various fix

pyriemann/stats.py

@@ -39,7 +39,7 @@ def unique_permutations(elements):
 class BasePermutation():
     """Base object for permutations test"""
 
-    def test(self, X, y, verbose=True):
+    def test(self, X, y, groups=None, verbose=True):
         """Performs the permutation test
 
         Parameters
@@ -48,7 +48,7 @@ def test(self, X, y, verbose=True):
             The data to fit. Can be, for example a list, or an array at
             least 2d.
 
-        y : array-like, optional, default: None
+        y : array-like
             The target variable to try to predict in the case of
             supervised learning.
 
@@ -64,15 +64,15 @@ def test(self, X, y, verbose=True):
         X = self._initial_transform(X)
 
         # get the non permuted score
-        self.scores_[0] = self.score(X, y)
+        self.scores_[0] = self.score(X, y, groups=groups)
 
         if Npe <= self.n_perms:
             print("Warning, number of unique permutations : %d" % Npe)
             perms = unique_permutations(y)
             ii = 0
             for perm in perms:
                 if not numpy.array_equal(perm, y):
-                    self.scores_[ii + 1] = self.score(X, perm)
+                    self.scores_[ii + 1] = self.score(X, perm, groups=groups)
                     ii += 1
                     if verbose:
                         self._print_progress(ii)
@@ -81,7 +81,7 @@ def test(self, X, y, verbose=True):
             rs = numpy.random.RandomState(self.random_state)
             for ii in range(self.n_perms - 1):
                 perm = rs.permutation(y)
-                self.scores_[ii + 1] = self.score(X, perm)
+                self.scores_[ii + 1] = self.score(X, perm, groups=groups)
                 if verbose:
                     self._print_progress(ii)
         if verbose:
@@ -127,7 +127,9 @@ def plot(self, nbins=10, range=None, axes=None):
         y_max = axes.get_ylim()[1]
         axes.plot([x_val, x_val], [0, y_max], '--r', lw=2)
         x_max = axes.get_xlim()[1]
-        axes.text(x_max * 0.5, y_max * 0.8, 'p-value: %.3f' % self.p_value_)
+        x_min = axes.get_xlim()[0]
+        x_pos = x_min + ((x_max - x_min) * 0.25)
+        axes.text(x_pos, y_max * 0.8, 'p-value: %.3f' % self.p_value_)
         axes.set_xlabel('Score')
         axes.set_ylabel('Count')
         return axes
@@ -197,7 +199,7 @@ def __init__(self, n_perms=100, model=MDM(), cv=3, scoring=None,
         self.n_jobs = n_jobs
         self.random_state = random_state
 
-    def score(self, X, y):
+    def score(self, X, y, groups=None):
         """Score one permutation.
 
         Parameters
@@ -206,12 +208,13 @@ def score(self, X, y):
             The data to fit. Can be, for example a list, or an array at
             least 2d.
 
-        y : array-like, optional, default: None
+        y : array-like
             The target variable to try to predict in the case of
             supervised learning.
         """
         score = cross_val_score(self.model, X, y, cv=self.cv,
-                                n_jobs=self.n_jobs, scoring=self.scoring)
+                                n_jobs=self.n_jobs, scoring=self.scoring,
+                                groups=groups)
         return score.mean()
 
 
@@ -300,7 +303,7 @@ def __init__(self, n_perms=100, metric='riemann', mode='pairwise',
         self.random_state = random_state
         self.estimator = estimator
 
-    def score(self, X, y):
+    def score(self, X, y, groups=None):
         """Score of a permutation.
 
         Parameters
@@ -309,7 +312,7 @@ def score(self, X, y):
             The data to fit. Can be, for example a list, or an array at
             least 2d.
 
-        y : array-like, optional, default: None
+        y : array-like
             The target variable to try to predict in the case of
             supervised learning.
         """

pyriemann/utils/viz.py

@@ -23,7 +23,8 @@ def plot_confusion_matrix(targets, predictions, target_names,
     return g
 
 
-def plot_embedding(X, y=None, metric='riemann'):
+def plot_embedding(X, y=None, metric='riemann',
+                   title='Spectral embedding of covariances'):
     """Plot 2D embedding of covariance matrices using Diffusion maps."""
     lapl = Embedding(n_components=2, metric=metric)
     embd = lapl.fit_transform(X)
@@ -38,9 +39,10 @@ def plot_embedding(X, y=None, metric='riemann'):
 
     ax.set_xlabel(r'$\varphi_1$', fontsize=16)
     ax.set_ylabel(r'$\varphi_2$', fontsize=16)
-    ax.set_title('Spectral embedding of ERP recordings', fontsize=16)
+    ax.set_title(title, fontsize=16)
     ax.grid(False)
     ax.set_xticks([-1.0, -0.5, 0.0, +0.5, 1.0])
     ax.set_yticks([-1.0, -0.5, 0.0, +0.5, 1.0])
+    ax.legend(list(np.unique(y)))
 
     return fig

requirements.txt

@@ -1,6 +1,6 @@
 numpy
 scipy
-scikit-learn
+scikit-learn>=0.18
 pandas
 joblib
 seaborn