ultralytics 8.3.53__py3-none-any.whl → 8.3.55__py3-none-any.whl

tests/__init__.py

@@ -17,7 +17,6 @@ __all__ = (
     "SOURCE",
     "SOURCES_LIST",
     "TMP",
-    "IS_TMP_WRITEABLE",
     "CUDA_IS_AVAILABLE",
     "CUDA_DEVICE_COUNT",
 )

tests/conftest.py

@@ -74,10 +74,10 @@ def pytest_terminal_summary(terminalreporter, exitstatus, config):
 
     # Remove files
     models = [path for x in ["*.onnx", "*.torchscript"] for path in WEIGHTS_DIR.rglob(x)]
-    for file in ["bus.jpg", "yolo11n.onnx", "yolo11n.torchscript"] + models:
+    for file in ["decelera_portrait_min.mov", "bus.jpg", "yolo11n.onnx", "yolo11n.torchscript"] + models:
         Path(file).unlink(missing_ok=True)
 
     # Remove directories
     models = [path for x in ["*.mlpackage", "*_openvino_model"] for path in WEIGHTS_DIR.rglob(x)]
-    for directory in [TMP.parents[1] / ".pytest_cache", TMP] + models:
+    for directory in [WEIGHTS_DIR / "path with spaces", TMP.parents[1] / ".pytest_cache", TMP] + models:
         shutil.rmtree(directory, ignore_errors=True)

tests/test_cli.py

@@ -59,7 +59,8 @@ def test_rtdetr(task="detect", model="yolov8n-rtdetr.yaml", data="coco8.yaml"):
     run(f"yolo train {task} model={model} data={data} --imgsz= 160 epochs =1, cache = disk fraction=0.25")
     run(f"yolo predict {task} model={model} source={ASSETS / 'bus.jpg'} imgsz=160 save save_crop save_txt")
     if TORCH_1_9:
-        run(f"yolo predict {task} model='rtdetr-l.pt' source={ASSETS / 'bus.jpg'} imgsz=160 save save_crop save_txt")
+        weights = WEIGHTS_DIR / "rtdetr-l.pt"
+        run(f"yolo predict {task} model={weights} source={ASSETS / 'bus.jpg'} imgsz=160 save save_crop save_txt")
 
 
 @pytest.mark.skipif(checks.IS_PYTHON_3_12, reason="MobileSAM with CLIP is not supported in Python 3.12")

tests/test_python.py

@@ -576,11 +576,11 @@ def test_model_embeddings():
 @pytest.mark.skipif(checks.IS_PYTHON_3_12, reason="YOLOWorld with CLIP is not supported in Python 3.12")
 def test_yolo_world():
     """Tests YOLO world models with CLIP support, including detection and training scenarios."""
-    model = YOLO("yolov8s-world.pt")  # no YOLO11n-world model yet
+    model = YOLO(WEIGHTS_DIR / "yolov8s-world.pt")  # no YOLO11n-world model yet
     model.set_classes(["tree", "window"])
     model(SOURCE, conf=0.01)
 
-    model = YOLO("yolov8s-worldv2.pt")  # no YOLO11n-world model yet
+    model = YOLO(WEIGHTS_DIR / "yolov8s-worldv2.pt")  # no YOLO11n-world model yet
     # Training from a pretrained model. Eval is included at the final stage of training.
     # Use dota8.yaml which has fewer categories to reduce the inference time of CLIP model
     model.train(

tests/test_solutions.py

@@ -3,18 +3,20 @@
 import cv2
 import pytest
 
+from tests import TMP
 from ultralytics import YOLO, solutions
+from ultralytics.utils import ASSETS_URL, WEIGHTS_DIR
 from ultralytics.utils.downloads import safe_download
 
-MAJOR_SOLUTIONS_DEMO = "https://github.com/ultralytics/assets/releases/download/v0.0.0/solutions_ci_demo.mp4"
-WORKOUTS_SOLUTION_DEMO = "https://github.com/ultralytics/assets/releases/download/v0.0.0/solution_ci_pose_demo.mp4"
+DEMO_VIDEO = "solutions_ci_demo.mp4"
+POSE_VIDEO = "solution_ci_pose_demo.mp4"
 
