'name': 'INVOICEITEMS',
      'status': 'PUBLIC_DRAFT',
      'catalog_name': 'quick start reusing features 20230511:0229'
    },
    {
      'name': 'GROCERYINVOICE',
      'status': 'PUBLIC_DRAFT',
      'catalog_name': 'quick start reusing features 20230511:0229'
    },
    {
      'name': 'GROCERYCUSTOMER',
      'status': 'PUBLIC_DRAFT',
      'catalog_name': 'quick start reusing features 20230511:0229'
    }
  ],
  'default_version_mode': 'AUTO',
  'version_count': 1,
  'catalog_name': 'quick start reusing features 20230511:0229',
  'dtype_distribution': [
    {
      'dtype': 'FLOAT',
      'count': 4
    },
    {
      'dtype': 'OBJECT',
      'count': 1
    }
  ],
  'status': 'DRAFT',
  'feature_count': 5,
  'version': {
    'this': 'V230511',
    'default': 'V230511'
  },
  'production_ready_fraction': {
    'this': 0.0,
    'default': 0.0
  },
  'versions_info': None,
  'deployed': False
}

In [14]:

            
                Copied!
                
# list the features in the feature list
state_features.list_features()
# list the features in the feature list
state_features.list_features()

Out[14]:

	id	name	version	dtype	readiness	online_enabled	tables	primary_tables	entities	primary_entities	created_at
0	645c533cb37fce40c0e3c8ae	StateMeanLongitude	V230511	FLOAT	DRAFT	False	[GROCERYCUSTOMER]	[GROCERYCUSTOMER]	[frenchstate]	[frenchstate]	2023-05-11 02:30:20.588
1	645c533bb37fce40c0e3c8ac	StateMeanLatitude	V230511	FLOAT	DRAFT	False	[GROCERYCUSTOMER]	[GROCERYCUSTOMER]	[frenchstate]	[frenchstate]	2023-05-11 02:30:19.863
2	645c533ab37fce40c0e3c8aa	StateAvgInvoiceAmount_28d	V230511	FLOAT	DRAFT	False	[GROCERYCUSTOMER, GROCERYINVOICE]	[GROCERYINVOICE]	[frenchstate]	[frenchstate]	2023-05-11 02:30:18.989
3	645c5337b37fce40c0e3c8a8	StateInventory_28d	V230511	OBJECT	DRAFT	False	[GROCERYCUSTOMER, GROCERYINVOICE, INVOICEITEMS...	[INVOICEITEMS]	[frenchstate]	[frenchstate]	2023-05-11 02:30:17.054
4	645c5336b37fce40c0e3c8a4	StatePopulation	V230511	FLOAT	DRAFT	False	[GROCERYCUSTOMER]	[GROCERYCUSTOMER]	[frenchstate]	[frenchstate]	2023-05-11 02:30:14.971

Search for Features¶

Learning Objectives

In this section, you will learn:

what a primary entity is
how to search for suitable features

Concept: Primary entity¶

Feature primary entity: The primary entity of a feature defines the level of analysis for that feature. When a feature is a result of an aggregation grouped by multiple entities, the primary entity is a tuple of those entities. For instance, if a feature quantifies the interaction between a customer entity and a merchant entity in the past, such as the sum of transaction amounts grouped by customer and merchant in the past 4 weeks, the primary entity is the tuple of customer and merchant.

When a feature is derived for features with different primary entities, the primary entity is determined by the entity relationships, and the lowest level entity is selected as the primary entity. If the underlying entities have no relationship, the primary entity becomes a tuple of those entities. For example, if a feature compares the basket of a customer with the average basket of customers in the same city, the primary entity is the customer since the customer entity is a child of the customer city entity. However, if the feature is the distance between the customer location and the merchant location, the primary entity becomes the tuple of customer and merchant since these entities do not have any child-parent relationship.

Feature List primary entity: The main focus of a feature list is determined by its primary entity, which typically corresponds to the primary entity of the Use Case that the feature list was created for.

If the features within the list pertain to different primary entities, the primary entity of the feature list is selected based on the entities relationships, with the lowest level entity chosen as the primary entity. In cases where there are no relationships between entities, the primary entity may become a tuple comprising those entities. To illustrate, consider a feature list comprising features related to card, customer, and customer city. In this case, the primary entity is the card entity since it is a child of both the customer and customer city entities. However, if the feature list also contains features for merchant and merchant city, the primary entity is a tuple of card and merchant.

