NLP Course documentation

Set up your Argilla instance

Hugging Face's logo
Join the Hugging Face community

and get access to the augmented documentation experience

to get started

Set up your Argilla instance

Ask a Question Open In Colab Open In Studio Lab

To start using Argilla, you will need to set up your own Argilla instance first. Then you will need to install the Python SDK so that you can manage Argilla using Python code.

Deploy the Argilla UI

The easiest way to set up your Argilla instance is through Hugging Face Spaces. To create your Argilla Space, simply follow this form. If you need further guidance, check the Argilla quickstart.

Space configuration form.

⚠️ You may want to enable Persistent storage so the data isn’t lost if the Space is paused or restarted. You can do that from the Settings of your Space.

Once Argilla is up and running, you can log in with your credentials.

Install and connect the Python SDK

Now you can go to your Python environment or notebook and install the argilla library:

!pip install argilla

Let’s connect with our Argilla instance. To do that you will need the following information:

  • Your API URL: This is the URL where Argilla is running. If you are using a Space, you can open the Space, click on the three dots in the top right corner, then “Embed this Space” and copy the Direct URL. It should look something like https://<your-username>.<space-name>.hf.space.
  • Your API key: To get your key, log in to your Argilla instance and go to “My Settings”, then copy the API key.
  • Your HF token: If your Space is private, you will need to an Access Token in your Hugging Face Hub account with writing permissions.
import argilla as rg

HF_TOKEN = "..."  # only for private spaces

client = rg.Argilla(
    api_url="...",
    api_key="...",
    headers={"Authorization": f"Bearer {HF_TOKEN}"},  # only for private spaces
)

To check that everything is working properly, we’ll call me. This should return our user:

client.me

If this worked, your Argilla instance is up and running and you’re connected to it! Congrats!

We can now get started with loading our first dataset to Argilla.

< > Update on GitHub