causact
causact copied to clipboard
flyaflya
causact: R package to accelerate computational Bayesian inference workflows in R through interactive visualization of models and their output.
Prepare for release: * [x] `git pull` * [x] Check [current CRAN check results](https://cran.rstudio.org/web/checks/check_results_causact.html) * [x] [Polish NEWS](https://style.tidyverse.org/news.html#news-release) * [x] `urlchecker::url_check()` * [x] `devtools::build_readme()` * [x] `devtools::check(remote = TRUE, manual...
ImportError: cannot import name 'gaussian' from 'scipy.signal' is caused by bug in scipy. Fix is forthcoming shortly
I suspect this is not a bug but rather either something that just can not be done, or I am doing something wrong. This first code block works fine. Basic...
See the issue described here: https://github.com/prabinov42/Greta_causact_question/blob/main/greta_causact_2.md If a child node is deterministic and not used as a parent node, causact incorrectly labels it as a distribution. Temporary fix (use mcmc=FALSE):...
See the important role of prior predictive checks as docuemnted in https://betanalpha.github.io/assets/case_studies/principled_bayesian_workflow.html It would be nice to support this workflow in the dag_ paradigm.
The dagp_plot() plotting will not scale well beyond around 60-100+ parameters. An intelligent way to show a subset of data when there are too many parameters is required.
Posterior predictive checks as advocated in https://betanalpha.github.io/assets/case_studies/principled_bayesian_workflow.html#step_eight:_analyze_the_posterior_predictive_distribution and http://causact.updog.co/parameter-estimation.html#posterior-predictive-checks can have an automated or easier workflow.
To add some aesthetic appeal, it would be nice to automate the animation of graph transitions. A proof of concept is available here: https://jsfiddle.net/x9kgu2cv/2/ but I cannot seem to reproduce...
There does not seem to be an obvious mechanism to define random vectors in a DAG, for example, to use in a multivariate regression. Is there an option/ability to apply...
Is there a plan to make causact more generic? By that I mean entertaining two scenarios: 1) Make causact able to generate static plots for any kind of generative DAG,...