Use Case primary entity: In a Use Case, the primary entity is the object or concept that defines its problem statement. Usually, this entity is singular, but in cases such as the recommendation engine use case, it can be a tuple of entities that interact with each other.

Case study: Predicting customer spend¶

Consider a use case to predict customer spend. The unit of analysis and primary entity is grocery customer. You can use features with primary entities of grocery customer or french state (because state is a parent entity of customer).

Example: Search for suitable features¶

In [15]:

            
                Copied!
                
                    
                    
                
                

        
# get a list of all the features in the catalog
all_features = catalog.list_features()

# filter to retain only those with grocery customer or state as their primary entity
child_entity = "groceryinvoice"
suitable_features = all_features.loc[
    [child_entity not in x for x in all_features.entities.values]
]
product_entity = "groceryproduct"
suitable_features = suitable_features.loc[
    [product_entity not in x for x in suitable_features.entities.values]
]

# show the features
display(suitable_features)
# get a list of all the features in the catalog
all_features = catalog.list_features()

# filter to retain only those with grocery customer or state as their primary entity
child_entity = "groceryinvoice"
suitable_features = all_features.loc[
    [child_entity not in x for x in all_features.entities.values]
]
product_entity = "groceryproduct"
suitable_features = suitable_features.loc[
    [product_entity not in x for x in suitable_features.entities.values]
]

# show the features
display(suitable_features)

	id	name	dtype	readiness	online_enabled	tables	primary_tables	entities	primary_entities	created_at
3	645c5343b37fce40c0e3c8b8	CustomerYearOfBirth	INT	DRAFT	False	[GROCERYCUSTOMER]	[GROCERYCUSTOMER]	[grocerycustomer]	[grocerycustomer]	2023-05-11 02:30:27.690
4	645c5341b37fce40c0e3c8b4	CustomerSpend_14d	FLOAT	DRAFT	False	[GROCERYINVOICE]	[GROCERYINVOICE]	[grocerycustomer]	[grocerycustomer]	2023-05-11 02:30:26.874
5	645c5340b37fce40c0e3c8b2	CustomerInventory_24w	OBJECT	DRAFT	False	[GROCERYINVOICE, INVOICEITEMS, GROCERYPRODUCT]	[INVOICEITEMS]	[grocerycustomer]	[grocerycustomer]	2023-05-11 02:30:25.298
6	645c533eb37fce40c0e3c8b0	CustomerInventory_28d	OBJECT	DRAFT	False	[GROCERYINVOICE, INVOICEITEMS, GROCERYPRODUCT]	[INVOICEITEMS]	[grocerycustomer]	[grocerycustomer]	2023-05-11 02:30:23.367
7	645c533cb37fce40c0e3c8ae	StateMeanLongitude	FLOAT	DRAFT	False	[GROCERYCUSTOMER]	[GROCERYCUSTOMER]	[frenchstate]	[frenchstate]	2023-05-11 02:30:20.604
8	645c533bb37fce40c0e3c8ac	StateMeanLatitude	FLOAT	DRAFT	False	[GROCERYCUSTOMER]	[GROCERYCUSTOMER]	[frenchstate]	[frenchstate]	2023-05-11 02:30:19.879
9	645c533ab37fce40c0e3c8aa	StateAvgInvoiceAmount_28d	FLOAT	DRAFT	False	[GROCERYCUSTOMER, GROCERYINVOICE]	[GROCERYINVOICE]	[frenchstate]	[frenchstate]	2023-05-11 02:30:19.007
10	645c5337b37fce40c0e3c8a8	StateInventory_28d	OBJECT	DRAFT	False	[GROCERYCUSTOMER, GROCERYINVOICE, INVOICEITEMS...	[INVOICEITEMS]	[frenchstate]	[frenchstate]	2023-05-11 02:30:17.076
11	645c5336b37fce40c0e3c8a4	StatePopulation	FLOAT	DRAFT	False	[GROCERYCUSTOMER]	[GROCERYCUSTOMER]	[frenchstate]	[frenchstate]	2023-05-11 02:30:14.988
12	645c5335b37fce40c0e3c8a2	StateName	VARCHAR	DRAFT	False	[GROCERYCUSTOMER]	[GROCERYCUSTOMER]	[grocerycustomer]	[grocerycustomer]	2023-05-11 02:30:13.971

In [16]:

            
                Copied!
                
# find suitable features that use the grocery invoice items table
grocery_items_features = suitable_features.loc[
    ["INVOICEITEMS" in x for x in suitable_features.tables.values]
]

# show the features
display(grocery_items_features)
# find suitable features that use the grocery invoice items table
grocery_items_features = suitable_features.loc[
    ["INVOICEITEMS" in x for x in suitable_features.tables.values]
]

# show the features
display(grocery_items_features)

	id	name	dtype	readiness	online_enabled	tables	primary_tables	entities	primary_entities	created_at
5	645c5340b37fce40c0e3c8b2	CustomerInventory_24w	OBJECT	DRAFT	False	[GROCERYINVOICE, INVOICEITEMS, GROCERYPRODUCT]	[INVOICEITEMS]	[grocerycustomer]	[grocerycustomer]	2023-05-11 02:30:25.298
6	645c533eb37fce40c0e3c8b0	CustomerInventory_28d	OBJECT	DRAFT	False	[GROCERYINVOICE, INVOICEITEMS, GROCERYPRODUCT]	[INVOICEITEMS]	[grocerycustomer]	[grocerycustomer]	2023-05-11 02:30:23.367
10	645c5337b37fce40c0e3c8a8	StateInventory_28d	OBJECT	DRAFT	False	[GROCERYCUSTOMER, GROCERYINVOICE, INVOICEITEMS...	[INVOICEITEMS]	[frenchstate]	[frenchstate]	2023-05-11 02:30:17.076

Understand an Existing feature¶

Learning Objectives

In this section you will learn how to:

load a feature from the catalog
view the metadata of a feature
materialize feature values
view feature lineage as a definition file

Example: Load a feature from the catalog¶

In [17]:

            
                Copied!
                
# get the CustomerInventory_28d feature
customer_inventory_28d = catalog.get_feature("CustomerInventory_28d")
# get the CustomerInventory_28d feature
customer_inventory_28d = catalog.get_feature("CustomerInventory_28d")

Example: View the metadata of a feature¶

In [18]:

            
                Copied!
                
# get a list of all the features in the catalog
all_features = catalog.list_features()

# display the current feature
display(all_features.loc[all_features.name == customer_inventory_28d.name])
# get a list of all the features in the catalog
all_features = catalog.list_features()

# display the current feature
display(all_features.loc[all_features.name == customer_inventory_28d.name])

	id	name	dtype	readiness	online_enabled	tables	primary_tables	entities	primary_entities	created_at
6	645c533eb37fce40c0e3c8b0	CustomerInventory_28d	OBJECT	DRAFT	False	[GROCERYINVOICE, INVOICEITEMS, GROCERYPRODUCT]	[INVOICEITEMS]	[grocerycustomer]	[grocerycustomer]	2023-05-11 02:30:23.367

In [19]:

            
                Copied!
                
# view the detailed metadata
customer_inventory_28d.info()
# view the detailed metadata
customer_inventory_28d.info()

Out[19]:

{
  'name': 'CustomerInventory_28d',
  'created_at': '2023-05-11T02:30:23.367000',
  'updated_at': None,
  'entities': [
    {
      'name': 'grocerycustomer',
      'serving_names': [
        'GROCERYCUSTOMERGUID'
      ],
      'catalog_name': 'quick start reusing features 20230511:0229'
    }
  ],
  'primary_entity': [
    {
      'name': 'grocerycustomer',
      'serving_names': [
        'GROCERYCUSTOMERGUID'
      ],
      'catalog_name': 'quick start reusing features 20230511:0229'
    }
  ],
  'tables': [
    {
      'name': 'GROCERYPRODUCT',
      'status': 'PUBLIC_DRAFT',
      'catalog_name': 'quick start reusing features 20230511:0229'
    },
    {
      'name': 'INVOICEITEMS',
      'status': 'PUBLIC_DRAFT',
      'catalog_name': 'quick start reusing features 20230511:0229'
    },
    {
      'name': 'GROCERYINVOICE',
      'status': 'PUBLIC_DRAFT',
      'catalog_name': 'quick start reusing features 20230511:0229'
    }
  ],
  'default_version_mode': 'AUTO',
  'version_count': 1,
  'catalog_name': 'quick start reusing features 20230511:0229',
  'dtype': 'OBJECT',
  'primary_table': [
    {
      'name': 'INVOICEITEMS',
      'status': 'PUBLIC_DRAFT',
      'catalog_name': 'quick start reusing features 20230511:0229'
    }
  ],
  'default_feature_id': '645c533eb37fce40c0e3c8b0',
  'version': {
    'this': 'V230511',
    'default': 'V230511'
  },
  'readiness': {
    'this': 'DRAFT',
    'default': 'DRAFT'
  },
  'table_feature_job_setting': {
    'this': [
      {
        'table_name': 'GROCERYINVOICE',
        'feature_job_setting': {
          'blind_spot': '0s',
          'frequency': '3600s',
          'time_modulo_frequency': '90s'
        }
      }
    ],
    'default': [
      {
        'table_name': 'GROCERYINVOICE',
        'feature_job_setting': {
          'blind_spot': '0s',
          'frequency': '3600s',
          'time_modulo_frequency': '90s'
        }
      }
    ]
  },
  'table_cleaning_operation': {
    'this': [],
    'default': []
  },
  'versions_info': None,
  'metadata': {
    'input_columns': {
      'Input0': {
        'data': 'GROCERYPRODUCT',
        'column_name': 'ProductGroup',
        'semantic': None
      }
    },
    'derived_columns': {},
    'aggregations': {
      'F0': {
        'name': 'CustomerInventory_28d',
        'column': None,
        'function': 'count',
        'keys': [
          'GroceryCustomerGuid'
        ],
        'window': '28d',
        'category': 'ProductGroup',
        'filter': False
      }
    },
    'post_aggregation': None
  }
}

