scorevision: push artifact

miner.py

@@ -1,4 +1,5 @@
-"""ScoreVision crime detector — YOLOv11s with flip TTA + per-class conf + cross-class NMS."""
+"""ScoreVision crime detector v24 — YOLOv11s trained on rival-consensus labels.
+Per-class conf thresholds tuned vs validator-PGT proxy; flip TTA; cross-class NMS."""
 from pathlib import Path
 import math
 
@@ -33,9 +34,9 @@ class Miner:
     cross_iou_thresh = 0.50
     max_aspect_ratio = 10.0
     max_det = 150
-    # tuned on held-out SAM3-labeled crime set
+    # per-class thresholds — tuned on rival-consensus GT (proxy for validator PGT)
     _conf_thres_array = np.array(
-        [0.50, 0.50, 0.30, 0.30, 0.40, 0.40], dtype=np.float32
+        [0.30, 0.70, 0.60, 0.50, 0.40, 0.40], dtype=np.float32
     )
 
     def __init__(self, path_hf_repo: Path) -> None:
@@ -167,7 +168,6 @@ class Miner:
         return np.array(keep_all, dtype=np.intp)
 
     def _cross_class_dedup(self, boxes, scores, cls_ids, iou_thr):
-        """If two boxes (any class) heavily overlap, keep only the higher-scoring one."""
         n = len(boxes)
         if n == 0:
             return np.array([], dtype=np.intp)
@@ -211,7 +211,6 @@ class Miner:
 
     def _decode(self, preds: ndarray, ratio: float, pad: tuple[float, float],
                 orig_size: tuple[int, int]) -> tuple[np.ndarray, np.ndarray, np.ndarray]:
-        """ONNX output is [1,300,6]: x1,y1,x2,y2,conf,cls in letterboxed coords."""
         if preds.ndim == 3 and preds.shape[0] == 1:
             preds = preds[0]
         if preds.ndim != 2 or preds.shape[1] < 6:
@@ -223,18 +222,15 @@ class Miner:
         boxes = preds[:, :4].astype(np.float32)
         scores = preds[:, 4].astype(np.float32)
         cls_ids = preds[:, 5].astype(np.int32)
-        # drop padded rows
         keep = (scores > 0) & (cls_ids >= 0) & (cls_ids < len(self.class_names))
         boxes, scores, cls_ids = boxes[keep], scores[keep], cls_ids[keep]
         if len(boxes) == 0:
             return boxes, scores, cls_ids
-        # per-class confidence
         thr = self._conf_thres_array[cls_ids]
         keep = scores >= thr
         boxes, scores, cls_ids = boxes[keep], scores[keep], cls_ids[keep]
         if len(boxes) == 0:
             return boxes, scores, cls_ids
-        # untransform: subtract pad, divide ratio
         pad_w, pad_h = pad
         boxes[:, [0, 2]] -= pad_w
         boxes[:, [1, 3]] -= pad_h
@@ -262,7 +258,6 @@ class Miner:
         cls_ids = np.concatenate([c0, cf], axis=0) if len(b0) or len(bf) else c0
         if len(boxes) == 0:
             return []
-        # filter + per-class NMS + cross-class dedup
         boxes, scores, cls_ids = self._filter_sane(
             boxes, scores, cls_ids, (image.shape[1], image.shape[0])
         )