In addition to providing out-of-the-box forests for quantile regression and instrumental variables, grf provides a framework for creating forests tailored to new statistical tasks. Certain components around splitting and prediction can be swapped out, within the general infrastructure for growing and predicting on trees.
The core forest implementation is written in C++, with an R interface powered by Rcpp. We recommend using a full-powered C++ IDE such as CLion, Xcode, or Visual Studio when working with the core code. To build the R package from source, cd into r-package and run the build_package.R script. Code style consistency is checked with the lintr package, and in RStudio you can make sure the code is consistent with the R standard by clicking Addins->Style.
The forest implementation is composed of two top-level components, ForestTrainer and ForestPredictor.
ForestTrainer drives the tree-growing process, and has two pluggable components.
- RelabelingStrategy is applied before every split, and produces a set of relabelled outcomes given the observations for a group of samples. In the case of quantile forests, for example, this strategy computes the quantiles for the group of samples, then relabels them with a factor corresponding to the quantile they belong to.
- SplittingRule is called to find the best split for a particular node, given a set of outcomes. There are currently implementations for standard regression and multinomial splitting.
The trained forest produces a Forest object. This can then be passed to the ForestPredictor to predict on test samples. The predictor has a pluggable 'prediction strategy', which computes a prediction given a test sample. Prediction strategies can be one of two types:
- DefaultPredictionStrategy computes a prediction given a weighted list of training sample IDs that share a leaf with the test sample. Taking quantile forests as an example, this strategy would compute the quantiles of the weighted leaf samples.
- OptimizedPredictionStrategy does not predict using a list of neighboring samples and weights, but instead precomputes summary values for each leaf during training, and uses these during prediction. This type of strategy will also be passed to ForestTrainer, so it can define how the summary values are computed.
Prediction strategies can also compute variance estimates for the predictions, given a forest trained with grouped trees. Because of performance constraints, only 'optimized' prediction strategies can provide variance estimates.
A particular type of forest is created by pulling together a set of pluggable components. As an example, a quantile forest is composed of a QuantileRelabelingStrategy, ProbabilitySplittingRule, and QuantilePredictionStrategy. The factory classes ForestTrainers and ForestPredictors define the common types of forests like regression, quantile, and causal forests.
When designing a forest for a new statistical task, we suggest following the format suggested in the paper: provide a custom RelabelingStrategy and PredictionStrategy, but use the standard RegressionSplittingRule. If modifications to the splitting rule must be made, they should be done carefully, as it is an extremely performance-sensitive piece of code.
To avoid the overhead of setting up new classes and Rcpp bindings, we provide a template for a 'custom' forest. To get started, fill in the implementation for CustomRelabelingStrategy and CustomPredictionStrategy. By default, the custom forest uses the standard regression splitting rule -- if you'd like to change this, consult the ForestTrainers.custom_trainer method.
This forest is made available in R as custom.forest (with predict.custom.forest). You can find a starter template for a test exercising the forest in testthat/test_custom_forest.R. Note that you'll need to re-run build_package.R after making changes to the C++ source.