Example: Materialize sample values¶

In [20]:

            
                Copied!
                
                    
                    
                
                

        
# get some invoice IDs and invoice event timestamps from Q4 2022
filter = (grocery_invoice_view["Timestamp"].dt.year == 2022) & (grocery_invoice_view["Timestamp"].dt.month >= 10)

observation_set = (
    grocery_invoice_view[filter].sample(10)[["GroceryCustomerGuid", "Timestamp"]]
    .rename({
        "Timestamp": "POINT_IN_TIME",
        "GroceryCustomerGuid": "GROCERYCUSTOMERGUID",
    }, axis=1)
)
display(observation_set)
# get some invoice IDs and invoice event timestamps from Q4 2022
filter = (grocery_invoice_view["Timestamp"].dt.year == 2022) & (grocery_invoice_view["Timestamp"].dt.month >= 10)

observation_set = (
    grocery_invoice_view[filter].sample(10)[["GroceryCustomerGuid", "Timestamp"]]
    .rename({
        "Timestamp": "POINT_IN_TIME",
        "GroceryCustomerGuid": "GROCERYCUSTOMERGUID",
    }, axis=1)
)
display(observation_set)

	GROCERYCUSTOMERGUID	POINT_IN_TIME
0	7a1bc5dc-e198-419e-b972-0abbdf8903c1	2022-12-23 10:41:08
1	6999ea3f-fc7e-4b48-b01f-02a71e0f474d	2022-11-30 18:25:02
2	33a71f26-f36f-423f-8ddb-a7d674102d3b	2022-10-29 09:29:31
3	adb23858-0ea8-4ec1-9d17-5ae5cb70d856	2022-11-29 11:54:38
4	09fbee0c-521e-40ee-a2ff-8ed4187dcbc4	2022-11-09 16:57:02
5	0a796b2c-db2d-4414-847b-a999557c4008	2022-11-28 03:10:50
6	be10bc87-b09e-49ec-a66a-d6a801a29abf	2022-10-21 21:05:13
7	08d9c64b-b5e1-40d3-9964-0b3e216ff0c7	2022-10-02 12:34:51
8	69f895f3-2677-47b1-a577-b048e5004d4d	2022-10-31 12:37:45
9	db2d5721-8869-40f7-984c-a94d614fdf69	2022-11-02 18:40:31

In [21]:

            
                Copied!
                
# display the feature values
display(customer_inventory_28d.preview(observation_set))
# display the feature values
display(customer_inventory_28d.preview(observation_set))

