1. Create Catalog

Create a new catalog¶

Every FeatureByte project starts with setting up a catalog. Think of the catalog as the central hub for storing metadata. It keeps track of details about the tables, columns, and features you and your teammates will share and work with.

We will reuse this catalog in all subsequent notebooks.

In [1]:

            
                Copied!
                
import featurebyte as fb

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

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

16:04:46 | WARNING  | Service endpoint is inaccessible: http://featurebyte-server:8088
16:04:46 | INFO     | Using profile: tutorial
16:04:47 | INFO     | Using configuration file at: /Users/gxav/.featurebyte/config.yaml
16:04:47 | INFO     | Active profile: tutorial (https://tutorials.featurebyte.com/api/v1)
16:04:47 | WARNING  | Remote SDK version (1.1.0.dev7) is different from local (1.1.0.dev1). Update local SDK to avoid unexpected behavior.
16:04:47 | INFO     | No catalog activated.

In [2]:

            
                Copied!
                
                    
                    
                
                

        
# Set the name of the feature store to associate with the catalog that you are creating
feature_store_name = "playground"
# if you change the catalog name, make sure it is consistent with the catalog name in other notebooks.
catalog_name = "Grocery Dataset Tutorial"
# create a catalog
catalog = fb.Catalog.create(catalog_name, feature_store_name=feature_store_name)
# Set the name of the feature store to associate with the catalog that you are creating
feature_store_name = "playground"
# if you change the catalog name, make sure it is consistent with the catalog name in other notebooks.
catalog_name = "Grocery Dataset Tutorial"
# create a catalog
catalog = fb.Catalog.create(catalog_name, feature_store_name=feature_store_name)

To create a catalog, we must provide the name of the feature store. Consider the feature store as an object that handles connections to your data warehouse and performs various optimizations to minimize feature serving delays.

Once catalog is created we are ready to work with it.

In [3]:

            
                Copied!
                
catalog.list_tables()
catalog.list_tables()

Out[3]:

	id	name	type	status	entities	created_at

It is empty for now, we will add more data into it in following notebooks.

1. Create Catalog

Create a new catalog¶

Fore more information on Feature stores and Catalogs refer to Glossary and SDK docs:¶

SDK reference for¶