ultralytics 8.3.43__tar.gz → 8.3.47__tar.gz

{ultralytics-8.3.43/ultralytics.egg-info → ultralytics-8.3.47}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: ultralytics
-Version: 8.3.43
+Version: 8.3.47
 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-8.3.43 → ultralytics-8.3.47}/ultralytics/__init__.py

@@ -1,6 +1,6 @@
 # Ultralytics YOLO 🚀, AGPL-3.0 license
 
-__version__ = "8.3.43"
+__version__ = "8.3.47"
 
 import os
 

{ultralytics-8.3.43 → ultralytics-8.3.47}/ultralytics/engine/exporter.py

@@ -73,7 +73,7 @@ from ultralytics.data import build_dataloader
 from ultralytics.data.dataset import YOLODataset
 from ultralytics.data.utils import check_cls_dataset, check_det_dataset
 from ultralytics.nn.autobackend import check_class_names, default_class_names
-from ultralytics.nn.modules import C2f, Detect, RTDETRDecoder
+from ultralytics.nn.modules import C2f, Classify, Detect, RTDETRDecoder
 from ultralytics.nn.tasks import DetectionModel, SegmentationModel, WorldModel
 from ultralytics.utils import (
     ARM64,
@@ -287,6 +287,8 @@ class Exporter:
 
             model = FXModel(model)
         for m in model.modules():
+            if isinstance(m, Classify):
+                m.export = True
             if isinstance(m, (Detect, RTDETRDecoder)):  # includes all Detect subclasses like Segment, Pose, OBB
                 m.dynamic = self.args.dynamic
                 m.export = True

{ultralytics-8.3.43 → ultralytics-8.3.47}/ultralytics/engine/model.py

@@ -136,6 +136,7 @@ class Model(nn.Module):
         # Check if Triton Server model
         elif self.is_triton_model(model):
             self.model_name = self.model = model
+            self.overrides["task"] = task or "detect"  # set `task=detect` if not explicitly set
             return
 
         # Load or create new YOLO model

{ultralytics-8.3.43 → ultralytics-8.3.47}/ultralytics/engine/predictor.py

@@ -155,7 +155,7 @@ class BasePredictor:
         same_shapes = len({x.shape for x in im}) == 1
         letterbox = LetterBox(
             self.imgsz,
-            auto=same_shapes and (self.model.pt or getattr(self.model, "dynamic", False)),
+            auto=same_shapes and (self.model.pt or (getattr(self.model, "dynamic", False) and not self.model.imx)),
             stride=self.model.stride,
         )
         return [letterbox(image=x) for x in im]

{ultralytics-8.3.43 → ultralytics-8.3.47}/ultralytics/models/sam/predict.py

@@ -1105,7 +1105,7 @@ class SAM2VideoPredictor(SAM2Predictor):
             for obj_temp_output_dict in temp_output_dict_per_obj.values():
                 temp_frame_inds.update(obj_temp_output_dict[storage_key].keys())
             consolidated_frame_inds[storage_key].update(temp_frame_inds)
-            # consolidate the temprary output across all objects on this frame
+            # consolidate the temporary output across all objects on this frame
             for frame_idx in temp_frame_inds:
                 consolidated_out = self._consolidate_temp_output_across_obj(
                     frame_idx, is_cond=is_cond, run_mem_encoder=True

{ultralytics-8.3.43 → ultralytics-8.3.47}/ultralytics/models/yolo/classify/predict.py

@@ -53,7 +53,8 @@ class ClassificationPredictor(BasePredictor):
         if not isinstance(orig_imgs, list):  # input images are a torch.Tensor, not a list
             orig_imgs = ops.convert_torch2numpy_batch(orig_imgs)
 
+        preds = preds[0] if isinstance(preds, (list, tuple)) else preds
         return [
-            Results(orig_img, path=img_path, names=self.model.names, probs=pred.softmax(0))
+            Results(orig_img, path=img_path, names=self.model.names, probs=pred)
             for pred, orig_img, img_path in zip(preds, orig_imgs, self.batch[0])
         ]

{ultralytics-8.3.43 → ultralytics-8.3.47}/ultralytics/models/yolo/classify/val.py

@@ -71,6 +71,10 @@ class ClassificationValidator(BaseValidator):
         self.metrics.confusion_matrix = self.confusion_matrix
         self.metrics.save_dir = self.save_dir
 
+    def postprocess(self, preds):
+        """Preprocesses the classification predictions."""
+        return preds[0] if isinstance(preds, (list, tuple)) else preds
+
     def get_stats(self):
         """Returns a dictionary of metrics obtained by processing targets and predictions."""
         self.metrics.process(self.targets, self.pred)

{ultralytics-8.3.43 → ultralytics-8.3.47}/ultralytics/nn/autobackend.py

@@ -96,7 +96,7 @@ class AutoBackend(nn.Module):
         Initialize the AutoBackend for inference.
 
