import threading
import numpy as np
from sklearn.ensemble import ExtraTreesRegressor as _sk_ExtraTreesRegressor
from sklearn.ensemble._forest import DecisionTreeRegressor, ForestRegressor
from deephyper.core.utils.joblib_utils import Parallel, delayed
def _accumulate_prediction_disentangled(tree, X, min_variance, out, lock):
"""
This is a utility function for joblib's Parallel.
It can't go locally in ForestClassifier or ForestRegressor, because joblib
complains that it cannot pickle it when placed there.
"""
mean_tree = tree.predict(X).T
var_tree = tree.tree_.impurity[tree.apply(X)]
# This rounding off is done in accordance with the
# adjustment done in section 4.3.3
# of http://arxiv.org/pdf/1211.0906v2.pdf to account
# for cases such as leaves with 1 sample in which there
# is zero variance.
var_tree = np.maximum(var_tree, min_variance)
with lock:
out[0] += mean_tree
out[1] += var_tree
out[2] += mean_tree**2
def _return_mean_and_std_distentangled(X, n_outputs, trees, min_variance, n_jobs):
"""
Returns `std(Y | X)`.
Can be calculated by E[Var(Y | Tree)] + Var(E[Y | Tree]) where
P(Tree) is `1 / len(trees)`.
Parameters
----------
X : array-like, shape=(n_samples, n_features)
Input data.
n_outputs: int.
Number of outputs.
trees : list, shape=(n_estimators,)
List of fit sklearn trees as obtained from the ``estimators_``
attribute of a fit RandomForestRegressor or ExtraTreesRegressor.
predictions : array-like, shape=(n_samples,)
Prediction of each data point as returned by RandomForestRegressor
or ExtraTreesRegressor.
Returns
-------
std : array-like, shape=(n_samples,)
Standard deviation of `y` at `X`. If criterion
is set to "mse", then `std[i] ~= std(y | X[i])`.
"""
# This derives std(y | x) as described in 4.3.2 of arXiv:1211.0906
mean = np.zeros((n_outputs, len(X)))
std_al = np.zeros((n_outputs, len(X)))
std_ep = np.zeros((n_outputs, len(X)))
# Parallel loop
lock = threading.Lock()
Parallel(n_jobs=n_jobs, verbose=0, require="sharedmem")(
delayed(_accumulate_prediction_disentangled)(
tree, X, min_variance, [mean, std_al, std_ep], lock
)
for tree in trees
)
mean, std_al, std_ep = mean.T, std_al.T, std_ep.T
mean /= len(trees)
std_al /= len(trees)
std_ep = std_ep / len(trees) - mean**2
std_al[std_al <= 0.0] = 0.0
std_al **= 0.5
std_ep[std_ep <= 0.0] = 0.0
std_ep **= 0.5
return mean.reshape(-1), std_al.reshape(-1), std_ep.reshape(-1)
def _accumulate_prediction(tree, X, min_variance, out, lock):
"""
This is a utility function for joblib's Parallel.
It can't go locally in ForestClassifier or ForestRegressor, because joblib
complains that it cannot pickle it when placed there.
"""
mean_tree = tree.predict(X).T
var_tree = tree.tree_.impurity[tree.apply(X)]
# This rounding off is done in accordance with the
# adjustment done in section 4.3.3
# of http://arxiv.org/pdf/1211.0906v2.pdf to account
# for cases such as leaves with 1 sample in which there
# is zero variance.
var_tree = np.maximum(var_tree, min_variance) + np.square(mean_tree)
with lock:
out[0] += mean_tree
out[1] += var_tree
def _return_mean_and_std(X, n_outputs, trees, min_variance, n_jobs):
"""
Returns `std(Y | X)`.
Can be calculated by E[Var(Y | Tree)] + Var(E[Y | Tree]) where
P(Tree) is `1 / len(trees)`.
Parameters
----------
X : array-like, shape=(n_samples, n_features)
Input data.
n_outputs: int.
Number of outputs.
trees : list, shape=(n_estimators,)
List of fit sklearn trees as obtained from the ``estimators_``
attribute of a fit RandomForestRegressor or ExtraTreesRegressor.
predictions : array-like, shape=(n_samples,)
Prediction of each data point as returned by RandomForestRegressor
or ExtraTreesRegressor.
