labelr 0.11.0__tar.gz → 0.11.1__tar.gz

{labelr-0.11.0/src/labelr.egg-info → labelr-0.11.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: labelr
-Version: 0.11.0
+Version: 0.11.1
 Summary: A command-line tool to manage labeling tasks with Label Studio.
 Requires-Python: >=3.10
 Description-Content-Type: text/markdown
@@ -137,7 +137,7 @@ where `PROJECT_ID` is the ID of the project you created.
 To accelerate annotation, you can pre-annotate the images with an object detection model. We support three pre-annotation backends:
 
 - `ultralytics`: use your own model or [Yolo-World](https://docs.ultralytics.com/models/yolo-world/), a zero-shot model that can detect any object using a text description of the object. You can specify the path or the name of the model with the `--model-name` option. If no model name is provided, the `yolov8x-worldv2.pt` model (Yolo-World) is used.
-- `ultralytics_sam3`: use [SAM3](https://docs.ultralytics.com/models/sam-3/), another zero-shot model. We advice to use this backend, as it's the most accurate. The `--model-name` option is ignored when this backend is used.
+- `ultralytics_sam3`: use [SAM3](https://docs.ultralytics.com/models/sam-3/), another zero-shot model. We advice to use this backend, as it's the most accurate. The `--model` option is ignored when this backend is used.
 - `robotoff`: the ML backend of Open Food Facts (specific to Open Food Facts projects).
 
 When using `ultralytics` or `ultralytics_sam3`, make sure you installed the labelr package with the `ultralytics` extra.

{labelr-0.11.0 → labelr-0.11.1}/README.md

@@ -107,7 +107,7 @@ where `PROJECT_ID` is the ID of the project you created.
 To accelerate annotation, you can pre-annotate the images with an object detection model. We support three pre-annotation backends:
 
 - `ultralytics`: use your own model or [Yolo-World](https://docs.ultralytics.com/models/yolo-world/), a zero-shot model that can detect any object using a text description of the object. You can specify the path or the name of the model with the `--model-name` option. If no model name is provided, the `yolov8x-worldv2.pt` model (Yolo-World) is used.
-- `ultralytics_sam3`: use [SAM3](https://docs.ultralytics.com/models/sam-3/), another zero-shot model. We advice to use this backend, as it's the most accurate. The `--model-name` option is ignored when this backend is used.
+- `ultralytics_sam3`: use [SAM3](https://docs.ultralytics.com/models/sam-3/), another zero-shot model. We advice to use this backend, as it's the most accurate. The `--model` option is ignored when this backend is used.
 - `robotoff`: the ML backend of Open Food Facts (specific to Open Food Facts projects).
 
 When using `ultralytics` or `ultralytics_sam3`, make sure you installed the labelr package with the `ultralytics` extra.

{labelr-0.11.0 → labelr-0.11.1}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "labelr"
-version = "0.11.0"
+version = "0.11.1"
 description = "A command-line tool to manage labeling tasks with Label Studio."
 readme = "README.md"
 requires-python = ">=3.10"

labelr-0.11.1/src/labelr/annotate.py

@@ -0,0 +1,57 @@
+import random
+import string
+
+from openfoodfacts.utils import get_logger
+
+from ultralytics import Results
+
+logger = get_logger(__name__)
+
+
+def format_annotation_results_from_ultralytics(
+    results: Results,
+    labels: list[str],
+    label_mapping: dict[str, str] | None = None,
+) -> list[dict]:
+    annotation_results = []
+    orig_height, orig_width = results.orig_shape
+    boxes = results.boxes
+    classes = boxes.cls.tolist()
+    for i, xyxyn in enumerate(boxes.xyxyn):
+        # Boxes found.
+        if len(xyxyn) > 0:
+            xyxyn = xyxyn.tolist()
+            x1 = xyxyn[0] * 100
+            y1 = xyxyn[1] * 100
+            x2 = xyxyn[2] * 100
+            y2 = xyxyn[3] * 100
+            width = x2 - x1
+            height = y2 - y1
+            label_id = int(classes[i])
+            label_name = labels[label_id]
+            if label_mapping:
+                label_name = label_mapping.get(label_name, label_name)
+            annotation_results.append(
+                {
+                    "id": generate_id(),
+                    "type": "rectanglelabels",
+                    "from_name": "label",
+                    "to_name": "image",
+                    "original_width": orig_width,
+                    "original_height": orig_height,
+                    "image_rotation": 0,
+                    "value": {
+                        "rotation": 0,
+                        "x": x1,
+                        "y": y1,
+                        "width": width,
+                        "height": height,
+                        "rectanglelabels": [label_name],
+                    },
+                },
+            )
+    return annotation_results
+
+
+def generate_id(length: int = 10) -> str:
+    return "".join(random.choices(string.ascii_letters + string.digits, k=length))