         Args:
-            weights (str): Path to the model weights file. Defaults to 'yolov8n.pt'.
+            weights (str | torch.nn.Module): Path to the model weights file or a module instance. Defaults to 'yolo11n.pt'.
             device (torch.device): Device to run the model on. Defaults to CPU.
             dnn (bool): Use OpenCV DNN module for ONNX inference. Defaults to False.
             data (str | Path | optional): Path to the additional data.yaml file containing class names. Optional.
@@ -462,6 +462,7 @@ class AutoBackend(nn.Module):
             from ultralytics.utils.triton import TritonRemoteModel
 
             model = TritonRemoteModel(w)
+            metadata = model.metadata
 
         # Any other format (unsupported)
         else:

{ultralytics-8.3.43 → ultralytics-8.3.47}/ultralytics/nn/modules/head.py

@@ -282,6 +282,8 @@ class Pose(Detect):
 class Classify(nn.Module):
     """YOLO classification head, i.e. x(b,c1,20,20) to x(b,c2)."""
 
+    export = False  # export mode
+
     def __init__(self, c1, c2, k=1, s=1, p=None, g=1):
         """Initializes YOLO classification head to transform input tensor from (b,c1,20,20) to (b,c2) shape."""
         super().__init__()
@@ -296,7 +298,10 @@ class Classify(nn.Module):
         if isinstance(x, list):
             x = torch.cat(x, 1)
         x = self.linear(self.drop(self.pool(self.conv(x)).flatten(1)))
-        return x
+        if self.training:
+            return x
+        y = x.softmax(1)  # get final output
+        return y if self.export else (y, x)
 
 
 class WorldDetect(Detect):

{ultralytics-8.3.43 → ultralytics-8.3.47}/ultralytics/trackers/utils/matching.py

@@ -13,7 +13,7 @@ try:
 except (ImportError, AssertionError, AttributeError):
     from ultralytics.utils.checks import check_requirements
 
-    check_requirements("lapx>=0.5.2")  # update to lap package from https://github.com/rathaROG/lapx
+    check_requirements("lap>=0.5.12")  # https://github.com/gatagat/lap
     import lap
 
 

{ultralytics-8.3.43 → ultralytics-8.3.47}/ultralytics/utils/checks.py

@@ -669,8 +669,22 @@ def check_amp(model):
     from ultralytics.utils.torch_utils import autocast
 
     device = next(model.parameters()).device  # get model device
+    prefix = colorstr("AMP: ")
     if device.type in {"cpu", "mps"}:
         return False  # AMP only used on CUDA devices
+    else:
+        # GPUs that have issues with AMP
+        pattern = re.compile(
+            r"(nvidia|geforce|quadro|tesla).*?(1660|1650|1630|t400|t550|t600|t1000|t1200|t2000|k40m)", re.IGNORECASE
+        )
+
+        gpu = torch.cuda.get_device_name(device)
+        if bool(pattern.search(gpu)):
+            LOGGER.warning(
+                f"{prefix}checks failed ❌. AMP training on {gpu} GPU may cause "
+                f"NaN losses or zero-mAP results, so AMP will be disabled during training."
+            )
+            return False
 
     def amp_allclose(m, im):
         """All close FP32 vs AMP results."""
@@ -683,7 +697,6 @@ def check_amp(model):
         return a.shape == b.shape and torch.allclose(a, b.float(), atol=0.5)  # close to 0.5 absolute tolerance
 
     im = ASSETS / "bus.jpg"  # image to check
-    prefix = colorstr("AMP: ")
     LOGGER.info(f"{prefix}running Automatic Mixed Precision (AMP) checks...")
     warning_msg = "Setting 'amp=True'. If you experience zero-mAP or NaN losses you can disable AMP with amp=False."
     try:

{ultralytics-8.3.43 → ultralytics-8.3.47}/ultralytics/utils/loss.py

@@ -604,6 +604,7 @@ class v8ClassificationLoss:
 
     def __call__(self, preds, batch):
         """Compute the classification loss between predictions and true labels."""
+        preds = preds[1] if isinstance(preds, (list, tuple)) else preds
         loss = F.cross_entropy(preds, batch["cls"], reduction="mean")
         loss_items = loss.detach()
         return loss, loss_items