 
 @pytest.mark.slow
 def test_major_solutions():
     """Test the object counting, heatmap, speed estimation and queue management solution."""
-    safe_download(url=MAJOR_SOLUTIONS_DEMO)
-    cap = cv2.VideoCapture("solutions_ci_demo.mp4")
+    safe_download(url=f"{ASSETS_URL}/{DEMO_VIDEO}", dir=TMP)
+    cap = cv2.VideoCapture(str(TMP / DEMO_VIDEO))
     assert cap.isOpened(), "Error reading video file"
     region_points = [(20, 400), (1080, 400), (1080, 360), (20, 360)]
     counter = solutions.ObjectCounter(region=region_points, model="yolo11n.pt", show=False)  # Test object counter
@@ -42,8 +44,8 @@ def test_major_solutions():
     cap.release()
 
     # Test workouts monitoring
-    safe_download(url=WORKOUTS_SOLUTION_DEMO)
-    cap1 = cv2.VideoCapture("solution_ci_pose_demo.mp4")
+    safe_download(url=f"{ASSETS_URL}/{POSE_VIDEO}", dir=TMP)
+    cap1 = cv2.VideoCapture(str(TMP / POSE_VIDEO))
     assert cap1.isOpened(), "Error reading video file"
     gym = solutions.AIGym(line_width=2, kpts=[5, 11, 13], show=False)
     while cap1.isOpened():
@@ -59,9 +61,9 @@ def test_instance_segmentation():
     """Test the instance segmentation solution."""
     from ultralytics.utils.plotting import Annotator, colors
 
-    model = YOLO("yolo11n-seg.pt")
+    model = YOLO(WEIGHTS_DIR / "yolo11n-seg.pt")
     names = model.names
-    cap = cv2.VideoCapture("solutions_ci_demo.mp4")
+    cap = cv2.VideoCapture(TMP / DEMO_VIDEO)
     assert cap.isOpened(), "Error reading video file"
     while cap.isOpened():
         success, im0 = cap.read()
@@ -82,4 +84,4 @@ def test_instance_segmentation():
 @pytest.mark.slow
 def test_streamlit_predict():
     """Test streamlit predict live inference solution."""
-    solutions.inference()
+    solutions.Inference().inference()

ultralytics/__init__.py

@@ -1,6 +1,6 @@
 # Ultralytics YOLO 🚀, AGPL-3.0 license
 
-__version__ = "8.3.53"
+__version__ = "8.3.55"
 
 import os
 

ultralytics/cfg/__init__.py

@@ -42,6 +42,7 @@ SOLUTION_MAP = {
     "workout": ("AIGym", "monitor"),
     "analytics": ("Analytics", "process_data"),
     "trackzone": ("TrackZone", "trackzone"),
+    "inference": ("Inference", "inference"),
     "help": None,
 }
 
@@ -85,7 +86,7 @@ SOLUTIONS_HELP_MSG = f"""
         yolo solutions count source="path/to/video/file.mp4" region=[(20, 400), (1080, 400), (1080, 360), (20, 360)]
 
     2. Call heatmaps solution
-        yolo solutions heatmap colormap=cv2.COLORMAP_PARAULA model=yolo11n.pt
+        yolo solutions heatmap colormap=cv2.COLORMAP_PARULA model=yolo11n.pt
 
     3. Call queue management solution
         yolo solutions queue region=[(20, 400), (1080, 400), (1080, 360), (20, 360)] model=yolo11n.pt
@@ -97,7 +98,10 @@ SOLUTIONS_HELP_MSG = f"""
         yolo solutions analytics analytics_type="pie"
     
     6. Track objects within specific zones
-        yolo solutions trackzone source="path/to/video/file.mp4" region=[(150, 150), (1130, 150), (1130, 570), (150, 570)] 
+        yolo solutions trackzone source="path/to/video/file.mp4" region=[(150, 150), (1130, 150), (1130, 570), (150, 570)]
+        
+    7. Streamlit real-time webcam inference GUI
+        yolo streamlit-predict
     """
 CLI_HELP_MSG = f"""
     Arguments received: {str(['yolo'] + ARGV[1:])}. Ultralytics 'yolo' commands use the following syntax:
@@ -121,13 +125,10 @@ CLI_HELP_MSG = f"""
     4. Export a YOLO11n classification model to ONNX format at image size 224 by 128 (no TASK required)
         yolo export model=yolo11n-cls.pt format=onnx imgsz=224,128
 
