ultralytics 8.3.123__py3-none-any.whl → 8.3.125__py3-none-any.whl

tests/test_python.py

@@ -10,7 +10,6 @@ import cv2
 import numpy as np
 import pytest
 import torch
-import yaml
 from PIL import Image
 
 from tests import CFG, MODEL, SOURCE, SOURCES_LIST, TMP
@@ -28,6 +27,7 @@ from ultralytics.utils import (
     ROOT,
     WEIGHTS_DIR,
     WINDOWS,
+    YAML,
     checks,
     is_dir_writeable,
     is_github_action_running,
@@ -190,13 +190,10 @@ def test_track_stream():
 
     # Test Global Motion Compensation (GMC) methods
     for gmc in "orb", "sift", "ecc":
-        with open(ROOT / "cfg/trackers/botsort.yaml", encoding="utf-8") as f:
-            data = yaml.safe_load(f)
-        tracker = TMP / f"botsort-{gmc}.yaml"
-        data["gmc_method"] = gmc
-        with open(tracker, "w", encoding="utf-8") as f:
-            yaml.safe_dump(data, f)
-        model.track(video_url, imgsz=160, tracker=tracker)
+        default_args = YAML.load(ROOT / "cfg/trackers/botsort.yaml")
+        custom_yaml = TMP / f"botsort-{gmc}.yaml"
+        YAML.save(custom_yaml, {**default_args, "gmc_method": gmc})
+        model.track(video_url, imgsz=160, tracker=custom_yaml)
 
 
 def test_val():

ultralytics/__init__.py

@@ -1,6 +1,6 @@
 # Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
 
-__version__ = "8.3.123"
+__version__ = "8.3.125"
 
 import os
 

ultralytics/cfg/__init__.py

@@ -15,7 +15,6 @@ from ultralytics.utils import (
     DEFAULT_CFG,
     DEFAULT_CFG_DICT,
     DEFAULT_CFG_PATH,
-    DEFAULT_SOL_DICT,
     IS_VSCODE,
     LOGGER,
     RANK,
@@ -24,13 +23,12 @@ from ultralytics.utils import (
     SETTINGS,
     SETTINGS_FILE,
     TESTS_RUNNING,
+    YAML,
     IterableSimpleNamespace,
     checks,
     colorstr,
     deprecation_warn,
     vscode_msg,
-    yaml_load,
-    yaml_print,
 )
 
 # Define valid solutions
@@ -271,7 +269,7 @@ def cfg2dict(cfg: Union[str, Path, Dict, SimpleNamespace]) -> Dict:
         - If cfg is already a dictionary, it's returned unchanged.
     """
     if isinstance(cfg, (str, Path)):
-        cfg = yaml_load(cfg)  # load dict
+        cfg = YAML.load(cfg)  # load dict
     elif isinstance(cfg, SimpleNamespace):
         cfg = vars(cfg)  # convert to dict
     return cfg
@@ -650,7 +648,6 @@ def handle_yolo_solutions(args: List[str]) -> None:
         >>> 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
         - Available solutions are defined in SOLUTION_MAP with their respective classes and methods
         - If an invalid solution is provided, defaults to 'count' solution
@@ -662,13 +659,9 @@ def handle_yolo_solutions(args: List[str]) -> None:
         - 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,
-        "blur_ratio": 0.5,
-        "vision_point": (20, 20),
-        "crop_dir": "cropped-detections",
-    }  # arguments dictionary
+    from ultralytics.solutions.config import SolutionConfig
+
+    full_args_dict = vars(SolutionConfig())  # arguments dictionary
     overrides = {}
 
