`ultralytics 8.0.120` CLI support for SAM, RTDETR (#3273)

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

docs/datasets/detect/index.md

@@ -40,13 +40,12 @@ In this example, the first object is of class 0 (person), with its center at (0.
 
 The Ultralytics framework uses a YAML file format to define the dataset and model configuration for training Detection Models. Here is an example of the YAML format used for defining a detection dataset:
 
-```
+```yaml
 train: <path-to-training-images>
 val: <path-to-validation-images>
 
 nc: <number-of-classes>
 names: [<class-1>, <class-2>, ..., <class-n>]
-
 ```
 
 The `train` and `val` fields specify the paths to the directories containing the training and validation images, respectively.
@@ -105,7 +104,7 @@ TODO
 
 ### COCO dataset format to YOLO format
 
-```
+```python
 from ultralytics.yolo.data.converter import convert_coco
 
 convert_coco(labels_dir='../coco/annotations/')

docs/datasets/segment/index.md

@@ -104,7 +104,7 @@ names: [ 'person', 'car' ]
 
 ### COCO dataset format to YOLO format
 
-```
+```python
 from ultralytics.yolo.data.converter import convert_coco
 
 convert_coco(labels_dir='../coco/annotations/', use_segments=True)

ultralytics/__init__.py

@@ -1,6 +1,6 @@
 # Ultralytics YOLO 🚀, AGPL-3.0 license
 
-__version__ = '8.0.119'
+__version__ = '8.0.120'
 
 from ultralytics.hub import start
 from ultralytics.vit.rtdetr import RTDETR

ultralytics/nn/tasks.py

@@ -243,6 +243,8 @@ class DetectionModel(BaseModel):
             m.stride = torch.tensor([s / x.shape[-2] for x in forward(torch.zeros(1, ch, s, s))])  # forward
             self.stride = m.stride
             m.bias_init()  # only run once
+        else:
+            self.stride = torch.Tensor([32])  # default stride for i.e. RTDETR
 
         # Init weights, biases
         initialize_weights(self)

ultralytics/vit/rtdetr/model.py

@@ -5,6 +5,8 @@ RT-DETR model interface
 
 from pathlib import Path
 
+import torch.nn as nn
+
 from ultralytics.nn.tasks import RTDETRDetectionModel, attempt_load_one_weight, yaml_model_load
 from ultralytics.yolo.cfg import get_cfg
 from ultralytics.yolo.engine.exporter import Exporter
@@ -37,7 +39,7 @@ class RTDETR:
         self.task = 'detect'
         self.model = RTDETRDetectionModel(cfg_dict, verbose=verbose)  # build model
 
-        # Below added to allow export from yamls
+        # Below added to allow export from YAMLs
         self.model.args = DEFAULT_CFG_DICT  # attach args to model
         self.model.task = self.task
 
@@ -125,6 +127,23 @@ class RTDETR:
         """Get model info"""
         return model_info(self.model, verbose=verbose)
 
+    def _check_is_pytorch_model(self):
+        """
+        Raises TypeError is model is not a PyTorch model
+        """
+        pt_str = isinstance(self.model, (str, Path)) and Path(self.model).suffix == '.pt'
+        pt_module = isinstance(self.model, nn.Module)
+        if not (pt_module or pt_str):
+            raise TypeError(f"model='{self.model}' must be a *.pt PyTorch model, but is a different type. "
+                            f'PyTorch models can be used to train, val, predict and export, i.e. '
+                            f"'yolo export model=yolov8n.pt', but exported formats like ONNX, TensorRT etc. only "
+                            f"support 'predict' and 'val' modes, i.e. 'yolo predict model=yolov8n.onnx'.")
+
+    def fuse(self):
+        """Fuse PyTorch Conv2d and BatchNorm2d layers."""
+        self._check_is_pytorch_model()
+        self.model.fuse()
+
     @smart_inference_mode()
     def export(self, **kwargs):
         """

ultralytics/yolo/cfg/default.yaml

@@ -19,7 +19,7 @@ workers: 8  # (int) number of worker threads for data loading (per RANK if DDP)
 project:  # (str, optional) project name
 name:  # (str, optional) experiment name, results saved to 'project/name' directory
 exist_ok: False  # (bool) whether to overwrite existing experiment
-pretrained: True  # (bool) whether to use a pretrained model
+pretrained: True  # (bool | str) whether to use a pretrained model (bool) or a model to load weights from (str)
 optimizer: auto  # (str) optimizer to use, choices=[SGD, Adam, Adamax, AdamW, NAdam, RAdam, RMSProp, auto]
 verbose: True  # (bool) whether to print verbose output
 seed: 0  # (int) random seed for reproducibility