Package index • randomPlantedForest

Fitting and prediction

Fit a model, predict with it, and print it.

rpf(): Random Planted Forest
predict(<rpf_pruned>): Predictions of a pruned Random Planted Forest
predict(<rpf>): Predict with a Random Planted Forest
print(<rpf_pruned>): Print a pruned Random Planted Forest
print(<rpf>): Print a Random Planted Forest

Reading the model

The components that add up to the prediction, and the purification that makes them unique.

predict_components(): The components of a Random Planted Forest's prediction
purify() is_purified(): Purify a Random Planted Forest

Pruning

Leave out the small components, by a rule chosen out of bag or by cross-validation.

prune_components(): Prune the components of a Random Planted Forest

The number of splits and tuning

The out-of-bag error along the splits, and the search over the greediness and the leaf size.

oob_path_checkpoints(): Checkpoints of the out-of-bag path
tune_control(): Settings of the out-of-bag tuning
greediness_ladder(): The greediness ladder of the out-of-bag tuning

Looking inside a tree family

Draw the trees of one family, or get its leaves as boxes.

plot(<rpf>): Plot a tree family
rpf_boxes(): Extract the boxes of a tree family

Saving and restoring

Models hold a C++ object, which needs marshalling to survive saveRDS() or a worker process.

rpf_marshal() rpf_unmarshal(): Serialize and restore Random Planted Forests
rpf_is_valid(): Check whether a model's C++ forest is still alive
bundle(<rpf>) bundle(<rpf_pruned>): Bundle a Random Planted Forest

Utilities

Compact printing of the flattened forest (export_forest = TRUE).

print(<rpf_forest>) str(<rpf_forest>): Compact printing of forest structures