Lace is a Bayesian tabular data analysis tool written in rust with a pyo3 wrapper for python use.
Conceptually, Lace sits somewhere between heuristic, optimization based methods like random forests and deep learning, and statistical methods like probabilistic programming languages. Lace builds a joint probability distribution over a data table from which users can make predictions, evaluate likelihood and uncertainty, simulate data, and more with no explicit model building required.
We’d love to have your feedback!
Thank you for sharing your crate. Looks pretty solid to me. I’ve tried Pyro about two years ago, but it simply wasn’t the right tool for the project back then.
In the meantime I’ve developed the deep causality crate to tackle structurally complex causality models.
I think there is a synergy to use Bayesian probability to model probable causes under uncertainty. Judea Pearl himself brought this up a while ago.
This topic was automatically closed 90 days after the last reply. We invite you to open a new topic if you have further questions or comments.