ultralytics 8.3.112__tar.gz → 8.3.113__tar.gz

{ultralytics-8.3.112/ultralytics.egg-info → ultralytics-8.3.113}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ultralytics
-Version: 8.3.112
+Version: 8.3.113
 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-8.3.112 → ultralytics-8.3.113}/ultralytics/__init__.py

@@ -1,6 +1,6 @@
 # Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
 
-__version__ = "8.3.112"
+__version__ = "8.3.113"
 
 import os
 

ultralytics-8.3.113/ultralytics/cfg/datasets/dota8-multispectral.yaml

@@ -0,0 +1,38 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# DOTA8-Multispectral dataset (DOTA8 interpolated across 10 channels in the visual spectrum) by Ultralytics
+# Documentation: https://docs.ultralytics.com/datasets/obb/dota8/
+# Example usage: yolo train model=yolov8n-obb.pt data=dota8-multispectral.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── dota8-multispectral  ← downloads here (1MB)
+
+# 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/dota8-multispectral # dataset root dir
+train: images/train # train images (relative to 'path') 4 images
+val: images/val # val images (relative to 'path') 4 images
+
+# Number of multispectral image channels
+channels: 10
+
+# Classes for DOTA 1.0
+names:
+  0: plane
+  1: ship
+  2: storage tank
+  3: baseball diamond
+  4: tennis court
+  5: basketball court
+  6: ground track field
+  7: harbor
+  8: bridge
+  9: large vehicle
+  10: small vehicle
+  11: helicopter
+  12: roundabout
+  13: soccer ball field
+  14: swimming pool
+
+# Download script/URL (optional)
+download: https://github.com/ultralytics/assets/releases/download/v0.0.0/dota8-multispectral.zip

{ultralytics-8.3.112 → ultralytics-8.3.113}/ultralytics/data/augment.py

@@ -2117,7 +2117,7 @@ class Format:
         if len(img.shape) < 3:
             img = np.expand_dims(img, -1)
         img = img.transpose(2, 0, 1)
-        img = np.ascontiguousarray(img[::-1] if random.uniform(0, 1) > self.bgr else img)
+        img = np.ascontiguousarray(img[::-1] if random.uniform(0, 1) > self.bgr and img.shape[0] == 3 else img)
         img = torch.from_numpy(img)
         return img
 

{ultralytics-8.3.112 → ultralytics-8.3.113}/ultralytics/data/converter.py

@@ -718,7 +718,9 @@ def convert_to_multispectral(path, n_channels=10, replace=False, zip=False):
 
     Examples:
         >>> # Convert a single image
-        >>> convert_to_multispectral("path/to/image.jpg", n_channels=12)
+        >>> convert_to_multispectral("path/to/image.jpg", n_channels=10)
+        >>> # Convert a dataset
+        >>> convert_to_multispectral("../datasets/coco8", n_channels=10)
     """
     from scipy.interpolate import interp1d
 

{ultralytics-8.3.112 → ultralytics-8.3.113}/ultralytics/data/loaders.py

@@ -492,7 +492,7 @@ class LoadPilAndNumpy:
         if isinstance(im, Image.Image):
             if im.mode != "RGB":
                 im = im.convert("RGB")
-            im = np.asarray(im)[:, :, ::-1]
+            im = np.asarray(im)[:, :, ::-1]  # RGB to BGR
             im = np.ascontiguousarray(im)  # contiguous
         return im
 

{ultralytics-8.3.112 → ultralytics-8.3.113}/ultralytics/engine/predictor.py

@@ -151,7 +151,9 @@ class BasePredictor:
         not_tensor = not isinstance(im, torch.Tensor)
         if not_tensor:
             im = np.stack(self.pre_transform(im))
-            im = im[..., ::-1].transpose((0, 3, 1, 2))  # BGR to RGB, BHWC to BCHW, (n, 3, h, w)
+            if im.shape[-1] == 3:
+                im = im[..., ::-1]  # BGR to RGB
+            im = im.transpose((0, 3, 1, 2))  # BHWC to BCHW, (n, 3, h, w)
             im = np.ascontiguousarray(im)  # contiguous
             im = torch.from_numpy(im)
 