{labelr-0.11.0 → labelr-0.11.1}/src/labelr/apps/label_studio.py

@@ -254,8 +254,20 @@ def annotate_from_prediction(
 
 class PredictorBackend(enum.StrEnum):
     ultralytics = enum.auto()
+    ultralytics_yolo_world = enum.auto()
     ultralytics_sam3 = enum.auto()
-    robotoff = enum.auto()
+
+
+YOLO_WORLD_MODELS = (
+    "yolov8s-world.pt",
+    "yolov8s-worldv2.pt",
+    "yolov8m-world.pt",
+    "yolov8m-worldv2.pt",
+    "yolov8l-world.pt",
+    "yolov8l-worldv2.pt",
+    "yolov8x-world.pt",
+    "yolov8x-worldv2.pt",
+)
 
 
 @app.command()
@@ -274,10 +286,16 @@ def add_prediction(
     model_name: Annotated[
         str | None,
         typer.Option(
-            help="Name of the object detection model to run (for Robotoff server) or "
-            "of the Ultralytics zero-shot model to run. If using Ultralytics backend "
-            "and no model name is provided, the default is yolov8x-worldv2.pt. "
-            "If using ultralytics_sam3 backend, the model name is ignored."
+            "--model",
+            help="Name or path of the object detection model to run. How this is used depends "
+            "on the backend. If using `ultralytics` backend, the option is required and is the "
+            "name of the model to download from the Ultralytics model zoo or the path to a local "
+            "model. "
+            "If using `ultralytics_yolo_world` backend, this is optional and is the name of the "
+            "`yolo-world` model to download from the Ultralytics model zoo or the path to a local "
+            "model (Defaults: `yolov8x-worldv2.pt`). "
+            "If using `ultralytics_sam3` backend, this option is ignored, as there is a single model. "
+            "The model is downloaded automatically from Hugging Face.",
         ),
     ] = None,
     skip_existing: Annotated[
@@ -286,17 +304,10 @@ def add_prediction(
             help="Skip tasks that already have predictions",
         ),
     ] = True,
-    server_url: Annotated[
-        str | None,
-        typer.Option(
-            help="The Robotoff URL if the backend is robotoff. If the backend is "
-            "different than robotoff, this option is ignored."
-        ),
-    ] = "https://robotoff.openfoodfacts.org",
     backend: Annotated[
         PredictorBackend,
         typer.Option(
-            help="The prediction backend, possible options are: `ultralytics`, `ultralytics_sam3` and `robotoff`"
+            help="The prediction backend, possible options are: `ultralytics` or `ultralytics_sam3`"
         ),
     ] = PredictorBackend.ultralytics,
     labels: Annotated[
@@ -320,9 +331,8 @@ def add_prediction(
     threshold: Annotated[
         float | None,
         typer.Option(
-            help="Confidence threshold for selecting bounding boxes. The default is 0.3 "
-            "for robotoff backend, 0.1 for ultralytics backend and 0.25 for "
-            "ultralytics_sam3 backend."
+            help="Confidence threshold for selecting bounding boxes. The default is 0.1 for "
+            "ultralytics backend and 0.25 for ultralytics_sam3 backend."
         ),
     ] = None,
     max_det: Annotated[int, typer.Option(help="Maximum numbers of detections")] = 300,
@@ -356,13 +366,10 @@ def add_prediction(
     import tqdm
     from huggingface_hub import hf_hub_download
     from label_studio_sdk.client import LabelStudio
-    from openfoodfacts.utils import get_image_from_url, http_session
+    from openfoodfacts.utils import get_image_from_url
     from PIL import Image
 
-    from ..annotate import (
-        format_annotation_results_from_robotoff,
-        format_annotation_results_from_ultralytics,
-    )
+    from ..annotate import format_annotation_results_from_ultralytics
 
     check_label_studio_api_key(api_key)
 
@@ -373,6 +380,13 @@ def add_prediction(
     if dry_run:
         logger.info("** Dry run mode enabled **")
 
+    if backend == PredictorBackend.ultralytics and not Path(model_name).is_file():
+        raise typer.BadParameter(
+            f"Model file '{model_name}' not found. When the backend is `ultralytics` "
+            "and the --model does not refer to a YOLO-World model, --model is expected "
+            "to be a local Ultralytics model file (`.pt`)."
+        )
+
     logger.info(
         "backend: %s, model_name: %s, labels: %s, threshold: %s, label mapping: %s",
         backend,
@@ -383,24 +397,26 @@ def add_prediction(
     )
     ls = LabelStudio(base_url=label_studio_url, api_key=api_key)
 
-    if backend == PredictorBackend.ultralytics:
+    if backend in (
+        PredictorBackend.ultralytics,
+        PredictorBackend.ultralytics_yolo_world,
+    ):
         from ultralytics import YOLO, YOLOWorld
 
-        if model_name is None:
-            model_name = "yolov8x-worldv2.pt"
-
         if labels is None:
-            raise typer.BadParameter("Labels are required for Ultralytics backend")
+            raise typer.BadParameter("Labels are required for `ultralytics` backend")
 
         if threshold is None:
             threshold = 0.1
 
-        model = YOLO(model_name)
-        if hasattr(model, "set_classes"):
-            model = typing.cast(YOLOWorld, model)
+        if backend == PredictorBackend.ultralytics:
+            model = YOLO(model_name)
+        elif backend == PredictorBackend.ultralytics_yolo_world:
+            if model_name is None:
+                model_name = "yolov8x-worldv2.pt"
+            model = YOLOWorld(model_name)
             model.set_classes(labels)
-        else:
-            logger.warning("The model does not support setting classes directly.")
+
     elif backend == PredictorBackend.ultralytics_sam3:
         from ultralytics.models.sam import SAM3SemanticPredictor
 
@@ -424,13 +440,6 @@ def add_prediction(
         if imgsz is not None:
             overrides["imgsz"] = imgsz
         model = SAM3SemanticPredictor(overrides=overrides)
-    elif backend == PredictorBackend.robotoff:
-        if server_url is None:
-            raise typer.BadParameter("--server-url is required for Robotoff backend")
-
-        if threshold is None:
-            threshold = 0.1
-            server_url = server_url.rstrip("/")
     else:
         raise typer.BadParameter(f"Unsupported backend: {backend}")
 
@@ -439,12 +448,14 @@ def add_prediction(
     ):
         if not (skip_existing and task.total_predictions > 0):
             image_url = task.data["image_url"]
-            image = typing.cast(
-                Image.Image,
-                get_image_from_url(image_url, error_raise=error_raise),
-            )
+            image = get_image_from_url(image_url, error_raise=error_raise)
+            if image is None:
+                continue
             min_score = None
-            if backend == PredictorBackend.ultralytics:
+            if backend in (
+                PredictorBackend.ultralytics,
+                PredictorBackend.ultralytics_yolo_world,
+            ):
                 predict_kwargs = {
                     "conf": threshold,
                     "max_det": max_det,
@@ -463,23 +474,6 @@ def add_prediction(
                     results, labels, label_mapping_dict
                 )
                 min_score = min(results.boxes.conf.tolist(), default=None)
-            elif backend == PredictorBackend.robotoff:
-                r = http_session.get(
-                    f"{server_url}/api/v1/images/predict",
-                    params={
-                        "models": model_name,
-                        "output_image": 0,
-                        "image_url": image_url,
-                    },
-                )
-                r.raise_for_status()
-                response = r.json()
-                label_studio_result = format_annotation_results_from_robotoff(
-                    response["predictions"][model_name],
-                    image.width,
-                    image.height,
-                    label_mapping_dict,
-                )
             if dry_run:
                 logger.info("image_url: %s", image_url)
                 logger.info("result: %s", label_studio_result)

