ultralytics 8.3.57__tar.gz → 8.3.59__tar.gz

{ultralytics-8.3.57/ultralytics.egg-info → ultralytics-8.3.59}/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.1
+Metadata-Version: 2.2
 Name: ultralytics
-Version: 8.3.57
+Version: 8.3.59
 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-8.3.57 → ultralytics-8.3.59}/ultralytics/__init__.py

@@ -1,6 +1,6 @@
 # Ultralytics YOLO 🚀, AGPL-3.0 license
 
-__version__ = "8.3.57"
+__version__ = "8.3.59"
 
 import os
 

{ultralytics-8.3.57 → ultralytics-8.3.59}/ultralytics/cfg/datasets/package-seg.yaml

@@ -9,8 +9,8 @@
 
 # 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/package-seg # dataset root dir
-train: images/train # train images (relative to 'path') 1920 images
-val: images/val # val images (relative to 'path') 89 images
+train: train/images # train images (relative to 'path') 1920 images
+val: valid/images # val images (relative to 'path') 89 images
 test: test/images # test images (relative to 'path') 188 images
 
 # Classes

ultralytics-8.3.59/ultralytics/cfg/models/11/yolo11-cls-resnet18.yaml

@@ -0,0 +1,21 @@
+# Ultralytics YOLO 🚀, AGPL-3.0 license
+# YOLO11-cls image classification model. For Usage examples see https://docs.ultralytics.com/tasks/classify
+
+# Parameters
+nc: 10 # number of classes
+scales: # model compound scaling constants, i.e. 'model=yolo11n-cls.yaml' will call yolo11-cls.yaml with scale 'n'
+  # [depth, width, max_channels]
+  n: [0.33, 0.25, 1024]
+  s: [0.33, 0.50, 1024]
+  m: [0.67, 0.75, 1024]
+  l: [1.00, 1.00, 1024]
+  x: [1.00, 1.25, 1024]
+
+# YOLO11n backbone
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, TorchVision, [512, "resnet18", "DEFAULT", True, 2]] # truncate two layers from the end
+
+# YOLO11n head
+head:
+  - [-1, 1, Classify, [nc]] # Classify

{ultralytics-8.3.57 → ultralytics-8.3.59}/ultralytics/data/converter.py

@@ -377,7 +377,7 @@ def convert_segment_masks_to_yolo_seg(masks_dir, output_dir, classes):
     """
     pixel_to_class_mapping = {i + 1: i for i in range(classes)}
     for mask_path in Path(masks_dir).iterdir():
-        if mask_path.suffix == ".png":
+        if mask_path.suffix in {".png", ".jpg"}:
             mask = cv2.imread(str(mask_path), cv2.IMREAD_GRAYSCALE)  # Read the mask image in grayscale
             img_height, img_width = mask.shape  # Get image dimensions
             LOGGER.info(f"Processing {mask_path} imgsz = {img_height} x {img_width}")

{ultralytics-8.3.57 → ultralytics-8.3.59}/ultralytics/engine/exporter.py

@@ -452,7 +452,11 @@ class Exporter:
             batch_size=batch,
         )
         n = len(dataset)
-        if n < 300:
+        if n < self.args.batch:
+            raise ValueError(
+                f"The calibration dataset ({n} images) must have at least as many images as the batch size ('batch={self.args.batch}')."
+            )
+        elif n < 300:
             LOGGER.warning(f"{prefix} WARNING ⚠️ >300 images recommended for INT8 calibration, found {n} images.")
         return build_dataloader(dataset, batch=batch, workers=0)  # required for batch loading
 

{ultralytics-8.3.57 → ultralytics-8.3.59}/ultralytics/nn/modules/__init__.py

@@ -56,6 +56,7 @@ from .block import (
     RepVGGDW,
     ResNetLayer,
     SCDown,
+    TorchVision,
 )
 from .conv import (
     CBAM,
@@ -68,6 +69,7 @@ from .conv import (
     DWConvTranspose2d,
     Focus,
     GhostConv,
+    Index,
     LightConv,
     RepConv,
     SpatialAttention,
@@ -156,4 +158,6 @@ __all__ = (
     "C2fCIB",
     "Attention",
     "PSA",
+    "TorchVision",
+    "Index",
 )

