dgenerate-ultralytics-headless 8.3.144__py3-none-any.whl → 8.3.146__py3-none-any.whl

ultralytics/solutions/solutions.py

@@ -169,11 +169,12 @@ class BaseSolution:
         with self.profilers[0]:
             self.tracks = self.model.track(
                 source=im0, persist=True, classes=self.classes, verbose=False, **self.track_add_args
-            )
-        self.track_data = self.tracks[0].obb or self.tracks[0].boxes  # Extract tracks for OBB or object detection
+            )[0]
+        is_obb = self.tracks.obb is not None
+        self.track_data = self.tracks.obb if is_obb else self.tracks.boxes  # Extract tracks for OBB or object detection
 
-        if self.track_data and self.track_data.id is not None:
-            self.boxes = self.track_data.xyxy.cpu()
+        if self.track_data and self.track_data.is_track:
+            self.boxes = (self.track_data.xyxyxyxy if is_obb else self.track_data.xyxy).cpu()
             self.clss = self.track_data.cls.cpu().tolist()
             self.track_ids = self.track_data.id.int().cpu().tolist()
             self.confs = self.track_data.conf.cpu().tolist()

ultralytics/trackers/bot_sort.py

@@ -260,11 +260,13 @@ class ReID:
         from ultralytics import YOLO
 
         self.model = YOLO(model)
-        self.model(embed=[len(self.model.model.model) - 2 if ".pt" in model else -1], verbose=False)  # initialize
+        self.model(embed=[len(self.model.model.model) - 2 if ".pt" in model else -1], verbose=False, save=False)  # init
 
     def __call__(self, img: np.ndarray, dets: np.ndarray) -> List[np.ndarray]:
         """Extract embeddings for detected objects."""
-        feats = self.model([save_one_box(det, img, save=False) for det in xywh2xyxy(torch.from_numpy(dets[:, :4]))])
+        feats = self.model.predictor(
+            [save_one_box(det, img, save=False) for det in xywh2xyxy(torch.from_numpy(dets[:, :4]))]
+        )
         if len(feats) != dets.shape[0] and feats[0].shape[0] == dets.shape[0]:
             feats = feats[0]  # batched prediction with non-PyTorch backend
         return [f.cpu().numpy() for f in feats]

ultralytics/utils/__init__.py

@@ -841,8 +841,7 @@ def is_docker() -> bool:
         (bool): True if the script is running inside a Docker container, False otherwise.
     """
     try:
-        with open("/proc/self/cgroup") as f:
-            return "docker" in f.read()
+        return os.path.exists("/.dockerenv")
     except Exception:
         return False
 

ultralytics/utils/benchmarks.py

@@ -59,6 +59,7 @@ def benchmark(
     verbose=False,
     eps=1e-3,
     format="",
+    **kwargs,
 ):
     """
     Benchmark a YOLO model across different formats for speed and accuracy.
@@ -73,6 +74,7 @@ def benchmark(
         verbose (bool | float): If True or a float, assert benchmarks pass with given metric.
         eps (float): Epsilon value for divide by zero prevention.
         format (str): Export format for benchmarking. If not supplied all formats are benchmarked.
+        **kwargs (Any): Additional keyword arguments for exporter.
 
     Returns:
         (pandas.DataFrame): A pandas DataFrame with benchmark results for each format, including file size, metric,
@@ -104,41 +106,41 @@ def benchmark(
     if format_arg:
         formats = frozenset(export_formats()["Argument"])
         assert format in formats, f"Expected format to be one of {formats}, but got '{format_arg}'."
-    for i, (name, format, suffix, cpu, gpu, _) in enumerate(zip(*export_formats().values())):
+    for name, format, suffix, cpu, gpu, _ in zip(*export_formats().values()):
         emoji, filename = "❌", None  # export defaults
         try:
             if format_arg and format_arg != format:
                 continue
 
