This repository hosts the source code for the project we did for PostNL. We created an application that translates data requests from natural language to a SQL Query.

First, the application needs a database. In order to get one, insert the raw anonimized .csv files to /data, then run /data/db_init/db_init.py which is a script that creates a database. Please note, this takes several minutes.

In order to run the application we set up a Dockerfile. The image requires an OPEN AI API key to be inserted as environment variable. Run the following commands to test the application locally:

Building the image

docker build -t databridge .

Running the image

docker run -d -p 8080:5000 -e OPENAI_API_KEY=<your-key> databridge

In this section you will find a high level explanation of the contents of this repository.

.
├── Dockerfile
├── README.MD
├── data
├── requirements.txt
├── services
│   ├── querier
│   └── ui
└── test

• data: In this directory all data and the related scripts are kept.
• querier: in this directory the source code of the backend is kept.
• ui: in this directory, you will find the source code of the front-end application
• Dockerfile: the blueprint for building a docker image that hosts the application.
• test: in this directory all scripts, results and data is kept related to testing the application.