search

TensorBFS / TensorInference.jl

Probabilistic inference using contraction of tensor networks
https://tensorbfs.github.io/TensorInference.jl/
MIT License
18 stars 2 forks source link

Asia network log partition is zero in the docs #62

Closed gdalle closed 1 year ago

gdalle commented 1 year ago

Is that expected?

https://github.com/openjournals/joss-reviews/issues/5700

mroavi commented 1 year ago

The partition function ( Z ) is basically the 'normalizer' for our model. It ensures that the probabilities sum to 1. If you've already got all your factors normalized, like in our toy example, then ( Z ) will be just 1.

We've added a comment in the example to clarify this point.