def apply(self):
common_args = dict()
common_args["n_estimators"] = self.n_estimators
if self.use_max_features:
common_args["max_features"] = self.max_features
if self.use_random_state:
common_args["random_state"] = self.random_state
if self.use_max_depth:
common_args["max_depth"] = self.max_depth
if self.use_max_leaf_nodes:
common_args["max_leaf_nodes"] = self.max_leaf_nodes
learner = RandomForestLearner(
preprocessors=self.preprocessors,**common_args
)
learner.name = self.learner_name
classifier = None
if self.data is not None:
self.error(0)
if not learner.check_learner_adequacy(self.data.domain):
self.error(0, learner.learner_adequacy_err_msg)
else:
classifier = learner(self.data)
classifier.name = self.learner_name
self.send("Learner", learner)
self.send("Classifier", classifier)
def test_forest_tree_table(self):
titanic_data = Table('titanic')[::50]
titanic = RandomForestLearner(n_estimators=3)(titanic_data)
titanic.instances = titanic_data
housing_data = Table('housing')[:10]
housing = RandomForestRegressionLearner(n_estimators=3)(housing_data)
housing.instances = housing_data
forest_w = self.create_widget(OWPythagoreanForest)
for data in (housing, titanic):
self.send_signal(forest_w.Inputs.random_forest,
data,
widget=forest_w)
tree = forest_w.forest_model[0].model
tree_w = self.widget
self.send_signal(tree_w.Inputs.tree, tree, widget=tree_w)
square = [
i for i in tree_w.scene.items()
if isinstance(i, SquareGraphicsItem)
][-1]
square.setSelected(True)
tab = self.get_output(tree_w.Outputs.selected_data, widget=tree_w)
self.assertGreater(len(tab), 0)
def apply(self):
common_args = dict()
common_args["n_estimators"] = self.n_estimators
if self.use_max_features:
common_args["max_features"] = self.max_features
if self.use_random_state:
common_args["random_state"] = self.random_state
if self.use_max_depth:
common_args["max_depth"] = self.max_depth
if self.use_max_leaf_nodes:
common_args["max_leaf_nodes"] = self.max_leaf_nodes
learner = RandomForestLearner(
preprocessors=self.preprocessors,**common_args
)
learner.name = self.learner_name
classifier = None
if self.data is not None:
self.error(0)
if not learner.check_learner_adequacy(self.data.domain):
self.error(0, learner.learner_adequacy_err_msg)
else:
classifier = learner(self.data)
classifier.name = self.learner_name
self.send("Learner", learner)
self.send("Classifier", classifier)
def test_changing_data_restores_depth_from_previous_settings(self):
titanic_data = Table("titanic")[::50]
forest = RandomForestLearner(n_estimators=3)(titanic_data)
forest.instances = titanic_data
self.send_signal(self.widget.Inputs.tree, forest.trees[0])
self.widget.controls.depth_limit.setValue(1)
# The domain is still the same, so restore the depth limit from before
self.send_signal(self.widget.Inputs.tree, forest.trees[1])
self.assertEqual(self.widget.ptree._depth_limit, 1)
def test_context(self):
iris = Table("iris")
iris_tree = RandomForestLearner()(iris)
iris_tree.instances = iris
self.send_signal(self.widget.Inputs.random_forest, self.titanic)
self.widget.target_class_index = 1
self.send_signal(self.widget.Inputs.random_forest, iris_tree)
self.assertEqual(0, self.widget.target_class_index)
self.widget.target_class_index = 2
self.send_signal(self.widget.Inputs.random_forest, self.titanic)
self.assertEqual(1, self.widget.target_class_index)
self.send_signal(self.widget.Inputs.random_forest, iris_tree)
self.assertEqual(2, self.widget.target_class_index)
def test_forest_tree_table(self):
titanic_data = Table('titanic')[::50]
titanic = RandomForestLearner(n_estimators=3)(titanic_data)
titanic.instances = titanic_data
housing_data = Table('housing')[:10]
housing = RandomForestRegressionLearner(n_estimators=3)(housing_data)
housing.instances = housing_data
forest_w = self.create_widget(OWPythagoreanForest)
for data in (housing, titanic):
self.send_signal(forest_w.Inputs.random_forest, data, widget=forest_w)
tree = forest_w.forest_model[0].model
tree_w = self.widget
self.send_signal(tree_w.Inputs.tree, tree, widget=tree_w)
square = [i for i in tree_w.scene.items() if isinstance(i, SquareGraphicsItem)][-1]
square.setSelected(True)
tab = self.get_output(tree_w.Outputs.selected_data, widget=tree_w)
self.assertGreater(len(tab), 0)
def setUpClass(cls):
super().setUpClass()
# Set up for widget tests
titanic_data = Table('titanic')[::50]
cls.titanic = RandomForestLearner(n_estimators=3)(titanic_data)
cls.titanic.instances = titanic_data
housing_data = Table('housing')[:10]
cls.housing = RandomForestRegressionLearner(
n_estimators=3)(housing_data)
cls.housing.instances = housing_data
return self._adapters
@property
def domain(self):
"""Get the domain."""
return self._domain
if __name__ == '__main__':
from AnyQt.QtWidgets import QApplication
import sys
app = QApplication(sys.argv)
ow = OWPythagoreanForest()
data = Table(sys.argv[1] if len(sys.argv) > 1 else 'iris')
if data.domain.has_discrete_class:
from Orange.classification.random_forest import \
RandomForestLearner as RFLearner
else:
from Orange.regression.random_forest import \
RandomForestRegressionLearner as RFLearner
rf = RFLearner(n_estimators=10, max_depth=1000)(data)
rf.instances = data
ow.set_rf(rf)
ow.show()
ow.raise_()
ow.handleNewSignals()
app.exec_()
return self._adapters
@property
def domain(self):
"""Get the domain."""
return self._domain
if __name__ == '__main__':
from AnyQt.QtWidgets import QApplication
import sys
app = QApplication(sys.argv)
ow = OWPythagoreanForest()
data = Table(sys.argv[1] if len(sys.argv) > 1 else 'iris')
if data.domain.has_discrete_class:
from Orange.classification.random_forest import \
RandomForestLearner as RFLearner
else:
from Orange.regression.random_forest import \
RandomForestRegressionLearner as RFLearner
rf = RFLearner(n_estimators=10, max_depth=1000)(data)
rf.instances = data
ow.set_rf(rf)
ow.show()
ow.raise_()
ow.handleNewSignals()
app.exec_()
]
return self._adapters
@property
def domain(self):
"""Get the domain."""
return self._domain
if __name__ == '__main__':
from AnyQt.QtWidgets import QApplication
import sys
app = QApplication(sys.argv)
data = Table(sys.argv[1] if len(sys.argv) > 1 else 'iris')
if data.domain.has_discrete_class:
from Orange.classification.random_forest import RandomForestLearner
else:
from Orange.regression.random_forest import \
RandomForestRegressionLearner as RandomForestLearner
rf = RandomForestLearner()(data)
rf.instances = data
ow = OWPythagoreanForest()
ow.set_rf(rf)
ow.show()
ow.handleNewSignals()
app.exec_()