{ultralytics-8.3.112 → ultralytics-8.3.113}/ultralytics/nn/autobackend.py

@@ -148,7 +148,7 @@ class AutoBackend(nn.Module):
         model, metadata, task = None, None, None
 
         # Set device
-        cuda = torch.cuda.is_available() and device.type != "cpu"  # use CUDA
+        cuda = isinstance(device, torch.device) and torch.cuda.is_available() and device.type != "cpu"  # use CUDA
         if cuda and not any([nn_module, pt, jit, engine, onnx, paddle]):  # GPU dataloader formats
             device = torch.device("cpu")
             cuda = False
@@ -264,6 +264,13 @@ class AutoBackend(nn.Module):
             import openvino as ov
 
             core = ov.Core()
+            device_name = "AUTO"
+            if isinstance(device, str) and device.startswith("intel"):
+                device_name = device.split(":")[1].upper()  # Intel OpenVINO device
+                device = torch.device("cpu")
+                if device_name not in core.available_devices:
+                    LOGGER.warning(f"OpenVINO device '{device_name}' not available. Using 'AUTO' instead.")
+                    device_name = "AUTO"
             w = Path(w)
             if not w.is_file():  # if not *.xml
                 w = next(w.glob("*.xml"))  # get *.xml file from *_openvino_model dir
@@ -276,7 +283,7 @@ class AutoBackend(nn.Module):
             LOGGER.info(f"Using OpenVINO {inference_mode} mode for batch={batch} inference...")
             ov_compiled_model = core.compile_model(
                 ov_model,
-                device_name="AUTO",  # AUTO selects best available device, do not modify
+                device_name=device_name,
                 config={"PERFORMANCE_HINT": inference_mode},
             )
             input_name = ov_compiled_model.input().get_any_name()

{ultralytics-8.3.112 → ultralytics-8.3.113}/ultralytics/solutions/region_counter.py

@@ -96,8 +96,8 @@ class RegionCounter(BaseSolution):
 
         # Process bounding boxes & check containment
         if points:
-            for (point, cls), box in zip(zip(points, self.clss), self.boxes):
-                annotator.box_label(box, label=self.names[cls], color=colors(cls))
+            for point, cls, track_id, box in zip(points, self.clss, self.track_ids, self.boxes):
+                annotator.box_label(box, label=self.names[cls], color=colors(track_id, True))
 
                 for region in self.counting_regions:
                     if region["prepared_polygon"].contains(point):
@@ -111,6 +111,7 @@ class RegionCounter(BaseSolution):
                 label=str(region["counts"]),
                 color=region["region_color"],
                 txt_color=region["text_color"],
+                margin=self.line_width * 4,
             )
             region["counts"] = 0  # Reset for next frame
         plot_im = annotator.result()

{ultralytics-8.3.112 → ultralytics-8.3.113}/ultralytics/utils/torch_utils.py

@@ -161,7 +161,7 @@ def select_device(device="", batch=0, newline=False, verbose=True):
     Note:
         Sets the 'CUDA_VISIBLE_DEVICES' environment variable for specifying which GPUs to use.
     """
-    if isinstance(device, torch.device) or str(device).startswith("tpu"):
+    if isinstance(device, torch.device) or str(device).startswith("tpu") or str(device).startswith("intel"):
         return device
 
     s = f"Ultralytics {__version__} 🚀 Python-{PYTHON_VERSION} torch-{torch.__version__} "

{ultralytics-8.3.112 → ultralytics-8.3.113/ultralytics.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ultralytics
-Version: 8.3.112
+Version: 8.3.113
 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-8.3.112 → ultralytics-8.3.113}/ultralytics.egg-info/SOURCES.txt

@@ -41,6 +41,7 @@ ultralytics/cfg/datasets/coco8-seg.yaml
 ultralytics/cfg/datasets/coco8.yaml
 ultralytics/cfg/datasets/crack-seg.yaml
 ultralytics/cfg/datasets/dog-pose.yaml
+ultralytics/cfg/datasets/dota8-multispectral.yaml
 ultralytics/cfg/datasets/dota8.yaml
 ultralytics/cfg/datasets/hand-keypoints.yaml
 ultralytics/cfg/datasets/lvis.yaml