Add estimator attribute _estimator_type

hypernets/experiment/compete.py

@@ -152,6 +152,7 @@ def __getstate__(self):
  state = super().__getstate__()
  # Don't pickle experiment
  if 'experiment' in state.keys():
+ state = state.copy()
  state['experiment'] = None
  return state
 

hypernets/model/estimator.py

@@ -6,10 +6,11 @@
 
 import numpy as np
 from sklearn.model_selection import KFold, StratifiedKFold
+from hypernets.utils import const
 
 
 class Estimator():
- def __init__(self, space_sample, task='binary', discriminator=None):
+ def __init__(self, space_sample, task=const.TASK_BINARY, discriminator=None):
  self.space_sample = space_sample
  self.task = task
  self.discriminator = discriminator
@@ -19,6 +20,15 @@ def __init__(self, space_sample, task='binary', discriminator=None):
  self.cv_ = None
  self.cv_models_ = None
 
+ @property
+ def _estimator_type(self):
+ if self.task in {const.TASK_BINARY, const.TASK_MULTICLASS, const.TASK_MULTILABEL}:
+ return 'classifier'
+ elif self.task in {const.TASK_REGRESSION, }:
+ return 'regressor'
+ else:
+ return None
+
  def set_discriminator(self, discriminator):
  self.discriminator = discriminator
 

hypernets/model/objectives.py

@@ -174,6 +174,16 @@ def __init__(self, estimators, x_vals, y_vals):
  def classes_(self):
  return self.estimators[0].classes_
 
+ @property
+ def _estimator_type(self):
+ try:
+ if len(self.classes_) > 1:
+ return 'classifier'
+ else:
+ return 'regressor'
+ except:
+ return 'regressor'
+
  def predict(self, X, **kwargs):
  rows = 0
  for x_val in self.x_vals:

hypernets/tests/model/test_objectives.py

@@ -77,6 +77,9 @@ def predict(self, *args, **kwargs):
  def predict_proba(self, *args, **kwargs):
  return self.cv_models_[0].predict_proba(*args, **kwargs)
 
+ @property
+ def _estimator_type(self):
+ return 'classifier'
 
 class TestPredictionPerformanceObjective(BaseTestWithinModel):