You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This was a first year University project I completed in 2013. It uses Python to insert records from a .txt document into an SQL database. Afterwards, it fetches data based on those records, specifically the best and worst ratings for a make of car, the average rating, and the most expensive cars.
This repo in the implementation of EMNLP'21 paper "SPARQLing Database Queries from Intermediate Question Decompositions" by Irina Saparina, Anton Osokin
[ACL 2021] This is the project containing source codes and pre-trained models about ACL2021 Long Paper ``LGESQL: Line Graph Enhanced Text-to-SQL Model with Mixed Local and Non-Local Relations".
sqlgpt-parser is a Python implementation of an SQL parser that effectively converts SQL statements into Abstract Syntax Trees (AST). By leveraging AST tree comparisons between two SQL queries, it becomes possible to achieve robust evaluation of text-to-SQL models.
Source code for grammar-based text-to-SQL parser using one variant of Transformer decoder called ASTormer. Implementations for prevalent cross-domain multi-table benchmarks Spider, SParC, CoSQL, DuSQL and Chase are all included.