{ultralytics-8.3.57 → ultralytics-8.3.59}/ultralytics/nn/modules/block.py

@@ -49,6 +49,7 @@ __all__ = (
     "Attention",
     "PSA",
     "SCDown",
+    "TorchVision",
 )
 
 
@@ -1107,3 +1108,51 @@ class SCDown(nn.Module):
     def forward(self, x):
         """Applies convolution and downsampling to the input tensor in the SCDown module."""
         return self.cv2(self.cv1(x))
+
+
+class TorchVision(nn.Module):
+    """
+    TorchVision module to allow loading any torchvision model.
+
+    This class provides a way to load a model from the torchvision library, optionally load pre-trained weights, and customize the model by truncating or unwrapping layers.
+
+    Attributes:
+        m (nn.Module): The loaded torchvision model, possibly truncated and unwrapped.
+
+    Args:
+        c1 (int): Input channels.
+        c2 (): Output channels.
+        model (str): Name of the torchvision model to load.
+        weights (str, optional): Pre-trained weights to load. Default is "DEFAULT".
+        unwrap (bool, optional): If True, unwraps the model to a sequential containing all but the last `truncate` layers. Default is True.
+        truncate (int, optional): Number of layers to truncate from the end if `unwrap` is True. Default is 2.
+        split (bool, optional): Returns output from intermediate child modules as list. Default is False.
+    """
+
+    def __init__(self, c1, c2, model, weights="DEFAULT", unwrap=True, truncate=2, split=False):
+        """Load the model and weights from torchvision."""
+        import torchvision
+
+        super().__init__()
+        if hasattr(torchvision.models, "get_model"):
+            self.m = torchvision.models.get_model(model, weights=weights)
+        else:
+            self.m = torchvision.models.__dict__[model](pretrained=bool(weights))
+        if unwrap:
+            layers = list(self.m.children())[:-truncate]
+            if isinstance(layers[0], nn.Sequential):  # Second-level for some models like EfficientNet, Swin
+                layers = [*list(layers[0].children()), *layers[1:]]
+            self.m = nn.Sequential(*layers)
+            self.split = split
+        else:
+            self.split = False
+            self.m.head = self.m.heads = nn.Identity()
+
+    def forward(self, x):
+        """Forward pass through the model."""
+        if self.split:
+            y = [x]
+            y.extend(m(y[-1]) for m in self.m)
+        else:
+            y = self.m(x)
+        return y

{ultralytics-8.3.57 → ultralytics-8.3.59}/ultralytics/nn/modules/conv.py

@@ -21,6 +21,7 @@ __all__ = (
     "CBAM",
     "Concat",
     "RepConv",
+    "Index",
 )
 
 
@@ -330,3 +331,20 @@ class Concat(nn.Module):
     def forward(self, x):
         """Forward pass for the YOLOv8 mask Proto module."""
         return torch.cat(x, self.d)
+
+
+class Index(nn.Module):
+    """Returns a particular index of the input."""
+
+    def __init__(self, c1, c2, index=0):
+        """Returns a particular index of the input."""
+        super().__init__()
+        self.index = index
+
+    def forward(self, x):
+        """
+        Forward pass.
+
+        Expects a list of tensors as input.
+        """
+        return x[self.index]

{ultralytics-8.3.57 → ultralytics-8.3.59}/ultralytics/nn/tasks.py

@@ -50,6 +50,7 @@ from ultralytics.nn.modules import (
     HGBlock,
     HGStem,
     ImagePoolingAttn,
+    Index,
     Pose,
     RepC3,
     RepConv,
@@ -59,6 +60,7 @@ from ultralytics.nn.modules import (
     RTDETRDecoder,
     SCDown,
     Segment,
+    TorchVision,
     WorldDetect,
     v10Detect,
 )
@@ -1052,7 +1054,7 @@ def parse_model(d, ch, verbose=True):  # model_dict, input_channels(3)
                 m.legacy = legacy
         elif m is RTDETRDecoder:  # special case, channels arg must be passed in index 1
             args.insert(1, [ch[x] for x in f])
-        elif m is CBLinear:
+        elif m in {CBLinear, TorchVision, Index}:
             c2 = args[0]
             c1 = ch[f]
             args = [c1, c2, *args[1:]]