-    5. Streamlit real-time webcam inference GUI
-        yolo streamlit-predict
-
-    6. Ultralytics solutions usage
+    5. Ultralytics solutions usage
         yolo solutions count or in {list(SOLUTION_MAP.keys())[1:-1]} source="path/to/video/file.mp4"
 
-    7. Run special commands:
+    6. Run special commands:
         yolo help
         yolo checks
         yolo version
@@ -636,6 +637,9 @@ def handle_yolo_solutions(args: List[str]) -> None:
         Run analytics with custom configuration:
         >>> handle_yolo_solutions(["analytics", "conf=0.25", "source=path/to/video/file.mp4"])
 
+        Run inference with custom configuration, requires Streamlit version 1.29.0 or higher.
+        >>> handle_yolo_solutions(["inference", "model=yolo11n.pt"])
+
     Notes:
         - Default configurations are merged from DEFAULT_SOL_DICT and DEFAULT_CFG_DICT
         - Arguments can be provided in the format 'key=value' or as boolean flags
@@ -645,7 +649,9 @@ def handle_yolo_solutions(args: List[str]) -> None:
         - For 'analytics' solution, frame numbers are tracked for generating analytical graphs
         - Video processing can be interrupted by pressing 'q'
         - Processes video frames sequentially and saves output in .avi format
-        - If no source is specified, downloads and uses a default sample video
+        - If no source is specified, downloads and uses a default sample video\
+        - The inference solution will be launched using the 'streamlit run' command.
+        - The Streamlit app file is located in the Ultralytics package directory.
     """
     full_args_dict = {**DEFAULT_SOL_DICT, **DEFAULT_CFG_DICT}  # arguments dictionary
     overrides = {}
@@ -678,60 +684,56 @@ def handle_yolo_solutions(args: List[str]) -> None:
     if args and args[0] == "help":  # Add check for return if user call `yolo solutions help`
         return
 
-    cls, method = SOLUTION_MAP[s_n]  # solution class name, method name and default source
-
-    from ultralytics import solutions  # import ultralytics solutions
-
-    solution = getattr(solutions, cls)(IS_CLI=True, **overrides)  # get solution class i.e ObjectCounter
-    process = getattr(solution, method)  # get specific function of class for processing i.e, count from ObjectCounter
-
-    cap = cv2.VideoCapture(solution.CFG["source"])  # read the video file
-
-    # extract width, height and fps of the video file, create save directory and initialize video writer
-    import os  # for directory creation
-    from pathlib import Path
-
-    from ultralytics.utils.files import increment_path  # for output directory path update
+    if s_n == "inference":
+        checks.check_requirements("streamlit>=1.29.0")
+        LOGGER.info("💡 Loading Ultralytics live inference app...")
+        subprocess.run(
+            [  # Run subprocess with Streamlit custom argument
+                "streamlit",
+                "run",
+                str(ROOT / "solutions/streamlit_inference.py"),
+                "--server.headless",
+                "true",
+                overrides.pop("model", "yolo11n.pt"),
+            ]
+        )
+    else:
+        cls, method = SOLUTION_MAP[s_n]  # solution class name, method name and default source
 
-    w, h, fps = (int(cap.get(x)) for x in (cv2.CAP_PROP_FRAME_WIDTH, cv2.CAP_PROP_FRAME_HEIGHT, cv2.CAP_PROP_FPS))
-    if s_n == "analytics":  # analytical graphs follow fixed shape for output i.e w=1920, h=1080
-        w, h = 1920, 1080
-    save_dir = increment_path(Path("runs") / "solutions" / "exp", exist_ok=False)
-    save_dir.mkdir(parents=True, exist_ok=True)  # create the output directory
-    vw = cv2.VideoWriter(os.path.join(save_dir, "solution.avi"), cv2.VideoWriter_fourcc(*"mp4v"), fps, (w, h))
+        from ultralytics import solutions  # import ultralytics solutions
 
