Getting Started with PromptQL

In this getting started guide, you’ll get hands on experience with PromptQL. In a few steps, you can have PromptQL running locally, connected to a sample dataset.

Setup

Install the DDN CLI

curl -L https://graphql-engine-cdn.hasura.io/ddn/cli/v4/get.sh | bash

Install Docker

Follow the instructions on the Docker website to install Docker on your machine.

Log in with the CLI

Logging in allows you to connect to the PromptQL runtime which is necessary for development. It also allows you to deploy your project to Hasura DDN.

ddn auth login

You will be redirected to DDN signup/login page.

Validate the installation

You can verify that the DDN CLI is installed correctly by running:

ddn doctor

Build your PromptQL app

Clone the example project

The example project is already set up to connect PromptQL to a Huggingface dataset

git clone [email protected]:hasura/huggingface-dataset-promptql.git
cd huggingface-dataset-promptql

Configure .env

Head to the app/connector/huggingface directory to now configure the Dataset.

cd app/connector/huggingface
cp .env.sample .env

Modify the values for the HUGGINGFACE_DATASET ENV. It is of the format: “user/dataset/file-path”.

For example:

HUGGINGFACE_DATASET="drossi/EDA_on_IMDB_Movies_Dataset/*.csv"

The IMDB example mentioned in the sample env is available as a sample dataset to choose. Feel free to configure the dataset that you would like to. Notice the usage of glob patterns for selecting all the CSV files. Replace this with the dataset of choice along with the path to the files. This should work for any “.csv”, “.parquet” or “.sqlite” files in Huggingface. Refer to this DuckDB blog for more examples of this format.

Introspect the Huggingface Connector

ddn connector introspect huggingface --log-level=DEBUG

Note: Depending on how big the dataset is, it should take sometime to fully import the data.

The above command runs in DEBUG mode to make it easy to catch errors for invalid files.

Add Models

Based on the dataset imported, a SQL schema would be generated. Let’s track all the models to get started quickly.

ddn model add huggingface "*"

Setup your DDN project

This will create a Hasura DDN cloud project and set up PromptQL keys to connect to the PromptQL runtime.

ddn project init

Fire up your PromptQL project

Build your supergraph.

ddn supergraph build local

Then bring up the PromptQL API server, the engine and the connector

ddn run docker-start

Act on your data

Open the PromptQL playground

In another terminal, run

ddn console --local

Browser support: PromptQL playground is supported on all browsers except Firefox and Safari. Support for these browsers should be available shortly.

Ask questions about your dataset

The console is a web app hosted at console.hasura.io that connects to your local PromptQL API and data sources. Your data is processed in the DDN PromptQL runtime but isn’t persisted externally.

Head over to the console and ask a few questions about your data.

> Hi, what are some questions that you can answer?

Overview Deploy and share