All posts

Truncated Back-propogation Through Time

How Lightning Helps You With Long Sequences Lightning is a lightweight PyTorch wrapper for high-performance AI research that reduces the boilerplate without limiting flexibility.

How Lightning Helps You With Long Sequences

Lightning is a lightweight PyTorch wrapper for high-performance AI research that reduces the boilerplate without limiting flexibility. In this series, we are covering all the tricks Lightning offers to supercharge your machine learning training.

Too many recursive steps can lead to unstable gradients. Truncating the sum of gradients of only the last N steps can eliminate this issue. In this video we explain how you can automatically truncate a time sequence with Lightning’s implementation of Truncated Back-propogation Through Time using the truncated_bptt_steps flag.

Follow along with this notebook: https://bit.ly/33YzC1P

Interested in building with Lightning?

Tell us what you need and we’ll reach out.

  • See a demo on workloads like yours
  • Get GPU pricing and capacity
  • Review security and compliance

Trusted by 450,000+ devs and AI teams

NVIDIACursorCiscoRunwaySupercellstability.aiBlack Forest Labs

You can also email us at [email protected]