Forthcoming fixes - cmu-phil/tetrad GitHub Wiki

New fixes

Interim up-to-date beta versions of the Tetrad launch JAR with the updates listed below can be downloaded using this link in py-tetrad:

https://github.com/cmu-phil/py-tetrad/blob/main/pytetrad/resources/tetrad-current.jar

To download the beta version, click the link above and then click the button with the downward-pointing arrow in the upper-right corner of the page. This will save a file called tetrad-current.jar, which is a launchable Tetrad interface JAR. You may rename this file if you wish to distinguish versions.

New features

  1. Adjusted RCIT implementation and added RCoT. The RCIT (Randomized Conditional Independence Test) code has been revised, and a true RCoT (Randomized Conditional Correlation Test) has been added as a general conditional independence test.
  2. Added experimental general scores. Two experimental score functions have been added. These are experimental and may be revised or removed prior to the next official release.
    • KCV BIC score - Full Graham matrix-based. Slow; small sample sizes are helpful.
    • RFF BIC score - Random Fourier Function-based. Faster.

Bug fixes / Technical improvements

  1. Changed default behavior for KCI conditional independence. The default KCI conditional independence test has been changed from using joint ((X,Z)) kernel features to using symmetrized, residualized features, improving stability in causal search. This change responds to a user-reported issue: https://github.com/cmu-phil/tetrad/issues/1946
  2. Fixed required-knowledge handling in FCI. A bug affecting the handling of required background knowledge in the FCI algorithm has been fixed in response to: https://github.com/cmu-phil/tetrad/issues/1947
  3. Added option for SP to output a DAG instead of a CPDAG.
  4. Added printing of all the highest scoring permutations to the console for SP.