     # check dictionary alignment
@@ -859,7 +852,7 @@ def entrypoint(debug: str = "") -> None:
         "checks": checks.collect_system_info,
         "version": lambda: LOGGER.info(__version__),
         "settings": lambda: handle_yolo_settings(args[1:]),
-        "cfg": lambda: yaml_print(DEFAULT_CFG_PATH),
+        "cfg": lambda: YAML.print(DEFAULT_CFG_PATH),
         "hub": lambda: handle_yolo_hub(args[1:]),
         "login": lambda: handle_yolo_hub(args),
         "logout": lambda: handle_yolo_hub(args),
@@ -886,7 +879,7 @@ def entrypoint(debug: str = "") -> None:
                 k, v = parse_key_value_pair(a)
                 if k == "cfg" and v is not None:  # custom.yaml passed
                     LOGGER.info(f"Overriding {DEFAULT_CFG_PATH} with {v}")
-                    overrides = {k: val for k, val in yaml_load(checks.check_yaml(v)).items() if k != "cfg"}
+                    overrides = {k: val for k, val in YAML.load(checks.check_yaml(v)).items() if k != "cfg"}
                 else:
                     overrides[k] = v
             except (NameError, SyntaxError, ValueError, AssertionError) as e:

ultralytics/cfg/default.yaml

@@ -35,7 +35,7 @@ resume: False # (bool) resume training from last checkpoint
 amp: True # (bool) Automatic Mixed Precision (AMP) training, choices=[True, False], True runs AMP check
 fraction: 1.0 # (float) dataset fraction to train on (default is 1.0, all images in train set)
 profile: False # (bool) profile ONNX and TensorRT speeds during training for loggers
-freeze: None # (int | list, optional) freeze first n layers, or freeze list of layer indices during training
+freeze: # (int | list, optional) freeze first n layers, or freeze list of layer indices during training
 multi_scale: False # (bool) Whether to use multiscale during training
 # Segmentation
 overlap_mask: True # (bool) merge object masks into a single image mask during training (segment train only)
@@ -84,7 +84,7 @@ int8: False # (bool) CoreML/TF INT8 quantization
 dynamic: False # (bool) ONNX/TF/TensorRT: dynamic axes
 simplify: True # (bool) ONNX: simplify model using `onnxslim`
 opset: # (int, optional) ONNX: opset version
-workspace: None # (float, optional) TensorRT: workspace size (GiB), `None` will let TensorRT auto-allocate memory
+workspace: # (float, optional) TensorRT: workspace size (GiB), `None` will let TensorRT auto-allocate memory
 nms: False # (bool) CoreML: add NMS
 
 # Hyperparameters ------------------------------------------------------------------------------------------------------

ultralytics/data/base.py

@@ -11,7 +11,6 @@ from typing import Optional
 
 import cv2
 import numpy as np
-import psutil
 from torch.utils.data import Dataset
 
 from ultralytics.data.utils import FORMATS_HELP_MSG, HELP_URL, IMG_FORMATS, check_file_speeds
@@ -336,7 +335,7 @@ class BaseDataset(Dataset):
             ratio = self.imgsz / max(im.shape[0], im.shape[1])  # max(h, w)  # ratio
             b += im.nbytes * ratio**2
         mem_required = b * self.ni / n * (1 + safety_margin)  # GB required to cache dataset into RAM
-        mem = psutil.virtual_memory()
+        mem = __import__("psutil").virtual_memory()
         if mem_required > mem.available:
             self.cache = None
             LOGGER.warning(

ultralytics/data/loaders.py

@@ -4,14 +4,13 @@ import glob
 import math
 import os
 import time
+import urllib
 from dataclasses import dataclass
 from pathlib import Path
 from threading import Thread
-from urllib.parse import urlparse
 
 import cv2
 import numpy as np
-import requests
 import torch
 from PIL import Image
 
@@ -111,7 +110,7 @@ class LoadStreams:
         for i, s in enumerate(sources):  # index, source
             # Start thread to read frames from video stream
             st = f"{i + 1}/{n}: {s}... "
-            if urlparse(s).hostname in {"www.youtube.com", "youtube.com", "youtu.be"}:  # if source is YouTube video
+            if urllib.parse.urlparse(s).hostname in {"www.youtube.com", "youtube.com", "youtu.be"}:  # YouTube video
                 # YouTube format i.e. 'https://www.youtube.com/watch?v=Jsn8D3aC840' or 'https://youtu.be/Jsn8D3aC840'
                 s = get_best_youtube_url(s)
             s = eval(s) if s.isnumeric() else s  # i.e. s = '0' local webcam
@@ -589,7 +588,7 @@ def autocast_list(source):
     files = []
     for im in source:
         if isinstance(im, (str, Path)):  # filename or uri
-            files.append(Image.open(requests.get(im, stream=True).raw if str(im).startswith("http") else im))
+            files.append(Image.open(urllib.request.urlopen(im) if str(im).startswith("http") else im))
         elif isinstance(im, (Image.Image, np.ndarray)):  # PIL or np Image
             files.append(im)
         else:

ultralytics/data/utils.py

@@ -1,6 +1,5 @@
 # Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
 
-import hashlib
 import json
 import os
 import random
@@ -19,16 +18,16 @@ from ultralytics.nn.autobackend import check_class_names
 from ultralytics.utils import (
     DATASETS_DIR,
     LOGGER,
+    MACOS,
     NUM_THREADS,
     ROOT,
     SETTINGS_FILE,
     TQDM,
+    YAML,
     clean_url,
     colorstr,
     emojis,
     is_dir_writeable,
-    yaml_load,
-    yaml_save,
 )
 from ultralytics.utils.checks import check_file, check_font, is_ascii
 from ultralytics.utils.downloads import download, safe_download, unzip_file
@@ -37,7 +36,7 @@ from ultralytics.utils.ops import segments2boxes
 HELP_URL = "See https://docs.ultralytics.com/datasets for dataset formatting guidance."
 IMG_FORMATS = {"bmp", "dng", "jpeg", "jpg", "mpo", "png", "tif", "tiff", "webp", "pfm", "heic"}  # image suffixes
 VID_FORMATS = {"asf", "avi", "gif", "m4v", "mkv", "mov", "mp4", "mpeg", "mpg", "ts", "wmv", "webm"}  # video suffixes
-PIN_MEMORY = str(os.getenv("PIN_MEMORY", True)).lower() == "true"  # global pin_memory for dataloaders
+PIN_MEMORY = str(os.getenv("PIN_MEMORY", not MACOS)).lower() == "true"  # global pin_memory for dataloaders
 FORMATS_HELP_MSG = f"Supported formats are:\nimages: {IMG_FORMATS}\nvideos: {VID_FORMATS}"
 
 
@@ -131,7 +130,7 @@ def get_hash(paths):
             size += os.stat(p).st_size
         except OSError:
             continue
-    h = hashlib.sha256(str(size).encode())  # hash sizes
+    h = __import__("hashlib").sha256(str(size).encode())  # hash sizes
     h.update("".join(paths).encode())  # hash paths
     return h.hexdigest()  # return hash
 
@@ -400,7 +399,7 @@ def check_det_dataset(dataset, autodownload=True):
         extract_dir, autodownload = file.parent, False
 
     # Read YAML
-    data = yaml_load(file, append_filename=True)  # dictionary
+    data = YAML.load(file, append_filename=True)  # dictionary
 
     # Checks
     for k in "train", "val":
@@ -493,7 +492,7 @@ def check_cls_dataset(dataset, split=""):
     # Download (optional if dataset=https://file.zip is passed directly)
     if str(dataset).startswith(("http:/", "https:/")):
         dataset = safe_download(dataset, dir=DATASETS_DIR, unzip=True, delete=False)
-    elif Path(dataset).suffix in {".zip", ".tar", ".gz"}:
+    elif str(dataset).endswith((".zip", ".tar", ".gz")):
         file = check_file(dataset)
         dataset = safe_download(file, dir=DATASETS_DIR, unzip=True, delete=False)
 
@@ -600,9 +599,9 @@ class HUBDatasetStats:
             _, data_dir, yaml_path = self._unzip(Path(path))
             try:
                 # Load YAML with checks
-                data = yaml_load(yaml_path)
+                data = YAML.load(yaml_path)
                 data["path"] = ""  # strip path since YAML should be in dataset root for all HUB datasets
-                yaml_save(yaml_path, data)
+                YAML.save(yaml_path, data)
                 data = check_det_dataset(yaml_path, autodownload)  # dict
                 data["path"] = data_dir  # YAML path should be set to '' (relative) or parent (absolute)
             except Exception as e:

