Below are some resources that you can make use of as a student, student instructor (TA, GSI), or educator.
We've included introductory content to help get you and your students started using Weights & Biases to enable collaborative, repeatable machine and deep learning in your classroom, research lab, or student-run organization.
The resources below are targeted at educators or instructors who are teaching machine learning, deep learning or reinforcement learning courses who want to make use of Weights & Biases in the classroom. We've curated a mix of videos, slide decks, and free resources for educators to help you and your students build reproducible, collaborative models with ease:
-
How to Use Weights & Biases in the Classroom - a simple introduction to Weights & Biases in classrooms.
-
Use W&B Reports to share your ML work with your class or lab colleagues - Weights & Biases Reports let you embed interactive charts alongside your own quick notes or more polished analysis, ready to share privately or publibly!
-
How to Use W&B Teams For Your University Machine Learning Projects For Free - Weights & Biases Teams allow you to privately share chosen parts of your research and group work with fellow team-mates, improving group collaboration and efficiency!
-
Collaborative Consistency: Crafting Repeatable Results slide deck – A W&B 101 slide deck introducing the 'how' and 'why' behind reproducible, collaborative ML and DL research; also learn about research groups like OpenAI, HarmonAI, and others who use W&B to power their research and explore their work on the W&B platform
-
Free GPU Credits and Cloud Credits for Academics, Researchers, and Students
Whether you use more traditional machine learning frameworks (such as scikit-learn or XGBoost) or deep learning frameworks (like TensorFlow, PyTorch, Jax, Keras, HuggingFace, etc.) Weights & Biases has you covered! Weights & Biases also integrates with SageMaker, Kubeflow Pipelines, Docker, Ray Tune, Databricks, and even OpenAI's Gym for reinforcement learning.
- Using your favorite ML or DL library with W&B: Integrations
- Creating your own Integration: if one of the dozens of W&B-authored integrations isn't quite what you're looking for we encourage you to create your own Integration using this guide: https://docs.wandb.ai/guides/integrations/add-wandb-to-any-library
If you've used Weights & Biases in your research we would love it if you cited us! Below is a BibTeX citation for you to use. Our whitepaper is available on this page.
If you'd like to explore papers by other researchers who use Weights & Biases in their machine and deep learning workflows, please check out the 500+ citations here on Google Scholar.
@misc{wandb,
title = {Experiment Tracking with Weights and Biases},
year = {2020},
note = {Software available from wandb.com},
url={https://www.wandb.com/},
author = {Biewald, Lukas},
}
Contact us at research [at] wandb [dot] ai if you'd like to learn more about W&B-hosted webinars for your class. With our 45-minute webinar one of our W&B staff will introduce your class or academic group to using Weights & Biases for your experiment tracking needs; participants will gain an understanding of the main functionalities of the W&B library and will be empowered to start using the tool in their daily machine or deep learning model training workflows. With the 1.5-hour webinar one of our W&B staff will introduce your class to our product for the first half of the webinar; to solidify student learning the second half of the webinar is devoted to an in-class Kaggle-like competition. At the end of either webinar - the 45-minute or 1.5-hour webinar - the participants will have developed
- an understanding of experiment tracking and reproducibility in machine learning: why it's important and how Weights & Biases can help you throughout your model-building lifecycle
- familiarity with the core features and functionalities of Weights & Biases
- comfort with system metrics: Am I using the right-sized GPU for the job? Have I under- or overprovisioned my GPUs?