ultralytics 8.3.51__tar.gz → 8.3.52__tar.gz

{ultralytics-8.3.51/ultralytics.egg-info → ultralytics-8.3.52}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: ultralytics
-Version: 8.3.51
+Version: 8.3.52
 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-8.3.51 → ultralytics-8.3.52}/ultralytics/__init__.py

@@ -1,6 +1,6 @@
 # Ultralytics YOLO 🚀, AGPL-3.0 license
 
-__version__ = "8.3.51"
+__version__ = "8.3.52"
 
 import os
 

ultralytics-8.3.52/ultralytics/cfg/solutions/default.yaml

@@ -0,0 +1,22 @@
+# Ultralytics YOLO 🚀, AGPL-3.0 license
+# Configuration for Ultralytics Solutions: https://docs.ultralytics.com/solutions/
+
+# Object counting settings  --------------------------------------------------------------------------------------------
+region: # list[tuple[int, int]] object counting, queue or speed estimation region points.
+show_in: True # (bool) flag to display objects moving *into* the defined region
+show_out: True # (bool) flag to display objects moving *out of* the defined region
+
+# Heatmaps settings ----------------------------------------------------------------------------------------------------
+colormap: #  (int | str) colormap for heatmap, Only OPENCV supported colormaps can be used.
+
+# Workouts monitoring settings -----------------------------------------------------------------------------------------
+up_angle: 145.0 # (float) Workouts up_angle for counts, 145.0 is default value.
+down_angle: 90 # (float) Workouts down_angle for counts, 90 is default value. Y
+kpts: [6, 8, 10] # (list[int]) keypoints for workouts monitoring, i.e. for pushups kpts have values of [6, 8, 10].
+
+# Analytics settings ---------------------------------------------------------------------------------------------------
+analytics_type: "line" # (str) analytics type i.e "line", "pie", "bar" or "area" charts.
+json_file: # (str) parking system regions file path.
+
+# Security alarm system settings ---------------------------------------------------------------------------------------
+records: 5 # (int) Total detections count to send an email about security

{ultralytics-8.3.51 → ultralytics-8.3.52}/ultralytics/utils/ops.py

@@ -75,8 +75,13 @@ def segment2box(segment, width=640, height=640):
         (np.ndarray): the minimum and maximum x and y values of the segment.
     """
     x, y = segment.T  # segment xy
-    x = x.clip(0, width)
-    y = y.clip(0, height)
+    # any 3 out of 4 sides are outside the image, clip coordinates first, https://github.com/ultralytics/ultralytics/pull/18294
+    if np.array([x.min() < 0, y.min() < 0, x.max() > width, y.max() > height]).sum() >= 3:
+        x = x.clip(0, width)
+        y = y.clip(0, height)
+    inside = (x >= 0) & (y >= 0) & (x <= width) & (y <= height)
+    x = x[inside]
+    y = y[inside]
     return (
         np.array([x.min(), y.min(), x.max(), y.max()], dtype=segment.dtype)
         if any(x)

{ultralytics-8.3.51 → ultralytics-8.3.52}/ultralytics/utils/torch_utils.py

@@ -617,6 +617,32 @@ def convert_optimizer_state_dict_to_fp16(state_dict):
     return state_dict
 
 
+@contextmanager
+def cuda_memory_usage(device=None):
+    """
+    Monitor and manage CUDA memory usage.
+
+    This function checks if CUDA is available and, if so, empties the CUDA cache to free up unused memory.
+    It then yields a dictionary containing memory usage information, which can be updated by the caller.
+    Finally, it updates the dictionary with the amount of memory reserved by CUDA on the specified device.
+
+    Args:
+        device (torch.device, optional): The CUDA device to query memory usage for. Defaults to None.
+
+    Yields:
+        (dict): A dictionary with a key 'memory' initialized to 0, which will be updated with the reserved memory.
+    """
+    cuda_info = dict(memory=0)
+    if torch.cuda.is_available():
+        torch.cuda.empty_cache()
+        try:
+            yield cuda_info
+        finally:
+            cuda_info["memory"] = torch.cuda.memory_reserved(device)
+    else:
+        yield cuda_info
+
+
 def profile(input, ops, n=10, device=None, max_num_obj=0):
     """
     Ultralytics speed, memory and FLOPs profiler.
@@ -653,27 +679,31 @@ def profile(input, ops, n=10, device=None, max_num_obj=0):
                 flops = 0
 