ultralytics/engine/exporter.py

@@ -89,10 +89,10 @@ from ultralytics.utils import (
     RKNN_CHIPS,
     ROOT,
     WINDOWS,
+    YAML,
     callbacks,
     colorstr,
     get_default_args,
-    yaml_save,
 )
 from ultralytics.utils.checks import (
     check_imgsz,
@@ -631,7 +631,7 @@ class Exporter:
                 ov_model.set_rt_info("fit_to_window_letterbox", ["model_info", "resize_type"])
 
             ov.save_model(ov_model, file, compress_to_fp16=self.args.half)
-            yaml_save(Path(file).parent / "metadata.yaml", self.metadata)  # add metadata.yaml
+            YAML.save(Path(file).parent / "metadata.yaml", self.metadata)  # add metadata.yaml
 
         if self.args.int8:
             fq = str(self.file).replace(self.file.suffix, f"_int8_openvino_model{os.sep}")
@@ -690,7 +690,7 @@ class Exporter:
         f = str(self.file).replace(self.file.suffix, f"_paddle_model{os.sep}")
 
         pytorch2paddle(module=self.model, save_dir=f, jit_type="trace", input_examples=[self.im])  # export
-        yaml_save(Path(f) / "metadata.yaml", self.metadata)  # add metadata.yaml
+        YAML.save(Path(f) / "metadata.yaml", self.metadata)  # add metadata.yaml
         return f, None
 
     @try_export
@@ -783,7 +783,7 @@ class Exporter:
         for f_debug in ("debug.bin", "debug.param", "debug2.bin", "debug2.param", *pnnx_files):
             Path(f_debug).unlink(missing_ok=True)
 
-        yaml_save(f / "metadata.yaml", self.metadata)  # add metadata.yaml
+        YAML.save(f / "metadata.yaml", self.metadata)  # add metadata.yaml
         return str(f), None
 
     @try_export
@@ -974,7 +974,7 @@ class Exporter:
             output_signaturedefs=True,  # fix error with Attention block group convolution
             optimization_for_gpu_delegate=True,
         )
-        yaml_save(f / "metadata.yaml", self.metadata)  # add metadata.yaml
+        YAML.save(f / "metadata.yaml", self.metadata)  # add metadata.yaml
 
         # Remove/rename TFLite models
         if self.args.int8:
@@ -1087,7 +1087,7 @@ class Exporter:
             LOGGER.warning(f"{prefix} your model may not work correctly with spaces in path '{f}'.")
 
         # Add metadata
-        yaml_save(Path(f) / "metadata.yaml", self.metadata)  # add metadata.yaml
+        YAML.save(Path(f) / "metadata.yaml", self.metadata)  # add metadata.yaml
         return f, None
 
     @try_export
@@ -1114,7 +1114,7 @@ class Exporter:
         rknn.build(do_quantization=self.args.int8)
         f = f.replace(".onnx", f"-{self.args.name}-int8.rknn" if self.args.int8 else f"-{self.args.name}-fp16.rknn")
         rknn.export_rknn(f"{export_path / f}")
-        yaml_save(export_path / "metadata.yaml", self.metadata)
+        YAML.save(export_path / "metadata.yaml", self.metadata)
         return export_path, None
 
     @try_export

ultralytics/engine/model.py

@@ -10,7 +10,6 @@ from PIL import Image
 
 from ultralytics.cfg import TASK2DATA, get_cfg, get_save_dir
 from ultralytics.engine.results import Results
-from ultralytics.hub import HUB_WEB_ROOT, HUBTrainingSession
 from ultralytics.nn.tasks import attempt_load_one_weight, guess_model_task, yaml_model_load
 from ultralytics.utils import (
     ARGV,
@@ -19,9 +18,9 @@ from ultralytics.utils import (
     LOGGER,
     RANK,
     SETTINGS,
+    YAML,
     callbacks,
     checks,
-    yaml_load,
 )
 
 
@@ -126,6 +125,8 @@ class Model(torch.nn.Module):
 