-    try:  # Process video frames
-        f_n = 0  # frame number, required for analytical graphs
-        while cap.isOpened():
-            success, frame = cap.read()
-            if not success:
-                break
-            frame = process(frame, f_n := f_n + 1) if s_n == "analytics" else process(frame)
-            vw.write(frame)
-            if cv2.waitKey(1) & 0xFF == ord("q"):
-                break
-    finally:
-        cap.release()
+        solution = getattr(solutions, cls)(IS_CLI=True, **overrides)  # get solution class i.e ObjectCounter
+        process = getattr(
+            solution, method
+        )  # get specific function of class for processing i.e, count from ObjectCounter
 
+        cap = cv2.VideoCapture(solution.CFG["source"])  # read the video file
 
-def handle_streamlit_inference():
-    """
-    Open the Ultralytics Live Inference Streamlit app for real-time object detection.
+        # extract width, height and fps of the video file, create save directory and initialize video writer
+        import os  # for directory creation
+        from pathlib import Path
 
-    This function initializes and runs a Streamlit application designed for performing live object detection using
-    Ultralytics models. It checks for the required Streamlit package and launches the app.
+        from ultralytics.utils.files import increment_path  # for output directory path update
 
-    Examples:
-        >>> handle_streamlit_inference()
+        w, h, fps = (int(cap.get(x)) for x in (cv2.CAP_PROP_FRAME_WIDTH, cv2.CAP_PROP_FRAME_HEIGHT, cv2.CAP_PROP_FPS))
+        if s_n == "analytics":  # analytical graphs follow fixed shape for output i.e w=1920, h=1080
+            w, h = 1920, 1080
+        save_dir = increment_path(Path("runs") / "solutions" / "exp", exist_ok=False)
+        save_dir.mkdir(parents=True, exist_ok=True)  # create the output directory
+        vw = cv2.VideoWriter(os.path.join(save_dir, "solution.avi"), cv2.VideoWriter_fourcc(*"mp4v"), fps, (w, h))
 
-    Notes:
-        - Requires Streamlit version 1.29.0 or higher.
-        - The app is launched using the 'streamlit run' command.
-        - The Streamlit app file is located in the Ultralytics package directory.
-    """
-    checks.check_requirements("streamlit>=1.29.0")
-    LOGGER.info("💡 Loading Ultralytics Live Inference app...")
-    subprocess.run(["streamlit", "run", ROOT / "solutions/streamlit_inference.py", "--server.headless", "true"])
+        try:  # Process video frames
+            f_n = 0  # frame number, required for analytical graphs
+            while cap.isOpened():
+                success, frame = cap.read()
+                if not success:
+                    break
+                frame = process(frame, f_n := f_n + 1) if s_n == "analytics" else process(frame)
+                vw.write(frame)
+                if cv2.waitKey(1) & 0xFF == ord("q"):
+                    break
+        finally:
+            cap.release()
 
 
 def parse_key_value_pair(pair: str = "key=value"):
@@ -853,7 +855,6 @@ def entrypoint(debug=""):
         "login": lambda: handle_yolo_hub(args),
         "logout": lambda: handle_yolo_hub(args),
         "copy-cfg": copy_default_cfg,
-        "streamlit-predict": lambda: handle_streamlit_inference(),
         "solutions": lambda: handle_yolo_solutions(args[1:]),
     }
     full_args_dict = {**DEFAULT_CFG_DICT, **{k: None for k in TASKS}, **{k: None for k in MODES}, **special}

ultralytics/cfg/datasets/coco-pose.yaml

@@ -1,5 +1,5 @@
 # Ultralytics YOLO 🚀, AGPL-3.0 license
-# COCO 2017 dataset https://cocodataset.org by Microsoft
+# COCO 2017 Keypoints dataset https://cocodataset.org by Microsoft
 # Documentation: https://docs.ultralytics.com/datasets/pose/coco/
 # Example usage: yolo train data=coco-pose.yaml
 # parent
@@ -9,9 +9,9 @@
 
 # Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
 path: ../datasets/coco-pose # dataset root dir
-train: train2017.txt # train images (relative to 'path') 118287 images
-val: val2017.txt # val images (relative to 'path') 5000 images
-test: test-dev2017.txt # 20288 of 40670 images, submit to https://competitions.codalab.org/competitions/20794
+train: train2017.txt # train images (relative to 'path') 56599 images
+val: val2017.txt # val images (relative to 'path') 2346 images
+test: test-dev2017.txt # 20288 of 40670 images, submit to https://codalab.lisn.upsaclay.fr/competitions/7403
 
 # Keypoints
 kpt_shape: [17, 3] # number of keypoints, number of dims (2 for x,y or 3 for x,y,visible)

ultralytics/cfg/datasets/lvis.yaml

@@ -11,7 +11,7 @@
 path: ../datasets/lvis # dataset root dir
 train: train.txt # train images (relative to 'path') 100170 images
 val: val.txt # val images (relative to 'path') 19809 images
-minival: minival.txt # minval images (relative to 'path') 5000 images
+minival: minival.txt # minival images (relative to 'path') 5000 images
 
 names:
   0: aerosol can/spray can

ultralytics/cfg/datasets/medical-pills.yaml

@@ -0,0 +1,21 @@
+# Ultralytics YOLO 🚀, AGPL-3.0 license
+# Medical-pills dataset by Ultralytics
+# Documentation: https://docs.ultralytics.com/datasets/detect/medical-pills/
+# Example usage: yolo train data=medical-pills.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── medical-pills  ← downloads here (8.19 MB)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/medical-pills # dataset root dir
+train: train/images # train images (relative to 'path') 92 images
+val: valid/images # val images (relative to 'path') 23 images
+test: # test images (relative to 'path')
+
+# Classes
+names:
+  0: pill
+
+# Download script/URL (optional)
+download: https://github.com/ultralytics/assets/releases/download/v0.0.0/medical-pills.zip

ultralytics/cfg/solutions/default.yaml

@@ -12,7 +12,7 @@ colormap: #  (int | str) colormap for heatmap, Only OPENCV supported colormaps c
 # 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].
+kpts: [6, 8, 10] # (list[int]) keypoints for workouts monitoring, i.e. for push-ups kpts have values of [6, 8, 10].
 
 # Analytics settings ---------------------------------------------------------------------------------------------------
 analytics_type: "line" # (str) analytics type i.e "line", "pie", "bar" or "area" charts.

ultralytics/data/augment.py

@@ -441,7 +441,8 @@ class BaseMixTransform:
         """
         raise NotImplementedError
 
-    def _update_label_text(self, labels):
+    @staticmethod
+    def _update_label_text(labels):
         """
         Updates label text and class IDs for mixed labels in image augmentation.
 
@@ -1259,7 +1260,8 @@ class RandomPerspective:
         labels["resized_shape"] = img.shape[:2]
         return labels
 
-    def box_candidates(self, box1, box2, wh_thr=2, ar_thr=100, area_thr=0.1, eps=1e-16):
+    @staticmethod
+    def box_candidates(box1, box2, wh_thr=2, ar_thr=100, area_thr=0.1, eps=1e-16):
         """
         Compute candidate boxes for further processing based on size and aspect ratio criteria.
 
@@ -1598,7 +1600,8 @@ class LetterBox:
         else:
             return img
 
-    def _update_labels(self, labels, ratio, padw, padh):
+    @staticmethod
+    def _update_labels(labels, ratio, padw, padh):
         """
         Updates labels after applying letterboxing to an image.
 

ultralytics/data/dataset.py

@@ -68,7 +68,7 @@ class YOLODataset(BaseDataset):
         Cache dataset labels, check images and read shapes.
 
         Args:
-            path (Path): Path where to save the cache file. Default is Path('./labels.cache').
+            path (Path): Path where to save the cache file. Default is Path("./labels.cache").
 