Returns
-------
std : array-like, shape=(n_samples,)
Standard deviation of `y` at `X`. If criterion
is set to "mse", then `std[i] ~= std(y | X[i])`.
"""
# This derives std(y | x) as described in 4.3.2 of arXiv:1211.0906
mean = np.zeros((n_outputs, len(X)))
std = np.zeros((n_outputs, len(X)))
# Parallel loop
lock = threading.Lock()
Parallel(n_jobs=n_jobs, verbose=0, require="sharedmem")(
delayed(_accumulate_prediction)(tree, X, min_variance, [mean, std], lock)
for tree in trees
)
mean, std = mean.T, std.T
mean /= len(trees)
std /= len(trees)
std = np.sqrt(np.maximum(std - np.square(mean), 0.0))
return mean.reshape(-1), std.reshape(-1)
[docs]class RandomForestRegressor(ForestRegressor):
"""
RandomForestRegressor that supports conditional std computation.
Parameters
----------
n_estimators : integer, optional (default=10)
The number of trees in the forest.
criterion : string, optional (default="mse")
The function to measure the quality of a split. Supported criteria
are "mse" for the mean squared error, which is equal to variance
reduction as feature selection criterion, and "mae" for the mean
absolute error.
max_features : int, float, string or None, optional (default="1.0")
The number of features to consider when looking for the best split:
- If int, then consider `max_features` features at each split.
- If float, then `max_features` is a percentage and
`int(max_features * n_features)` features are considered at each
split.
- If "sqrt", then `max_features=sqrt(n_features)`.
- If "log2", then `max_features=log2(n_features)`.
- If None, then `max_features=n_features`.
.. note::
The search for a split does not stop until at least one
valid partition of the node samples is found, even if it
requires to effectively inspect more than ``max_features``
features.
max_depth : integer or None, optional (default=None)
The maximum depth of the tree. If None, then nodes are expanded until
all leaves are pure or until all leaves contain less than
min_samples_split samples.
min_samples_split : int, float, optional (default=2)
The minimum number of samples required to split an internal node:
- If int, then consider `min_samples_split` as the minimum number.
- If float, then `min_samples_split` is a percentage and
`ceil(min_samples_split * n_samples)` are the minimum
number of samples for each split.
min_samples_leaf : int, float, optional (default=1)
The minimum number of samples required to be at a leaf node:
- If int, then consider `min_samples_leaf` as the minimum number.
- If float, then `min_samples_leaf` is a percentage and
`ceil(min_samples_leaf * n_samples)` are the minimum
number of samples for each node.
min_weight_fraction_leaf : float, optional (default=0.)
The minimum weighted fraction of the sum total of weights (of all
the input samples) required to be at a leaf node. Samples have
equal weight when sample_weight is not provided.
max_leaf_nodes : int or None, optional (default=None)
Grow trees with ``max_leaf_nodes`` in best-first fashion.
Best nodes are defined as relative reduction in impurity.
If None then unlimited number of leaf nodes.
min_impurity_decrease : float, optional (default=0.)
A node will be split if this split induces a decrease of the impurity
greater than or equal to this value.
The weighted impurity decrease equation is the following::
N_t / N * (impurity - N_t_R / N_t * right_impurity
- N_t_L / N_t * left_impurity)
where ``N`` is the total number of samples, ``N_t`` is the number of
samples at the current node, ``N_t_L`` is the number of samples in the
left child, and ``N_t_R`` is the number of samples in the right child.
``N``, ``N_t``, ``N_t_R`` and ``N_t_L`` all refer to the weighted sum,
if ``sample_weight`` is passed.
bootstrap : boolean, optional (default=True)
Whether bootstrap samples are used when building trees.
oob_score : bool, optional (default=False)
whether to use out-of-bag samples to estimate
the R^2 on unseen data.
n_jobs : integer, optional (default=1)
The number of jobs to run in parallel for both `fit` and `predict`.
If -1, then the number of jobs is set to the number of cores.
random_state : int, RandomState instance or None, optional (default=None)
If int, random_state is the seed used by the random number generator;
If RandomState instance, random_state is the random number generator;
If None, the random number generator is the RandomState instance used
by `np.random`.
verbose : int, optional (default=0)
Controls the verbosity of the tree building process.
warm_start : bool, optional (default=False)
When set to ``True``, reuse the solution of the previous call to fit
and add more estimators to the ensemble, otherwise, just fit a whole
new forest.
