-
Notifications
You must be signed in to change notification settings - Fork 146
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Text-to-SQL #24
Comments
Hope Embedding pg_catalo can improve text to sql be more nice that traditional zeroshot or one-shot mod. I based on zero shot idea develop a shiny llm app. It has 5 steps from convert catalog to final get data.
For example, I want to know "find the top 5 sold music name and total revenue and quantity as well as average unit price", the ai will generate sql for me basedon schema and question, I extract data using the ai sql. Below is the detail steps.
"select
'CREATE TABLE ' || nspname || '.' || relname || ' (' || chr(10)||
array_to_string(
array_agg(attname ||
' ' ||
atttypid::regtype::text||
CASE attnotnull WHEN true THEN ' NOT NULL' ELSE ' NULL' end ||
coalesce(' -- COMMENT ' || quote_literal(description), '')||
',' || chr(10)
), ' '
) || chr(10)|| ');' as definition
from
pg_attribute
join
pg_class on
pg_class.oid = pg_attribute.attrelid
join
pg_namespace on
pg_namespace.oid = pg_class.relnamespace
left join
pg_description on
pg_description.objoid = pg_class.oid
and pg_description.objsubid = pg_attribute.attnum
where
nspname not in ('pg_catalog', 'information_schema')
and relkind in ('r', 'v')
group by
nspname, relname;"
=======================================================
Attache my shiny_llm_map git link and demo line for your reference
|
Experiment with text-to-sql from within the database. Can we embed the pg_catalog and use it to power text-to-sql?
The text was updated successfully, but these errors were encountered: