ultralytics 8.3.202__tar.gz → 8.3.203__tar.gz

{ultralytics-8.3.202 → ultralytics-8.3.203}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ultralytics
-Version: 8.3.202
+Version: 8.3.203
 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-8.3.202 → ultralytics-8.3.203}/pyproject.toml

@@ -158,6 +158,9 @@ source = ["ultralytics/"]
 data_file = "tests/.coverage"
 omit = ["ultralytics/utils/callbacks/*"]
 
+[tool.uv]
+required-version = "==0.8.19"
+
 [tool.isort]
 line_length = 120
 multi_line_output = 0

{ultralytics-8.3.202 → ultralytics-8.3.203}/tests/test_cli.py

@@ -1,13 +1,14 @@
 # Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
 
 import subprocess
+from pathlib import Path
 
 import pytest
 from PIL import Image
 
 from tests import CUDA_DEVICE_COUNT, CUDA_IS_AVAILABLE, MODELS, TASK_MODEL_DATA
 from ultralytics.utils import ARM64, ASSETS, LINUX, WEIGHTS_DIR, checks
-from ultralytics.utils.torch_utils import TORCH_1_9
+from ultralytics.utils.torch_utils import TORCH_1_11
 
 
 def run(cmd: str) -> None:
@@ -48,15 +49,12 @@ def test_export(model: str) -> None:
     run(f"yolo export model={model} format=torchscript imgsz=32")
 
 
-def test_rtdetr(task: str = "detect", model: str = "yolov8n-rtdetr.yaml", data: str = "coco8.yaml") -> None:
+@pytest.mark.skipif(not TORCH_1_11, reason="RTDETR requires torch>=1.11")
+def test_rtdetr(task: str = "detect", model: Path = WEIGHTS_DIR / "rtdetr-l.pt", data: str = "coco8.yaml") -> None:
     """Test the RTDETR functionality within Ultralytics for detection tasks using specified model and data."""
-    # Warning: must use imgsz=640 (note also add comma, spaces, fraction=0.25 args to test single-image training)
-    run(f"yolo train {task} model={model} data={data} --imgsz= 160 epochs =1, cache = disk fraction=0.25")  # spaces
+    # Add comma, spaces, fraction=0.25 args to test single-image training
     run(f"yolo predict {task} model={model} source={ASSETS / 'bus.jpg'} imgsz=160 save save_crop save_txt")
-    if TORCH_1_9:
-        weights = WEIGHTS_DIR / "rtdetr-l.pt"
-        run(f"yolo predict {task} model={weights} source={ASSETS / 'bus.jpg'} imgsz=160 save save_crop save_txt")
-        run(f"yolo train {task} model={weights} epochs=1 imgsz=160 cache=disk data=coco8.yaml")
+    run(f"yolo train {task} model={model} data={data} --imgsz= 160 epochs =1, cache = disk fraction=0.25")
 
 
 @pytest.mark.skipif(checks.IS_PYTHON_3_12, reason="MobileSAM with CLIP is not supported in Python 3.12")

{ultralytics-8.3.202 → ultralytics-8.3.203}/tests/test_exports.py

@@ -20,7 +20,7 @@ from ultralytics.utils import (
     WINDOWS,
     checks,
 )
-from ultralytics.utils.torch_utils import TORCH_1_9, TORCH_1_13
+from ultralytics.utils.torch_utils import TORCH_1_11, TORCH_1_13, TORCH_2_1
 
 
 def test_export_torchscript():
@@ -35,7 +35,7 @@ def test_export_onnx():
     YOLO(file)(SOURCE, imgsz=32)  # exported model inference
 
 
-@pytest.mark.skipif(not TORCH_1_13, reason="OpenVINO requires torch>=1.13")
+@pytest.mark.skipif(not TORCH_2_1, reason="OpenVINO requires torch>=2.1")
 def test_export_openvino():
     """Test YOLO export to OpenVINO format for model inference compatibility."""
     file = YOLO(MODEL).export(format="openvino", imgsz=32)
@@ -43,7 +43,7 @@ def test_export_openvino():
 
 
 @pytest.mark.slow
-@pytest.mark.skipif(not TORCH_1_13, reason="OpenVINO requires torch>=1.13")
+@pytest.mark.skipif(not TORCH_2_1, reason="OpenVINO requires torch>=2.1")
 @pytest.mark.parametrize(
     "task, dynamic, int8, half, batch, nms",
     [  # generate all combinations except for exclusion cases
@@ -117,7 +117,7 @@ def test_export_torchscript_matrix(task, dynamic, int8, half, batch, nms):
 
 @pytest.mark.slow
 @pytest.mark.skipif(not MACOS, reason="CoreML inference only supported on macOS")
-@pytest.mark.skipif(not TORCH_1_9, reason="CoreML>=7.2 not supported with PyTorch<=1.8")
+@pytest.mark.skipif(not TORCH_1_11, reason="CoreML export requires torch>=1.11")
 @pytest.mark.skipif(checks.IS_PYTHON_3_13, reason="CoreML not supported in Python 3.13")
 @pytest.mark.parametrize(
     "task, dynamic, int8, half, nms, batch",
@@ -169,7 +169,7 @@ def test_export_tflite_matrix(task, dynamic, int8, half, batch, nms):
     Path(file).unlink()  # cleanup
 
 
-@pytest.mark.skipif(not TORCH_1_9, reason="CoreML>=7.2 not supported with PyTorch<=1.8")
+@pytest.mark.skipif(not TORCH_1_11, reason="CoreML export requires torch>=1.11")
 @pytest.mark.skipif(WINDOWS, reason="CoreML not supported on Windows")  # RuntimeError: BlobWriter not loaded
 @pytest.mark.skipif(LINUX and ARM64, reason="CoreML not supported on aarch64 Linux")
 @pytest.mark.skipif(checks.IS_PYTHON_3_13, reason="CoreML not supported in Python 3.13")

{ultralytics-8.3.202 → ultralytics-8.3.203}/tests/test_python.py

@@ -34,7 +34,7 @@ from ultralytics.utils import (
     is_github_action_running,
 )
 from ultralytics.utils.downloads import download
-from ultralytics.utils.torch_utils import TORCH_1_9
+from ultralytics.utils.torch_utils import TORCH_1_11, TORCH_1_13
 
 IS_TMP_WRITEABLE = is_dir_writeable(TMP)  # WARNING: must be run once tests start as TMP does not exist on tests/init
 
@@ -125,7 +125,9 @@ def test_predict_img(model_name):
     batch = [
         str(SOURCE),  # filename
         Path(SOURCE),  # Path
-        "https://github.com/ultralytics/assets/releases/download/v0.0.0/zidane.jpg" if ONLINE else SOURCE,  # URI
+        "https://github.com/ultralytics/assets/releases/download/v0.0.0/zidane.jpg?token=123"
+        if ONLINE
+        else SOURCE,  # URI
         im,  # OpenCV
         Image.open(SOURCE),  # PIL
         np.zeros((320, 640, channels), dtype=np.uint8),  # numpy
@@ -246,7 +248,7 @@ def test_all_model_yamls():
     """Test YOLO model creation for all available YAML configurations in the `cfg/models` directory."""
     for m in (ROOT / "cfg" / "models").rglob("*.yaml"):
         if "rtdetr" in m.name:
-            if TORCH_1_9:  # torch<=1.8 issue - TypeError: __init__() got an unexpected keyword argument 'batch_first'
+            if TORCH_1_11:
                 _ = RTDETR(m.name)(SOURCE, imgsz=640)  # must be 640
         else:
             YOLO(m.name)
@@ -634,7 +636,8 @@ def test_yolo_world():
     )
 
 
-@pytest.mark.skipif(checks.IS_PYTHON_3_12 or not TORCH_1_9, reason="YOLOE with CLIP is not supported in Python 3.12")
+@pytest.mark.skipif(not TORCH_1_13, reason="YOLOE with CLIP requires torch>=1.13")
+@pytest.mark.skipif(checks.IS_PYTHON_3_12, reason="YOLOE with CLIP is not supported in Python 3.12")
 @pytest.mark.skipif(
     checks.IS_PYTHON_3_8 and LINUX and ARM64,
     reason="YOLOE with CLIP is not supported in Python 3.8 and aarch64 Linux",
@@ -648,16 +651,12 @@ def test_yoloe():
     model.set_classes(names, model.get_text_pe(names))
     model(SOURCE, conf=0.01)
 
