ultralytics 8.2.14__tar.gz → 8.2.16__tar.gz

{ultralytics-8.2.14 → ultralytics-8.2.16}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: ultralytics
-Version: 8.2.14
+Version: 8.2.16
 Summary: Ultralytics YOLOv8 for SOTA object detection, multi-object tracking, instance segmentation, pose estimation and image classification.
 Author: Glenn Jocher, Ayush Chaurasia, Jing Qiu
 Maintainer: Glenn Jocher, Ayush Chaurasia, Jing Qiu

{ultralytics-8.2.14 → ultralytics-8.2.16}/ultralytics/__init__.py

@@ -1,6 +1,6 @@
 # Ultralytics YOLO 🚀, AGPL-3.0 license
 
-__version__ = "8.2.14"
+__version__ = "8.2.16"
 
 from ultralytics.data.explorer.explorer import Explorer
 from ultralytics.models import RTDETR, SAM, YOLO, YOLOWorld

{ultralytics-8.2.14 → ultralytics-8.2.16}/ultralytics/engine/trainer.py

@@ -527,13 +527,13 @@ class BaseTrainer:
         if isinstance(self.model, torch.nn.Module):  # if model is loaded beforehand. No setup needed
             return
 
-        model, weights = self.model, None
+        cfg, weights = self.model, None
         ckpt = None
-        if str(model).endswith(".pt"):
-            weights, ckpt = attempt_load_one_weight(model)
+        if str(self.model).endswith(".pt"):
+            weights, ckpt = attempt_load_one_weight(self.model)
             cfg = weights.yaml
-        else:
-            cfg = model
+        elif isinstance(self.args.pretrained, (str, Path)):
+            weights, _ = attempt_load_one_weight(self.args.pretrained)
         self.model = self.get_model(cfg=cfg, weights=weights, verbose=RANK == -1)  # calls Model(cfg, weights)
         return ckpt
 

{ultralytics-8.2.14 → ultralytics-8.2.16}/ultralytics/utils/__init__.py

@@ -102,13 +102,15 @@ HELP_MSG = """
     GitHub: https://github.com/ultralytics/ultralytics
     """
 
-# Settings
+# Settings and Environment Variables
 torch.set_printoptions(linewidth=320, precision=4, profile="default")
 np.set_printoptions(linewidth=320, formatter={"float_kind": "{:11.5g}".format})  # format short g, %precision=5
 cv2.setNumThreads(0)  # prevent OpenCV from multithreading (incompatible with PyTorch DataLoader)
 os.environ["NUMEXPR_MAX_THREADS"] = str(NUM_THREADS)  # NumExpr max threads
 os.environ["CUBLAS_WORKSPACE_CONFIG"] = ":4096:8"  # for deterministic training
-os.environ["TF_CPP_MIN_LOG_LEVEL"] = "2"  # suppress verbose TF compiler warnings in Colab
+os.environ["TF_CPP_MIN_LOG_LEVEL"] = "3"  # suppress verbose TF compiler warnings in Colab
+os.environ["TORCH_CPP_LOG_LEVEL"] = "ERROR"  # suppress "NNPACK.cpp could not initialize NNPACK" warnings
+os.environ["KINETO_LOG_LEVEL"] = "5"  # suppress verbose PyTorch profiler output when computing FLOPs
 
 
 class TQDM(tqdm_original):

{ultralytics-8.2.14 → ultralytics-8.2.16}/ultralytics/utils/torch_utils.py

@@ -331,19 +331,28 @@ def get_flops(model, imgsz=640):
 
 
 def get_flops_with_torch_profiler(model, imgsz=640):
-    """Compute model FLOPs (thop alternative)."""
-    if TORCH_2_0:
-        model = de_parallel(model)
-        p = next(model.parameters())
+    """Compute model FLOPs (thop package alternative, but 2-10x slower unfortunately)."""
+    if not TORCH_2_0:  # torch profiler implemented in torch>=2.0
+        return 0.0
+    model = de_parallel(model)
+    p = next(model.parameters())
+    if not isinstance(imgsz, list):
+        imgsz = [imgsz, imgsz]  # expand if int/float
+    try:
+        # Use stride size for input tensor
         stride = (max(int(model.stride.max()), 32) if hasattr(model, "stride") else 32) * 2  # max stride
-        im = torch.zeros((1, p.shape[1], stride, stride), device=p.device)  # input image in BCHW format
+        im = torch.empty((1, p.shape[1], stride, stride), device=p.device)  # input image in BCHW format
         with torch.profiler.profile(with_flops=True) as prof:
             model(im)
         flops = sum(x.flops for x in prof.key_averages()) / 1e9
-        imgsz = imgsz if isinstance(imgsz, list) else [imgsz, imgsz]  # expand if int/float
         flops = flops * imgsz[0] / stride * imgsz[1] / stride  # 640x640 GFLOPs
-        return flops
-    return 0
+    except Exception:
+        # Use actual image size for input tensor (i.e. required for RTDETR models)
+        im = torch.empty((1, p.shape[1], *imgsz), device=p.device)  # input image in BCHW format
+        with torch.profiler.profile(with_flops=True) as prof:
+            model(im)
+        flops = sum(x.flops for x in prof.key_averages()) / 1e9
+    return flops
 
 
 def initialize_weights(model):
@@ -390,8 +399,13 @@ def copy_attr(a, b, include=(), exclude=()):
 
 
 def get_latest_opset():
-    """Return second-most (for maturity) recently supported ONNX opset by this version of torch."""
-    return max(int(k[14:]) for k in vars(torch.onnx) if "symbolic_opset" in k) - 1  # opset
+    """Return the second-most recent ONNX opset version supported by this version of PyTorch, adjusted for maturity."""
+    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=()):

{ultralytics-8.2.14 → ultralytics-8.2.16}/ultralytics.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: ultralytics
-Version: 8.2.14
+Version: 8.2.16
 Summary: Ultralytics YOLOv8 for SOTA object detection, multi-object tracking, instance segmentation, pose estimation and image classification.
 Author: Glenn Jocher, Ayush Chaurasia, Jing Qiu
 Maintainer: Glenn Jocher, Ayush Chaurasia, Jing Qiu