	GROCERYCUSTOMERGUID	POINT_IN_TIME	CustomerInventory_28d
0	7a1bc5dc-e198-419e-b972-0abbdf8903c1	2022-12-23 10:41:08	{"Aide à la Pâtisserie":1,"Beurre":1,"Biscuits...
1	6999ea3f-fc7e-4b48-b01f-02a71e0f474d	2022-11-30 18:25:02	{"Biscuits":2,"Chips et Tortillas":1,"Colas, T...
2	33a71f26-f36f-423f-8ddb-a7d674102d3b	2022-10-29 09:29:31	{"Adoucissants et Soin du linge":1,"Autres Pro...
3	adb23858-0ea8-4ec1-9d17-5ae5cb70d856	2022-11-29 11:54:38	NaN
4	09fbee0c-521e-40ee-a2ff-8ed4187dcbc4	2022-11-09 16:57:02	{"Bières et Cidres":1,"Chips et Tortillas":2,"...
5	0a796b2c-db2d-4414-847b-a999557c4008	2022-11-28 03:10:50	{"Chocolats en Poudre":1,"Céréales":1,"Lait UH...
6	be10bc87-b09e-49ec-a66a-d6a801a29abf	2022-10-21 21:05:13	{"Bières et Cidres":2,"Colas, Thés glacés et S...
7	08d9c64b-b5e1-40d3-9964-0b3e216ff0c7	2022-10-02 12:34:51	{"Autres Produits Laitiers":1,"Biscuits apérit...
8	69f895f3-2677-47b1-a577-b048e5004d4d	2022-10-31 12:37:45	{"Fromages":2,"Lessives":1,"Nettoyants Vaissel...
9	db2d5721-8869-40f7-984c-a94d614fdf69	2022-11-02 18:40:31	{"Biscuits apéritifs":1,"Chat":5,"Colas, Thés ...

Example: View the feature lineage¶

In [22]:

            
                Copied!
                
# display the feature lineage for the feature we just loaded from the feature store
display(customer_inventory_28d.definition)
# display the feature lineage for the feature we just loaded from the feature store
display(customer_inventory_28d.definition)

# Generated by SDK version: 0.2.2
from bson import ObjectId
from featurebyte import DimensionTable
from featurebyte import FeatureJobSetting
from featurebyte import ItemTable


# item_table name: "INVOICEITEMS", event_table name: "GROCERYINVOICE"
item_table = ItemTable.get_by_id(ObjectId("645c532db37fce40c0e3c89c"))
item_view = item_table.get_view(
    event_suffix=None,
    view_mode="manual",
    drop_column_names=[],
    column_cleaning_operations=[],
    event_drop_column_names=["record_available_at"],
    event_column_cleaning_operations=[],
    event_join_column_names=[
        "Timestamp",
        "GroceryInvoiceGuid",
        "GroceryCustomerGuid",
        "tz_offset",
    ],
)

# dimension_table name: "GROCERYPRODUCT"
dimension_table = DimensionTable.get_by_id(ObjectId("645c5330b37fce40c0e3c89d"))
dimension_view = dimension_table.get_view(
    view_mode="manual", drop_column_names=[], column_cleaning_operations=[]
)
joined_view = item_view.join(dimension_view, on=None, how="left", rsuffix="")
grouped = joined_view.groupby(
    by_keys=["GroceryCustomerGuid"], category="ProductGroup"
).aggregate_over(
    value_column=None,
    method="count",
    windows=["28d"],
    feature_names=["CustomerInventory_28d"],
    feature_job_setting=FeatureJobSetting(
        blind_spot="0s", frequency="3600s", time_modulo_frequency="90s"
    ),
    skip_fill_na=True,
)
feat = grouped["CustomerInventory_28d"]
output = feat

Create New Features from Existing Features¶

You can use existing features as inputs to new features.

Learning objectives

In this section you wil learn how to:

create a new feature from two existing features

Example: Create a new similarity feature from two existing features¶

In [23]:

            
                Copied!
                
                    
                    
                
                

        
# get the StateInventory_28d feature
state_inventory_28d = catalog.get_feature("StateInventory_28d")

# get the CustomerInventory_28d feature
customer_inventory_28d = catalog.get_feature("CustomerInventory_28d")

# create a new feature that is the cosine similarity of the two features
customer_state_items_similarity_28d = customer_inventory_28d.cd.cosine_similarity(
    state_inventory_28d
)
customer_state_items_similarity_28d.name = "CustomerStateItemsSimilarity_28d"
customer_state_items_similarity_28d.save()

# display the feature lineage for the feature we just created
display(customer_state_items_similarity_28d.definition)
# get the StateInventory_28d feature
state_inventory_28d = catalog.get_feature("StateInventory_28d")

# get the CustomerInventory_28d feature
customer_inventory_28d = catalog.get_feature("CustomerInventory_28d")

# create a new feature that is the cosine similarity of the two features
customer_state_items_similarity_28d = customer_inventory_28d.cd.cosine_similarity(
    state_inventory_28d
)
customer_state_items_similarity_28d.name = "CustomerStateItemsSimilarity_28d"
customer_state_items_similarity_28d.save()