             # Checks
-            if i == 7:  # TF GraphDef
+            if format == "pb":
                 assert model.task != "obb", "TensorFlow GraphDef not supported for OBB task"
-            elif i == 9:  # Edge TPU
+            elif format == "edgetpu":
                 assert LINUX and not ARM64, "Edge TPU export only supported on non-aarch64 Linux"
-            elif i in {5, 10}:  # CoreML and TF.js
+            elif format in {"coreml", "tfjs"}:
                 assert MACOS or (LINUX and not ARM64), (
                     "CoreML and TF.js export only supported on macOS and non-aarch64 Linux"
                 )
-            if i in {5}:  # CoreML
+            if format == "coreml":
                 assert not IS_PYTHON_3_13, "CoreML not supported on Python 3.13"
-            if i in {6, 7, 8, 9, 10}:  # TF SavedModel, TF GraphDef, and TFLite, TF EdgeTPU and TF.js
+            if format in {"saved_model", "pb", "tflite", "edgetpu", "tfjs"}:
                 assert not isinstance(model, YOLOWorld), "YOLOWorldv2 TensorFlow exports not supported by onnx2tf yet"
                 # assert not IS_PYTHON_MINIMUM_3_12, "TFLite exports not supported on Python>=3.12 yet"
-            if i == 11:  # Paddle
+            if format == "paddle":
                 assert not isinstance(model, YOLOWorld), "YOLOWorldv2 Paddle exports not supported yet"
                 assert model.task != "obb", "Paddle OBB bug https://github.com/PaddlePaddle/Paddle/issues/72024"
                 assert not is_end2end, "End-to-end models not supported by PaddlePaddle yet"
                 assert (LINUX and not IS_JETSON) or MACOS, "Windows and Jetson Paddle exports not supported yet"
-            if i == 12:  # MNN
+            if format == "mnn":
                 assert not isinstance(model, YOLOWorld), "YOLOWorldv2 MNN exports not supported yet"
-            if i == 13:  # NCNN
+            if format == "ncnn":
                 assert not isinstance(model, YOLOWorld), "YOLOWorldv2 NCNN exports not supported yet"
-            if i == 14:  # IMX
+            if format == "imx":
                 assert not is_end2end
                 assert not isinstance(model, YOLOWorld), "YOLOWorldv2 IMX exports not supported"
                 assert model.task == "detect", "IMX only supported for detection task"
                 assert "C2f" in model.__str__(), "IMX only supported for YOLOv8"  # TODO: enable for YOLO11
-            if i == 15:  # RKNN
+            if format == "rknn":
                 assert not isinstance(model, YOLOWorld), "YOLOWorldv2 RKNN exports not supported yet"
                 assert not is_end2end, "End-to-end models not supported by RKNN yet"
                 assert LINUX, "RKNN only supported on Linux"
@@ -154,17 +156,17 @@ def benchmark(
                 exported_model = model  # PyTorch format
             else:
                 filename = model.export(
-                    imgsz=imgsz, format=format, half=half, int8=int8, data=data, device=device, verbose=False
+                    imgsz=imgsz, format=format, half=half, int8=int8, data=data, device=device, verbose=False, **kwargs
                 )
                 exported_model = YOLO(filename, task=model.task)
                 assert suffix in str(filename), "export failed"
             emoji = "❎"  # indicates export succeeded
 
             # Predict
-            assert model.task != "pose" or i != 7, "GraphDef Pose inference is not supported"
-            assert i not in {9, 10}, "inference not supported"  # Edge TPU and TF.js are unsupported
-            assert i != 5 or platform.system() == "Darwin", "inference only supported on macOS>=10.13"  # CoreML
-            if i in {13}:
+            assert model.task != "pose" or format != "pb", "GraphDef Pose inference is not supported"
+            assert format not in {"edgetpu", "tfjs"}, "inference not supported"
+            assert format != "coreml" or platform.system() == "Darwin", "inference only supported on macOS>=10.13"
+            if format == "ncnn":
                 assert not is_end2end, "End-to-end torch.topk operation is not supported for NCNN prediction yet"
             exported_model.predict(ASSETS / "bus.jpg", imgsz=imgsz, device=device, half=half, verbose=False)
 

ultralytics/utils/checks.py

@@ -401,11 +401,16 @@ def check_requirements(requirements=ROOT.parent / "requirements.txt", exclude=()
     def attempt_install(packages, commands, use_uv):
         """Attempt package installation with uv if available, falling back to pip."""
         if use_uv:
-            # Note requires --break-system-packages on ARM64 dockerfile
-            cmd = f"uv pip install --system --no-cache-dir {packages} {commands} --index-strategy=unsafe-best-match --break-system-packages --prerelease=allow"
-        else:
-            cmd = f"pip install --no-cache-dir {packages} {commands}"
-        return subprocess.check_output(cmd, shell=True).decode()
+            base = f"uv pip install --no-cache-dir {packages} {commands} --index-strategy=unsafe-best-match --break-system-packages --prerelease=allow"
+            try:
+                return subprocess.check_output(base, shell=True, stderr=subprocess.PIPE).decode()
+            except subprocess.CalledProcessError as e:
+                if e.stderr and "No virtual environment found" in e.stderr.decode():
+                    return subprocess.check_output(
+                        base.replace("uv pip install", "uv pip install --system"), shell=True
+                    ).decode()
+                raise
+        return subprocess.check_output(f"pip install --no-cache-dir {packages} {commands}", shell=True).decode()
 