Attributes
----------
estimators_ : list of DecisionTreeRegressor
The collection of fitted sub-estimators.
feature_importances_ : array of shape = [n_features]
The feature importances (the higher, the more important the feature).
n_features_ : int
The number of features when ``fit`` is performed.
n_outputs_ : int
The number of outputs when ``fit`` is performed.
oob_score_ : float
Score of the training dataset obtained using an out-of-bag estimate.
oob_prediction_ : array of shape = [n_samples]
Prediction computed with out-of-bag estimate on the training set.
Notes
-----
The default values for the parameters controlling the size of the trees
(e.g. ``max_depth``, ``min_samples_leaf``, etc.) lead to fully grown and
unpruned trees which can potentially be very large on some data sets. To
reduce memory consumption, the complexity and size of the trees should be
controlled by setting those parameter values.
The features are always randomly permuted at each split. Therefore,
the best found split may vary, even with the same training data,
``max_features=n_features`` and ``bootstrap=False``, if the improvement
of the criterion is identical for several splits enumerated during the
search of the best split. To obtain a deterministic behaviour during
fitting, ``random_state`` has to be fixed.
References
----------
.. [1] L. Breiman, "Random Forests", Machine Learning, 45(1), 5-32, 2001.
"""
def __init__(
self,
n_estimators=100,
*,
criterion="squared_error",
max_depth=None,
min_samples_split=10,
min_samples_leaf=1,
min_weight_fraction_leaf=0.0,
max_features=1.0,
max_leaf_nodes=None,
min_impurity_decrease=0.0,
bootstrap=True,
oob_score=False,
n_jobs=None,
random_state=None,
verbose=0,
warm_start=False,
ccp_alpha=0.0,
max_samples=None,
min_variance=0.0,
splitter="random"
):
super().__init__(
# !keyword-argument changing from sklearn==1.2.0, positional fixed it!
DecisionTreeRegressor(),
n_estimators=n_estimators,
estimator_params=(
"criterion",
"max_depth",
"min_samples_split",
"min_samples_leaf",
"min_weight_fraction_leaf",
"max_features",
"max_leaf_nodes",
"min_impurity_decrease",
"random_state",
"ccp_alpha",
"splitter",
),
bootstrap=bootstrap,
oob_score=oob_score,
n_jobs=n_jobs,
random_state=random_state,
verbose=verbose,
warm_start=warm_start,
max_samples=max_samples,
)
self.criterion = criterion
self.max_depth = max_depth
self.min_samples_split = min_samples_split
self.min_samples_leaf = min_samples_leaf
self.min_weight_fraction_leaf = min_weight_fraction_leaf
self.max_features = max_features
self.max_leaf_nodes = max_leaf_nodes
self.min_impurity_decrease = min_impurity_decrease
self.ccp_alpha = ccp_alpha
self.min_variance = min_variance
self.splitter = splitter
[docs] def predict(self, X, return_std=False, disentangled_std=False):
"""Predict continuous output for X.
Parameters
----------
X : array of shape = (n_samples, n_features)
Input data.
return_std : boolean
Whether or not to return the standard deviation.
Returns
-------
predictions : array-like of shape = (n_samples,)
Predicted values for X. If criterion is set to "mse",
then `predictions[i] ~= mean(y | X[i])`.
std : array-like of shape=(n_samples,)
Standard deviation of `y` at `X`. If criterion
is set to "mse", then `std[i] ~= std(y | X[i])`.
disentangled_std : the std is returned disentangled between aleatoric and epistemic.
"""
if return_std:
if self.criterion != "squared_error":
raise ValueError(
"Expected impurity to be 'squared_error', got %s instead"
% self.criterion
)
if disentangled_std:
mean, std_al, std_ep = _return_mean_and_std_distentangled(
X, self.n_outputs_, self.estimators_, self.min_variance, self.n_jobs
)
return mean, std_al, std_ep
else:
mean, std = _return_mean_and_std(
X, self.n_outputs_, self.estimators_, self.min_variance, self.n_jobs
)
return mean, std
else:
mean = super(RandomForestRegressor, self).predict(X)
return mean