Prepare the 0.10.4 release. (grf-labs#518)

* Bump version to 0.10.4 and fix CRAN checks.
* Update the changelog.

README.md

@@ -26,7 +26,7 @@ install.packages("grf")
 Any published release can also be installed from source:
 
 ```R
-install.packages("https://raw.github.com/grf-labs/grf/master/releases/grf_0.10.3.tar.gz", repos = NULL, type = "source")
+install.packages("https://raw.github.com/grf-labs/grf/master/releases/grf_0.10.4.tar.gz", repos = NULL, type = "source")
 ```
 
 `conda` users can install from the [conda-forge](https://anaconda.org/conda-forge/r-grf) channel:

r-package/grf/DESCRIPTION

@@ -1,12 +1,14 @@
 Package: grf
 Title: Generalized Random Forests
-Version: 0.10.3
+Version: 0.10.4
 Authors@R:c(
     person("Julie", "Tibshirani", role = c("aut", "cre"), email = "[email protected]"),
     person("Susan", "Athey", role = "aut"),
     person("Rina", "Friedberg", role = "ctb"),
     person("Vitor", "Hadad", role = "ctb"),
+    person("David", "Hirshberg", role = "ctb"),
     person("Luke", "Miner", role = "ctb"),
+    person("Erik", "Sverdrup", role = "ctb"),
     person("Stefan", "Wager", role = "aut"),
     person("Marvin", "Wright", role = "ctb")
     )

r-package/grf/NAMESPACE

@@ -43,6 +43,7 @@ importFrom(stats,lm)
 importFrom(stats,predict)
 importFrom(stats,quantile)
 importFrom(stats,runif)
+importFrom(stats,sd)
 importFrom(stats,var)
 importFrom(stats,weighted.mean)
 importFrom(utils,capture.output)

r-package/grf/R/causal_tuning.R

@@ -83,6 +83,7 @@
 #' }
 #'
 #' @importFrom stats runif
+#' @importFrom stats sd
 #' @importFrom utils capture.output
 #' @export
 tune_causal_forest <- function(X, Y, W, Y.hat, W.hat,

r-package/grf/R/regression_tuning.R

@@ -71,6 +71,7 @@
 #' }
 #'
 #' @importFrom stats runif
+#' @importFrom stats sd
 #' @importFrom utils capture.output
 #' @export
 tune_regression_forest <- function(X, Y,

releases/CHANGELOG.md

@@ -4,11 +4,30 @@ All notable changes to grf will be documented in this file.
 The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/)
 and this project adheres to [Semantic Versioning](http://semver.org/spec/v2.0.0.html).
 
+## [0.10.4] - 2019-09-01
+
+### Changed (breaking)
+**IMPORTANT** These changes might cause small differences in results compared to previous releases, even if the same random seed is used.
+- Ensure forest estimates are consistent across platforms. [#469](https://github.com/grf-labs/grf/pull/469), [#492](https://github.com/grf-labs/grf/pull/492)
+- The number of trees used for orthogonalization was changed from `min(500, num.trees)` to `max(50, num.trees / 4)`. [#439](https://github.com/grf-labs/grf/pull/439)
+- Solidify the parameter tuning procedure. If the optimization procedure fails, or if the selected parameters perform worse than defaults, we now return default parameters instead. [#455](https://github.com/grf-labs/grf/pull/455)
+- Introduce parameters `honesty.fraction` and `prune.empty.leaves` to help mitigate the effect of honesty on small datasets, and tune over them when `tune.parameters=TRUE`. [#456](https://github.com/grf-labs/grf/pull/456), [#484](https://github.com/grf-labs/grf/pull/484)
+
+### Added
+- Add variance estimates for local linear forests. [#442](https://github.com/grf-labs/grf/pull/442)
+- Include information about leaf samples in plotting and printing. [#460](https://github.com/grf-labs/grf/pull/460)
+- Add example of saving a plot with DiagrammeRsvg. [#478](https://github.com/grf-labs/grf/pull/478)
+- Support average effect estimates for instrumental forests (ACLATE). [#490](https://github.com/grf-labs/grf/pull/490)
+
+### Fixed
+- Performance improvements to forest training. [#514](https://github.com/grf-labs/grf/pull/514)
+
 ## [0.10.3] - 2019-06-01
 
-### Breaking Changes
-- Fix two bugs in the termination criterion for tree splitting. **IMPORTANT:** these bug fixes may cause results to change compared to previous releases, even if the same random seed is used.
-  - Remove the purity condition on outcomes during splitting. For all tree types, we used to stop splitting if all outcomes in a leaf are the same. This behavior does not make sense for causal forests (which incorporates other observations besides the outcome), so it was removed. [#362](https://github.com/grf-labs/grf/pull/362).
+### Changed (breaking)
+**IMPORTANT** These changes might cause small differences in results compared to previous releases, even if the same random seed is used.
+- Fix two bugs in the termination criterion for tree splitting.
+  - Remove the purity condition on outcomes during splitting. For all tree types, we used to stop splitting if all outcomes in a leaf are the same. This behavior does not make sense for causal forests (which incorporates other observations besides the outcome), so it was removed. [#362](https://github.com/grf-labs/grf/pull/362)
   - Stop splitting if the objective can no longer be improved. With this change, `causal_forest` may split slightly less aggressively. [#415](https://github.com/grf-labs/grf/pull/415)
 
 ### Added