From the course: Building Generative AI Apps to Talk to Your Data
Unlock this course with a free trial
Join today to access over 24,700 courses taught by industry experts.
Understanding the semantic model - Snowflake Tutorial
From the course: Building Generative AI Apps to Talk to Your Data
Understanding the semantic model
- Welcome back. Now that everything is set up and ready to go in your notebook, we can get started. But before we move on, I want to talk about semantic models in a little more detail. So let's discuss semantic model structure. A semantic model represents a collection of tables, each of which contains descriptions of specific aspects of the table. Each table described in the semantic model maps to a base table in Snowflake. The semantic model addresses issues related to language differences between business users and database definitions, and this provides semantic details like descriptive names and synonyms that allow cortex analysts to answer questions about data the way you intend. Let's take a look at an example. We'll include information that sales and income are synonyms for revenue. Although sales and income are not the same in general accounting and finance terms, imagine in this particular case, that they happen to be used synonymously by individuals in the business unit…
Contents
-
-
-
-
-
(Locked)
Avoiding death by dashboard4m 51s
-
(Locked)
Setting up the app4m 58s
-
(Locked)
Understanding the semantic model8m 18s
-
(Locked)
Working with the semantic model4m 37s
-
(Locked)
Using Cortex Analyst5m 56s
-
(Locked)
From text-to-SQL to TAG: Creating table-assisted generation3m 17s
-
(Locked)
Expanding the scope of the semantic model12m 11s
-
(Locked)
Building the Streamlit app11m 41s
-
(Locked)
Congratulations!3m 6s
-
(Locked)