         # Check if Ultralytics HUB model from https://hub.ultralytics.com
         if self.is_hub_model(model):
+            from ultralytics.hub import HUBTrainingSession
+
             # Fetch model from HUB
             checks.check_requirements("hub-sdk>=0.0.12")
             session = HUBTrainingSession.create_session(model)
@@ -141,7 +142,7 @@ class Model(torch.nn.Module):
 
         # Load or create new YOLO model
         __import__("os").environ["CUBLAS_WORKSPACE_CONFIG"] = ":4096:8"  # to avoid deterministic warnings
-        if Path(model).suffix in {".yaml", ".yml"}:
+        if str(model).endswith((".yaml", ".yml")):
             self._new(model, task=task, verbose=verbose)
         else:
             self._load(model, task=task)
@@ -225,6 +226,8 @@ class Model(torch.nn.Module):
             >>> Model.is_hub_model("yolo11n.pt")
             False
         """
+        from ultralytics.hub import HUB_WEB_ROOT
+
         return model.startswith(f"{HUB_WEB_ROOT}/models/")
 
     def _new(self, cfg: str, task=None, model=None, verbose=False) -> None:
@@ -770,7 +773,7 @@ class Model(torch.nn.Module):
 
         checks.check_pip_update_available()
 
-        overrides = yaml_load(checks.check_yaml(kwargs["cfg"])) if kwargs.get("cfg") else self.overrides
+        overrides = YAML.load(checks.check_yaml(kwargs["cfg"])) if kwargs.get("cfg") else self.overrides
         custom = {
             # NOTE: handle the case when 'cfg' includes 'data'.
             "data": overrides.get("data") or DEFAULT_CFG_DICT["data"] or TASK2DATA[self.task],

ultralytics/engine/trainer.py

@@ -31,11 +31,11 @@ from ultralytics.utils import (
     LOGGER,
     RANK,
     TQDM,
+    YAML,
     callbacks,
     clean_url,
     colorstr,
     emojis,
-    yaml_save,
 )
 from ultralytics.utils.autobatch import check_train_batch_size
 from ultralytics.utils.checks import check_amp, check_file, check_imgsz, check_model_file_from_stem, print_args
@@ -117,7 +117,7 @@ class BaseTrainer:
         if RANK in {-1, 0}:
             self.wdir.mkdir(parents=True, exist_ok=True)  # make dir
             self.args.save_dir = str(self.save_dir)
-            yaml_save(self.save_dir / "args.yaml", vars(self.args))  # save run args
+            YAML.save(self.save_dir / "args.yaml", vars(self.args))  # save run args
         self.last, self.best = self.wdir / "last.pt", self.wdir / "best.pt"  # checkpoint paths
         self.save_period = self.args.save_period
 

ultralytics/engine/tuner.py

@@ -23,7 +23,7 @@ import numpy as np
 import torch
 
 from ultralytics.cfg import get_cfg, get_save_dir
-from ultralytics.utils import DEFAULT_CFG, LOGGER, callbacks, colorstr, remove_colorstr, yaml_print, yaml_save
+from ultralytics.utils import DEFAULT_CFG, LOGGER, YAML, callbacks, colorstr, remove_colorstr
 from ultralytics.utils.plotting import plot_tune_results
 
 
@@ -235,9 +235,9 @@ class Tuner:
             )
             LOGGER.info("\n" + header)
             data = {k: float(x[best_idx, i + 1]) for i, k in enumerate(self.space.keys())}
-            yaml_save(
+            YAML.save(
                 self.tune_dir / "best_hyperparameters.yaml",
                 data=data,
                 header=remove_colorstr(header.replace(self.prefix, "# ")) + "\n",
             )
-            yaml_print(self.tune_dir / "best_hyperparameters.yaml")
+            YAML.print(self.tune_dir / "best_hyperparameters.yaml")

ultralytics/hub/session.py

@@ -201,7 +201,7 @@ class HUBTrainingSession:
             HUBModelError: If the identifier format is not recognized.
         """
         api_key, model_id, filename = None, None, None
-        if Path(identifier).suffix in {".pt", ".yaml"}:
+        if str(identifier).endswith((".pt", ".yaml")):
             filename = identifier
         elif identifier.startswith(f"{HUB_WEB_ROOT}/models/"):
             parsed_url = urlparse(identifier)

ultralytics/models/sam/model.py

@@ -19,7 +19,6 @@ from pathlib import Path
 from ultralytics.engine.model import Model
 from ultralytics.utils.torch_utils import model_info
 