     s = " ".join(f'"{x}"' for x in pkgs)  # console string
     if s:

ultralytics/utils/downloads.py

@@ -36,6 +36,7 @@ GITHUB_ASSETS_NAMES = frozenset(
     + [
         "mobile_sam.pt",
         "mobileclip_blt.ts",
+        "yolo11n-grayscale.pt",
         "calibration_image_sample_data_20x128x128x3_float32.npy.zip",
     ]
 )

ultralytics/utils/metrics.py

@@ -520,7 +520,7 @@ def plot_pr_curve(
     py: np.ndarray,
     ap: np.ndarray,
     save_dir: Path = Path("pr_curve.png"),
-    names: dict = {},
+    names: Dict[int, str] = {},
     on_plot=None,
 ):
     """
@@ -531,7 +531,7 @@ def plot_pr_curve(
         py (np.ndarray): Y values for the PR curve.
         ap (np.ndarray): Average precision values.
         save_dir (Path, optional): Path to save the plot.
-        names (dict, optional): Dictionary mapping class indices to class names.
+        names (Dict[int, str], optional): Dictionary mapping class indices to class names.
         on_plot (callable, optional): Function to call after plot is saved.
     """
     import matplotlib.pyplot as plt  # scope for faster 'import ultralytics'
@@ -563,7 +563,7 @@ def plot_mc_curve(
     px: np.ndarray,
     py: np.ndarray,
     save_dir: Path = Path("mc_curve.png"),
-    names: dict = {},
+    names: Dict[int, str] = {},
     xlabel: str = "Confidence",
     ylabel: str = "Metric",
     on_plot=None,
@@ -575,7 +575,7 @@ def plot_mc_curve(
         px (np.ndarray): X values for the metric-confidence curve.
         py (np.ndarray): Y values for the metric-confidence curve.
         save_dir (Path, optional): Path to save the plot.
-        names (dict, optional): Dictionary mapping class indices to class names.
+        names (Dict[int, str], optional): Dictionary mapping class indices to class names.
         xlabel (str, optional): X-axis label.
         ylabel (str, optional): Y-axis label.
         on_plot (callable, optional): Function to call after plot is saved.
@@ -645,7 +645,7 @@ def ap_per_class(
     plot: bool = False,
     on_plot=None,
     save_dir: Path = Path(),
-    names: dict = {},
+    names: Dict[int, str] = {},
     eps: float = 1e-16,
     prefix: str = "",
 ) -> Tuple:
@@ -660,7 +660,7 @@ def ap_per_class(
         plot (bool, optional): Whether to plot PR curves or not.
         on_plot (callable, optional): A callback to pass plots path and data when they are rendered.
         save_dir (Path, optional): Directory to save the PR curves.
-        names (dict, optional): Dict of class names to plot PR curves.
+        names (Dict[int, str], optional): Dictionary of class names to plot PR curves.
         eps (float, optional): A small value to avoid division by zero.
         prefix (str, optional): A prefix string for saving the plot files.
 
@@ -720,8 +720,7 @@ def ap_per_class(
 
     # Compute F1 (harmonic mean of precision and recall)
     f1_curve = 2 * p_curve * r_curve / (p_curve + r_curve + eps)
-    names = [v for k, v in names.items() if k in unique_classes]  # list: only classes that have data
-    names = dict(enumerate(names))  # to dict
+    names = {i: names[k] for i, k in enumerate(unique_classes) if k in names}  # dict: only classes that have data
     if plot:
         plot_pr_curve(x, prec_values, ap, save_dir / f"{prefix}PR_curve.png", names, on_plot=on_plot)
         plot_mc_curve(x, f1_curve, save_dir / f"{prefix}F1_curve.png", names, ylabel="F1", on_plot=on_plot)
@@ -915,20 +914,20 @@ class DetMetrics(SimpleClass, DataExportMixin):
     Attributes:
         save_dir (Path): A path to the directory where the output plots will be saved.
         plot (bool): A flag that indicates whether to plot precision-recall curves for each class.
-        names (dict): A dictionary of class names.
+        names (Dict[int, str]): A dictionary of class names.
         box (Metric): An instance of the Metric class for storing detection results.
-        speed (dict): A dictionary for storing execution times of different parts of the detection process.
+        speed (Dict[str, float]): A dictionary for storing execution times of different parts of the detection process.
         task (str): The task type, set to 'detect'.
     """
 
-    def __init__(self, save_dir: Path = Path("."), plot: bool = False, names: dict = {}) -> None:
+    def __init__(self, save_dir: Path = Path("."), plot: bool = False, names: Dict[int, str] = {}) -> None:
         """
         Initialize a DetMetrics instance with a save directory, plot flag, and class names.
 
