Incompatibility with scikit-learn model selection module

[nguigs]

It is currently not possible to use the scikit-learn GridSearchCV module with estimators from Parsimony because the method get_params is not compatible:

import parsimony.estimators as estimators
from sklearn.model_selection import GridSearchCV

rr = estimators.RidgeRegression(l=1.0)
param = {'ridgeregression__l': [0.2 * k for k in range(6)]}
grid_search = GridSearchCV(estimator=rr, param_grid=param, cv=5)

grid_search.fit(Xtr, ytr)

Traceback (most recent call last):

  File "<ipython-input-51-4f6ebd276276>", line 8, in <module>
    grid_search.fit(Xtr, ytr)

  File "/home/python-environments/env/lib/python2.7/site-packages/sklearn/model_selection/_search.py", line 625, in fit
    base_estimator = clone(self.estimator)

  File "/home/python-environments/env/lib/python2.7/site-packages/sklearn/base.py", line 60, in clone
    new_object_params = estimator.get_params(deep=False)

TypeError: get_params() got an unexpected keyword argument 'deep'

This could be fixed adding the parameter deep and as in scikit-learn:

https://github.com/scikit-learn/scikit-learn/blob/f0ab589f1541b1ca4570177d93fd7979613497e3/sklearn/base.py#L244-L248