         Returns:
             (dict): labels.
@@ -219,7 +219,7 @@ class YOLODataset(BaseDataset):
         segment_resamples = 100 if self.use_obb else 1000
         if len(segments) > 0:
             # make sure segments interpolate correctly if original length is greater than segment_resamples
-            max_len = max([len(s) for s in segments])
+            max_len = max(len(s) for s in segments)
             segment_resamples = (max_len + 1) if segment_resamples < max_len else segment_resamples
             # list[np.array(segment_resamples, 2)] * num_samples
             segments = np.stack(resample_segments(segments, n=segment_resamples), axis=0)

ultralytics/engine/exporter.py

@@ -102,19 +102,19 @@ def export_formats():
     """Ultralytics YOLO export formats."""
     x = [
         ["PyTorch", "-", ".pt", True, True, []],
-        ["TorchScript", "torchscript", ".torchscript", True, True, ["optimize", "batch"]],
-        ["ONNX", "onnx", ".onnx", True, True, ["half", "dynamic", "simplify", "opset", "batch"]],
-        ["OpenVINO", "openvino", "_openvino_model", True, False, ["half", "int8", "batch"]],
-        ["TensorRT", "engine", ".engine", False, True, ["half", "dynamic", "simplify", "int8", "batch"]],
-        ["CoreML", "coreml", ".mlpackage", True, False, ["half", "int8", "nms", "batch"]],
-        ["TensorFlow SavedModel", "saved_model", "_saved_model", True, True, ["keras", "int8", "batch"]],
+        ["TorchScript", "torchscript", ".torchscript", True, True, ["batch", "optimize"]],
+        ["ONNX", "onnx", ".onnx", True, True, ["batch", "dynamic", "half", "opset", "simplify"]],
+        ["OpenVINO", "openvino", "_openvino_model", True, False, ["batch", "dynamic", "half", "int8"]],
+        ["TensorRT", "engine", ".engine", False, True, ["batch", "dynamic", "half", "int8", "simplify"]],
+        ["CoreML", "coreml", ".mlpackage", True, False, ["batch", "half", "int8", "nms"]],
+        ["TensorFlow SavedModel", "saved_model", "_saved_model", True, True, ["batch", "int8", "keras"]],
         ["TensorFlow GraphDef", "pb", ".pb", True, True, ["batch"]],
-        ["TensorFlow Lite", "tflite", ".tflite", True, False, ["half", "int8", "batch"]],
+        ["TensorFlow Lite", "tflite", ".tflite", True, False, ["batch", "half", "int8"]],
         ["TensorFlow Edge TPU", "edgetpu", "_edgetpu.tflite", True, False, []],
-        ["TensorFlow.js", "tfjs", "_web_model", True, False, ["half", "int8", "batch"]],
+        ["TensorFlow.js", "tfjs", "_web_model", True, False, ["batch", "half", "int8"]],
         ["PaddlePaddle", "paddle", "_paddle_model", True, True, ["batch"]],
-        ["MNN", "mnn", ".mnn", True, True, ["batch", "int8", "half"]],
-        ["NCNN", "ncnn", "_ncnn_model", True, True, ["half", "batch"]],
+        ["MNN", "mnn", ".mnn", True, True, ["batch", "half", "int8"]],
+        ["NCNN", "ncnn", "_ncnn_model", True, True, ["batch", "half"]],
         ["IMX", "imx", "_imx_model", True, True, ["int8"]],
     ]
     return dict(zip(["Format", "Argument", "Suffix", "CPU", "GPU", "Arguments"], zip(*x)))
@@ -813,7 +813,7 @@ class Exporter:
         workspace = int(self.args.workspace * (1 << 30)) if self.args.workspace is not None else 0
         if is_trt10 and workspace > 0:
             config.set_memory_pool_limit(trt.MemoryPoolType.WORKSPACE, workspace)
-        elif workspace > 0 and not is_trt10:  # TensorRT versions 7, 8
+        elif workspace > 0:  # TensorRT versions 7, 8
             config.max_workspace_size = workspace
         flag = 1 << int(trt.NetworkDefinitionCreationFlag.EXPLICIT_BATCH)
         network = builder.create_network(flag)

ultralytics/engine/model.py

@@ -2,7 +2,7 @@
 
 import inspect
 from pathlib import Path
-from typing import Dict, List, Union
+from typing import Any, Dict, List, Union
 
 import numpy as np
 import torch
@@ -152,7 +152,7 @@ class Model(nn.Module):
         self,
         source: Union[str, Path, int, Image.Image, list, tuple, np.ndarray, torch.Tensor] = None,
         stream: bool = False,
-        **kwargs,
+        **kwargs: Any,
     ) -> list:
         """
         Alias for the predict method, enabling the model instance to be callable for predictions.
@@ -165,7 +165,7 @@ class Model(nn.Module):
                 the image(s) to make predictions on. Can be a file path, URL, PIL image, numpy array, PyTorch
                 tensor, or a list/tuple of these.
             stream (bool): If True, treat the input source as a continuous stream for predictions.
-            **kwargs (Any): Additional keyword arguments to configure the prediction process.
+            **kwargs: Additional keyword arguments to configure the prediction process.
 
