Tutorials

Get started using Weights & Biases with interactive tutorials.

  2 minute read  

Fundamentals

The following tutorials take you through the fundamentals of W&B for machine learning experiment tracking, model evaluation, hyperparameter tuning, model and dataset versioning, and more.

Track experiments

Use W&B for machine learning experiment tracking, model checkpointing, collaboration with your team and more.

Visualize predictions

Track, visualize, and compare model predictions over the course of training, using PyTorch on MNIST data.

Tune hyperparameters

Use W&B Sweeps to create an organized way to automatically search through combinations of hyperparameter values such as the learning rate, batch size, number of hidden layers, and more.

Track models and datasets

Track your ML experiment pipelines using W&B Artifacts.

See the following tutorials for step by step information on how to use popular ML frameworks and libraries with W&B:

PyTorch

Integrate W&B with your PyTorch code to add experiment tracking to your pipeline.

HuggingFace Transformers

Visualize your Hugging Face model’s performance quickly with the W&B integration.

Keras

Use W&B and Keras for machine learning experiment tracking, dataset versioning, and project collaboration.

XGBoost

Use W&B and XGBoost for machine learning experiment tracking, dataset versioning, and project collaboration.

Other resources

Visit the W&B AI Academy to learn how to train, fine-tune and use LLMs in your applications. Implement MLOps and LLMOps solutions. Tackle real-world ML challenges with W&B courses.


Last modified March 6, 2025: 05435e8