Another attempt to fix circle.

circle.yml

@@ -37,9 +37,8 @@ dependencies:
       else
         echo "Conda already set up.";
       fi
-    - conda install -n circleenv --yes numpy scipy scikit-learn matplotlib sphinx=1.4.1 pillow six IPython pandas;
-    - conda install -n circleenv --yes traits=4.6.0 traitsui=5.1.0;
-    - conda install -n circleenv --yes mayavi;
+    - conda install -n circleenv --yes numpy scipy scikit-learn matplotlib sphinx=1.4.1 pillow six IPython pandas apptools=4.4.0 envisage=4.5.1 pyface=5.1.0 traits=4.6.0 traitsui=5.1.0 vtk=5.10.1;
+    - pip install mayavi==4.4.4;
     - ls -al /home/ubuntu/miniconda;
     - ls -al /home/ubuntu/miniconda/bin;
     - echo $PATH;

examples/connectivity/plot_mne_inverse_coherence_epochs.py

@@ -33,20 +33,20 @@
 event_id, tmin, tmax = 1, -0.2, 0.5
 method = "dSPM"  # use dSPM method (could also be MNE or sLORETA)
 
-# Load data
+# Load data.
 inverse_operator = read_inverse_operator(fname_inv)
 label_lh = mne.read_label(fname_label_lh)
 raw = mne.io.read_raw_fif(fname_raw)
 events = mne.read_events(fname_event)
 
-# Add a bad channel
+# Add a bad channel.
 raw.info['bads'] += ['MEG 2443']
 
-# pick MEG channels
+# pick MEG channels.
 picks = mne.pick_types(raw.info, meg=True, eeg=False, stim=False, eog=True,
                        exclude='bads')
 
-# Read epochs
+# Read epochs.
 epochs = mne.Epochs(raw, events, event_id, tmin, tmax, picks=picks,
                     baseline=(None, 0),
                     reject=dict(mag=4e-12, grad=4000e-13, eog=150e-6))

mne/viz/misc.py

@@ -110,6 +110,8 @@ def plot_cov(cov, info, exclude=[], colorbar=True, proj=False, show_svd=True,
 
     fig_cov, axes = plt.subplots(1, len(idx_names),
                                  figsize=(2.5 * len(idx_names), 2.7))
+    if isinstance(axes, plt.Axes):
+        axes = [axes]
     for k, (idx, name, _, _) in enumerate(idx_names):
         axes[k].imshow(C[idx][:, idx], interpolation="nearest", cmap='RdBu_r')
         axes[k].set(title=name)
@@ -119,6 +121,8 @@ def plot_cov(cov, info, exclude=[], colorbar=True, proj=False, show_svd=True,
     fig_svd = None
     if show_svd:
         fig_svd, axes = plt.subplots(1, len(idx_names))
+        if isinstance(axes, plt.Axes):
+            axes = [axes]
         for k, (idx, name, unit, scaling) in enumerate(idx_names):
             s = linalg.svd(C[idx][:, idx], compute_uv=False)
             # Protect against true zero singular values