docs improvements

docs/how_to/detect_and_annotate.md

@@ -102,7 +102,7 @@ Now that we have predictions from a model, we can load them into Supervision.
 
     We can do so using the [`sv.Detections.from_transformers`](detection/core/#supervision.detection.core.Detections.from_transformers) method, which accepts model results from both detection and segmentation models.
 
-    ```{ .py hl_lines="2 19" }
+    ```{ .py hl_lines="2 19-21" }
     import torch
     import supervision as sv
     from PIL import Image
@@ -121,7 +121,9 @@ Now that we have predictions from a model, we can load them into Supervision.
     target_size = torch.tensor([[height, width]])
     results = processor.post_process_object_detection(
         outputs=outputs, target_sizes=target_size)[0]
-    detections = sv.Detections.from_transformers(results)
+    detections = sv.Detections.from_transformers(
+        transformers_results=results, 
+        id2label=model.config.id2label)
     ```
 
 You can load predictions from other computer vision frameworks and libraries using:
@@ -180,7 +182,7 @@ Finally, we can annotate the image with the predictions. Since we are working wi
 
 === "Transformers"
 
-    ```{ .py hl_lines="21-27" }
+    ```{ .py hl_lines="23-30" }
     import torch
     import supervision as sv
     from PIL import Image
@@ -199,7 +201,9 @@ Finally, we can annotate the image with the predictions. Since we are working wi
     target_size = torch.tensor([[height, width]])
     results = processor.post_process_object_detection(
         outputs=outputs, target_sizes=target_size)[0]
-    detections = sv.Detections.from_transformers(results)
+    detections = sv.Detections.from_transformers(
+        transformers_results=results, 
+        id2label=model.config.id2label)
 
     bounding_box_annotator = sv.BoundingBoxAnnotator()
     label_annotator = sv.LabelAnnotator()
@@ -274,7 +278,7 @@ override this behavior by passing a list of custom `labels` to the `annotate` me
 
 === "Transformers"
 
-    ```{ .py hl_lines="24-28 33" }
+    ```{ .py hl_lines="26-30 35" }
     import torch
     import supervision as sv
     from PIL import Image
@@ -293,15 +297,17 @@ override this behavior by passing a list of custom `labels` to the `annotate` me
     target_size = torch.tensor([[height, width]])
     results = processor.post_process_object_detection(
         outputs=outputs, target_sizes=target_size)[0]
-    detections = sv.Detections.from_transformers(results)
+    detections = sv.Detections.from_transformers(
+        transformers_results=results, 
+        id2label=model.config.id2label)
 
     bounding_box_annotator = sv.BoundingBoxAnnotator()
     label_annotator = sv.LabelAnnotator()
 
     labels = [
-        f"{model.config.id2label[class_id]} {confidence:.2f}"
-        for class_id, confidence
-        in zip(detections.class_id, detections.confidence)
+        f"{class_name} {confidence:.2f}"
+        for class_name, confidence
+        in zip(detections['class_name'], detections.confidence)
     ]
 
     annotated_image = bounding_box_annotator.annotate(
@@ -383,15 +389,17 @@ that will allow you to draw masks instead of boxes.
     target_size = torch.tensor([[height, width]])
     results = processor.post_process_segmentation(
         outputs=outputs, target_sizes=target_size)[0]
-    detections = sv.Detections.from_transformers(results)
+    detections = sv.Detections.from_transformers(
+        transformers_results=results, 
+        id2label=model.config.id2label)
 
     mask_annotator = sv.MaskAnnotator()
     label_annotator = sv.LabelAnnotator(text_position=sv.Position.CENTER)
 
     labels = [
-        f"{model.config.id2label[class_id]} {confidence:.2f}"
-        for class_id, confidence
-        in zip(detections.class_id, detections.confidence)
+        f"{class_name} {confidence:.2f}"
+        for class_name, confidence
+        in zip(detections['class_name'], detections.confidence)
     ]
 
     annotated_image = mask_annotator.annotate(

supervision/detection/core.py

@@ -189,7 +189,7 @@ class Detections:
 
         Args:
             ultralytics_results (ultralytics.yolo.engine.results.Results):
-                The output Results instance from YOLOv8
+                The output Results instance from Ultralytics
 
         Returns:
             Detections: A new Detections object.
@@ -394,12 +394,19 @@ class Detections:
     ) -> Detections:
         """
         Creates a Detections instance from object detection or segmentation
-        [transformer](https://github.com/huggingface/transformers) inference result.
+        [Transformer](https://github.com/huggingface/transformers) inference result.
+        
+        !!! note
+
+            Class names can be accessed using the key `class_name` in the returned
+            object's data attribute.
 
         Args:
-            transformers_results (dict): A dictionary containing the scores, labels, and boxes for an image
-                as predicted by the Transformers model.
-            id2label (Optional[Dict[int, str]]): A dictionary mapping class IDs to class names. If None, class names will not be included.
+            transformers_results (dict): The output of Transformers model inference. A 
+                dictionary containing the `scores`, `labels`, `boxes` and `masks` keys.
+            id2label (Optional[Dict[int, str]]): A dictionary mapping class IDs to 
+                class names. If provided, the resulting Detections object will contain
+                `class_name` data field with the class names.
 
         Returns:
             Detections: A new Detections object.
@@ -424,7 +431,11 @@ class Detections:
             target_size = torch.tensor([[height, width]])
             results = processor.post_process_object_detection(
                 outputs=outputs, target_sizes=target_size)[0]
-            detections = sv.Detections.from_transformers(results)
+                
+            detections = sv.Detections.from_transformers(
+                transformers_results=results, 
+                id2label=model.config.id2label
+            )
             ```
         """  # noqa: E501 // docs
 
@@ -507,7 +518,7 @@ class Detections:
 
         !!! note
 
-            Class names can be accessed using the key 'class_name' in the returned
+            Class names can be accessed using the key `class_name` in the returned
             object's data attribute.
 
         Args: