pairwise-comparison-framework 1.0.0__py3-none-any.whl

pairwise_comparison_framework/__init__.py

@@ -0,0 +1,4 @@
+from .clusterizer import Clusterizer
+from .predictor import Predictor
+
+__all__ = ["Clusterizer", "Predictor"]

pairwise_comparison_framework/_base.py

@@ -0,0 +1,59 @@
+import json
+
+from .models import ItemLabel, UserComparisons
+
+
+class BaseDataLoader:
+    """Shared data-loading logic for classes that consume comparison data.
+
+    Provides `_load_comparisons` and `_load_labels` together with the
+    `comparisons` and `labels` attributes they populate. Subclasses
+    should call `super().__init__()` and then perform their own
+    initialisation.
+    """
+
+    comparisons: list[UserComparisons]
+    labels: list[ItemLabel]
+
+    def __init__(self) -> None:
+        self.comparisons = []
+        self.labels = []
+
+    def _load_comparisons(self, filename: str) -> None:
+        """Load pairwise comparisons from a JSON file.
+
+        The file must contain a JSON array of objects, each with a
+        `user_id` and a `comparisons` list.
+
+        Args:
+            filename: Path to the JSON file.
+        """
+        with open(filename, "r") as f:
+            data = json.load(f)
+
+        if not isinstance(data, list):
+            raise ValueError("comparisons must be a list")
+
+        self.comparisons = [UserComparisons(**item) for item in data]
+
+        print(f"Loaded pairwise comparisons from {len(self.comparisons)} users")
+        print(
+            f"Total amount of pairwise comparisons: "
+            f"{sum(len(user.comparisons) for user in self.comparisons)}"
+        )
+
+    def _load_labels(self, filename: str) -> None:
+        """Load item labels from a JSON file.
+
+        Args:
+            filename: Path to the JSON file containing a list of
+                `{"item_id": ..., "label": ...}` objects.
+        """
+        with open(filename, "r") as f:
+            data = json.load(f)
+
+        if not isinstance(data, list):
+            raise ValueError("labels must be a list")
+
+        self.labels = [ItemLabel(**item) for item in data]
+        print(f"Loaded labels from {len(self.labels)} items")