FastRGF_Regressor: A Fast Regularized Greedy Forest regressor

In mlampros/RGF: Regularized Greedy Forest

Description

A Fast Regularized Greedy Forest regressor

A Fast Regularized Greedy Forest regressor

Usage

# init <- FastRGF_Regressor$new(n_estimators = 500, max_depth = 6,
#                                      max_leaf = 50, tree_gain_ratio = 1.0,
#                                      min_samples_leaf = 5, l1 = 1.0,
#                                      l2 = 1000.0, opt_algorithm = "rgf",
#                                      learning_rate = 0.001, max_bin = NULL,
#                                      min_child_weight = 5.0, data_l2 = 2.0,
#                                      sparse_max_features = 80000,
#                                      sparse_min_occurences = 5,
#                                      n_jobs = 1, verbose = 0)

Details

the fit function builds a regressor from the training set (x, y).

the predict function predicts the regression target for x.

the cleanup function removes tempfiles used by this model. See the issue https://github.com/RGF-team/rgf/issues/75, which explains in which cases the cleanup function applies.

the get_params function returns the parameters of the model.

the score function returns the coefficient of determination ( R^2 ) for the predictions.

Methods

FastRGF_Regressor$new(n_estimators = 500, max_depth = 6, max_leaf = 50, tree_gain_ratio = 1.0, min_samples_leaf = 5, l1 = 1.0, l2 = 1000.0, opt_algorithm = "rgf", learning_rate = 0.001, max_bin = NULL, min_child_weight = 5.0, data_l2 = 2.0, sparse_max_features = 80000, sparse_min_occurences = 5, n_jobs = 1, verbose = 0)

--------------

fit(x, y, sample_weight = NULL)

--------------

predict(x)

--------------

cleanup()

--------------

get_params(deep = TRUE)

--------------

score(x, y, sample_weight = NULL)

--------------

Super class

RGF::Internal_class -> FastRGF_Regressor

Methods

Public methods

• FastRGF_Regressor$new()

• FastRGF_Regressor$clone()

Inherited methods

• RGF::Internal_class$cleanup()

• RGF::Internal_class$dump_model()

• RGF::Internal_class$feature_importances()

• RGF::Internal_class$fit()

• RGF::Internal_class$get_params()

• RGF::Internal_class$predict()

• RGF::Internal_class$predict_proba()

• RGF::Internal_class$save_model()

• RGF::Internal_class$score()

---

Method new()

Usage

FastRGF_Regressor$new(
  n_estimators = 500,
  max_depth = 6,
  max_leaf = 50,
  tree_gain_ratio = 1,
  min_samples_leaf = 5,
  l1 = 1,
  l2 = 1000,
  opt_algorithm = "rgf",
  learning_rate = 0.001,
  max_bin = NULL,
  min_child_weight = 5,
  data_l2 = 2,
  sparse_max_features = 80000,
  sparse_min_occurences = 5,
  n_jobs = 1,
  verbose = 0
)

Arguments

n_estimators

an integer. The number of trees in the forest (Original name: forest.ntrees.)

max_depth

an integer. Maximum tree depth (Original name: dtree.max_level.)

max_leaf

an integer. Maximum number of leaf nodes in best-first search (Original name: dtree.max_nodes.)

tree_gain_ratio

a float. New tree is created when leaf-nodes gain < this value * estimated gain of creating new tree (Original name: dtree.new_tree_gain_ratio.)

min_samples_leaf

an integer or float. Minimum number of training data points in each leaf node. If an integer, then consider min_samples_leaf as the minimum number. If a float, then min_samples_leaf is a percentage and ceil(min_samples_leaf * n_samples) are the minimum number of samples for each node (Original name: dtree.min_sample.)

l1

a float. Used to control the degree of L1 regularization (Original name: dtree.lamL1.)

l2

a float. Used to control the degree of L2 regularization (Original name: dtree.lamL2.)

opt_algorithm

a character string. Either "rgf" or "epsilon-greedy". Optimization method for training forest (Original name: forest.opt.)

learning_rate

a float. Step size of epsilon-greedy boosting. Meant for being used with opt_algorithm = "epsilon-greedy" (Original name: forest.stepsize.)

max_bin

an integer or NULL. Maximum number of discretized values (bins). If NULL, 65000 is used for dense data and 200 for sparse data (Original name: discretize.(sparse/dense).max_buckets.)

min_child_weight

a float. Minimum sum of data weights for each discretized value (bin) (Original name: discretize.(sparse/dense).min_bucket_weights.)

data_l2

a float. Used to control the degree of L2 regularization for discretization (Original name: discretize.(sparse/dense).lamL2.)

sparse_max_features

an integer. Maximum number of selected features. Meant for being used with sparse data (Original name: discretize.sparse.max_features.)

sparse_min_occurences

an integer. Minimum number of occurrences for a feature to be selected. Meant for being used with sparse data (Original name: discretize.sparse.min_occrrences.)

n_jobs

an integer. The number of jobs to run in parallel for both fit and predict. If -1, all CPUs are used. If -2, all CPUs but one are used. If < -1, (n_cpus + 1 + n_jobs) are used (Original name: set.nthreads.)

verbose

an integer. Controls the verbosity of the tree building process (Original name: set.verbose.)

---

Method clone()

The objects of this class are cloneable with this method.

Usage

FastRGF_Regressor$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

References

https://github.com/RGF-team/rgf/tree/master/python-package, Tong Zhang, FastRGF: Multi-core Implementation of Regularized Greedy Forest (https://github.com/RGF-team/rgf/tree/master/FastRGF)

Examples

try({
    if (reticulate::py_available() && reticulate::py_module_available("rgf.sklearn")) {

        library(RGF)

        set.seed(1)
        x = matrix(runif(100000), nrow = 100, ncol = 1000)

        y = runif(100)

        fast_RGF_regr = FastRGF_Regressor$new(max_leaf = 50)

        fast_RGF_regr$fit(x, y)

        preds = fast_RGF_regr$predict(x)
    }
}, silent=TRUE)

mlampros/RGF documentation built on March 17, 2021, 1:50 p.m.