-    import numpy as np
-
     from ultralytics import YOLOE
     from ultralytics.models.yolo.yoloe import YOLOEVPSegPredictor
 
     # visual-prompts
     visuals = dict(
-        bboxes=np.array(
-            [[221.52, 405.8, 344.98, 857.54], [120, 425, 160, 445]],
-        ),
+        bboxes=np.array([[221.52, 405.8, 344.98, 857.54], [120, 425, 160, 445]]),
         cls=np.array([0, 1]),
     )
     model.predict(

{ultralytics-8.3.202 → ultralytics-8.3.203}/ultralytics/__init__.py

@@ -1,6 +1,6 @@
 # Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
 
-__version__ = "8.3.202"
+__version__ = "8.3.203"
 
 import importlib
 import os

{ultralytics-8.3.202 → ultralytics-8.3.203}/ultralytics/data/build.py

@@ -7,6 +7,7 @@ import random
 from collections.abc import Iterator
 from pathlib import Path
 from typing import Any
+from urllib.parse import urlsplit
 
 import numpy as np
 import torch
@@ -247,8 +248,10 @@ def check_source(source):
     if isinstance(source, (str, int, Path)):  # int for local usb camera
         source = str(source)
         source_lower = source.lower()
-        is_file = source_lower.rpartition(".")[-1] in (IMG_FORMATS | VID_FORMATS)
         is_url = source_lower.startswith(("https://", "http://", "rtsp://", "rtmp://", "tcp://"))
+        is_file = (urlsplit(source_lower).path if is_url else source_lower).rpartition(".")[-1] in (
+            IMG_FORMATS | VID_FORMATS
+        )
         webcam = source.isnumeric() or source.endswith(".streams") or (is_url and not is_file)
         screenshot = source_lower == "screen"
         if is_url and is_file:

{ultralytics-8.3.202 → ultralytics-8.3.203}/ultralytics/engine/exporter.py

@@ -112,7 +112,7 @@ from ultralytics.utils.metrics import batch_probiou
 from ultralytics.utils.nms import TorchNMS
 from ultralytics.utils.ops import Profile
 from ultralytics.utils.patches import arange_patch
-from ultralytics.utils.torch_utils import TORCH_1_13, get_latest_opset, select_device
+from ultralytics.utils.torch_utils import TORCH_1_11, TORCH_1_13, TORCH_2_1, select_device
 
 
 def export_formats():
@@ -152,6 +152,31 @@ def export_formats():
     return dict(zip(["Format", "Argument", "Suffix", "CPU", "GPU", "Arguments"], zip(*x)))
 
 
+def best_onnx_opset(onnx) -> int:
+    """Return max ONNX opset for this torch version with ONNX fallback."""
+    if TORCH_1_13:  # not supported by torch<1.13
+        opset = torch.onnx.utils._constants.ONNX_MAX_OPSET - 1  # use second-latest version for safety
+    else:
+        opset = {
+            "1.8": 12,
+            "1.9": 12,
+            "1.10": 13,
+            "1.11": 14,
+            "1.12": 15,
+            "1.13": 17,
+            "2.0": 18,
+            "2.1": 19,
+            "2.2": 19,
+            "2.3": 19,
+            "2.4": 20,
+            "2.5": 20,
+            "2.6": 20,
+            "2.7": 20,
+            "2.8": 23,
+        }.get(".".join(TORCH_VERSION.split(".")[:2]), 12)
+    return min(opset, onnx.defs.onnx_opset_version())
+
+
 def validate_args(format, passed_args, valid_args):
     """
     Validate arguments based on the export format.
@@ -586,8 +611,8 @@ class Exporter:
         check_requirements(requirements)
         import onnx  # noqa
 
