Error with the Autonet

[aaronkl]

If I let autosklearn (from autonet branch) run I get the following error message. I use SMAC = 0.3.0. with shared_mode = True.

  File "/ihome/kleinaa/devel/git/autonet_experiments/experiments/run_comparison.py", line 165, in <module>
    automl.fit(X_train, y_train, metric='bac_metric', dataset_name=dataset_name)
  File "/home/kleinaa/virtualenv/local_gpu/lib/python3.5/site-packages/auto_sklearn-0.1.3-py3.5-linux-x86_64.egg/autosklearn/estimators.py", line 358, in fit
    dataset_name=dataset_name)
  File "/home/kleinaa/virtualenv/local_gpu/lib/python3.5/site-packages/auto_sklearn-0.1.3-py3.5-linux-x86_64.egg/autosklearn/estimators.py", line 301, in fit
    super(AutoSklearnEstimator, self).fit(*args, **kwargs)
  File "/home/kleinaa/virtualenv/local_gpu/lib/python3.5/site-packages/auto_sklearn-0.1.3-py3.5-linux-x86_64.egg/autosklearn/estimators.py", line 21, in fit
    self._automl.fit(*args, **kwargs)
  File "/home/kleinaa/virtualenv/local_gpu/lib/python3.5/site-packages/auto_sklearn-0.1.3-py3.5-linux-x86_64.egg/autosklearn/estimators.py", line 486, in fit
    return self._automl.fit(X, y, task, metric, feat_type, dataset_name)
  File "/home/kleinaa/virtualenv/local_gpu/lib/python3.5/site-packages/auto_sklearn-0.1.3-py3.5-linux-x86_64.egg/autosklearn/automl.py", line 172, in fit
    return self._fit(loaded_data_manager)
  File "/home/kleinaa/virtualenv/local_gpu/lib/python3.5/site-packages/auto_sklearn-0.1.3-py3.5-linux-x86_64.egg/autosklearn/automl.py", line 421, in _fit
    for entry in self.trajectory_]
  File "/home/kleinaa/virtualenv/local_gpu/lib/python3.5/site-packages/auto_sklearn-0.1.3-py3.5-linux-x86_64.egg/autosklearn/automl.py", line 421, in <listcomp>
    for entry in self.trajectory_]
TypeError: can only concatenate tuple (not "list") to tuple