package orf

You can search for identifiers within the package.

in-package search v0.2.0

package orf

orf
- Orf
  - Feature_vector
  - RFC
  - RFR
  - Utls

Legend:
Library
Module
Module type
Parameter
Class
Class type

type features = int Feature_vector.t

type dep_var = float

type sample = features * dep_var

type tree =

| Leaf of dep_var
| Node of tree * int * int * tree

type metric =

| MSE
| MAE
| MAD

type forest = (tree * int array) array

val train : 
  int ->
  Stdlib.Random.State.t ->
  metric ->
  int ->
  RFC.int_or_float ->
  int ->
  RFC.int_or_float ->
  int ->
  sample array ->
  forest

train ncores rng metric ntrees max_features card_features max_samples min_node_size training_set

val predict_one : int -> forest -> sample -> dep_var * float

(pred_avg, pred_std_dev) = predict_one ncores trained_forest sample

val predict_many : int -> forest -> sample array -> (dep_var * float) array

like predict_one but for an array of samples

val predict_OOB : forest -> sample array -> (dep_var * dep_var) array

use a trained forest to predict on the Out Of Bag (OOB) training set of each tree. The training_set must be provided in the same order than when the model was trained. Can be used to get a reliable model performance estimate, even if you don't have a left out test set. truth_preds = predict_OOB forest training_set

val r2 : (dep_var * dep_var) array -> float

r2 truth_preds: coefficient of determination R^2

val drop_OOB : forest -> forest

make trained model forget OOB samples (reduce model size)

type filename = string

val save : filename -> forest -> unit

Save model to file (Marshal). OOB samples are dropped prior to saving the model.

val restore : filename -> forest

Restore model from file (Marshal)