Ludwig

Ludwig logo

Declarative machine learning: End-to-end machine learning pipelines using simple and flexible data-driven configurations.

What is Ludwig?¶

Ludwig is a declarative machine learning framework that makes it easy to define machine learning pipelines using a simple and flexible data-driven configuration system. Ludwig is suitable for a wide variety of AI tasks, and is hosted by the Linux Foundation AI & Data.

The configuration declares the input and output features, with their respective data types. Users can also specify additional parameters to preprocess, encode, and decode features, load from pre-trained models, compose the internal model architecture, set training parameters, or run hyperparameter optimization.

Ludwig will build an end-to-end machine learning pipeline automatically, using whatever is explicitly specified in the configuration, while falling back to smart defaults for any parameters that are not.

Declarative Machine Learning¶

Ludwig’s declarative approach to machine learning empowers you to have full control of the components of the machine learning pipeline that you care about, while leaving it up to Ludwig to make reasonable decisions for the rest.

Analysts, scientists, engineers, and researchers use Ludwig to explore state-of-the-art model architectures, run hyperparameter search, scale up to larger than available memory datasets and multi-node clusters, and finally serve the best model in production.

Finally, the use of abstract interfaces throughout the codebase makes it easy for users to extend Ludwig by adding new models, metrics, losses, and preprocessing functions that can be registered to make them immediately useable in the same unified configuration system.

Main Features¶

Data-Driven configuration system

A config YAML file that describes the schema of your data (input features, output features, and their types) is all you need to start training deep learning models. Ludwig uses declared features to compose a deep learning model accordingly.

input_features:
    - name: data_column_1
      type: number
    - name: data_column_2
      type: category
    - name: data_column_3
      type: text
    - name: data_column_4
      type: image
    ...

output_features:
    - name: data_column_5
      type: number
    - name: data_column_6
      type: category
    ...

Training, prediction, and evaluation from the command line

Simple commands can be used to train models and predict new data.

ludwig train --config config.yaml --dataset data.csv
ludwig predict --model_path results/experiment_run/model --dataset test.csv
ludwig eval --model_path results/experiment_run/model --dataset test.csv

Programmatic API

Ludwig also provides a simple programmatic API for all of the functionality described above and more.

from ludwig.api import LudwigModel

# train a model
config = {
    "input_features": [...],
    "output_features": [...],
}
model = LudwigModel(config)
data = pd.read_csv("data.csv")
train_stats, _, model_dir = model.train(data)

# or load a model
model = LudwigModel.load(model_dir)

# obtain predictions
predictions = model.predict(data)

Distributed training

Scale to very large datasets, train on multiple GPUs and multiple machines in a distributed setting using Ray, with no code or config changes required. Runs natively in the cloud on Kubernetes using KubeRay.

Serving

Serve models using FastAPI with a single command:

ludwig serve --model_path ./results/experiment_run/model
curl http://0.0.0.0:8000/predict -X POST -F "movie_title=Friends With Money" -F "content_rating=R" -F "genres=Art House & International, Comedy, Drama" -F "runtime=88.0" -F "top_critic=TRUE" -F "review_content=The cast is terrific, the movie isn't."

For optimized performance, compile end-to-end models with TorchScript and serve with Nviida Triton.

Hyperparameter optimization

Run hyperparameter optimization locally or using Ray Tune.

ludwig hyperopt --config config.yaml --dataset data.csv

AutoML

Ludwig AutoML takes a dataset, the target column, and a time budget, and returns a trained Ludwig model.

Third-Party integrations

Ludwig provides an extendable interface to integrate with third-party systems for tracking experiments. Third-party integrations exist for Comet ML, Weights & Biases, WhyLabs, and MLFlow.

Extensibility

Ludwig is built from the ground up with extensibility in mind. It is easy to add new data types by implementing clear, well-documented abstract classes that define functions to preprocess, encode, and decode data.

Furthermore, new torch nn.Module models can be easily added by them to a registry. This encourages reuse and sharing new models with the community. Refer to the Developer Guide for further details.

Quick Start¶

For a full tutorial, check out the official getting started guide, or take a look at end-to-end Examples.

Step 1: Install¶

Install from PyPi. Be aware that Ludwig requires Python 3.7+.

pip install ludwig

Step 2: Define a configuration¶

Create a config that describes the schema of your data.

Assume we have a text classification task, with data containing a sentence and class column like the following.

sentence	class
Former president Barack Obama ...	politics
Juventus hired Cristiano Ronaldo ...	sport
LeBron James joins the Lakers ...	sport
...	...