         Returns:
             (List[ultralytics.engine.results.Results]): A list of prediction results, each encapsulated in a
@@ -466,7 +466,7 @@ class Model(nn.Module):
         self,
         source: Union[str, Path, int, list, tuple, np.ndarray, torch.Tensor] = None,
         stream: bool = False,
-        **kwargs,
+        **kwargs: Any,
     ) -> list:
         """
         Generates image embeddings based on the provided source.
@@ -478,7 +478,7 @@ class Model(nn.Module):
             source (str | Path | int | List | Tuple | np.ndarray | torch.Tensor): The source of the image for
                 generating embeddings. Can be a file path, URL, PIL image, numpy array, etc.
             stream (bool): If True, predictions are streamed.
-            **kwargs (Any): Additional keyword arguments for configuring the embedding process.
+            **kwargs: Additional keyword arguments for configuring the embedding process.
 
         Returns:
             (List[torch.Tensor]): A list containing the image embeddings.
@@ -501,7 +501,7 @@ class Model(nn.Module):
         source: Union[str, Path, int, Image.Image, list, tuple, np.ndarray, torch.Tensor] = None,
         stream: bool = False,
         predictor=None,
-        **kwargs,
+        **kwargs: Any,
     ) -> List[Results]:
         """
         Performs predictions on the given image source using the YOLO model.
@@ -517,7 +517,7 @@ class Model(nn.Module):
             stream (bool): If True, treats the input source as a continuous stream for predictions.
             predictor (BasePredictor | None): An instance of a custom predictor class for making predictions.
                 If None, the method uses a default predictor.
-            **kwargs (Any): Additional keyword arguments for configuring the prediction process.
+            **kwargs: Additional keyword arguments for configuring the prediction process.
 
         Returns:
             (List[ultralytics.engine.results.Results]): A list of prediction results, each encapsulated in a
@@ -562,7 +562,7 @@ class Model(nn.Module):
         source: Union[str, Path, int, list, tuple, np.ndarray, torch.Tensor] = None,
         stream: bool = False,
         persist: bool = False,
-        **kwargs,
+        **kwargs: Any,
     ) -> List[Results]:
         """
         Conducts object tracking on the specified input source using the registered trackers.
@@ -576,7 +576,7 @@ class Model(nn.Module):
                 tracking. Can be a file path, URL, or video stream.
             stream (bool): If True, treats the input source as a continuous video stream. Defaults to False.
             persist (bool): If True, persists trackers between different calls to this method. Defaults to False.
-            **kwargs (Any): Additional keyword arguments for configuring the tracking process.
+            **kwargs: Additional keyword arguments for configuring the tracking process.
 
         Returns:
             (List[ultralytics.engine.results.Results]): A list of tracking results, each a Results object.
@@ -607,7 +607,7 @@ class Model(nn.Module):
     def val(
         self,
         validator=None,
-        **kwargs,
+        **kwargs: Any,
     ):
         """
         Validates the model using a specified dataset and validation configuration.
@@ -619,7 +619,7 @@ class Model(nn.Module):
         Args:
             validator (ultralytics.engine.validator.BaseValidator | None): An instance of a custom validator class for
                 validating the model.
-            **kwargs (Any): Arbitrary keyword arguments for customizing the validation process.
+            **kwargs: Arbitrary keyword arguments for customizing the validation process.
 
         Returns:
             (ultralytics.utils.metrics.DetMetrics): Validation metrics obtained from the validation process.
@@ -642,7 +642,7 @@ class Model(nn.Module):
 
     def benchmark(
         self,
-        **kwargs,
+        **kwargs: Any,
     ):
         """
         Benchmarks the model across various export formats to evaluate performance.
@@ -653,7 +653,7 @@ class Model(nn.Module):
         defaults, and any additional user-provided keyword arguments.
 
