Step-by-step tutorial on creating a feature table model.
9 minute read
Feature table model is now deprecated. You can use the Feature View model to extract the required features for an entity from your data warehouse.
Once you have done identity stitching to unify the identity of your users across all the cross-platforms, you can evaluate and maintain the required features/traits for each identified user in a feature table.
This guide provides a detailed walkthrough on how to use a PB project and create output tables in a warehouse for a feature table model.
Features using Stripe tables like total fees, days since first sale, sales i n past 365 days, has credit card, etc.
Sample project
This sample project uses the output of an identity stitching model as an input to create a feature table. The following sections describe how to define your PB project files:
Project detail
The pb_project.yaml file defines the project details such as name, schema version, connection name and the entities which represent different identifiers.
You can define all the identifiers from different input sources you want to stitch together as a user_main_id:
You need to add main_id to the list only if you have defined main_id_type: main_id in the ID stitcher buildspec.
# Project namename:sample_id_stitching# Project's yaml schema versionschema_version:84# Warehouse connectionconnection:test# Folder containing modelsmodel_folders:- models# Entities in this project and their ids.entities:- name:userid_types:- main_id # You need to add ``main_id`` to the list only if you have defined ``main_id_type:main_id`` in the id stitcher buildspec.- user_id# one of the identifier from your data source.- emailid_types:- name:user_id- name:emailfilters:- type:includeregex:".+@.+"- type:excludevalue:"test@company.com"
Input
The input file file includes the input table references and corresponding SQL for the above-mentioned entities:
inputs:- name:rsIdentifiescontract:# constraints that a model adheres tois_optional:falseis_event_stream:truewith_entity_ids:- userwith_columns:- name:timestamp- name:user_id- name:anonymous_id- name:emailapp_defaults:table:rudder_events_production.web.identifies# one of the WH table RudderStack generates when processing identify or track events.occurred_at_col:timestampids:- select:"user_id"# kind of identity sql to pick this column from above table.type:user_identity:user# as defined in project fileto_default_stitcher:true- select:"anonymous_id"type:anonymous_identity:userto_default_stitcher:true- select:"lower(email)"# can use sql.type:emailentity:userto_default_stitcher:true- name:rsTrackscontract:is_optional:falseis_event_stream:truewith_entity_ids:- userwith_columns:- name:timestamp- name:user_id- name:anonymous_idapp_defaults:table:rudder_events_production.web.tracks# another table in WH maintained by RudderStack processing track events.occurred_at_col:timestampids:- select:"user_id"type:user_identity:userto_default_stitcher:true- select:"anonymous_id"type:anonymous_identity:userto_default_stitcher:true
Model
Profiles Feature Table model lets you define the specific features/traits you want to evaluate from the huge spread of scattered data in your warehouse tables.
A sample profiles.yaml file specifying a feature table model (user_profile):
models:- name:user_profilemodel_type:feature_table_modelmodel_spec:entity_key:userfeatures:- user_lifespan- days_active- min_num_c_rank_num_b_partitionvar_groups:- name:user_varsentity_key:uservars:- entity_var:name:first_seenselect:min(timestamp::date)from:inputs/rsTrackswhere:properties_country is not null and properties_country != ''- entity_var:name:last_seenselect:max(timestamp::date)from:inputs/rsTracks- entity_var:name:user_lifespanselect:'{{user.last_seen}} - {{user.first_seen}}'description:Life Time Value of a customer- entity_var:name:days_activeselect:count(distinct timestamp::date)from:inputs/rsTracksdescription:No. of days a customer was active- entity_var:name:campaign_sourcedefault_value:"'organic'"- entity_var:name:user_rankdefault_value:-1- entity_var:name:campaign_source_first_touchselect:first_value(context_campaign_source)window:order_by:- timestamp ascfrom:inputs/rsIdentifieswhere:context_campaign_source is not null and context_campaign_source != ''- input_var:name:num_c_rank_num_b_partitionselect:rank()from:inputs/tbl_cdefault_value:-1window:partition_by:- "{{tbl_c}}.num_b"order_by:- "{{tbl_c}}.num_c asc"where:"{{tbl_c}}.num_b >= 10"- entity_var:name:min_num_c_rank_num_b_partitionselect:min(num_c_rank_num_b_partition)from:inputs/tbl_c- entity_var:name:first_billselect:min({{tbl_billing.Var("payment")}})from:inputs/tbl_billingcolumn_data_type:'{{warehouse.DataType("float")}}'
Model specification fields
Field
Data type
Description
entity_key
String
Specifies the relevant entity from your input.yaml file.
features
String
Specifies the list of name in entity_var, that must act as a feature.
entity_var
The entity_var field defines the features which act as an input for the feature table model. This variable stores the data temporarily, however, you can choose to store its data permanently by specifying the name in it as a feature in the features key.
