ultralytics 8.3.104__tar.gz → 8.3.105__tar.gz

{ultralytics-8.3.104/ultralytics.egg-info → ultralytics-8.3.105}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ultralytics
-Version: 8.3.104
+Version: 8.3.105
 Summary: Ultralytics YOLO 🚀 for SOTA object detection, multi-object tracking, instance segmentation, pose estimation and image classification.
 Author-email: Glenn Jocher <glenn.jocher@ultralytics.com>, Jing Qiu <jing.qiu@ultralytics.com>
 Maintainer-email: Ultralytics <hello@ultralytics.com>

{ultralytics-8.3.104 → ultralytics-8.3.105}/tests/test_python.py

@@ -199,7 +199,7 @@ def test_track_stream():
 
 def test_val():
     """Test the validation mode of the YOLO model."""
-    YOLO(MODEL).val(data="coco8.yaml", imgsz=32, save_hybrid=True)
+    YOLO(MODEL).val(data="coco8.yaml", imgsz=32)
 
 
 def test_train_scratch():

{ultralytics-8.3.104 → ultralytics-8.3.105}/tests/test_solutions.py

@@ -20,7 +20,7 @@ PARKING_VIDEO = "solution_ci_parking_demo.mp4"  # only for parking management so
 PARKING_AREAS_JSON = "solution_ci_parking_areas.json"  # only for parking management solution
 PARKING_MODEL = "solutions_ci_parking_model.pt"  # only for parking management solution
 MODEL_FILE = "yolo11n.pt"  # model file used for solutions, except parking management and instance segmentation
-REGION = [(20, 400), (1080, 400), (1080, 360), (20, 360)]  # for object counting, speed estimation and queue management
+REGION = [(10, 200), (540, 200), (540, 180), (10, 180)]  # for object counting, speed estimation and queue management
 
 # Test configs for each solution : (name, class, needs_frame_count, video, kwargs)
 SOLUTIONS = [

{ultralytics-8.3.104 → ultralytics-8.3.105}/ultralytics/__init__.py

@@ -1,6 +1,6 @@
 # Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
 
-__version__ = "8.3.104"
+__version__ = "8.3.105"
 
 import os
 

{ultralytics-8.3.104 → ultralytics-8.3.105}/ultralytics/cfg/__init__.py

@@ -214,7 +214,6 @@ CFG_BOOL_KEYS = frozenset(
         "overlap_mask",
         "val",
         "save_json",
-        "save_hybrid",
         "half",
         "dnn",
         "plots",
@@ -437,22 +436,25 @@ def _handle_deprecation(custom: Dict) -> Dict:
         equivalents. It also handles value conversions where necessary, such as inverting boolean values for
         'hide_labels' and 'hide_conf'.
     """
-    for key in custom.copy().keys():
-        if key == "boxes":
-            deprecation_warn(key, "show_boxes")
-            custom["show_boxes"] = custom.pop("boxes")
-        if key == "hide_labels":
-            deprecation_warn(key, "show_labels")
-            custom["show_labels"] = custom.pop("hide_labels") == "False"
-        if key == "hide_conf":
-            deprecation_warn(key, "show_conf")
-            custom["show_conf"] = custom.pop("hide_conf") == "False"
-        if key == "line_thickness":
-            deprecation_warn(key, "line_width")
-            custom["line_width"] = custom.pop("line_thickness")
-        if key == "label_smoothing":
-            deprecation_warn(key)
-            custom.pop("label_smoothing")
+    deprecated_mappings = {
+        "boxes": ("show_boxes", lambda v: v),
+        "hide_labels": ("show_labels", lambda v: not bool(v)),
+        "hide_conf": ("show_conf", lambda v: not bool(v)),
+        "line_thickness": ("line_width", lambda v: v),
+    }
+    removed_keys = {"label_smoothing", "save_hybrid"}
+
+    for old_key, (new_key, transform) in deprecated_mappings.items():
+        if old_key not in custom:
+            continue
+        deprecation_warn(old_key, new_key)
+        custom[new_key] = transform(custom.pop(old_key))
+
+    for key in removed_keys:
+        if key not in custom:
+            continue
+        deprecation_warn(key)
+        custom.pop(key)
 
     return custom
 

{ultralytics-8.3.104 → ultralytics-8.3.105}/ultralytics/cfg/default.yaml

@@ -47,7 +47,6 @@ dropout: 0.0 # (float) use dropout regularization (classify train only)
 val: True # (bool) validate/test during training
 split: val # (str) dataset split to use for validation, i.e. 'val', 'test' or 'train'
 save_json: False # (bool) save results to JSON file
-save_hybrid: False # (bool) save hybrid version of labels (labels + additional predictions)
 conf: # (float, optional) object confidence threshold for detection (default 0.25 predict, 0.001 val)
 iou: 0.7 # (float) intersection over union (IoU) threshold for NMS
 max_det: 300 # (int) maximum number of detections per image

{ultralytics-8.3.104 → ultralytics-8.3.105}/ultralytics/models/yolo/detect/val.py

@@ -62,12 +62,6 @@ class DetectionValidator(BaseValidator):
         self.metrics = DetMetrics(save_dir=self.save_dir)
         self.iouv = torch.linspace(0.5, 0.95, 10)  # IoU vector for mAP@0.5:0.95
         self.niou = self.iouv.numel()
-        self.lb = []  # for autolabelling
-        if self.args.save_hybrid and self.args.task == "detect":
-            LOGGER.warning(
-                "WARNING ⚠️ 'save_hybrid=True' will append ground truth to predictions for autolabelling.\n"
-                "WARNING ⚠️ 'save_hybrid=True' will cause incorrect mAP.\n"
-            )
 
     def preprocess(self, batch):
         """
@@ -84,15 +78,6 @@ class DetectionValidator(BaseValidator):
         for k in ["batch_idx", "cls", "bboxes"]:
             batch[k] = batch[k].to(self.device)
 
