Update metrics.py

Author: Oppie2333

src/patchcore/metrics.py

@@ -1,13 +1,9 @@
-"""Anomaly metrics."""
 import numpy as np
-from sklearn import metrics
+from sklearn.metrics import average_precision_score
 
-
-def compute_imagewise_retrieval_metrics(
-    anomaly_prediction_weights, anomaly_ground_truth_labels
-):
+def compute_image_ap(anomaly_prediction_weights, anomaly_ground_truth_labels):
     """
-    Computes retrieval statistics (AUROC, FPR, TPR).
+    Computes the average precision (AP) for image-wise anomaly detection.
 
     Args:
         anomaly_prediction_weights: [np.array or list] [N] Assignment weights
@@ -16,19 +12,11 @@ def compute_imagewise_retrieval_metrics(
         anomaly_ground_truth_labels: [np.array or list] [N] Binary labels - 1
                                     if image is an anomaly, 0 if not.
     """
-    fpr, tpr, thresholds = metrics.roc_curve(
-        anomaly_ground_truth_labels, anomaly_prediction_weights
-    )
-    auroc = metrics.roc_auc_score(
-        anomaly_ground_truth_labels, anomaly_prediction_weights
-    )
-    return {"auroc": auroc, "fpr": fpr, "tpr": tpr, "threshold": thresholds}
-
+    return average_precision_score(anomaly_ground_truth_labels, anomaly_prediction_weights)
 
-def compute_pixelwise_retrieval_metrics(anomaly_segmentations, ground_truth_masks):
+def compute_pixel_ap(anomaly_segmentations, ground_truth_masks):
     """
-    Computes pixel-wise statistics (AUROC, FPR, TPR) for anomaly segmentations
-    and ground truth segmentation masks.
+    Computes the average precision (AP) for pixel-wise anomaly detection.
 
     Args:
         anomaly_segmentations: [list of np.arrays or np.array] [NxHxW] Contains
@@ -44,33 +32,17 @@ def compute_pixelwise_retrieval_metrics(anomaly_segmentations, ground_truth_mask
     flat_anomaly_segmentations = anomaly_segmentations.ravel()
     flat_ground_truth_masks = ground_truth_masks.ravel()
 
-    fpr, tpr, thresholds = metrics.roc_curve(
-        flat_ground_truth_masks.astype(int), flat_anomaly_segmentations
-    )
-    auroc = metrics.roc_auc_score(
-        flat_ground_truth_masks.astype(int), flat_anomaly_segmentations
-    )
-
-    precision, recall, thresholds = metrics.precision_recall_curve(
-        flat_ground_truth_masks.astype(int), flat_anomaly_segmentations
-    )
-    F1_scores = np.divide(
-        2 * precision * recall,
-        precision + recall,
-        out=np.zeros_like(precision),
-        where=(precision + recall) != 0,
-    )
+    return average_precision_score(flat_ground_truth_masks.astype(int), flat_anomaly_segmentations)
 
-    optimal_threshold = thresholds[np.argmax(F1_scores)]
-    predictions = (flat_anomaly_segmentations >= optimal_threshold).astype(int)
-    fpr_optim = np.mean(predictions > flat_ground_truth_masks)
-    fnr_optim = np.mean(predictions < flat_ground_truth_masks)
+def compute_pro(anomaly_segmentations, ground_truth_masks):
+    """
+    Computes the PRO score. This is a placeholder function and needs to be implemented.
 
-    return {
-        "auroc": auroc,
-        "fpr": fpr,
-        "tpr": tpr,
-        "optimal_threshold": optimal_threshold,
-        "optimal_fpr": fpr_optim,
-        "optimal_fnr": fnr_optim,
-    }
+    Args:
+        anomaly_segmentations: [list of np.arrays or np.array] [NxHxW] Contains
+                                generated segmentation masks.
+        ground_truth_masks: [list of np.arrays or np.array] [NxHxW] Contains
+                            predefined ground truth segmentation masks
+    """
+    # Placeholder implementation, replace with actual PRO calculation
+    return np.random.rand()