         Args:
             save_dir (Path, optional): Directory to save plots.
             plot (bool, optional): Whether to plot precision-recall curves.
-            names (dict, optional): Dictionary mapping class indices to names.
+            names (Dict[int, str], optional): Dictionary of class names.
         """
         self.save_dir = save_dir
         self.plot = plot
@@ -1033,21 +1032,21 @@ class SegmentMetrics(SimpleClass, DataExportMixin):
     Attributes:
         save_dir (Path): Path to the directory where the output plots should be saved.
         plot (bool): Whether to save the detection and segmentation plots.
-        names (dict): Dictionary of class names.
-        box (Metric): An instance of the Metric class to calculate box detection metrics.
+        names (Dict[int, str]): Dictionary of class names.
+        box (Metric): An instance of the Metric class for storing detection results.
         seg (Metric): An instance of the Metric class to calculate mask segmentation metrics.
-        speed (dict): Dictionary to store the time taken in different phases of inference.
+        speed (Dict[str, float]): A dictionary for storing execution times of different parts of the detection process.
         task (str): The task type, set to 'segment'.
     """
 
-    def __init__(self, save_dir: Path = Path("."), plot: bool = False, names: tuple = ()) -> None:
+    def __init__(self, save_dir: Path = Path("."), plot: bool = False, names: Dict[int, str] = {}) -> None:
         """
         Initialize a SegmentMetrics instance with a save directory, plot flag, and class names.
 
         Args:
             save_dir (Path, optional): Directory to save plots.
             plot (bool, optional): Whether to plot precision-recall curves.
-            names (tuple, optional): Tuple mapping class indices to names.
+            names (Dict[int, str], optional): Dictionary of class names.
         """
         self.save_dir = save_dir
         self.plot = plot
@@ -1196,10 +1195,10 @@ class PoseMetrics(SegmentMetrics):
     Attributes:
         save_dir (Path): Path to the directory where the output plots should be saved.
         plot (bool): Whether to save the detection and pose plots.
-        names (dict): Dictionary of class names.
-        box (Metric): An instance of the Metric class to calculate box detection metrics.
+        names (Dict[int, str]): Dictionary of class names.
         pose (Metric): An instance of the Metric class to calculate pose metrics.
-        speed (dict): Dictionary to store the time taken in different phases of inference.
+        box (Metric): An instance of the Metric class for storing detection results.
+        speed (Dict[str, float]): A dictionary for storing execution times of different parts of the detection process.
         task (str): The task type, set to 'pose'.
 
     Methods:
@@ -1212,14 +1211,14 @@ class PoseMetrics(SegmentMetrics):
         results_dict: Return the dictionary containing all the detection and segmentation metrics and fitness score.
     """
 
-    def __init__(self, save_dir: Path = Path("."), plot: bool = False, names: tuple = ()) -> None:
+    def __init__(self, save_dir: Path = Path("."), plot: bool = False, names: Dict[int, str] = {}) -> None:
         """
         Initialize the PoseMetrics class with directory path, class names, and plotting options.
 
         Args:
             save_dir (Path, optional): Directory to save plots.
             plot (bool, optional): Whether to plot precision-recall curves.
-            names (tuple, optional): Tuple mapping class indices to names.
+            names (Dict[int, str], optional): Dictionary of class names.
         """
         super().__init__(save_dir, plot, names)
         self.save_dir = save_dir
@@ -1420,23 +1419,23 @@ class OBBMetrics(SimpleClass, DataExportMixin):
     Attributes:
         save_dir (Path): Path to the directory where the output plots should be saved.
         plot (bool): Whether to save the detection plots.
-        names (dict): Dictionary of class names.
+        names (Dict[int, str]): Dictionary of class names.
         box (Metric): An instance of the Metric class for storing detection results.
-        speed (dict): A dictionary for storing execution times of different parts of the detection process.
+        speed (Dict[str, float]): A dictionary for storing execution times of different parts of the detection process.
         task (str): The task type, set to 'obb'.
 
     References:
         https://arxiv.org/pdf/2106.06072.pdf
     """
 
-    def __init__(self, save_dir: Path = Path("."), plot: bool = False, names: tuple = ()) -> None:
+    def __init__(self, save_dir: Path = Path("."), plot: bool = False, names: Dict[int, str] = {}) -> None:
         """
         Initialize an OBBMetrics instance with directory, plotting, and class names.
 