-        opset_version = self.args.opset or get_latest_opset()
-        LOGGER.info(f"\n{prefix} starting export with onnx {onnx.__version__} opset {opset_version}...")
+        opset = self.args.opset or best_onnx_opset(onnx)
+        LOGGER.info(f"\n{prefix} starting export with onnx {onnx.__version__} opset {opset}...")
         f = str(self.file.with_suffix(".onnx"))
         output_names = ["output0", "output1"] if isinstance(self.model, SegmentationModel) else ["output0"]
         dynamic = self.args.dynamic
@@ -601,14 +626,14 @@ class Exporter:
             if self.args.nms:  # only batch size is dynamic with NMS
                 dynamic["output0"].pop(2)
         if self.args.nms and self.model.task == "obb":
-            self.args.opset = opset_version  # for NMSModel
+            self.args.opset = opset  # for NMSModel
 
         with arange_patch(self.args):
             torch2onnx(
                 NMSModel(self.model, self.args) if self.args.nms else self.model,
                 self.im,
                 f,
-                opset=opset_version,
+                opset=opset,
                 input_names=["images"],
                 output_names=output_names,
                 dynamic=dynamic or None,
@@ -633,6 +658,11 @@ class Exporter:
             meta = model_onnx.metadata_props.add()
             meta.key, meta.value = k, str(v)
 
+        # IR version
+        if getattr(model_onnx, "ir_version", 0) > 10:
+            LOGGER.info(f"{prefix} limiting IR version {model_onnx.ir_version} to 10 for ONNXRuntime compatibility...")
+            model_onnx.ir_version = 10
+
         onnx.save(model_onnx, f)
         return f
 
@@ -644,7 +674,7 @@ class Exporter:
         import openvino as ov
 
         LOGGER.info(f"\n{prefix} starting export with openvino {ov.__version__}...")
-        assert TORCH_1_13, f"OpenVINO export requires torch>=1.13.0 but torch=={TORCH_VERSION} is installed"
+        assert TORCH_2_1, f"OpenVINO export requires torch>=2.1 but torch=={TORCH_VERSION} is installed"
         ov_model = ov.convert_model(
             NMSModel(self.model, self.args) if self.args.nms else self.model,
             input=None if self.args.dynamic else [self.im.shape],
@@ -837,6 +867,7 @@ class Exporter:
 
         LOGGER.info(f"\n{prefix} starting export with coremltools {ct.__version__}...")
         assert not WINDOWS, "CoreML export is not supported on Windows, please run on macOS or Linux."
+        assert TORCH_1_11, "CoreML export requires torch>=1.11"
         assert self.args.batch == 1, "CoreML batch sizes > 1 are not supported. Please retry at 'batch=1'."
         f = self.file.with_suffix(".mlmodel" if mlmodel else ".mlpackage")
         if f.is_dir():

{ultralytics-8.3.202 → ultralytics-8.3.203}/ultralytics/engine/trainer.py

@@ -415,9 +415,12 @@ class BaseTrainer:
                 # Forward
                 with autocast(self.amp):
                     batch = self.preprocess_batch(batch)
-                    # decouple inference and loss calculations for torch.compile convenience
-                    preds = self.model(batch["img"])
-                    loss, self.loss_items = unwrap_model(self.model).loss(batch, preds)
+                    if self.args.compile:
+                        # Decouple inference and loss calculations for improved compile performance
+                        preds = self.model(batch["img"])
+                        loss, self.loss_items = unwrap_model(self.model).loss(batch, preds)
+                    else:
+                        loss, self.loss_items = self.model(batch)
                     self.loss = loss.sum()
                     if RANK != -1:
                         self.loss *= self.world_size
@@ -581,6 +584,7 @@ class BaseTrainer:
                 "ema": deepcopy(unwrap_model(self.ema.ema)).half(),
                 "updates": self.ema.updates,
                 "optimizer": convert_optimizer_state_dict_to_fp16(deepcopy(self.optimizer.state_dict())),
+                "scaler": self.scaler.state_dict(),
                 "train_args": vars(self.args),  # save as dict
                 "train_metrics": {**self.metrics, **{"fitness": self.fitness}},
                 "train_results": self.read_results_csv(),
@@ -809,9 +813,11 @@ class BaseTrainer:
             return
         best_fitness = 0.0
         start_epoch = ckpt.get("epoch", -1) + 1
-        if ckpt.get("optimizer", None) is not None:
+        if ckpt.get("optimizer") is not None:
             self.optimizer.load_state_dict(ckpt["optimizer"])  # optimizer
             best_fitness = ckpt["best_fitness"]
+        if ckpt.get("scaler") is not None:
+            self.scaler.load_state_dict(ckpt["scaler"])
         if self.ema and ckpt.get("ema"):
             self.ema.ema.load_state_dict(ckpt["ema"].float().state_dict())  # EMA
             self.ema.updates = ckpt["updates"]

