This repository hosts the source code for the project we conducted in the context of the Data Consultancy in Action course at JADS. PostNL presented us with a business problem for which we presented a solution 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 OPENAI 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.