-        if self.args.save_hybrid and self.args.task == "detect":
-            height, width = batch["img"].shape[2:]
-            nb = len(batch["img"])
-            bboxes = batch["bboxes"] * torch.tensor((width, height, width, height), device=self.device)
-            self.lb = [
-                torch.cat([batch["cls"][batch["batch_idx"] == i], bboxes[batch["batch_idx"] == i]], dim=-1)
-                for i in range(nb)
-            ]
-
         return batch
 
     def init_metrics(self, model):
@@ -139,7 +124,6 @@ class DetectionValidator(BaseValidator):
             preds,
             self.args.conf,
             self.args.iou,
-            labels=self.lb,
             nc=self.nc,
             multi_label=True,
             agnostic=self.args.single_cls or self.args.agnostic_nms,

{ultralytics-8.3.104 → ultralytics-8.3.105}/ultralytics/nn/modules/head.py

@@ -369,8 +369,8 @@ class LRPCHead(nn.Module):
         if self.enabled:
             pf_score = self.pf(cls_feat)[0, 0].flatten(0)
             mask = pf_score.sigmoid() > conf
-
-            cls_feat = self.vocab(cls_feat.flatten(2).transpose(-1, -2)[:, mask])
+            cls_feat = cls_feat.flatten(2).transpose(-1, -2)
+            cls_feat = self.vocab(cls_feat * mask.unsqueeze(-1).int() if not conf else cls_feat[:, mask])
             return (self.loc(loc_feat), cls_feat.transpose(-1, -2)), mask
         else:
             cls_feat = self.vocab(cls_feat)
@@ -478,7 +478,9 @@ class YOLOEDetect(Detect):
             cls_feat = self.cv3[i](x[i])
             loc_feat = self.cv2[i](x[i])
             assert isinstance(self.lrpc[i], LRPCHead)
-            x[i], mask = self.lrpc[i](cls_feat, loc_feat, getattr(self, "conf", 0.001))
+            x[i], mask = self.lrpc[i](
+                cls_feat, loc_feat, 0 if self.export and not self.dynamic else getattr(self, "conf", 0.001)
+            )
             masks.append(mask)
         shape = x[0][0].shape
         if self.dynamic or self.shape != shape:
@@ -499,7 +501,7 @@ class YOLOEDetect(Detect):
             dbox = self.decode_bboxes(self.dfl(box), self.anchors.unsqueeze(0)) * self.strides
 
         mask = torch.cat(masks)
-        y = torch.cat((dbox[:, :, mask], cls.sigmoid()), 1)
+        y = torch.cat((dbox if self.export and not self.dynamic else dbox[..., mask], cls.sigmoid()), 1)
 
         if return_mask:
             return (y, mask) if self.export else ((y, x), mask)
@@ -560,7 +562,7 @@ class YOLOESegment(YOLOEDetect):
             return x, mc, p
 
         if has_lrpc:
-            mc = mc[:, :, mask]
+            mc = (mc * mask.int()) if self.export and not self.dynamic else mc[..., mask]
 
         return (torch.cat([x, mc], 1), p) if self.export else (torch.cat([x[0], mc], 1), (x[1], mc, p))
 

{ultralytics-8.3.104 → ultralytics-8.3.105}/ultralytics/solutions/object_counter.py

@@ -45,6 +45,7 @@ class ObjectCounter(BaseSolution):
 
         self.show_in = self.CFG["show_in"]
         self.show_out = self.CFG["show_out"]
+        self.margin = self.line_width * 2  # Scales the background rectangle size to display counts properly
 
     def count_objects(self, current_centroid, track_id, prev_position, cls):
         """
@@ -144,7 +145,7 @@ class ObjectCounter(BaseSolution):
             if value["IN"] != 0 or value["OUT"] != 0
         }
         if labels_dict:
-            self.annotator.display_analytics(plot_im, labels_dict, (104, 31, 17), (255, 255, 255), 10)
+            self.annotator.display_analytics(plot_im, labels_dict, (104, 31, 17), (255, 255, 255), self.margin)
 
     def process(self, im0):
         """

{ultralytics-8.3.104 → ultralytics-8.3.105}/ultralytics/solutions/solutions.py

@@ -79,9 +79,7 @@ class BaseSolution:
         self.LOGGER.info(f"Ultralytics Solutions: ✅ {DEFAULT_SOL_DICT}")
 
         self.region = self.CFG["region"]  # Store region data for other classes usage
-        self.line_width = (
-            self.CFG["line_width"] if self.CFG["line_width"] is not None else 2
-        )  # Store line_width for usage
+        self.line_width = self.CFG["line_width"] if self.CFG["line_width"] not in (None, 0) else 2  # Store line_width
 
         # Load Model and store classes names
         if self.CFG["model"] is None:

{ultralytics-8.3.104 → ultralytics-8.3.105/ultralytics.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ultralytics
-Version: 8.3.104
+Version: 8.3.105
 Summary: Ultralytics YOLO 🚀 for SOTA object detection, multi-object tracking, instance segmentation, pose estimation and image classification.
 Author-email: Glenn Jocher <glenn.jocher@ultralytics.com>, Jing Qiu <jing.qiu@ultralytics.com>
 Maintainer-email: Ultralytics <hello@ultralytics.com>