-from .build import build_sam
 from .predict import Predictor, SAM2Predictor
 
 
@@ -78,6 +77,8 @@ class SAM(Model):
             >>> sam = SAM("sam_b.pt")
             >>> sam._load("path/to/custom_weights.pt")
         """
+        from .build import build_sam  # slow import
+
         self.model = build_sam(weights)
 
     def predict(self, source, stream=False, bboxes=None, points=None, labels=None, **kwargs):

ultralytics/models/sam/modules/tiny_encoder.py

@@ -15,7 +15,6 @@ from typing import Tuple
 import torch
 import torch.nn as nn
 import torch.nn.functional as F
-import torch.utils.checkpoint as checkpoint
 
 from ultralytics.nn.modules import LayerNorm2d
 from ultralytics.utils.instance import to_2tuple
@@ -308,7 +307,7 @@ class ConvLayer(nn.Module):
     def forward(self, x):
         """Processes input through convolutional layers, applying MBConv blocks and optional downsampling."""
         for blk in self.blocks:
-            x = checkpoint.checkpoint(blk, x) if self.use_checkpoint else blk(x)
+            x = torch.utils.checkpoint(blk, x) if self.use_checkpoint else blk(x)  # warn: checkpoint is slow import
         return x if self.downsample is None else self.downsample(x)
 
 
@@ -751,7 +750,7 @@ class BasicLayer(nn.Module):
     def forward(self, x):
         """Processes input through TinyViT blocks and optional downsampling."""
         for blk in self.blocks:
-            x = checkpoint.checkpoint(blk, x) if self.use_checkpoint else blk(x)
+            x = torch.utils.checkpoint(blk, x) if self.use_checkpoint else blk(x)  # warn: checkpoint is slow import
         return x if self.downsample is None else self.downsample(x)
 
     def extra_repr(self) -> str:

ultralytics/models/sam/predict.py

@@ -31,7 +31,6 @@ from .amg import (
     uncrop_boxes_xyxy,
     uncrop_masks,
 )
-from .build import build_sam
 
 
 class Predictor(BasePredictor):
@@ -439,6 +438,8 @@ class Predictor(BasePredictor):
 
     def get_model(self):
         """Retrieves or builds the Segment Anything Model (SAM) for image segmentation tasks."""
+        from .build import build_sam  # slow import
+
         return build_sam(self.args.model)
 
     def postprocess(self, preds, img, orig_imgs):
@@ -658,6 +659,8 @@ class SAM2Predictor(Predictor):
 
     def get_model(self):
         """Retrieves and initializes the Segment Anything Model 2 (SAM2) for image segmentation tasks."""
+        from .build import build_sam  # slow import
+
         return build_sam(self.args.model)
 
     def prompt_inference(

ultralytics/models/yolo/model.py

@@ -15,7 +15,7 @@ from ultralytics.nn.tasks import (
     YOLOEModel,
     YOLOESegModel,
 )
-from ultralytics.utils import ROOT, yaml_load
+from ultralytics.utils import ROOT, YAML
 
 
 class YOLO(Model):
@@ -107,7 +107,7 @@ class YOLOWorld(Model):
 
         # Assign default COCO class names when there are no custom names
         if not hasattr(self.model, "names"):
-            self.model.names = yaml_load(ROOT / "cfg/datasets/coco8.yaml").get("names")
+            self.model.names = YAML.load(ROOT / "cfg/datasets/coco8.yaml").get("names")
 
     @property
     def task_map(self):
@@ -156,7 +156,7 @@ class YOLOE(Model):
 
         # Assign default COCO class names when there are no custom names
         if not hasattr(self.model, "names"):
-            self.model.names = yaml_load(ROOT / "cfg/datasets/coco8.yaml").get("names")
+            self.model.names = YAML.load(ROOT / "cfg/datasets/coco8.yaml").get("names")
 
