chore: change naming

wildlife_tools/similarity/wildfusion.py

@@ -129,49 +129,49 @@ class WildFusion:
 
     WildFusion can be used with a limited computational budget by applying it only B times per query
     image. It uses a fast-to-compute similarity score (e.g., cosine similarity of deep features) provided
-    by the priority_matcher to construct a shortlist of the most promising matches for a given query.
+    by the priority_pipeline to construct a shortlist of the most promising matches for a given query.
     Final ranking is then based on WildFusion scores calculated for the pairs in the shortlist.
     """
 
     def __init__(
         self,
-        calibrated_matchers: list[SimilarityPipeline],
-        priority_matcher: SimilarityPipeline | None = None,
+        calibrated_pipelines: list[SimilarityPipeline],
+        priority_pipeline: SimilarityPipeline | None = None,
     ):
         """
         Args:
-            calibrated_matchers (list[SimilarityPipeline]): List of SimilarityPipeline objects.
-            priority_matcher (SimilarityPipeline, optional): Fast-to-compute similarity matcher
+            calibrated_pipelines (list[SimilarityPipeline]): List of SimilarityPipeline objects.
+            priority_pipeline (SimilarityPipeline, optional): Fast-to-compute similarity matcher
                 used for shortlisting.
         """
 
-        self.calibrated_matchers = calibrated_matchers
-        self.priority_matcher = priority_matcher
+        self.calibrated_pipelines = calibrated_pipelines
+        self.priority_pipeline = priority_pipeline
 
     def fit_calibration(self, dataset0: ImageDataset, dataset1: ImageDataset):
         """
-        Fit the all calibration models for all matchers in `calibrated_matchers`.
+        Fit the all calibration models for all matchers in `calibrated_pipelines`.
 
         Args:
             dataset0 (ImageDataset): The first dataset (e.g., part of training set).
             dataset1 (ImageDataset): The second dataset (e.g., part of training set).
         """
 
-        for matcher in self.calibrated_matchers:
+        for matcher in self.calibrated_pipelines:
             matcher.fit_calibration(dataset0, dataset1)
 
-        if self.priority_matcher is not None:
-            self.priority_matcher.fit_calibration(dataset0, dataset1)
+        if self.priority_pipeline is not None:
+            self.priority_pipeline.fit_calibration(dataset0, dataset1)
 
     def get_priority_pairs(
         self, dataset0: ImageDataset, dataset1: ImageDataset, B: int
     ) -> np.ndarray:
         """Implements shortlisting strategy for selection of most relevant pairs."""
 
-        if self.priority_matcher is None:
+        if self.priority_pipeline is None:
             raise ValueError("Priority matcher is not assigned.")
 
-        priority = self.priority_matcher(dataset0, dataset1)
+        priority = self.priority_pipeline(dataset0, dataset1)
         _, idx1 = torch.topk(torch.tensor(priority), min(B, priority.shape[1]))
         idx0 = np.indices(idx1.numpy().shape)[0]
         grid_indices = np.stack([idx0.flatten(), idx1.flatten()]).T
@@ -197,7 +197,7 @@ def __call__(
             pairs (list of tuples, optional): Specific pairs of images to compute similarity scores.
                                               If None, compute similarity scores for all pairs.
                                               Is ignored if `B` is provided.
-            B (int, optional): Number of pairs to compute similarity scores for. Required `priority_matcher` to be assigned.
+            B (int, optional): Number of pairs to compute similarity scores for. Required `priority_pipeline` to be assigned.
                                               If None, compute similarity scores for all pairs.
 
         Returns:
@@ -209,7 +209,7 @@ def __call__(
             pairs = self.get_priority_pairs(dataset0, dataset1, B=B)
 
         scores = []
-        for matcher in self.calibrated_matchers:
+        for matcher in self.calibrated_pipelines:
             scores.append(matcher(dataset0, dataset1, pairs=pairs))
 
         score_combined = np.mean(scores, axis=0)