ran ruff formatter again

Author: MechaCritter

examples/m_ap_and_top_k_accuracy.ipynb

@@ -80,7 +80,14 @@
     }
    },
    "outputs": [],
-   "source": "train_imgs, train_labels = zip(\n    *[(img, label) for img, label, _ in train_dataset], strict=True\n)\nval_imgs, val_labels = zip(*[(img, label) for img, label, _ in val_dataset], strict=True)"
+   "source": [
+    "train_imgs, train_labels = zip(\n",
+    "    *[(img, label) for img, label, _ in train_dataset], strict=True\n",
+    ")\n",
+    "val_imgs, val_labels = zip(\n",
+    "    *[(img, label) for img, label, _ in val_dataset], strict=True\n",
+    ")"
+   ]
   },
   {
    "cell_type": "markdown",
@@ -176,7 +183,15 @@
     }
    },
    "outputs": [],
-   "source": "train_paths, train_labels = zip(\n    *[(path, label) for _, label, path in train_dataset], strict=True\n)\nencodings_vlad = vlad_encoder.generate_encoding_map(train_paths)\nencodings_fisher = fisher_vector_encoder.generate_encoding_map(train_paths)\nencodings_pipeline = pipeline_with_pca.generate_encoding_map(train_paths)\ndataset_labels_dict = dict(zip(train_paths, train_labels, strict=True))"
+   "source": [
+    "train_paths, train_labels = zip(\n",
+    "    *[(path, label) for _, label, path in train_dataset], strict=True\n",
+    ")\n",
+    "encodings_vlad = vlad_encoder.generate_encoding_map(train_paths)\n",
+    "encodings_fisher = fisher_vector_encoder.generate_encoding_map(train_paths)\n",
+    "encodings_pipeline = pipeline_with_pca.generate_encoding_map(train_paths)\n",
+    "dataset_labels_dict = dict(zip(train_paths, train_labels, strict=True))"
+   ]
   },
   {
    "cell_type": "markdown",
@@ -570,4 +585,4 @@
  },
  "nbformat": 4,
  "nbformat_minor": 5
-}
+}

pyvisim/_utils.py

@@ -29,7 +29,9 @@ def cosine_similarity(x: np.ndarray, y: np.ndarray) -> np.ndarray:
     x = x.reshape(1, -1) if len(x.shape) == 1 else x
     y = y.reshape(1, -1) if len(y.shape) == 1 else y
     if x.shape[-1] <= 1 or y.shape[-1] <= 1:
-        raise ValueError(f"Cosine similarity requires at least 2 features. Got {x.shape[-1]} features for x and {y.shape[-1]} features for y.")
+        raise ValueError(
+            f"Cosine similarity requires at least 2 features. Got {x.shape[-1]} features for x and {y.shape[-1]} features for y."
+        )
 
     return cs(x, y)
 
@@ -83,7 +85,10 @@ def cluster_and_return_labels(
         if n_clusters is None:
             raise ValueError("n_clusters must be specified for Spectral Clustering.")
         model = SpectralClustering(
-            n_clusters=n_clusters, affinity="nearest_neighbors", random_state=42, **kwargs
+            n_clusters=n_clusters,
+            affinity="nearest_neighbors",
+            random_state=42,
+            **kwargs,
         )
         return model.fit_predict(data)
 
@@ -153,9 +158,13 @@ def plot_and_save_heatmap(
         matrix = matrix.detach().cpu().numpy()
 
     figsize = (
-        matrix.shape[1] * 0.7,
-        matrix.shape[0] * 0.7,
-    ) if figsize is None else figsize
+        (
+            matrix.shape[1] * 0.7,
+            matrix.shape[0] * 0.7,
+        )
+        if figsize is None
+        else figsize
+    )
     plt.figure(figsize=figsize)
     sns.heatmap(
         matrix,

pyvisim/encoders/vlad.py

@@ -103,7 +103,9 @@ def encode(self, images: Iterable[np.ndarray] | np.ndarray) -> np.ndarray:
                 descriptors = self.pca.transform(descriptors.astype(np.float32))
 
             if descriptors is None or descriptors.shape[0] == 0:
-                raise ValueError("No descriptors found in the image. Cannot compute VLAD encoding.")
+                raise ValueError(
+                    "No descriptors found in the image. Cannot compute VLAD encoding."
+                )
 
             labels = self.clustering_model.predict(descriptors.astype(np.float32))
             centroids = self.clustering_model.cluster_centers_