print_model_weights fails for multi-class logistic regression models

[desilinguist]

Traceback (most recent call last):
  File "/Users/nmadnani/anaconda/envs/rsmtool/bin/print_model_weights", line 6, in <module>
    sys.exit(main())
  File "/Users/nmadnani/anaconda/envs/rsmtool/lib/python3.4/site-packages/skll/utilities/print_model_weights.py", line 56, in main
    (weights, intercept) = learner.model_params
  File "/Users/nmadnani/anaconda/envs/rsmtool/lib/python3.4/site-packages/skll/learner.py", line 712, in model_params
    if self.model.intercept_:
ValueError: The truth value of an array with more than one element is ambiguous. Use a.any() or a.all()

It fails because self.model.intercept_ is an array and, as the error says, you can't treat an array as a bool. We don't have a test for this case - we check binary LogisticRegression, 2-class LinearRegression, and multi-class LinearSVR but not multi-class LogisticRegression. It's still surprising that no one else has run into this error though.

[desilinguist]

this will be addressed by #267

[desilinguist]

Addressed by #267.