{ultralytics-8.3.202 → ultralytics-8.3.203}/ultralytics/engine/tuner.py

@@ -435,7 +435,7 @@ class Tuner:
                 best_metrics = {k: round(v, 5) for k, v in metrics.items()}
                 for ckpt in weights_dir.glob("*.pt"):
                     shutil.copy2(ckpt, self.tune_dir / "weights")
-            elif cleanup:
+            elif cleanup and best_save_dir:
                 shutil.rmtree(best_save_dir, ignore_errors=True)  # remove iteration dirs to reduce storage space
 
             # Plot tune results

{ultralytics-8.3.202 → ultralytics-8.3.203}/ultralytics/models/fastsam/predict.py

@@ -7,6 +7,7 @@ from ultralytics.models.yolo.segment import SegmentationPredictor
 from ultralytics.utils import DEFAULT_CFG, checks
 from ultralytics.utils.metrics import box_iou
 from ultralytics.utils.ops import scale_masks
+from ultralytics.utils.torch_utils import TORCH_1_10
 
 from .utils import adjust_bboxes_to_image_border
 
@@ -135,7 +136,7 @@ class FastSAMPredictor(SegmentationPredictor):
                 crop_ims, filter_idx = [], []
                 for i, b in enumerate(result.boxes.xyxy.tolist()):
                     x1, y1, x2, y2 = (int(x) for x in b)
-                    if masks[i].sum() <= 100:
+                    if (masks[i].sum() if TORCH_1_10 else masks[i].sum(0).sum()) <= 100:  # torch 1.9 bug workaround
                         filter_idx.append(i)
                         continue
                     crop_ims.append(Image.fromarray(result.orig_img[y1:y2, x1:x2, ::-1]))

{ultralytics-8.3.202 → ultralytics-8.3.203}/ultralytics/models/rtdetr/model.py

@@ -11,6 +11,7 @@ References:
 
 from ultralytics.engine.model import Model
 from ultralytics.nn.tasks import RTDETRDetectionModel
+from ultralytics.utils.torch_utils import TORCH_1_11
 
 from .predict import RTDETRPredictor
 from .train import RTDETRTrainer
@@ -44,6 +45,7 @@ class RTDETR(Model):
         Args:
             model (str): Path to the pre-trained model. Supports .pt, .yaml, and .yml formats.
         """
+        assert TORCH_1_11, "RTDETR requires torch>=1.11"
         super().__init__(model=model, task="detect")
 
     @property

{ultralytics-8.3.202 → ultralytics-8.3.203}/ultralytics/models/yolo/world/train.py

@@ -64,6 +64,7 @@ class WorldTrainer(DetectionTrainer):
         """
         if overrides is None:
             overrides = {}
+        assert not overrides.get("compile"), f"Training with 'model={overrides['model']}' requires 'compile=False'"
         super().__init__(cfg, overrides, _callbacks)
         self.text_embeddings = None
 

{ultralytics-8.3.202 → ultralytics-8.3.203}/ultralytics/models/yolo/yoloe/train.py

@@ -46,6 +46,7 @@ class YOLOETrainer(DetectionTrainer):
         """
         if overrides is None:
             overrides = {}
+        assert not overrides.get("compile"), f"Training with 'model={overrides['model']}' requires 'compile=False'"
         overrides["overlap_mask"] = False
         super().__init__(cfg, overrides, _callbacks)
 

