Semantic SQL

SQL queries that support semantic predicates as well to recommand next 10 best songs.

Setup

• Clone this repository RECURSIVELY. It has a submodule.

git clone --recursive git@github.com:rjvkr2021/CS315_Project.git

• Install the requirements
• Python version required 3.8

pip install -r requirements.txt

• Run the command to start

streamlit run main.py

This command starts the semantic-sql at the following address http://localhost:8501/

• Click Setup DB button to create and populate the database table and build the vector index. Wait for sometime, this may take upto ~30 seconds. This step is only required for the first time. You don't need to run this again ever in the future unless you click Reset DB
• Type in the sql query and click Execute Query

Features

• SQL Expansion: In addition to the standard SQL syntax, user can specify SEMANTIC predicate at the end of the sql query as shown in the example queries below.
• Results: At the end, top 10 recommanded songs will be shown.