FAQ

Frequently Asked Questions

What is VAME?

VAME is a framework to cluster behavioral signals obtained from pose-estimation tools. It is a PyTorch-based deep learning framework which leverages the power of recurrent neural networks (RNN) to model sequential data. In order to learn the underlying complex data distribution, we use the RNN in a variational autoencoder setting to extract the latent state of the animal in every step of the input time series.

Where I can find the original VAME Publication

The original VAME publication can be found at the following link: VAME

[Advanced][Community Analysis] Why is the default value of 'cut_tree' set to 3?

Starting in VAME 0.12, we have added a default that the communities are assigned by cutting the dendrogram at node level 3. In a purely bifurcating dendrogram, this would result in 8 communities (A-H).
See [image] example of a dendrogram before and after node-3 cut.
This approach is recommended for VAME analysis of murine exploration of an open field for >20 minutes. Other species or behavioral conditions may result in dendrogram that are differently structured than the above example. In those cases, users may consider selecting a higher or lower cut level (passed as an argument), or making custom motif->community assignments (post-VAME, done manually).