{ultralytics-8.3.202 → ultralytics-8.3.203}/ultralytics/nn/modules/head.py

@@ -12,8 +12,8 @@ import torch.nn.functional as F
 from torch.nn.init import constant_, xavier_uniform_
 
 from ultralytics.utils import NOT_MACOS14
-from ultralytics.utils.tal import TORCH_1_10, dist2bbox, dist2rbox, make_anchors
-from ultralytics.utils.torch_utils import fuse_conv_and_bn, smart_inference_mode
+from ultralytics.utils.tal import dist2bbox, dist2rbox, make_anchors
+from ultralytics.utils.torch_utils import TORCH_1_11, fuse_conv_and_bn, smart_inference_mode
 
 from .block import DFL, SAVPE, BNContrastiveHead, ContrastiveHead, Proto, Residual, SwiGLUFFN
 from .conv import Conv, DWConv
@@ -1052,7 +1052,7 @@ class RTDETRDecoder(nn.Module):
         for i, (h, w) in enumerate(shapes):
             sy = torch.arange(end=h, dtype=dtype, device=device)
             sx = torch.arange(end=w, dtype=dtype, device=device)
-            grid_y, grid_x = torch.meshgrid(sy, sx, indexing="ij") if TORCH_1_10 else torch.meshgrid(sy, sx)
+            grid_y, grid_x = torch.meshgrid(sy, sx, indexing="ij") if TORCH_1_11 else torch.meshgrid(sy, sx)
             grid_xy = torch.stack([grid_x, grid_y], -1)  # (h, w, 2)
 
             valid_WH = torch.tensor([w, h], dtype=dtype, device=device)

{ultralytics-8.3.202 → ultralytics-8.3.203}/ultralytics/nn/modules/transformer.py

@@ -10,6 +10,8 @@ import torch.nn as nn
 import torch.nn.functional as F
 from torch.nn.init import constant_, xavier_uniform_
 
+from ultralytics.utils.torch_utils import TORCH_1_11
+
 from .conv import Conv
 from .utils import _get_clones, inverse_sigmoid, multi_scale_deformable_attn_pytorch
 
@@ -236,7 +238,7 @@ class AIFI(TransformerEncoderLayer):
         assert embed_dim % 4 == 0, "Embed dimension must be divisible by 4 for 2D sin-cos position embedding"
         grid_w = torch.arange(w, dtype=torch.float32)
         grid_h = torch.arange(h, dtype=torch.float32)
-        grid_w, grid_h = torch.meshgrid(grid_w, grid_h, indexing="ij")
+        grid_w, grid_h = torch.meshgrid(grid_w, grid_h, indexing="ij") if TORCH_1_11 else torch.meshgrid(grid_w, grid_h)
         pos_dim = embed_dim // 4
         omega = torch.arange(pos_dim, dtype=torch.float32) / pos_dim
         omega = 1.0 / (temperature**omega)

{ultralytics-8.3.202 → ultralytics-8.3.203}/ultralytics/utils/checks.py

@@ -669,6 +669,9 @@ def check_yolo(verbose=True, device=""):
     else:
         s = ""
 
+    if GIT.is_repo:
+        check_multiple_install()  # check conflicting installation if using local clone
+
     select_device(device=device, newline=False)
     LOGGER.info(f"Setup complete ✅ {s}")
 
@@ -807,6 +810,30 @@ def check_amp(model):
     return True
 
 
+def check_multiple_install():
+    """Check if there are multiple Ultralytics installations."""
+    import sys
+
+    try:
+        result = subprocess.run([sys.executable, "-m", "pip", "show", "ultralytics"], capture_output=True, text=True)
+        install_msg = (
+            f"Install your local copy in editable mode with 'pip install -e {ROOT.parent}' to avoid "
+            "issues. See https://docs.ultralytics.com/quickstart/"
+        )
+        if result.returncode != 0:
+            if "not found" in result.stderr.lower():  # Package not pip-installed but locally imported
+                LOGGER.warning(f"Ultralytics not found via pip but importing from: {ROOT}. {install_msg}")
+            return
+        yolo_path = (Path(re.findall(r"location:\s+(.+)", result.stdout, flags=re.I)[-1]) / "ultralytics").resolve()
+        if not yolo_path.samefile(ROOT.resolve()):
+            LOGGER.warning(
+                f"Multiple Ultralytics installations detected. The `yolo` command uses: {yolo_path}, "
+                f"but current session imports from: {ROOT}. This may cause version conflicts. {install_msg}"
+            )
+    except Exception:
+        return
+
+
 def print_args(args: dict | None = None, show_file=True, show_func=False):
     """
     Print function arguments (optional args dict).

