HN Showcase
Back to browse
GitHub Repository

Create SQL that match your selection (with explainable AI), not the other way around

15 starsPython

Construct SQL from table records by breaking down decision tree

by rentruewang·Jun 10, 2026·3 points·1 comment
Visit ProjectView on HN

AI Analysis

●●SolidBig BrainNiche Gem

Decision tree overfitting for SQL generation is genuinely clever explainable AI.

Strengths
  • •Decision tree overfitting creates explainable query generation paths
  • •sympy simplification keeps generated boolean logic readable
  • •Streamlit demo actually works for quick testing
Weaknesses
  • •Only handles single-table CSV workflows, no multi-table joins
  • •Early stage with 12 stars and limited real-world validation
Target Audience

Data analysts and developers working with tabular data

Similar To

Text-to-SQL tools · PandasAI · LangChain SQL agents

Post Description

Create SQL by over fitting decision tree on data

Then I simplify the boolean representation.

The demo is hosted in streamlit (https://inversql.streamlit.app).

Sorry if this is a repost, my previous post had the wrong tag.

Similar Projects