# display the feature lineage for the feature we just created
display(customer_state_items_similarity_28d.definition)

# Generated by SDK version: 0.2.2
from bson import ObjectId
from featurebyte import DimensionTable
from featurebyte import FeatureJobSetting
from featurebyte import ItemTable
from featurebyte import SCDTable


# scd_table name: "GROCERYCUSTOMER"
scd_table = SCDTable.get_by_id(ObjectId("645c5328b37fce40c0e3c89a"))
scd_view = scd_table.get_view(
    view_mode="manual",
    drop_column_names=["record_available_at", "CurrentRecord"],
    column_cleaning_operations=[],
)

# item_table name: "INVOICEITEMS", event_table name: "GROCERYINVOICE"
item_table = ItemTable.get_by_id(ObjectId("645c532db37fce40c0e3c89c"))
item_view = item_table.get_view(
    event_suffix=None,
    view_mode="manual",
    drop_column_names=[],
    column_cleaning_operations=[],
    event_drop_column_names=["record_available_at"],
    event_column_cleaning_operations=[],
    event_join_column_names=[
        "Timestamp",
        "GroceryInvoiceGuid",
        "GroceryCustomerGuid",
        "tz_offset",
    ],
)

# dimension_table name: "GROCERYPRODUCT"
dimension_table = DimensionTable.get_by_id(ObjectId("645c5330b37fce40c0e3c89d"))
dimension_view = dimension_table.get_view(
    view_mode="manual", drop_column_names=[], column_cleaning_operations=[]
)
joined_view = item_view.join(dimension_view, on=None, how="left", rsuffix="")
joined_view_1 = joined_view.join(scd_view, on=None, how="left", rsuffix="")
grouped = joined_view_1.groupby(
    by_keys=["State"], category="ProductGroup"
).aggregate_over(
    value_column=None,
    method="count",
    windows=["28d"],
    feature_names=["StateInventory_28d"],
    feature_job_setting=FeatureJobSetting(
        blind_spot="0s", frequency="3600s", time_modulo_frequency="90s"
    ),
    skip_fill_na=True,
)
feat = grouped["StateInventory_28d"]
grouped_1 = joined_view.groupby(
    by_keys=["GroceryCustomerGuid"], category="ProductGroup"
).aggregate_over(
    value_column=None,
    method="count",
    windows=["28d"],
    feature_names=["CustomerInventory_28d"],
    feature_job_setting=FeatureJobSetting(
        blind_spot="0s", frequency="3600s", time_modulo_frequency="90s"
    ),
    skip_fill_na=True,
)
feat_1 = grouped_1["CustomerInventory_28d"]
feat_2 = feat_1.cd.cosine_similarity(other=feat)
feat_2.name = "CustomerStateItemsSimilarity_28d"
output = feat_2

Create a New Feature List From Existing Features¶

Learning objectives

In this section you will learn how to:

create a feature list with a primary entity suited to your use case

Example: Create a customer level feature list¶

In [24]:

            
                Copied!
                
                    
                    
                
                

        
# get a list of all the features in the catalog
all_features = catalog.list_features()

# filter to retain only those with grocery customer or state as their primary entity
child_entity = "groceryinvoice"
suitable_features = all_features.loc[
    [child_entity not in x for x in all_features.entities.values]
]
product_entity = "groceryproduct"
suitable_features = suitable_features.loc[
    [product_entity not in x for x in suitable_features.entities.values]
]

# show the features
display(suitable_features)
# get a list of all the features in the catalog
all_features = catalog.list_features()

# filter to retain only those with grocery customer or state as their primary entity
child_entity = "groceryinvoice"
suitable_features = all_features.loc[
    [child_entity not in x for x in all_features.entities.values]
]
product_entity = "groceryproduct"
suitable_features = suitable_features.loc[
    [product_entity not in x for x in suitable_features.entities.values]
]

