Add option to restrict LassoCV to positive-only coefficients

bwignall · agramfort · commit 0b20feb5b894 · 2014-01-11T21:39:24.000+01:00
diff --git a/sklearn/linear_model/coordinate_descent.py b/sklearn/linear_model/coordinate_descent.py
@@ -861,7 +861,7 @@ class LinearModelCV(six.with_metaclass(ABCMeta, LinearModel)):
     @abstractmethod
     def __init__(self, eps=1e-3, n_alphas=100, alphas=None, fit_intercept=True,
                  normalize=False, precompute='auto', max_iter=1000, tol=1e-4,
-                 copy_X=True, cv=None, verbose=False):
+                 copy_X=True, cv=None, verbose=False, positive=False):
         self.eps = eps
         self.n_alphas = n_alphas
         self.alphas = alphas
@@ -873,6 +873,7 @@ def __init__(self, eps=1e-3, n_alphas=100, alphas=None, fit_intercept=True,
         self.copy_X = copy_X
         self.cv = cv
         self.verbose = verbose
+        self.positive = positive
 
     def fit(self, X, y):
         """Fit linear model with coordinate descent
@@ -1051,6 +1052,9 @@ class LassoCV(LinearModelCV, RegressorMixin):
     verbose : bool or integer
         amount of verbosity
 
+    positive : bool, optional
+        If positive, restrict regression coefficients to be positive
+
     Attributes
     ----------
     ``alpha_`` : float
@@ -1101,12 +1105,12 @@ class LassoCV(LinearModelCV, RegressorMixin):
 
     def __init__(self, eps=1e-3, n_alphas=100, alphas=None, fit_intercept=True,
                  normalize=False, precompute='auto', max_iter=1000, tol=1e-4,
-                 copy_X=True, cv=None, verbose=False):
+                 copy_X=True, cv=None, verbose=False, positive=False):
         super(LassoCV, self).__init__(
             eps=eps, n_alphas=n_alphas, alphas=alphas,
             fit_intercept=fit_intercept, normalize=normalize,
             precompute=precompute, max_iter=max_iter, tol=tol, copy_X=copy_X,
-            cv=cv, verbose=verbose)
+            cv=cv, verbose=verbose, positive=positive)
 
 
 class ElasticNetCV(LinearModelCV, RegressorMixin):