{ultralytics-8.3.43 → ultralytics-8.3.47}/ultralytics/utils/ops.py

@@ -400,7 +400,7 @@ def xyxy2xywh(x):
         y (np.ndarray | torch.Tensor): The bounding box coordinates in (x, y, width, height) format.
     """
     assert x.shape[-1] == 4, f"input shape last dimension expected 4 but input shape is {x.shape}"
-    y = torch.empty_like(x) if isinstance(x, torch.Tensor) else np.empty_like(x)  # faster than clone/copy
+    y = empty_like(x)  # faster than clone/copy
     y[..., 0] = (x[..., 0] + x[..., 2]) / 2  # x center
     y[..., 1] = (x[..., 1] + x[..., 3]) / 2  # y center
     y[..., 2] = x[..., 2] - x[..., 0]  # width
@@ -420,7 +420,7 @@ def xywh2xyxy(x):
         y (np.ndarray | torch.Tensor): The bounding box coordinates in (x1, y1, x2, y2) format.
     """
     assert x.shape[-1] == 4, f"input shape last dimension expected 4 but input shape is {x.shape}"
-    y = torch.empty_like(x) if isinstance(x, torch.Tensor) else np.empty_like(x)  # faster than clone/copy
+    y = empty_like(x)  # faster than clone/copy
     xy = x[..., :2]  # centers
     wh = x[..., 2:] / 2  # half width-height
     y[..., :2] = xy - wh  # top left xy
@@ -443,7 +443,7 @@ def xywhn2xyxy(x, w=640, h=640, padw=0, padh=0):
             x1,y1 is the top-left corner, x2,y2 is the bottom-right corner of the bounding box.
     """
     assert x.shape[-1] == 4, f"input shape last dimension expected 4 but input shape is {x.shape}"
-    y = torch.empty_like(x) if isinstance(x, torch.Tensor) else np.empty_like(x)  # faster than clone/copy
+    y = empty_like(x)  # faster than clone/copy
     y[..., 0] = w * (x[..., 0] - x[..., 2] / 2) + padw  # top left x
     y[..., 1] = h * (x[..., 1] - x[..., 3] / 2) + padh  # top left y
     y[..., 2] = w * (x[..., 0] + x[..., 2] / 2) + padw  # bottom right x
@@ -469,7 +469,7 @@ def xyxy2xywhn(x, w=640, h=640, clip=False, eps=0.0):
     if clip:
         x = clip_boxes(x, (h - eps, w - eps))
     assert x.shape[-1] == 4, f"input shape last dimension expected 4 but input shape is {x.shape}"
-    y = torch.empty_like(x) if isinstance(x, torch.Tensor) else np.empty_like(x, dtype=float)  # faster than clone/copy
+    y = empty_like(x)  # faster than clone/copy
     y[..., 0] = ((x[..., 0] + x[..., 2]) / 2) / w  # x center
     y[..., 1] = ((x[..., 1] + x[..., 3]) / 2) / h  # y center
     y[..., 2] = (x[..., 2] - x[..., 0]) / w  # width
@@ -625,8 +625,9 @@ def resample_segments(segments, n=1000):
     """
     for i, s in enumerate(segments):
         s = np.concatenate((s, s[0:1, :]), axis=0)
-        x = np.linspace(0, len(s) - 1, n)
+        x = np.linspace(0, len(s) - 1, n - len(s) if len(s) < n else n)
         xp = np.arange(len(s))
+        x = np.insert(x, np.searchsorted(x, xp), xp) if len(s) < n else x
         segments[i] = (
             np.concatenate([np.interp(x, xp, s[:, i]) for i in range(2)], dtype=np.float32).reshape(2, -1).T
         )  # segment xy
@@ -837,3 +838,10 @@ def clean_str(s):
         (str): a string with special characters replaced by an underscore _
     """
     return re.sub(pattern="[|@#!¡·$€%&()=?¿^*;:,¨´><+]", repl="_", string=s)
+
+
+def empty_like(x):
+    """Creates empty torch.Tensor or np.ndarray with same shape as input and float32 dtype."""
+    return (
+        torch.empty_like(x, dtype=torch.float32) if isinstance(x, torch.Tensor) else np.empty_like(x, dtype=np.float32)
+    )

{ultralytics-8.3.43 → ultralytics-8.3.47}/ultralytics/utils/triton.py

@@ -66,6 +66,7 @@ class TritonRemoteModel:
         self.np_input_formats = [type_map[x] for x in self.input_formats]
         self.input_names = [x["name"] for x in config["input"]]
         self.output_names = [x["name"] for x in config["output"]]
+        self.metadata = eval(config.get("parameters", {}).get("metadata", {}).get("string_value", "None"))
 
     def __call__(self, *inputs: np.ndarray) -> List[np.ndarray]:
         """

{ultralytics-8.3.43 → ultralytics-8.3.47/ultralytics.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: ultralytics
-Version: 8.3.43
+Version: 8.3.47
 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>