A configuration will look like this.

input_features:
  - name: sentence
    type: text

output_features:
  - name: class
    type: category

Starting from a simple config like the one above, any and all aspects of the model architecture, training loop, hyperparameter search, and backend infrastructure can be modified as additional fields in the declarative configuration to customize the pipeline to meet your requirements.

input_features:
  - name: sentence
    type: text
    encoder: transformer
    layers: 6
    embedding_size: 512

output_features:
  - name: class
    type: category
    loss: cross_entropy

trainer:
  epochs: 50
  batch_size: 64
  optimizer:
    type: adamw
    beat1: 0.9
  learning_rate: 0.001

backend:
  type: ray
  cache_format: parquet
  processor:
    type: dask
  trainer:
    use_gpu: true
    num_workers: 4
    resources_per_worker:
      CPU: 4
      GPU: 1

hyperopt:
  metric: f1
  sampler: random
  parameters:
    title.num_layers:
      lower: 1
      upper: 5
    trainer.learning_rate:
      values: [0.01, 0.003, 0.001]

For details on what can be configured, check out Ludwig Configuration docs.

Step 3: Train a model¶

Simple commands can be used to train models and predict new data.

ludwig train --config config.yaml --dataset data.csv

Step 4: Predict and evaluate¶

The training process will produce a model that can be used for evaluating on and obtaining predictions for new data.

ludwig predict --model path/to/trained/model --dataset heldout.csv
ludwig evaluate --model path/to/trained/model --dataset heldout.csv

Step 5: Visualize¶

Ludwig provides a suite of visualization tools allows you to analyze models' training and test performance and to compare them.

ludwig visualize --visualization compare_performance --test_statistics path/to/test_statistics_model_1.json path/to/test_statistics_model_2.json

For the full set of visualization see the Visualization Guide.

Step 6: Happy modeling¶

Try applying Ludwig to your data. Reach out if you have any questions.

Advantages¶

Minimal machine learning boilerplate

Ludwig takes care of the engineering complexity of machine learning out of the box, enabling research scientists to focus on building models at the highest level of abstraction. Data preprocessing, hyperparameter optimization, device management, and distributed training for torch.nn.Module models come completely free.

Easily build your benchmarks

Creating a state-of-the-art baseline and comparing it with a new model is a simple config change.

Easily apply new architectures to multiple problems and datasets

Apply new models across the extensive set of tasks and datasets that Ludwig supports. Ludwig includes a full benchmarking toolkit accessible to any user, for running experiments with multiple models across multiple datasets with just a simple configuration.

Highly configurable data preprocessing, modeling, and metrics

Any and all aspects of the model architecture, training loop, hyperparameter search, and backend infrastructure can be modified as additional fields in the declarative configuration to customize the pipeline to meet your requirements. For details on what can be configured, check out Ludwig Configuration docs.

Multi-modal, multi-task learning out-of-the-box

Mix and match tabular data, text, images, and even audio into complex model configurations without writing code.

Rich model exporting and tracking

Automatically track all trials and metrics with tools like Tensorboard, Comet ML, Weights & Biases, MLFlow, and Aim Stack.

Automatically scale training to multi-GPU, multi-node clusters

Go from training on your local machine to the cloud without code changes.

Low-code interface for state-of-the-art models, including pre-trained Huggingface Transformers

Ludwig also natively integrates with pre-trained models, such as the ones available in Huggingface Transformers. Users can choose from a vast collection of state-of-the-art pre-trained PyTorch models to use without needing to write any code at all. For example, training a BERT-based sentiment analysis model with Ludwig is as simple as:

ludwig train --dataset sst5 --config_str “{input_features: [{name: sentence, type: text, encoder: bert}], output_features: [{name: label, type: category}]}”

Low-code interface for AutoML

Ludwig AutoML allows users to obtain trained models by providing just a dataset, the target column, and a time budget.

auto_train_results = ludwig.automl.auto_train(dataset=my_dataset_df, target=target_column_name, time_limit_s=7200)

Easy productionisation

Ludwig makes it easy to serve deep learning models, including on GPUs. Launch a REST API for your trained Ludwig model.

ludwig serve --model_path=/path/to/model

Ludwig supports exporting models to efficient Torschscript bundles.

ludwig export_torchscript -–model_path=/path/to/model

Tutorials¶

Example Use Cases¶

More Information¶

Read our publications on Ludwig, declarative ML, and Ludwig’s SoTA benchmarks.

Learn more about how Ludwig works, how to get started, and work through more examples.

If you are interested in contributing, have questions, comments, or thoughts to share, or if you just want to be in the know, please consider joining the Ludwig Slack and follow us on Twitter!