     @property
     def task_map(self):

ultralytics/nn/autobackend.py

@@ -14,7 +14,7 @@ import torch
 import torch.nn as nn
 from PIL import Image
 
-from ultralytics.utils import ARM64, IS_JETSON, LINUX, LOGGER, PYTHON_VERSION, ROOT, yaml_load
+from ultralytics.utils import ARM64, IS_JETSON, LINUX, LOGGER, PYTHON_VERSION, ROOT, YAML
 from ultralytics.utils.checks import check_requirements, check_suffix, check_version, check_yaml, is_rockchip
 from ultralytics.utils.downloads import attempt_download_asset, is_url
 
@@ -33,7 +33,7 @@ def check_class_names(names):
                 f"{min(names.keys())}-{max(names.keys())} defined in your dataset YAML."
             )
         if isinstance(names[0], str) and names[0].startswith("n0"):  # imagenet class codes, i.e. 'n01440764'
-            names_map = yaml_load(ROOT / "cfg/datasets/ImageNet.yaml")["map"]  # human-readable names
+            names_map = YAML.load(ROOT / "cfg/datasets/ImageNet.yaml")["map"]  # human-readable names
             names = {k: names_map[v] for k, v in names.items()}
     return names
 
@@ -42,7 +42,7 @@ def default_class_names(data=None):
     """Applies default class names to an input YAML file or returns numerical class names."""
     if data:
         try:
-            return yaml_load(check_yaml(data))["names"]
+            return YAML.load(check_yaml(data))["names"]
         except Exception:
             pass
     return {i: f"class{i}" for i in range(999)}  # return default if above errors
@@ -536,7 +536,7 @@ class AutoBackend(nn.Module):
 
         # Load external metadata YAML
         if isinstance(metadata, (str, Path)) and Path(metadata).exists():
-            metadata = yaml_load(metadata)
+            metadata = YAML.load(metadata)
         if metadata and isinstance(metadata, dict):
             for k, v in metadata.items():
                 if k in {"stride", "batch", "channels"}:

ultralytics/nn/tasks.py

@@ -69,7 +69,7 @@ from ultralytics.nn.modules import (
     YOLOESegment,
     v10Detect,
 )
-from ultralytics.utils import DEFAULT_CFG_DICT, DEFAULT_CFG_KEYS, LOGGER, colorstr, emojis, yaml_load
+from ultralytics.utils import DEFAULT_CFG_DICT, DEFAULT_CFG_KEYS, LOGGER, YAML, colorstr, emojis
 from ultralytics.utils.checks import check_requirements, check_suffix, check_yaml
 from ultralytics.utils.loss import (
     E2EDetectLoss,
@@ -92,11 +92,6 @@ from ultralytics.utils.torch_utils import (
     time_sync,
 )
 
-try:
-    import thop
-except ImportError:
-    thop = None  # conda support without 'ultralytics-thop' installed
-
 
 class BaseModel(torch.nn.Module):
     """The BaseModel class serves as a base class for all the models in the Ultralytics YOLO family."""
@@ -183,6 +178,11 @@ class BaseModel(torch.nn.Module):
             x (torch.Tensor): The input data to the layer.
             dt (list): A list to store the computation time of the layer.
         """
+        try:
+            import thop
+        except ImportError:
+            thop = None  # conda support without 'ultralytics-thop' installed
+
         c = m == self.model[-1] and isinstance(x, list)  # is final layer list, copy input as inplace fix
         flops = thop.profile(m, inputs=[x.copy() if c else x], verbose=False)[0] / 1e9 * 2 if thop else 0  # GFLOPs
         t = time_sync()
@@ -1523,7 +1523,7 @@ def yaml_model_load(path):
 
     unified_path = re.sub(r"(\d+)([nslmx])(.+)?$", r"\1\3", str(path))  # i.e. yolov8x.yaml -> yolov8.yaml
     yaml_file = check_yaml(unified_path, hard=False) or check_yaml(path)
-    d = yaml_load(yaml_file)  # model dict
+    d = YAML.load(yaml_file)  # model dict
     d["scale"] = guess_model_scale(path)
     d["yaml_file"] = str(path)
     return d