109 lines
3.4 KiB
Python
109 lines
3.4 KiB
Python
"""
|
|
Base class for ensemble-based estimators.
|
|
"""
|
|
|
|
# Authors: Gilles Louppe
|
|
# License: BSD 3 clause
|
|
|
|
import numpy as np
|
|
|
|
from ..base import clone
|
|
from ..base import BaseEstimator
|
|
from ..base import MetaEstimatorMixin
|
|
from ..externals.joblib import cpu_count
|
|
|
|
|
|
class BaseEnsemble(BaseEstimator, MetaEstimatorMixin):
|
|
"""Base class for all ensemble classes.
|
|
|
|
Warning: This class should not be used directly. Use derived classes
|
|
instead.
|
|
|
|
Parameters
|
|
----------
|
|
base_estimator : object, optional (default=None)
|
|
The base estimator from which the ensemble is built.
|
|
|
|
n_estimators : integer
|
|
The number of estimators in the ensemble.
|
|
|
|
estimator_params : list of strings
|
|
The list of attributes to use as parameters when instantiating a
|
|
new base estimator. If none are given, default parameters are used.
|
|
|
|
Attributes
|
|
----------
|
|
`base_estimator_`: list of estimators
|
|
The base estimator from which the ensemble is grown.
|
|
|
|
`estimators_`: list of estimators
|
|
The collection of fitted base estimators.
|
|
"""
|
|
|
|
def __init__(self, base_estimator, n_estimators=10,
|
|
estimator_params=tuple()):
|
|
# Set parameters
|
|
self.base_estimator = base_estimator
|
|
self.n_estimators = n_estimators
|
|
self.estimator_params = estimator_params
|
|
|
|
# Don't instantiate estimators now! Parameters of base_estimator might
|
|
# still change. Eg., when grid-searching with the nested object syntax.
|
|
# This needs to be filled by the derived classes.
|
|
self.estimators_ = []
|
|
|
|
def _validate_estimator(self, default=None):
|
|
"""Check the estimator and set the `base_estimator_` attribute."""
|
|
if self.base_estimator is not None:
|
|
self.base_estimator_ = self.base_estimator
|
|
else:
|
|
self.base_estimator_ = default
|
|
|
|
if self.base_estimator_ is None:
|
|
raise ValueError("base_estimator cannot be None")
|
|
|
|
def _make_estimator(self, append=True):
|
|
"""Make and configure a copy of the `base_estimator_` attribute.
|
|
|
|
Warning: This method should be used to properly instantiate new
|
|
sub-estimators.
|
|
"""
|
|
estimator = clone(self.base_estimator_)
|
|
estimator.set_params(**dict((p, getattr(self, p))
|
|
for p in self.estimator_params))
|
|
|
|
if append:
|
|
self.estimators_.append(estimator)
|
|
|
|
return estimator
|
|
|
|
def __len__(self):
|
|
"""Returns the number of estimators in the ensemble."""
|
|
return len(self.estimators_)
|
|
|
|
def __getitem__(self, index):
|
|
"""Returns the index'th estimator in the ensemble."""
|
|
return self.estimators_[index]
|
|
|
|
def __iter__(self):
|
|
"""Returns iterator over estimators in the ensemble."""
|
|
return iter(self.estimators_)
|
|
|
|
|
|
def _partition_estimators(ensemble):
|
|
"""Private function used to partition estimators between jobs."""
|
|
# Compute the number of jobs
|
|
if ensemble.n_jobs == -1:
|
|
n_jobs = min(cpu_count(), ensemble.n_estimators)
|
|
|
|
else:
|
|
n_jobs = min(ensemble.n_jobs, ensemble.n_estimators)
|
|
|
|
# Partition estimators between jobs
|
|
n_estimators = (ensemble.n_estimators // n_jobs) * np.ones(n_jobs,
|
|
dtype=np.int)
|
|
n_estimators[:ensemble.n_estimators % n_jobs] += 1
|
|
starts = np.cumsum(n_estimators)
|
|
|
|
return n_jobs, n_estimators.tolist(), [0] + starts.tolist()
|