{ultralytics-8.3.57 → ultralytics-8.3.59}/ultralytics/solutions/streamlit_inference.py

@@ -184,12 +184,8 @@ class Inference:
 if __name__ == "__main__":
     import sys  # Import the sys module for accessing command-line arguments
 
-    model = None  # Initialize the model variable as None
-
     # Check if a model name is provided as a command-line argument
     args = len(sys.argv)
-    if args > 1:
-        model = sys.argv[1]  # Assign the first argument as the model name
-
+    model = sys.argv[1] if args > 1 else None  # assign first argument as the model name
     # Create an instance of the Inference class and run inference
     Inference(model=model).inference()

{ultralytics-8.3.57 → ultralytics-8.3.59}/ultralytics/utils/benchmarks.py

@@ -464,7 +464,7 @@ class ProfileModels:
 
         # Model and input
         model = YOLO(engine_file)
-        input_data = np.random.rand(self.imgsz, self.imgsz, 3).astype(np.float32)  # must be FP32
+        input_data = np.zeros((self.imgsz, self.imgsz, 3), dtype=np.uint8)  # use uint8 for Classify
 
         # Warmup runs
         elapsed = 0.0

{ultralytics-8.3.57 → ultralytics-8.3.59}/ultralytics/utils/metrics.py

@@ -73,11 +73,16 @@ def box_iou(box1, box2, eps=1e-7):
 
 def bbox_iou(box1, box2, xywh=True, GIoU=False, DIoU=False, CIoU=False, eps=1e-7):
     """
-    Calculate Intersection over Union (IoU) of box1(1, 4) to box2(n, 4).
+    Calculates the Intersection over Union (IoU) between bounding boxes.
+
+    This function supports various shapes for `box1` and `box2` as long as the last dimension is 4.
+    For instance, you may pass tensors shaped like (4,), (N, 4), (B, N, 4), or (B, N, 1, 4).
+    Internally, the code will split the last dimension into (x, y, w, h) if `xywh=True`,
+    or (x1, y1, x2, y2) if `xywh=False`.
 
     Args:
-        box1 (torch.Tensor): A tensor representing a single bounding box with shape (1, 4).
-        box2 (torch.Tensor): A tensor representing n bounding boxes with shape (n, 4).
+        box1 (torch.Tensor): A tensor representing one or more bounding boxes, with the last dimension being 4.
+        box2 (torch.Tensor): A tensor representing one or more bounding boxes, with the last dimension being 4.
         xywh (bool, optional): If True, input boxes are in (x, y, w, h) format. If False, input boxes are in
                                (x1, y1, x2, y2) format. Defaults to True.
         GIoU (bool, optional): If True, calculate Generalized IoU. Defaults to False.

{ultralytics-8.3.57 → ultralytics-8.3.59/ultralytics.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.1
+Metadata-Version: 2.2
 Name: ultralytics
-Version: 8.3.57
+Version: 8.3.59
 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-8.3.57 → ultralytics-8.3.59}/ultralytics.egg-info/SOURCES.txt

@@ -58,6 +58,7 @@ ultralytics/cfg/datasets/package-seg.yaml
 ultralytics/cfg/datasets/signature.yaml
 ultralytics/cfg/datasets/tiger-pose.yaml
 ultralytics/cfg/datasets/xView.yaml
+ultralytics/cfg/models/11/yolo11-cls-resnet18.yaml
 ultralytics/cfg/models/11/yolo11-cls.yaml
 ultralytics/cfg/models/11/yolo11-obb.yaml
 ultralytics/cfg/models/11/yolo11-pose.yaml