Python framework to explore, label, and monitor data for NLP

Usage example ยท Get started ยท Quick links ยท Docs



Example: Named Entity Recognition data exploration and annotation with spaCy and the IMDB dataset

What is Rubrix?

Rubrix is a production-ready Python framework for exploring, annotating, and managing data in NLP projects.

Key features:

  • Open: Rubrix is free, open-source, and 100% compatible with major NLP libraries (Hugging Face transformers, spaCy, Stanford Stanza, Flair, etc.). In fact, you can use and combine your preferred libraries without implementing any specific interface.

  • End-to-end: Most annotation tools treat data collection as a one-off activity at the beginning of each project. In real-world projects, data collection is a key activity of the iterative process of ML model development. Once a model goes into production, you want to monitor and analyze its predictions, and collect more data to improve your model over time. Rubrix is designed to close this gap, enabling you to iterate as much as you need.

  • User and Developer Experience: The key to sustainable NLP solutions is to make it easier for everyone to contribute to projects. Domain experts should feel comfortable interpreting and annotating data. Data scientists should feel free to experiment and iterate. Engineers should feel in control of data pipelines. Rubrix optimizes the experience for these core users to make your teams more productive.

  • Beyond hand-labeling: Classical hand labeling workflows are costly and inefficient, but having humans-in-the-loop is essential. Easily combine hand-labeling with active learning, bulk-labeling, zero-shot models, and weak-supervision in novel data annotation workflows.


Let’s see Rubrix in action with a quick example: Bootstraping data annotation with a zero-shot classifier


  • The availability of pre-trained language models with zero-shot capabilities means you can, sometimes, accelerate your data annotation tasks by pre-annotating your corpus with a pre-trained zeroshot model.
  • The same workflow can be applied if there is a pre-trained “supervised” model that fits your categories but needs fine-tuning for your own use case. For example, fine-tuning a sentiment classifier for a very specific type of message.


  • A zero-shot classifier from the
    Hub: typeform/distilbert-base-uncased-mnli
  • A dataset containing news
  • A set of target categories: Business, Sports, etc.

What are we going to do:

  1. Make predictions and log them into a Rubrix dataset.
  2. Use the Rubrix web app to explore, filter, and annotate some examples.
  3. Load the annotated examples and create a training set, which you can then use to train a supervised classifier.

1. Predict and log

Let’s load the zero-shot pipeline and the dataset (we are using the AGNews dataset for demonstration, but this could be your own dataset). Then, let’s go over the dataset records and log them using rb.log(). This will create a Rubrix dataset, accesible from the web app.

from transformers import pipeline
from datasets import load_dataset
import rubrix as rb

model = pipeline('zero-shot-classification', model="typeform/distilbert-base-uncased-mnli")

dataset = load_dataset("ag_news", split='test[0:100]')

labels = ['World', 'Sports', 'Business', 'Sci/Tech']

for item in dataset:
    prediction = model(item['text'], labels)

    record = rb.TextClassificationRecord(
        prediction=list(zip(prediction['labels'], prediction['scores']))

    rb.log(record, name="news_zeroshot")

2. Explore, Filter and Label

Now let’s access our Rubrix dataset and start annotating data. Let’s filter the records predicted as Business with high probability and use the bulk-labeling feature for labeling 15 records as Business:


3. Load and create a training set

After a few iterations of data annotation, we can load the Rubrix dataset and create a training set to train or fine-tune a supervised model.

# load the Rubrix dataset as a pandas DataFrame
rb_df = rb.load(name='news_zeroshot')

# filter annotated records
rb_df = rb_df[rb_df.status == "Validated"]

# select text input and the annotated label
train_df = pd.DataFrame({
    "text": rb_df.inputs.transform(lambda r: r["text"]),
    "label": rb_df.annotation,


Rubrix main components are:

  • Rubrix Python client: Python client to log, load, copy and delete Rubrix datasets.
  • Rubrix server: FastAPI REST service for reading and writing data.
  • Elasticsearch: The storage layer and search engine powering the API and the web app.
  • Rubrix web app: Easy-to-use web application for data exploration and annotation.

Quick links

Doc Description

First steps

New to Rubrix and want to get started?


Want to know more about Rubrix concepts?

๐Ÿ› ๏ธ
Setup and install

How to configure and install Rubrix


What can you use Rubrix for?

Web app reference

How to use the web-app for data exploration and annotation

Python client API

How to use the Python classes and methods

Rubrix cookbook

How to use Rubrix with your favourite libraries (flair, stanza…)

Community forum

Ask questions, share feedback, ideas and suggestions

Hugging Face tutorial

Using Rubrix with
transformers and datasets

spaCy tutorial

Using spaCy with Rubrix for NER projects

Weak supervision tutorial

How to leverage weak supervision with snorkel & Rubrix

Active learning tutorial

How to use active learning with modAL & Rubrix

Knowledge graph tutorial

How to use Rubrix with kglab & pytorch_geometric

Get started

To get started you need to follow three steps:

  1. Install the Python client
  2. Launch the web app
  3. Start logging data

1. Install the Python client

You can install the Python client with pip:

pip install rubrix

2. Launch the web app

There are two ways to launch the web app:

  • a) Using docker-compose (recommended).
  • b) Executing the server code manually

a) Using docker-compose (recommended)

Create a folder:

mkdir rubrix && cd rubrix

and launch the docker-contained web app with the following command:

wget -O docker-compose.yml https://git.io/rb-docker && docker-compose up

This is the recommended way because it automatically includes an Elasticsearch instance, Rubrix’s main persistence layer.

b) Executing the server code manually

When executing the server code manually you need to provide an Elasticsearch instance yourself.

  1. First you need to install Elasticsearch (we recommend version 7.10) and launch an Elasticsearch instance. For MacOS and Windows there are Homebrew formulae and a msi package, respectively.
  2. Install the Python client together with its server dependencies:

pip install rubrix[server]
  1. Launch a local instance of the web app

python -m rubrix.server

By default, the Rubrix server will look for your Elasticsearch endpoint at http://localhost:9200. But you can customize this by setting the ELASTICSEARCH environment variable.

3. Start logging data

The following code will log one record into a data set called example-dataset:

import rubrix as rb

    rb.TextClassificationRecord(inputs="My first Rubrix example"),

If you go to your Rubrix web app at http://localhost:6900/, you should see your first dataset. The default username and password are rubrix and 1234. You can also check the REST API docs at http://localhost:6900/api/docs.

Congratulations! You are ready to start working with Rubrix.

To better understand what’s possible take a look at Rubrix’s Cookbook


As a new open-source project, we are eager to hear your thoughts, fix bugs, and help you get started. Feel free to use the Discussion forum or the Issues and we’ll be pleased to help out.