{ultralytics-8.3.202 → ultralytics-8.3.203}/ultralytics/utils/tal.py

@@ -3,12 +3,10 @@
 import torch
 import torch.nn as nn
 
-from . import LOGGER, TORCH_VERSION
-from .checks import check_version
+from . import LOGGER
 from .metrics import bbox_iou, probiou
 from .ops import xywhr2xyxyxyxy
-
-TORCH_1_10 = check_version(TORCH_VERSION, "1.10.0")
+from .torch_utils import TORCH_1_11
 
 
 class TaskAlignedAssigner(nn.Module):
@@ -373,7 +371,7 @@ def make_anchors(feats, strides, grid_cell_offset=0.5):
         h, w = feats[i].shape[2:] if isinstance(feats, list) else (int(feats[i][0]), int(feats[i][1]))
         sx = torch.arange(end=w, device=device, dtype=dtype) + grid_cell_offset  # shift x
         sy = torch.arange(end=h, device=device, dtype=dtype) + grid_cell_offset  # shift y
-        sy, sx = torch.meshgrid(sy, sx, indexing="ij") if TORCH_1_10 else torch.meshgrid(sy, sx)
+        sy, sx = torch.meshgrid(sy, sx, indexing="ij") if TORCH_1_11 else torch.meshgrid(sy, sx)
         anchor_points.append(torch.stack((sx, sy), -1).view(-1, 2))
         stride_tensor.append(torch.full((h * w, 1), stride, dtype=dtype, device=device))
     return torch.cat(anchor_points), torch.cat(stride_tensor)

{ultralytics-8.3.202 → ultralytics-8.3.203}/ultralytics/utils/torch_utils.py

@@ -38,8 +38,11 @@ from ultralytics.utils.patches import torch_load
 
 # Version checks (all default to version>=min_version)
 TORCH_1_9 = check_version(TORCH_VERSION, "1.9.0")
+TORCH_1_10 = check_version(TORCH_VERSION, "1.10.0")
+TORCH_1_11 = check_version(TORCH_VERSION, "1.11.0")
 TORCH_1_13 = check_version(TORCH_VERSION, "1.13.0")
 TORCH_2_0 = check_version(TORCH_VERSION, "2.0.0")
+TORCH_2_1 = check_version(TORCH_VERSION, "2.1.0")
 TORCH_2_4 = check_version(TORCH_VERSION, "2.4.0")
 TORCHVISION_0_10 = check_version(TORCHVISION_VERSION, "0.10.0")
 TORCHVISION_0_11 = check_version(TORCHVISION_VERSION, "0.11.0")
@@ -534,21 +537,6 @@ def copy_attr(a, b, include=(), exclude=()):
             setattr(a, k, v)
 
 
-def get_latest_opset():
-    """
-    Return the second-most recent ONNX opset version supported by this version of PyTorch, adjusted for maturity.
-
-    Returns:
-        (int): The ONNX opset version.
-    """
-    if TORCH_1_13:
-        # If the PyTorch>=1.13, dynamically compute the latest opset minus one using 'symbolic_opset'
-        return max(int(k[14:]) for k in vars(torch.onnx) if "symbolic_opset" in k) - 1
-    # Otherwise for PyTorch<=1.12 return the corresponding predefined opset
-    version = torch.onnx.producer_version.rsplit(".", 1)[0]  # i.e. '2.3'
-    return {"1.12": 15, "1.11": 14, "1.10": 13, "1.9": 12, "1.8": 12}.get(version, 12)
-
-
 def intersect_dicts(da, db, exclude=()):
     """
     Return a dictionary of intersecting keys with matching shapes, excluding 'exclude' keys, using da values.

{ultralytics-8.3.202 → ultralytics-8.3.203}/ultralytics.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ultralytics
-Version: 8.3.202
+Version: 8.3.203
 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>