def testCatboostNoEarlyStop(self):
estimator_list = [create_catboost(), create_catboost()]
config = self.base_params(estimator_list=estimator_list)
config["early_stopping"] = False
trainable = _Trainable(config)
trainable.train()
assert not any(trainable.saved_models)
trainable.stop()
def testWarmStart(self):
# Hard to get introspection so we just test that it runs.
config = self.base_params([LogisticRegression(), LogisticRegression()])
config["early_stopping"] = True
trainable = _Trainable(config)
trainable.train()
trainable.train()
trainable.stop()
def testNoPartialFit(self):
config = self.base_params([SGDClassifier(), SGDClassifier()])
config["early_stopping"] = False
trainable = _Trainable(config)
trainable.train()
assert not hasattr(trainable.estimator_list[0], "t_")
trainable.train()
assert not hasattr(trainable.estimator_list[0], "t_")
trainable.stop()
def testLGBMNoEarlyStop(self):
config = self.base_params(
estimator_list=[create_lightgbm(),
create_lightgbm()])
config["early_stopping"] = False
trainable = _Trainable(config)
trainable.train()
assert not any(trainable.saved_models)
trainable.stop()
def testPartialFit(self):
config = self.base_params([SGDClassifier(), SGDClassifier()])
config["early_stopping"] = True
trainable = _Trainable(config)
trainable.train()
assert trainable.estimator_list[0].t_ > 0
previous_t = trainable.estimator_list[0].t_
trainable.train()
assert trainable.estimator_list[0].t_ > previous_t
trainable.stop()
def testWarmStart(self):
# Hard to get introspection so we just test that it runs.
estimator_list = [LogisticRegression(), LogisticRegression()]
config = self.base_params(estimator_list)
config["early_stopping"] = True
config["early_stop_type"] = get_early_stop_type(
estimator_list[0], True)
trainable = _Trainable(config)
trainable.train()
trainable.train()
trainable.stop()
def testXGBoostEarlyStop(self):
config = self.base_params(
estimator_list=[create_xgboost(),
create_xgboost()])
config["early_stopping"] = True
trainable = _Trainable(config)
trainable.train()
assert all(trainable.saved_models)
trainable.train()
assert all(trainable.saved_models)
trainable.stop()
def testXGBoostEarlyStop(self):
estimator_list = [create_xgboost(), create_xgboost()]
config = self.base_params(estimator_list=estimator_list)
config["early_stopping"] = True
config["early_stop_type"] = get_early_stop_type(
estimator_list[0], True)
trainable = _Trainable(config)
trainable.train()
assert all(trainable.saved_models)
trainable.train()
assert all(trainable.saved_models)
trainable.stop()
def testLGBMEarlyStop(self):
config = self.base_params(
estimator_list=[create_lightgbm(),
create_lightgbm()])
config["early_stopping"] = True
config["early_stop_type"] = get_early_stop_type(
config["estimator_list"][0], True)
trainable = _Trainable(config)
trainable.train()
assert all(trainable.saved_models)
trainable.train()
assert all(trainable.saved_models)
trainable.stop()
def testPartialFit(self):
estimator_list = [SGDClassifier(), SGDClassifier()]
config = self.base_params(estimator_list)
config["early_stopping"] = True
config["early_stop_type"] = get_early_stop_type(
estimator_list[0], True)
trainable = _Trainable(config)
trainable.train()
assert trainable.estimator_list[0].t_ > 0
previous_t = trainable.estimator_list[0].t_
trainable.train()
assert trainable.estimator_list[0].t_ > previous_t
trainable.stop()
def test_basic_train(self):
config = self.base_params(estimator_list=[SVC(), SVC()])
trainable = _Trainable(config)
trainable.train()
trainable.stop()