# show the features
display(suitable_features)

	id	name	dtype	readiness	online_enabled	tables	primary_tables	entities	primary_entities	created_at
0	645c5363b37fce40c0e3c8c9	CustomerStateItemsSimilarity_28d	FLOAT	DRAFT	False	[GROCERYCUSTOMER, GROCERYINVOICE, INVOICEITEMS...	[INVOICEITEMS]	[grocerycustomer, frenchstate]	[grocerycustomer]	2023-05-11 02:31:01.994
4	645c5343b37fce40c0e3c8b8	CustomerYearOfBirth	INT	DRAFT	False	[GROCERYCUSTOMER]	[GROCERYCUSTOMER]	[grocerycustomer]	[grocerycustomer]	2023-05-11 02:30:27.690
5	645c5341b37fce40c0e3c8b4	CustomerSpend_14d	FLOAT	DRAFT	False	[GROCERYINVOICE]	[GROCERYINVOICE]	[grocerycustomer]	[grocerycustomer]	2023-05-11 02:30:26.874
6	645c5340b37fce40c0e3c8b2	CustomerInventory_24w	OBJECT	DRAFT	False	[GROCERYINVOICE, INVOICEITEMS, GROCERYPRODUCT]	[INVOICEITEMS]	[grocerycustomer]	[grocerycustomer]	2023-05-11 02:30:25.298
7	645c533eb37fce40c0e3c8b0	CustomerInventory_28d	OBJECT	DRAFT	False	[GROCERYINVOICE, INVOICEITEMS, GROCERYPRODUCT]	[INVOICEITEMS]	[grocerycustomer]	[grocerycustomer]	2023-05-11 02:30:23.367
8	645c533cb37fce40c0e3c8ae	StateMeanLongitude	FLOAT	DRAFT	False	[GROCERYCUSTOMER]	[GROCERYCUSTOMER]	[frenchstate]	[frenchstate]	2023-05-11 02:30:20.604
9	645c533bb37fce40c0e3c8ac	StateMeanLatitude	FLOAT	DRAFT	False	[GROCERYCUSTOMER]	[GROCERYCUSTOMER]	[frenchstate]	[frenchstate]	2023-05-11 02:30:19.879
10	645c533ab37fce40c0e3c8aa	StateAvgInvoiceAmount_28d	FLOAT	DRAFT	False	[GROCERYCUSTOMER, GROCERYINVOICE]	[GROCERYINVOICE]	[frenchstate]	[frenchstate]	2023-05-11 02:30:19.007
11	645c5337b37fce40c0e3c8a8	StateInventory_28d	OBJECT	DRAFT	False	[GROCERYCUSTOMER, GROCERYINVOICE, INVOICEITEMS...	[INVOICEITEMS]	[frenchstate]	[frenchstate]	2023-05-11 02:30:17.076
12	645c5336b37fce40c0e3c8a4	StatePopulation	FLOAT	DRAFT	False	[GROCERYCUSTOMER]	[GROCERYCUSTOMER]	[frenchstate]	[frenchstate]	2023-05-11 02:30:14.988
13	645c5335b37fce40c0e3c8a2	StateName	VARCHAR	DRAFT	False	[GROCERYCUSTOMER]	[GROCERYCUSTOMER]	[grocerycustomer]	[grocerycustomer]	2023-05-11 02:30:13.971

In [25]:

            
                Copied!
                
                    
                    
                
                

        
# create a new feature list from the 12 features we just searched for
customer_features = fb.FeatureList([
    catalog.get_feature(x)
    for x in suitable_features.name.values
], name="CustomerFeatures")
customer_features.save()

# display a sample of the feature list values
display(customer_features.preview(observation_set))
# create a new feature list from the 12 features we just searched for
customer_features = fb.FeatureList([
    catalog.get_feature(x)
    for x in suitable_features.name.values
], name="CustomerFeatures")
customer_features.save()

# display a sample of the feature list values
display(customer_features.preview(observation_set))

Saving Feature(s) |████████████████████████████████████████| 11/11 [100%] in 3.6
Loading Feature(s) |████████████████████████████████████████| 11/11 [100%] in 2.

