Behavioural parameters: upper bound #124
Answered
by
MMathisLab
sofiagilardini
asked this question in
Q&A
-
|
Hello, I was wondering if there was a specified upper bound for the number of behavioural inputs (continuous) for training a CEBRA model? Thank you! |
Beta Was this translation helpful? Give feedback.
Answered by
MMathisLab
Jan 25, 2024
Replies: 1 comment
-
|
Hi there! You should not have more labels than input (neurons). The goal of cebra is to effectively do nonlinear ICA, so to separate labels into independent components ; in practice you should consider the input data, as it might be very simple or complex and need more or less inputs- if that makes sense! |
Beta Was this translation helpful? Give feedback.
0 replies
Answer selected by
stes
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Hi there! You should not have more labels than input (neurons). The goal of cebra is to effectively do nonlinear ICA, so to separate labels into independent components ; in practice you should consider the input data, as it might be very simple or complex and need more or less inputs- if that makes sense!