Changelog • ranger

New CRAN version

Add handling of missing values for classification and regression

Fix a bug for always.split.variables (for some settings)

Add Poisson splitting rule for regression trees

Set num.threads=2 as default; respect environment variables and options
Add hierarchical shrinkage
Allow vector min.node.size and min.bucket for class-specific limits

New CRAN version

Add node.stats option to save node statistics of all nodes
Add time.interest option to restrict unique survival times (faster and saves memory)

Fix min bucket option in C++ version

Fix memory error for always.split.variables in certain settings

Fix quantile regression for factor variables in “order” mode

New CRAN version

Switch to C++14 standard
Add min.bucket parameter to restrict terminal node size
Fix a bug with always.split.variables selecting the wrong variables

URL fix for CRAN

New CRAN version

Fix weight calculation in case-specific RF (csrf())
Add as.data.frame() method for predictions

Allow split.select.weights and always.split.variables together

Faster permutation variable importance for high dimensional data (thanks to Roman Hornung)

Add deforest() function to remove trees from ensemble
Fix cross compiling for Windows

Fix https URLs

New CRAN version

Bug fixes

Add … argument to ranger()
Bug fixes

Bug fixes

Bug fixes

New CRAN version

Add regularization
Faster computation (in some cases)

Add local variable importance

Add “hellinger” splitrule for binary classification

Add x/y interface
Internal changes (seed differences possible, prediction incompatible with older versions)

Add “beta” splitrule for bounded outcomes

Accept user-specified function in quantile prediction

Bug fixes

Bug fixes

New CRAN version

Internal changes (slightly improved computation speed)
Warning: Possible seed differences compared to older versions
Bug fixes

Add support of splitting weights for corrected impurity importance
Bug fixes

Add inbag argument for manual selection of observations in trees

Bug fixes

Add max.depth parameter to limit tree depth

Bug fixes

New CRAN version

Remove GenABEL from suggested packages (removed from CRAN). GenABEL data is still supported

Improve memory management (internal changes)

Add impurity importance for the maxstat splitting rule
Bug fixes

Add ‘order’ mode for unordered factors for GenABEL SNP data (binary classification and regression)

Bug fixes

Change license of C++ core to MIT (R package is still GPL3)

Better ‘order’ mode for unordered factors for multiclass and survival

Bug fixes

Add class-weighted Gini splitting

Bug fixes

Add fixed proportion sampling

Bug fixes

New CRAN version

Faster aggregation of predictions
Fix memory issues on Windows 7
Add treeInfo() function to extract human readable tree structure

Add quantile prediction as in quantile regression forests

Add standard error estimation with the infinitesimal jackknife (now the default)

Add bias-corrected impurity importance (actual impurity reduction, AIR)
Add impurity importance for survival forests

Bug fixes

New CRAN version

Handle sparse data of class Matrix::dgCMatrix
Add prediction of standard errors to predict()

Allow devtools::install_github() without subdir and on Windows
Bug fixes

New CRAN version

Improvements in holdoutRF and importance p-value estimation

Split at mid-point between candidate values

Better formula interface: Support interactions terms and faster computation

Add randomized splitting (extraTrees)

Bug fixes

Drop unused factor levels in outcome before growing
Add predict.all for probability and survival prediction

Bug fixes

New CRAN version

Faster version of getTerminalNodeIDs(), included in predict()

Handle new factor levels in ‘order’ mode
Bug fixes

Set write.forest=TRUE by default
Add num.trees option to predict()
Bug fixes

Bug fixes

Use unadjusted p-value for 2 categories in maxstat splitting

Bug fixes

New CRAN version

Add splitting by maximally selected rank statistics for regression forests

Bug fixes

Use faster method for unordered factor splitting

Add p-values for variable importance
Bug fixes

Add splitting by maximally selected rank statistics for survival forests
Bug fixes

Add Windows multithreading support for new toolchain

Runtime improvement for regression forests on classification data

New CRAN version. New CRAN versions will be 0.x.0, development versions 0.x.y

Reduce memory usage of savest forest objects (changed child.nodeIDs interface)

Remove tuning functions, please use mlr or caret

Fix bug with alternative interface and prediction
Small fixes

Add keep.inbag option to track in-bag counts
Add option sample.fraction for fraction of sampled observations

Add tree-wise split.select.weights

Add predict.all option in predict() to get individual predictions for each tree for classification and regression
Small changes in documentation

Add case-specific random forests

Add case weights (weighted bootstrapping or subsampling)

Catch error of outdated gcc not supporting C++11 completely

Allow the user to interrupt computation from R
Transpose classification.table and rename to confusion.matrix
Respect R seed for prediction
Memory improvements for variable importance computation
Fix bug: Probability prediction for single observations
Fix bug: Results not identical when using alternative interface

Small fixes for Solaris compiler

Add C-index splitting
Fix NA SNP handling

Fix matrix and gwaa alternative survival interface
Version submitted to JSS

Small changes in documentation

Preallocate memory for splitting

Remove recursive splitting

Allow matrix as input data in R version

Fix prediction of classification forests in R

Speedup growing for continuous covariates
Add memory save option to save memory for very large datasets (but slower)
Remove memory mode option from R version since no performance gain

Fix problems when using Rcpp <0.11.4

Add option to split on unordered categorical covariates

Optimize memory management for very large survival forests

Set required Rcpp version to 0.11.2
Fix large $call objects when using BatchJobs
Add details and example on GenABEL usage to documentation
Minor changes to documentation

Speedup for survival forests with continuous covariates
R version: Generate seed from R. It is no longer necessary to set the seed argument in ranger calls.

Windows support for R version (without multithreading)

Speedup growing of regression and probability prediction forests
Prediction forests are now handled like regression forests: MSE used for prediction error and permutation importance
Fixed name conflict with randomForest package for “importance”
Fixed a bug: prediction function is now working for probability prediction forests
Slot “predictions” for probability forests now contains class probabilities
importance function is now working even if randomForest package is loaded after ranger
Fixed a bug: Split selection weights are now working as expected
Small changes in documentation