             try:
+                mem = 0
                 for _ in range(n):
-                    t[0] = time_sync()
-                    y = m(x)
-                    t[1] = time_sync()
-                    try:
-                        (sum(yi.sum() for yi in y) if isinstance(y, list) else y).sum().backward()
-                        t[2] = time_sync()
-                    except Exception:  # no backward method
-                        # print(e)  # for debug
-                        t[2] = float("nan")
+                    with cuda_memory_usage(device) as cuda_info:
+                        t[0] = time_sync()
+                        y = m(x)
+                        t[1] = time_sync()
+                        try:
+                            (sum(yi.sum() for yi in y) if isinstance(y, list) else y).sum().backward()
+                            t[2] = time_sync()
+                        except Exception:  # no backward method
+                            # print(e)  # for debug
+                            t[2] = float("nan")
+                    mem += cuda_info["memory"] / 1e9  # (GB)
                     tf += (t[1] - t[0]) * 1000 / n  # ms per op forward
                     tb += (t[2] - t[1]) * 1000 / n  # ms per op backward
                     if max_num_obj:  # simulate training with predictions per image grid (for AutoBatch)
-                        torch.randn(
-                            x.shape[0],
-                            max_num_obj,
-                            int(sum((x.shape[-1] / s) * (x.shape[-2] / s) for s in m.stride.tolist())),
-                            device=device,
-                            dtype=torch.float32,
-                        )
-                mem = torch.cuda.memory_reserved() / 1e9 if torch.cuda.is_available() else 0  # (GB)
+                        with cuda_memory_usage(device) as cuda_info:
+                            torch.randn(
+                                x.shape[0],
+                                max_num_obj,
+                                int(sum((x.shape[-1] / s) * (x.shape[-2] / s) for s in m.stride.tolist())),
+                                device=device,
+                                dtype=torch.float32,
+                            )
+                        mem += cuda_info["memory"] / 1e9  # (GB)
                 s_in, s_out = (tuple(x.shape) if isinstance(x, torch.Tensor) else "list" for x in (x, y))  # shapes
                 p = sum(x.numel() for x in m.parameters()) if isinstance(m, nn.Module) else 0  # parameters
                 LOGGER.info(f"{p:12}{flops:12.4g}{mem:>14.3f}{tf:14.4g}{tb:14.4g}{str(s_in):>24s}{str(s_out):>24s}")

{ultralytics-8.3.51 → ultralytics-8.3.52/ultralytics.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: ultralytics
-Version: 8.3.51
+Version: 8.3.52
 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-8.3.51/ultralytics/cfg/solutions/default.yaml

@@ -1,22 +0,0 @@
-# Ultralytics YOLO 🚀, AGPL-3.0 license
-# Configuration for Ultralytics Solutions
-
-# Object counting settings
-region: # Object counting, queue or speed estimation region points. Default region points are [(20, 400), (1080, 400), (1080, 360), (20, 360)]
-show_in: True # Flag to display objects moving *into* the defined region
-show_out: True # Flag to display objects moving *out of* the defined region
-
-# Heatmaps settings
-colormap: # Colormap for heatmap, Only OPENCV supported colormaps can be used. By default COLORMAP_PARULA will be used for visualization.
-
-# Workouts monitoring settings
-up_angle: 145.0 # Workouts up_angle for counts, 145.0 is default value. You can adjust it for different workouts, based on position of keypoints.
-down_angle: 90 # Workouts down_angle for counts, 90 is default value. You can change it for different workouts, based on position of keypoints.
-kpts: [6, 8, 10] # Keypoints for workouts monitoring, i.e. If you want to consider keypoints for pushups that have mostly values of [6, 8, 10].
-
-# Analytics settings
-analytics_type: "line" # Analytics type i.e "line", "pie", "bar" or "area" charts. By default, "line" analytics will be used for processing.
-json_file: # parking system regions file path.
-
-# Security alarm system
-records: 5 # Total detections count to send an email about security