ultralytics 8.3.100__tar.gz → 8.3.101__tar.gz

{ultralytics-8.3.100/ultralytics.egg-info → ultralytics-8.3.101}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ultralytics
-Version: 8.3.100
+Version: 8.3.101
 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-8.3.100 → ultralytics-8.3.101}/ultralytics/__init__.py

@@ -1,6 +1,6 @@
 # Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
 
-__version__ = "8.3.100"
+__version__ = "8.3.101"
 
 import os
 

{ultralytics-8.3.100 → ultralytics-8.3.101}/ultralytics/engine/model.py

@@ -534,7 +534,7 @@ class Model(torch.nn.Module):
             x in ARGV for x in ("predict", "track", "mode=predict", "mode=track")
         )
 
-        custom = {"conf": 0.25, "batch": 1, "save": is_cli, "mode": "predict"}  # method defaults
+        custom = {"conf": 0.25, "batch": 1, "save": is_cli, "mode": "predict", "rect": True}  # method defaults
         args = {**self.overrides, **custom, **kwargs}  # highest priority args on the right
         prompts = args.pop("prompts", None)  # for SAM-type models
 

{ultralytics-8.3.100 → ultralytics-8.3.101}/ultralytics/engine/predictor.py

@@ -183,7 +183,9 @@ 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) and not self.model.imx)),
+            auto=same_shapes
+            and self.args.rect
+            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-8.3.100 → ultralytics-8.3.101}/ultralytics/models/yolo/model.py

@@ -2,6 +2,7 @@
 
 from pathlib import Path
 
+from ultralytics.data.build import load_inference_source
 from ultralytics.engine.model import Model
 from ultralytics.models import yolo
 from ultralytics.nn.tasks import (
@@ -267,7 +268,14 @@ class YOLOE(Model):
                 f"{len(visual_prompts['cls'])} respectively"
             )
         self.predictor = (predictor or self._smart_load("predictor"))(
-            overrides={"task": "segment", "mode": "predict", "save": False, "verbose": False}, _callbacks=self.callbacks
+            overrides={
+                "task": self.model.task,
+                "mode": "predict",
+                "save": False,
+                "verbose": refer_image is None,
+                "batch": 1,
+            },
+            _callbacks=self.callbacks,
         )
 
         if len(visual_prompts):
@@ -281,6 +289,12 @@ class YOLOE(Model):
             self.predictor.set_prompts(visual_prompts.copy())
 
         self.predictor.setup_model(model=self.model)
+
+        if refer_image is None:
+            dataset = load_inference_source(source)
+            if dataset.mode in {"video", "stream"}:
+                # NOTE: set the first frame as refer image for videos/streams inference
+                refer_image = next(iter(dataset))[1][0]
         if refer_image is not None and len(visual_prompts):
             vpe = self.predictor.get_vpe(refer_image)
             self.model.set_classes(self.model.names, vpe)

{ultralytics-8.3.100 → ultralytics-8.3.101}/ultralytics/nn/tasks.py

@@ -972,8 +972,8 @@ class YOLOEModel(DetectionModel):
                     assert not self.training
                     return vpe
                 cls_pe = self.get_cls_pe(m.get_tpe(tpe), vpe).to(device=x[0].device, dtype=x[0].dtype)
-                if len(cls_pe) != b:
-                    cls_pe = cls_pe.repeat(b, 1, 1)
+                if cls_pe.shape[0] != b or m.export:
+                    cls_pe = cls_pe.expand(b, -1, -1)
                 x = m(x, cls_pe)
             else:
                 x = m(x)  # run

{ultralytics-8.3.100 → ultralytics-8.3.101/ultralytics.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ultralytics
-Version: 8.3.100
+Version: 8.3.101
 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>