def apply_ridge(self):
if self.use_ridge:
learner = linear.LinearRegressionLearner(
name=self.name, ridgeLambda=self.ridge_lambda)
else:
learner = linear.LinearRegressionLearner(name=self.name)
predictor = None
if self.preprocessor:
learner = self.preprocessor.wrapLearner(learner)
self.error(0)
if self.data is not None:
try:
predictor = learner(self.data)
predictor.name = self.name
except Exception, ex:
self.error(0, "An error during learning: %r" % ex)
def apply(self):
if self.reg_type == OWLinearRegression.OLS:
learner = linear.LinearRegressionLearner()
elif self.reg_type == OWLinearRegression.Ridge:
learner = linear.RidgeRegressionLearner(alpha=self.ridgealpha)
elif self.reg_type == OWLinearRegression.Lasso:
learner = linear.RidgeRegressionLearner(alpha=self.lassoalpha)
else:
assert False
learner.name = self.learner_name
predictor = None
if self.data is not None:
predictor = learner(self.data)
predictor.name = self.learner_name
self.send("Learner", learner)
self.send("Predictor", predictor)
import Orange
from Orange.regression import linear
from Orange.data import Table
from Orange.evaluation import scoring, testing
#data operations
data = Table("auto-mpg.tab")
#set up linear regressions
##regularizations
Linear = linear.LinearRegressionLearner(preprocessors=None)
Linear.name = "No regularization"
Ridge = linear.RidgeRegressionLearner(alpha=0.0001,
fit_intercept=True,
normalize=False,
copy_X=True,
max_iter=None,
tol=0.001,
solver='auto',
preprocessors=None)
Ridge.name = "Ridge Regression(L2)"
Lasso = linear.LassoRegressionLearner(alpha=1.0,
fit_intercept=True,
normalize=False,
precompute=False,
copy_X=True,
max_iter=1000,
tol=0.0001,
warm_start=False,
positive=False,
def setUp(self):
self.learner = linear.LinearRegressionLearner(ridge_lambda=2)
def setUp(self):
self.learner = linear.LinearRegressionLearner()