Skip to content
anyfp

Cebra is a latent embedding that can be learned and utilized for both behavioral and neural analysis, offering valuable insights for research purposes.

About Cebra

Cebra is a machine learning tool designed for researchers in neuroscience and related fields to analyze high-dimensional behavioral and neural data. It provides a method for compressing time series data to uncover hidden structures and relationships, particularly excelling when behavioral and neural recordings are collected simultaneously. The tool supports both supervised and self-supervised approaches, enabling the creation of interpretable latent spaces for tasks such as decoding neural activity or reconstructing behavioral trajectories. Cebra has been validated on various datasets, including calcium imaging and electrophysiology, and is available as open-source software. Pricing information is not specified.

Best for: neuroscience researchers, computational neuroscientists, machine learning researchers

Key features

  • Compresses time series data to reveal hidden structures in neural and behavioral recordings
  • Supports both supervised and self-supervised learning for latent space creation
  • Enables decoding of neural activity and behavioral trajectories
  • Works with single and multi-session datasets across species
  • Provides consistent latent spaces for various neural recording modalities

Use cases

  • Analyzing neural activity to reconstruct visual stimuli or behavioral trajectories
  • Mapping animal position and movement direction from hippocampal data
  • Decoding neural data from motor and somatosensory cortices

Pros

  • Open-source implementation available on GitHub

Limitations

  • Patent pending may restrict non-academic use

Screenshots

Cebra screenshot

Reviews

Sign in to share your experience with this tool.

Sign in

No reviews yet. Be the first to review this tool.

Cebra alternatives

More Education tools →

Related tools