From 38aec17b8189b1d7560f7bac6d5b8af14c5bfe24 Mon Sep 17 00:00:00 2001 From: Drew Herren Date: Thu, 20 Nov 2025 14:06:07 -0600 Subject: [PATCH 1/2] Update release notes --- NEWS.md | 22 ++++++++++++++++------ 1 file changed, 16 insertions(+), 6 deletions(-) diff --git a/NEWS.md b/NEWS.md index c794b856..efdacd71 100644 --- a/NEWS.md +++ b/NEWS.md @@ -1,4 +1,4 @@ -# stochtree 0.1.2 +# stochtree 0.2.0 ## New Features @@ -17,14 +17,24 @@ * Fixed indexing bug in cleanup of grow-from-root (GFR) samples in BART and BCF models * Avoid using covariate preprocessor in `computeForestLeafIndices` function when a `ForestSamples` object is provided (rather than a `bartmodel` or `bcfmodel` object) +* Correctly compute feature-specific split counts in R and Python ([#220](https://github.com/StochasticTree/stochtree/issues/220)) +* Avoid override of user-specified `num_burnin` parameter in BCF models with an internal propensity score ([#222](https://github.com/StochasticTree/stochtree/issues/222)) +* Outcome predictions correctly incorporate adaptive coding of untreated observations in BCF with binary treatment ([#231](https://github.com/StochasticTree/stochtree/issues/231)) + +## Documentation Improvements + +* Clarify structure / layout of samples when users request multiple chains in BART and BCF models ([#220](https://github.com/StochasticTree/stochtree/issues/220)) ## Other Changes -* Standardized naming conventions for out of sample data in prediction and posterior computation routines (we raise warnings when data are passed through `y`, `X`, `Z`, etc... arguments) - * Covariates / features are always referred to as "covariates" rather than "X" - * Treatment is referred to as "treatment" rather than "Z" - * Propensity scores are referred to as "propensity" rather than "pi_X" - * Outcomes are referred to as "outcome" rather than "Y" +* Standardized naming conventions for data elements of BART and BCF models across R and Python interfaces + * Covariates / features are always referred to as "`X`" + * Treatment is always referred to as "`Z`" + * Propensity scores are referred to as "`propensity`" (rather than "`pi`") + * Outcomes are referred to as "`y`" + * Basis vectors for leaf-wise regression models in forest terms are referred to as "`leaf_basis`" + * Group labels for additive random effects models are referred to as "`rfx_group_ids`" + * Basis vectors for additive random effects models are referred to as "`rfx_basis`" # stochtree 0.1.1 From 3d3669942f597d37402239a4561fb67791c94bad Mon Sep 17 00:00:00 2001 From: Drew Herren Date: Fri, 21 Nov 2025 11:58:37 -0600 Subject: [PATCH 2/2] Updated changelog --- CHANGELOG.md | 32 ++++++++++++++++++++++++++++++-- NEWS.md | 1 + 2 files changed, 31 insertions(+), 2 deletions(-) diff --git a/CHANGELOG.md b/CHANGELOG.md index d5cf26c5..5b06dba0 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -1,15 +1,43 @@ # Changelog -# stochtree 0.1.2 +# stochtree 0.2.0 ## New Features +* Support for multithreading in various elements of the GFR and MCMC algorithms ([#182](https://github.com/StochasticTree/stochtree/pull/182)) * Support for binary outcomes in BART and BCF with a probit link ([#164](https://github.com/StochasticTree/stochtree/pull/164)) +* Enable "restricted sweep" of tree algorithms over a handful of trees ([#173](https://github.com/StochasticTree/stochtree/pull/173)) +* Support for multivariate treatment in R ([#183](https://github.com/StochasticTree/stochtree/pull/183)) +* Enable modification of dataset variables (weights, etc...) via low-level interface ([#194](https://github.com/StochasticTree/stochtree/pull/194)) + +## Computational Improvements + +* Modified default random effects initialization ([#190](https://github.com/StochasticTree/stochtree/pull/190)) +* Avoid double prediction on training set ([#178](https://github.com/StochasticTree/stochtree/pull/178)) ## Bug Fixes * Fixed indexing bug in cleanup of grow-from-root (GFR) samples in BART and BCF models -* Avoid using covariate preprocessor in `computeForestLeafIndices` R function when a `ForestSamples` object is provided (instead of a `bartmodel` or `bcfmodel` object) +* Avoid using covariate preprocessor in `computeForestLeafIndices` function when a `ForestSamples` object is provided (rather than a `bartmodel` or `bcfmodel` object) +* Correctly compute feature-specific split counts in R and Python ([#220](https://github.com/StochasticTree/stochtree/issues/220)) +* Avoid override of user-specified `num_burnin` parameter in BCF models with an internal propensity score ([#222](https://github.com/StochasticTree/stochtree/issues/222)) +* Outcome predictions correctly incorporate adaptive coding of untreated observations in BCF with binary treatment ([#231](https://github.com/StochasticTree/stochtree/issues/231)) + +## Documentation Improvements + +* Clarify structure / layout of samples when users request multiple chains in BART and BCF models ([#220](https://github.com/StochasticTree/stochtree/issues/220)) + +## Other Changes + +* Standardized naming conventions for data elements of BART and BCF models across R and Python interfaces + * Covariates / features are always referred to as "`X`" + * Treatment is always referred to as "`Z`" + * Propensity scores are referred to as "`propensity`" (rather than "`pi`") + * Outcomes are referred to as "`y`" + * Basis vectors for leaf-wise regression models in forest terms are referred to as "`leaf_basis`" + * Group labels for additive random effects models are referred to as "`rfx_group_ids`" + * Basis vectors for additive random effects models are referred to as "`rfx_basis`" +* Run-time checks for variables that are treated as continuous but have many "ties" (which presents issues with the current GFR algorithm) when only GFR samples are requested ([#243](https://github.com/StochasticTree/stochtree/pull/243)) # stochtree 0.1.1 diff --git a/NEWS.md b/NEWS.md index efdacd71..a5489c4a 100644 --- a/NEWS.md +++ b/NEWS.md @@ -35,6 +35,7 @@ * Basis vectors for leaf-wise regression models in forest terms are referred to as "`leaf_basis`" * Group labels for additive random effects models are referred to as "`rfx_group_ids`" * Basis vectors for additive random effects models are referred to as "`rfx_basis`" +* Run-time checks for variables that are treated as continuous but have many "ties" (which presents issues with the current GFR algorithm) when only GFR samples are requested ([#243](https://github.com/StochasticTree/stochtree/pull/243)) # stochtree 0.1.1