FIX : omp default param could fail

Author: agramfort

sklearn/decomposition/dict_learning.py

@@ -228,7 +228,7 @@ def sparse_encode(X, dictionary, gram=None, cov=None, algorithm='lasso_lars',
     if algorithm in ('lars', 'omp'):
         regularization = n_nonzero_coefs
         if regularization is None:
-            regularization = max(n_features / 10, 1)
+            regularization = min(max(n_features / 10, 1), n_components)
     else:
         regularization = alpha
         if regularization is None:

sklearn/decomposition/tests/test_dict_learning.py

@@ -176,6 +176,13 @@ def test_sparse_encode_error():
     assert_less(np.sqrt(np.sum((np.dot(code, V) - X) ** 2)), 0.1)
 
 
+def test_sparse_encode_error_default_sparsity():
+    X = np.random.randn(100, 64)
+    D = np.random.randn(2, 64)
+    code = sparse_encode(X, D, algorithm='omp', n_nonzero_coefs=None)
+    assert_equal(code.shape, (100, 2))
+
+
 def test_unknown_method():
     n_components = 12
     rng = np.random.RandomState(0)