         Args:
             save_dir (Path, optional): Directory to save plots.
             plot (bool, optional): Whether to plot precision-recall curves.
-            names (tuple, optional): Tuple mapping class indices to names.
+            names (Dict[int, str], optional): Dictionary of class names.
         """
         self.save_dir = save_dir
         self.plot = plot

ultralytics/utils/plotting.py

@@ -201,6 +201,11 @@ class Annotator:
         input_is_pil = isinstance(im, Image.Image)
         self.pil = pil or non_ascii or input_is_pil
         self.lw = line_width or max(round(sum(im.size if input_is_pil else im.shape) / 2 * 0.003), 2)
+        if not input_is_pil:
+            if im.shape[2] == 1:  # handle grayscale
+                im = cv2.cvtColor(im, cv2.COLOR_GRAY2BGR)
+            elif im.shape[2] > 3:  # multispectral
+                im = np.ascontiguousarray(im[..., :3])
         if self.pil:  # use PIL
             self.im = im if input_is_pil else Image.fromarray(im)
             if self.im.mode not in {"RGB", "RGBA"}:  # multispectral
@@ -216,10 +221,6 @@ class Annotator:
             if check_version(pil_version, "9.2.0"):
                 self.font.getsize = lambda x: self.font.getbbox(x)[2:4]  # text width, height
         else:  # use cv2
-            if im.shape[2] == 1:  # handle grayscale
-                im = cv2.cvtColor(im, cv2.COLOR_GRAY2BGR)
-            elif im.shape[2] > 3:  # multispectral
-                im = np.ascontiguousarray(im[..., :3])
             assert im.data.contiguous, "Image not contiguous. Apply np.ascontiguousarray(im) to Annotator input images."
             self.im = im if im.flags.writeable else im.copy()
             self.tf = max(self.lw - 1, 1)  # font thickness
@@ -644,7 +645,7 @@ def save_one_box(
         gain (float, optional): A multiplicative factor to increase the size of the bounding box.
         pad (int, optional): The number of pixels to add to the width and height of the bounding box.
         square (bool, optional): If True, the bounding box will be transformed into a square.
-        BGR (bool, optional): If True, the image will be saved in BGR format, otherwise in RGB.
+        BGR (bool, optional): If True, the image will be returned in BGR format, otherwise in RGB.
         save (bool, optional): If True, the cropped image will be saved to disk.
 
     Returns:
@@ -664,12 +665,14 @@ def save_one_box(
     b[:, 2:] = b[:, 2:] * gain + pad  # box wh * gain + pad
     xyxy = ops.xywh2xyxy(b).long()
     xyxy = ops.clip_boxes(xyxy, im.shape)
-    crop = im[int(xyxy[0, 1]) : int(xyxy[0, 3]), int(xyxy[0, 0]) : int(xyxy[0, 2]), :: (1 if BGR else -1)]
+    grayscale = im.shape[2] == 1  # grayscale image
+    crop = im[int(xyxy[0, 1]) : int(xyxy[0, 3]), int(xyxy[0, 0]) : int(xyxy[0, 2]), :: (1 if BGR or grayscale else -1)]
     if save:
         file.parent.mkdir(parents=True, exist_ok=True)  # make directory
         f = str(increment_path(file).with_suffix(".jpg"))
         # cv2.imwrite(f, crop)  # save BGR, https://github.com/ultralytics/yolov5/issues/7007 chroma subsampling issue
-        Image.fromarray(crop[..., ::-1]).save(f, quality=95, subsampling=0)  # save RGB
+        crop = crop.squeeze(-1) if grayscale else crop[..., ::-1] if BGR else crop
+        Image.fromarray(crop).save(f, quality=95, subsampling=0)  # save RGB
     return crop
 
 

ultralytics/utils/torch_utils.py

@@ -10,7 +10,7 @@ from contextlib import contextmanager
 from copy import deepcopy
 from datetime import datetime
 from pathlib import Path
-from typing import Union
+from typing import Any, Dict, Union
 
 import numpy as np
 import torch
@@ -704,7 +704,7 @@ class ModelEMA:
             copy_attr(self.ema, model, include, exclude)
 
 
-def strip_optimizer(f: Union[str, Path] = "best.pt", s: str = "", updates: dict = None) -> dict:
+def strip_optimizer(f: Union[str, Path] = "best.pt", s: str = "", updates: Dict[str, Any] = None) -> Dict[str, Any]:
     """
     Strip optimizer from 'f' to finalize training, optionally save as 's'.