braindecode 0.4.8__tar.gz → 0.4.82__tar.gz

{Braindecode-0.4.8 → Braindecode-0.4.82}/Braindecode.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 1.1
 Name: Braindecode
-Version: 0.4.8
+Version: 0.4.82
 Summary: A deep learning toolbox to decode raw time-domain EEG.
 Home-page: https://github.com/robintibor/braindecode
 Author: Robin Tibor Schirrmeister

{Braindecode-0.4.8 → Braindecode-0.4.82}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 1.1
 Name: Braindecode
-Version: 0.4.8
+Version: 0.4.82
 Summary: A deep learning toolbox to decode raw time-domain EEG.
 Home-page: https://github.com/robintibor/braindecode
 Author: Robin Tibor Schirrmeister

{Braindecode-0.4.8 → Braindecode-0.4.82}/braindecode/models/base.py

@@ -69,7 +69,12 @@ class BaseModel(object):
             self.compiled = False
 
     def compile(
-        self, loss, optimizer, extra_monitors=None, cropped=False, iterator_seed=0
+        self,
+        loss,
+        optimizer,
+        extra_monitors=None,
+        cropped=False,
+        iterator_seed=0,
     ):
         """
         Setup training for this model.
@@ -179,7 +184,8 @@ class BaseModel(object):
             self.network.eval()
             test_input = np_to_var(
                 np.ones(
-                    (1, train_X.shape[1], input_time_length) + train_X.shape[3:],
+                    (1, train_X[0].shape[0], input_time_length)
+                    + train_X[0].shape[2:],
                     dtype=np.float32,
                 )
             )
@@ -207,7 +213,9 @@ class BaseModel(object):
         train_set = SignalAndTarget(train_X, train_y)
         optimizer = self.optimizer
         if scheduler is not None:
-            assert scheduler == "cosine", "Supply either 'cosine' or None as scheduler."
+            assert (
+                scheduler == "cosine"
+            ), "Supply either 'cosine' or None as scheduler."
             n_updates_per_epoch = sum(
                 [1 for _ in self.iterator.get_batches(train_set, shuffle=True)]
             )
@@ -218,7 +226,9 @@ class BaseModel(object):
             if optimizer.__class__.__name__ == "AdamW":
                 schedule_weight_decay = True
             optimizer = ScheduledOptimizer(
-                scheduler, self.optimizer, schedule_weight_decay=schedule_weight_decay
+                scheduler,
+                self.optimizer,
+                schedule_weight_decay=schedule_weight_decay,
             )
         loss_function = self.loss
         if self.cropped:
@@ -351,7 +361,9 @@ class BaseModel(object):
             pred_labels = np.array(pred_labels)
         return pred_labels
 
-    def predict_outs(self, X, threshold_for_binary_case=None, individual_crops=False):
+    def predict_outs(
+        self, X, threshold_for_binary_case=None, individual_crops=False
+    ):
         """
         Predict raw outputs of the network for given input.
 
@@ -373,7 +385,10 @@ class BaseModel(object):
             assert self.cropped, "Cropped labels only for cropped decoding"
         all_preds = []
         with th.no_grad():
-            for b_X, _ in self.iterator.get_batches(SignalAndTarget(X, X), False):
+            dummy_y = np.ones(len(X), dtype=np.int64)
+            for b_X, _ in self.iterator.get_batches(
+                SignalAndTarget(X, dummy_y), False
+            ):
                 b_X_var = np_to_var(b_X)
                 if self.cuda:
                     b_X_var = b_X_var.cuda()
@@ -383,7 +398,9 @@ class BaseModel(object):
                 all_preds, self.iterator.input_time_length, X
             )
             if not individual_crops:
-                outs_per_trial = np.array([np.mean(o, axis=1) for o in outs_per_trial])
+                outs_per_trial = np.array(
+                    [np.mean(o, axis=1) for o in outs_per_trial]
+                )
         else:
             outs_per_trial = np.concatenate(all_preds)
         return outs_per_trial

{Braindecode-0.4.8 → Braindecode-0.4.82}/braindecode/models/deep4.py

@@ -75,7 +75,10 @@ class Deep4Net(BaseModel):
             model.add_module(
                 "conv_time",
                 nn.Conv2d(
-                    1, self.n_filters_time, (self.filter_time_length, 1), stride=1
+                    1,
+                    self.n_filters_time,
+                    (self.filter_time_length, 1),
+                    stride=1,
                 ),
             )
             model.add_module(
@@ -139,7 +142,10 @@ class Deep4Net(BaseModel):
                 model.add_module(
                     "bnorm" + suffix,
                     nn.BatchNorm2d(
-                        n_filters, momentum=self.batch_norm_alpha, affine=True, eps=1e-5
+                        n_filters,
+                        momentum=self.batch_norm_alpha,
+                        affine=True,
+                        eps=1e-5,
                     ),
                 )
             model.add_module("nonlin" + suffix, Expression(self.later_nonlin))
@@ -147,10 +153,13 @@ class Deep4Net(BaseModel):
             model.add_module(
                 "pool" + suffix,
                 later_pool_class(
-                    kernel_size=(self.pool_time_length, 1), stride=(pool_stride, 1)
+                    kernel_size=(self.pool_time_length, 1),
+                    stride=(pool_stride, 1),
                 ),
             )
-            model.add_module("pool_nonlin" + suffix, Expression(self.later_pool_nonlin))
+            model.add_module(
+                "pool_nonlin" + suffix, Expression(self.later_pool_nonlin)
+            )
 
         add_conv_pool_block(
             model, n_filters_conv, self.n_filters_2, self.filter_length_2, 2
@@ -168,7 +177,8 @@ class Deep4Net(BaseModel):
             out = model(
                 np_to_var(
                     np.ones(
-                        (1, self.in_chans, self.input_time_length, 1), dtype=np.float32
+                        (1, self.in_chans, self.input_time_length, 1),
+                        dtype=np.float32,
                     )
                 )
             )
@@ -177,7 +187,10 @@ class Deep4Net(BaseModel):
         model.add_module(
             "conv_classifier",
             nn.Conv2d(
-                self.n_filters_4, self.n_classes, (self.final_conv_length, 1), bias=True
+                self.n_filters_4,
+                self.n_classes,
+                (self.final_conv_length, 1),
+                bias=True,
             ),
         )
         model.add_module("softmax", nn.LogSoftmax(dim=1))

Braindecode-0.4.82/braindecode/version.py

@@ -0,0 +1 @@
+__version__ = "0.4.82"

Braindecode-0.4.8/braindecode/version.py

@@ -1 +0,0 @@
-__version__ = "0.4.8"