Field
Data type
Description
name
String
Name of the entity_var to identify it uniquely.
select
String
Column name/value you want to select from the table. This defines the actual value that will be stored in the variable. You can use simple SQL expressions or select an entity_var as {{entityName.Var(\"entity_var\")}}. It has to be an aggregate operation that ensures the output is a unique value for a given main_id. For example: min(timestamp), count(*), sum(amount) etc. This holds true even when a window function (optional) is used. For example:: first_value(), last_value() etc are valid while rank(), row_number(), etc. are not valid and give unpredictable results.
from
List
Reference to the source table from where data is to be fetched. You can either refer to another model from the same YAML or some other table specified in input YAML.
where
String
Any filters you want to apply on the input table before selecting a value. This must be SQL compatible and should consider the data type of the table.
default_value
String
Default value in case no data matches the filter. When defining default values, make sure to enclose the string values in single quotes followed by double quotes to avoid SQL failure. However, you can use the non-string values without any quotes.
description
String
Textual description of the entity_var.
window
Object
Specifies the window function. Window functions in SQL usually have both partition_by and order_by properties. But for entity_var, partition_by is added with main_id as default; so, adding partition_by manually is not supported. If you need partitioning on other columns too, check out input_var where partition_by on arbitrary and multiple columns is supported.
column_data_type
String
(Optional) Data type for the entity_var. Supported data types are: integer, variant, float, varchar, text, and timestamp.
input_var
The syntax of input_var is similar to entity_var, with the only difference that instead of each value being associated to a row of the feature table, it’s associated with a row of the specified input. While you can think of an entity_var as adding a helper column to the feature table, you can consider an input_var as adding a helper column to the input.
Field
Data type
Description
name
String
Name to store the retrieved data.
select
String
Data to be stored in the name.
from
List
Reference to the source table from where data is to be fetched.
where
String
(Optional) Applies conditions for fetching data.
default_value
String
(Optional) Default value for any entity for which the calculated value would otherwise be NULL.
description
String
(Optional) Textual description.
column_data_type
String
(Optional) Data type for the input_var. Supported data types are: integer, variant, float, varchar, text, and timestamp.
window
Object
(Optional) Specifies a window over which the value should be calculated.
window
Field
Data type
Description
partition_by
String
(Optional) List of SQL expressions to use in partitioning the data.
order_by
String
(Optional) List of SQL expressions to use in ordering the data.
In window option, main_id is not added by default, it can be any arbitrary list of columns from the input table. So if a feature should be partitioned by main_id, you must add it in the partition_by key.
In the left sidebar, click Database and the corresponding Schema to view the list of all tables. You can hover over a table to see the full table name along with its creation date and time.
Write a SQL query like select * from <table_name> and execute it to see the results:
Open Postico2. If required, create a new connection by entering the relevant details. Click Test Connection followed by Connect.
Click the + icon next to Queries in the left sidebar.
You can click Database and the corresponding schema to view the list of all tables/views.
Double click on the appropriate view name to paste the name on an empty worksheet.
You can prefix SELECT * from the view name pasted previously and suffix LIMIT 10; at the end.
Press Cmd+Enter keys, or click the Run button to execute the query.
Enter your Databricks workspace URL in the web browser and log in with your username and password.
Click the Catalog icon in left sidebar.
Choose the appropriate catalog from the list and click on it to view contents.
You will see list of tables/views. Click the appropriate table/view name to paste the name on worksheet.
You can prefix SELECT * FROM before the pasted view name and suffix LIMIT 10; at the end.
Select the query text. Press Cmd+Enter, or click the Run button to execute the query.
Select Bigquery from Product and Pages to open the Bigquery Explorer.
Select the correct project from top left drop down menu.
In the left sidebar, click the project ID, then the corresponding dataset view list of all the tables and views.
Write a SQL query like select * from <table_name> limit 10; and execute it to see the results.
A sample output containing the results in Snowflake:
Partial feature tables
Partial feature tables are created when only a few input sources are available.
For example, lets say that you import a library package and some of the input models assumed in the package are not present in your warehouse.
When you remap some of these input models to nil, those inputs and the features directly or indirectly dependent upon those inputs are disabled. In such cases, a partial feature table is created from the rest of the available inputs. Similarly, ID stitcher also runs even if a few of the edge sources are not present in the warehouse or remapped to nil.
Feature table for cohort
To create feature table for a specific cohort, you can pass the cohort’s path in the entity_cohort field:
This site uses cookies to improve your experience while you navigate through the website. Out of
these
cookies, the cookies that are categorized as necessary are stored on your browser as they are as
essential
for the working of basic functionalities of the website. We also use third-party cookies that
help
us
analyze and understand how you use this website. These cookies will be stored in your browser
only
with
your
consent. You also have the option to opt-out of these cookies. But opting out of some of these
cookies
may
have an effect on your browsing experience.
Necessary
Always Enabled
Necessary cookies are absolutely essential for the website to function properly. This
category only includes cookies that ensures basic functionalities and security
features of the website. These cookies do not store any personal information.
This site uses cookies to improve your experience. If you want to
learn more about cookies and why we use them, visit our cookie
policy. We'll assume you're ok with this, but you can opt-out if you wish Cookie Settings.