Merge remote-tracking branch 'origin/main'

jameschapman19 · Nov 13, 2023 · ec348d7 · ec348d7
2 parents 1894e3f + c588ecb
commit ec348d7
Show file tree

Hide file tree

Showing 3 changed files with 21 additions and 21 deletions.
diff --git a/cca_zoo/deep/objectives.py b/cca_zoo/deep/objectives.py
@@ -267,11 +267,11 @@ def loss(self, representations, independent_representations=None):
         }
 
     def derivative(
-            self,
-            views,
-            representations,
-            independent_views=None,
-            independent_representations=None,
+        self,
+        views,
+        representations,
+        independent_views=None,
+        independent_representations=None,
     ):
         A, B = self.get_AB(representations)
         sum_representations = torch.sum(torch.stack(representations), dim=0)
@@ -318,11 +318,11 @@ def loss(self, representations, independent_representations=None):
         }
 
     def derivative(
-            self,
-            views,
-            representations,
-            independent_views=None,
-            independent_representations=None,
+        self,
+        views,
+        representations,
+        independent_views=None,
+        independent_representations=None,
     ):
         A, B = self.get_AB(representations)
         sum_representations = torch.sum(torch.stack(representations), dim=0)
@@ -367,11 +367,11 @@ def loss(self, representations, independent_representations=None):
         }
 
     def derivative(
-            self,
-            views,
-            representations,
-            independent_views=None,
-            independent_representations=None,
+        self,
+        views,
+        representations,
+        independent_views=None,
+        independent_representations=None,
     ):
         C = torch.cov(torch.hstack(representations).T)
         latent_dims = representations[0].shape[1]
@@ -421,10 +421,12 @@ class _PLS_PowerLoss(_PLSAB):
     def loss(self, representations):
         cov = torch.cov(torch.hstack(representations).T)
         return {
-            "objective": torch.trace(cov[: representations[0].shape[1], representations[0].shape[1]:])
+            "objective": torch.trace(
+                cov[: representations[0].shape[1], representations[0].shape[1] :]
+            )
         }
 
     @staticmethod
-    def derivative( views, representations):
+    def derivative(views, representations):
         grads = [views[0].T @ representations[1], views[1].T @ representations[0]]
         return grads
diff --git a/cca_zoo/linear/_gradient/_stochasticpls.py b/cca_zoo/linear/_gradient/_stochasticpls.py
@@ -25,9 +25,7 @@ def training_step(self, batch, batch_idx):
                 on_epoch=True,
                 batch_size=batch["views"][0].shape[0],
             )
-        manual_grads = self.objective.derivative(
-            batch["views"], representations
-        )
+        manual_grads = self.objective.derivative(batch["views"], representations)
         for i, weights in enumerate(self.torch_weights):
             weights.grad = manual_grads[i]
         opt.step()

diff --git a/tests/test_gradient.py b/tests/test_gradient.py
@@ -52,7 +52,7 @@ def test_batch_pls():
         random_state=random_state,
         trainer_kwargs=trainer_kwargs,
     ).fit((X, Y))
-    spls=PLSStochasticPower(
+    spls = PLSStochasticPower(
         latent_dimensions=latent_dims,
         epochs=epochs,
         random_state=random_state,