         Args:
-            **kwargs (Any): Arbitrary keyword arguments to customize the benchmarking process. These are combined with
+            **kwargs: Arbitrary keyword arguments to customize the benchmarking process. These are combined with
                 default configurations, model-specific arguments, and method defaults. Common options include:
                 - data (str): Path to the dataset for benchmarking.
                 - imgsz (int | List[int]): Image size for benchmarking.
@@ -691,7 +691,7 @@ class Model(nn.Module):
 
     def export(
         self,
-        **kwargs,
+        **kwargs: Any,
     ) -> str:
         """
         Exports the model to a different format suitable for deployment.
@@ -701,7 +701,7 @@ class Model(nn.Module):
         defaults, and any additional arguments provided.
 
         Args:
-            **kwargs (Dict): Arbitrary keyword arguments to customize the export process. These are combined with
+            **kwargs: Arbitrary keyword arguments to customize the export process. These are combined with
                 the model's overrides and method defaults. Common arguments include:
                 format (str): Export format (e.g., 'onnx', 'engine', 'coreml').
                 half (bool): Export model in half-precision.
@@ -740,7 +740,7 @@ class Model(nn.Module):
     def train(
         self,
         trainer=None,
-        **kwargs,
+        **kwargs: Any,
     ):
         """
         Trains the model using the specified dataset and training configuration.
@@ -755,7 +755,7 @@ class Model(nn.Module):
 
         Args:
             trainer (BaseTrainer | None): Custom trainer instance for model training. If None, uses default.
-            **kwargs (Any): Arbitrary keyword arguments for training configuration. Common options include:
+            **kwargs: Arbitrary keyword arguments for training configuration. Common options include:
                 data (str): Path to dataset configuration file.
                 epochs (int): Number of training epochs.
                 batch_size (int): Batch size for training.
@@ -816,8 +816,8 @@ class Model(nn.Module):
         self,
         use_ray=False,
         iterations=10,
-        *args,
-        **kwargs,
+        *args: Any,
+        **kwargs: Any,
     ):
         """
         Conducts hyperparameter tuning for the model, with an option to use Ray Tune.
@@ -830,8 +830,8 @@ class Model(nn.Module):
         Args:
             use_ray (bool): If True, uses Ray Tune for hyperparameter tuning. Defaults to False.
             iterations (int): The number of tuning iterations to perform. Defaults to 10.
-            *args (List): Variable length argument list for additional arguments.
-            **kwargs (Dict): Arbitrary keyword arguments. These are combined with the model's overrides and defaults.
+            *args: Variable length argument list for additional arguments.
+            **kwargs: Arbitrary keyword arguments. These are combined with the model's overrides and defaults.
 
         Returns:
             (Dict): A dictionary containing the results of the hyperparameter search.
@@ -1170,6 +1170,4 @@ class Model(nn.Module):
             >>> print(model.stride)
             >>> print(model.task)
         """
-        if name == "model":
-            return self._modules["model"]
-        return getattr(self.model, name)
+        return self._modules["model"] if name == "model" else getattr(self.model, name)

ultralytics/engine/validator.py

@@ -245,7 +245,7 @@ class BaseValidator:
 
                 cost_matrix = iou * (iou >= threshold)
                 if cost_matrix.any():
-                    labels_idx, detections_idx = scipy.optimize.linear_sum_assignment(cost_matrix, maximize=True)
+                    labels_idx, detections_idx = scipy.optimize.linear_sum_assignment(cost_matrix)
                     valid = cost_matrix[labels_idx, detections_idx] > 0
                     if valid.any():
                         correct[detections_idx[valid], i] = True

ultralytics/models/sam/modules/tiny_encoder.py

@@ -955,7 +955,8 @@ class TinyViT(nn.Module):
 
         self.apply(_check_lr_scale)
 
-    def _init_weights(self, m):
+    @staticmethod
+    def _init_weights(m):
         """Initializes weights for linear and normalization layers in the TinyViT model."""
         if isinstance(m, nn.Linear):
             # NOTE: This initialization is needed only for training.