	GROCERYCUSTOMERGUID	POINT_IN_TIME	CustomerStateItemsSimilarity_28d	CustomerYearOfBirth	CustomerSpend_14d	CustomerInventory_24w	CustomerInventory_28d	StateMeanLongitude	StateMeanLatitude	StateAvgInvoiceAmount_28d	StateInventory_28d	StatePopulation	StateName
0	7a1bc5dc-e198-419e-b972-0abbdf8903c1	2022-12-23 10:41:08	0.811556	1943	82.98	{"Adoucissants et Soin du linge":1,"Aide à la ...	{"Aide à la Pâtisserie":1,"Beurre":1,"Biscuits...	5.887195	43.456104	16.133960	{"Adoucissants et Soin du linge":5,"Aide à la ...	53	Provence-Alpes-Côte d'Azur
1	6999ea3f-fc7e-4b48-b01f-02a71e0f474d	2022-11-30 18:25:02	0.685521	1950	8.96	{"Adoucissants et Soin du linge":2,"Biscuits a...	{"Biscuits":2,"Chips et Tortillas":1,"Colas, T...	2.241215	48.738384	19.683605	{"Adoucissants et Soin du linge":13,"Aide à la...	181	Île-de-France
2	33a71f26-f36f-423f-8ddb-a7d674102d3b	2022-10-29 09:29:31	0.623732	1955	182.03	{"Adoucissants et Soin du linge":2,"Aide à la ...	{"Adoucissants et Soin du linge":1,"Autres Pro...	2.241215	48.738384	18.578355	{"Adoucissants et Soin du linge":21,"Aide à la...	181	Île-de-France
3	adb23858-0ea8-4ec1-9d17-5ae5cb70d856	2022-11-29 11:54:38	NaN	1977	0.00	{"Cave à Vins":2,"Colas, Thés glacés et Sodas"...	NaN	2.241215	48.738384	19.542023	{"Adoucissants et Soin du linge":14,"Aide à la...	181	Île-de-France
4	09fbee0c-521e-40ee-a2ff-8ed4187dcbc4	2022-11-09 16:57:02	0.601944	1997	7.50	{"Adoucissants et Soin du linge":7,"Aide à la ...	{"Bières et Cidres":1,"Chips et Tortillas":2,"...	5.855939	48.789776	21.820370	{"Adoucissants et Soin du linge":1,"Aide à la ...	8	Lorraine
5	0a796b2c-db2d-4414-847b-a999557c4008	2022-11-28 03:10:50	0.268576	1963	11.24	{"Biscuits apéritifs":1,"Biscuits":1,"Boucheri...	{"Chocolats en Poudre":1,"Céréales":1,"Lait UH...	2.241215	48.738384	19.911254	{"Adoucissants et Soin du linge":14,"Aide à la...	181	Île-de-France
6	be10bc87-b09e-49ec-a66a-d6a801a29abf	2022-10-21 21:05:13	0.550035	1972	34.26	{"Adoucissants et Soin du linge":1,"Bières et ...	{"Bières et Cidres":2,"Colas, Thés glacés et S...	2.241215	48.738384	18.652529	{"Adoucissants et Soin du linge":25,"Aide à la...	181	Île-de-France
7	08d9c64b-b5e1-40d3-9964-0b3e216ff0c7	2022-10-02 12:34:51	0.836615	1971	21.59	{"Aide à la Pâtisserie":2,"Animalerie, Soins e...	{"Autres Produits Laitiers":1,"Biscuits apérit...	7.573264	48.177401	17.236800	{"Aide à la Pâtisserie":5,"Autres Produits Lai...	12	Alsace
8	69f895f3-2677-47b1-a577-b048e5004d4d	2022-10-31 12:37:45	0.442991	1992	7.08	{"Beurre":2,"Biscuits":1,"Bières et Cidres":2,...	{"Fromages":2,"Lessives":1,"Nettoyants Vaissel...	5.054081	45.500198	17.619500	{"Adoucissants et Soin du linge":5,"Aide à la ...	33	Rhône-Alpes
9	db2d5721-8869-40f7-984c-a94d614fdf69	2022-11-02 18:40:31	0.428169	1957	7.52	{"Aide à la Pâtisserie":1,"Biscuits apéritifs"...	{"Biscuits apéritifs":1,"Chat":5,"Colas, Thés ...	2.241215	48.738384	18.723496	{"Adoucissants et Soin du linge":14,"Aide à la...	181	Île-de-France

In [26]:

            
                Copied!
                
# list the feature lists in the catalog
catalog.list_feature_lists()
# list the feature lists in the catalog
catalog.list_feature_lists()

Out[26]:

	id	name	num_feature	status	deployed	readiness_frac	online_frac	tables	entities	created_at
0	645c536eb37fce40c0e3c8cb	CustomerFeatures	11	DRAFT	False	0.0	0.0	[GROCERYCUSTOMER, GROCERYINVOICE, INVOICEITEMS...	[grocerycustomer, frenchstate]	2023-05-11 02:31:16.244
1	645c5349b37fce40c0e3c8c2	StateFeatureList	5	DRAFT	False	0.0	0.0	[GROCERYCUSTOMER, GROCERYINVOICE, INVOICEITEMS...	[frenchstate]	2023-05-11 02:30:36.305

Next Steps¶

Now that you've completed the quick-start reusing features tutorial, you can put your knowledge into practice or learn more:

Learn more about materializing features via the "Deep Dive Materializing Features" tutorial
Learn more about feature engineering via the "Deep Dive Feature Engineering" tutorial
Learn about data modeling via the "Deep Dive Data Modeling" tutorial