monitors for binary case, experiment works with lists of outputs also

braindecode/experiments/experiment.py

@@ -320,7 +320,11 @@ class Experiment(object):
             target_vars = target_vars.cuda()
         outputs = self.model(input_vars)
         loss = self.loss_function(outputs, target_vars)
-        outputs = outputs.cpu().data.numpy()
+        if hasattr(outputs, 'cpu'):
+            outputs = outputs.cpu().data.numpy()
+        else:
+            # assume it is iterable
+            outputs = [o.cpu().data.numpy() for o in outputs]
         loss = loss.cpu().data.numpy()
         return outputs, loss
 

braindecode/experiments/monitors.py

@@ -10,10 +10,15 @@ class MisclassMonitor(object):
     ----------
     col_suffix: str, optional
         Name of the column in the monitoring output.
+    threshold_for_binary_case: bool, optional
+        In case of binary classification with only one output prediction
+        per target, define the threshold for separating the classes, i.e.
+        0.5 for sigmoid outputs, or np.log(0.5) for log sigmoid outputs
     """
 
-    def __init__(self, col_suffix='misclass'):
+    def __init__(self, col_suffix='misclass', threshold_for_binary_case=None):
         self.col_suffix = col_suffix
+        self.threshold_for_binary_case = threshold_for_binary_case
 
     def monitor_epoch(self, ):
         return
@@ -28,11 +33,19 @@ class MisclassMonitor(object):
             # or just
             # examples x classes
             # make sure not to remove first dimension if it only has size one
-            only_one_row = preds.shape[0] == 1
-            pred_labels = np.argmax(preds, axis=1).squeeze()
-            # add first dimension again if needed
-            if only_one_row:
-                pred_labels = pred_labels[None]
+            if preds.ndim > 1:
+                only_one_row = preds.shape[0] == 1
+
+                pred_labels = np.argmax(preds, axis=1).squeeze()
+                # add first dimension again if needed
+                if only_one_row:
+                    pred_labels = pred_labels[None]
+            else:
+                assert self.threshold_for_binary_case is not None, (
+                    "In case of only one output, please supply the "
+                    "threshold_for_binary_case parameter")
+                # binary classification case... assume logits
+                pred_labels = np.int32(preds > self.threshold_for_binary_case)
             # now examples x time or examples
             all_pred_labels.extend(pred_labels)
             targets = all_targets[i_batch]

braindecode/visualization/plot.py

@@ -115,8 +115,6 @@ def ax_scalp(v, channels,
     image = ax.imshow(zz, vmin=vmin, vmax=vmax, cmap=cmap,
                       extent=[min(x), max(x), min(y), max(y)], origin='lower',
                       interpolation=interpolation)
-    # image = ax.contourf(xx, yy, zz, 100, vmin=vmin, vmax=vmax,
-    #    cmap=colormap)
     if scalp_line_width > 0:
         # paint the head
         ax.add_artist(plt.Circle((0, 0), 1, linestyle=scalp_line_style,