{labelr-0.11.0 → labelr-0.11.1}/src/labelr/config.py

@@ -3,7 +3,7 @@ from pathlib import Path
 from pydantic import BaseModel, Field
 import os
 
-CONFIG_PATH = Path("~").expanduser() / ".config/.labelr/config.json"
+CONFIG_PATH = Path("~").expanduser() / ".config/labelr/config.json"
 
 
 # validate_assignment allows to validate the model everytime it is updated

{labelr-0.11.0 → labelr-0.11.1/src/labelr.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: labelr
-Version: 0.11.0
+Version: 0.11.1
 Summary: A command-line tool to manage labeling tasks with Label Studio.
 Requires-Python: >=3.10
 Description-Content-Type: text/markdown
@@ -137,7 +137,7 @@ where `PROJECT_ID` is the ID of the project you created.
 To accelerate annotation, you can pre-annotate the images with an object detection model. We support three pre-annotation backends:
 
 - `ultralytics`: use your own model or [Yolo-World](https://docs.ultralytics.com/models/yolo-world/), a zero-shot model that can detect any object using a text description of the object. You can specify the path or the name of the model with the `--model-name` option. If no model name is provided, the `yolov8x-worldv2.pt` model (Yolo-World) is used.
-- `ultralytics_sam3`: use [SAM3](https://docs.ultralytics.com/models/sam-3/), another zero-shot model. We advice to use this backend, as it's the most accurate. The `--model-name` option is ignored when this backend is used.
+- `ultralytics_sam3`: use [SAM3](https://docs.ultralytics.com/models/sam-3/), another zero-shot model. We advice to use this backend, as it's the most accurate. The `--model` option is ignored when this backend is used.
 - `robotoff`: the ML backend of Open Food Facts (specific to Open Food Facts projects).
 
 When using `ultralytics` or `ultralytics_sam3`, make sure you installed the labelr package with the `ultralytics` extra.

labelr-0.11.0/src/labelr/annotate.py

@@ -1,108 +0,0 @@
-import random
-import string
-
-from openfoodfacts.types import JSONType
-from openfoodfacts.utils import get_logger
-
-logger = get_logger(__name__)
-
-
-def format_annotation_results_from_robotoff(
-    objects: list[JSONType],
-    image_width: int,
-    image_height: int,
-    label_mapping: dict[str, str] | None = None,
-) -> list[JSONType]:
-    """Format annotation results from Robotoff prediction endpoint into
-    Label Studio format."""
-    annotation_results = []
-    for object_ in objects:
-        bounding_box = object_["bounding_box"]
-        label_name = object_["label"]
-
-        if label_mapping:
-            label_name = label_mapping.get(label_name, label_name)
-
-        # These are relative coordinates (between 0.0 and 1.0)
-        y_min, x_min, y_max, x_max = bounding_box
-        # Make sure the coordinates are within the image boundaries,
-        # and convert them to percentages
-        y_min = min(max(0, y_min), 1.0) * 100
-        x_min = min(max(0, x_min), 1.0) * 100
-        y_max = min(max(0, y_max), 1.0) * 100
-        x_max = min(max(0, x_max), 1.0) * 100
-        x = x_min
-        y = y_min
-        width = x_max - x_min
-        height = y_max - y_min
-
-        id_ = generate_id()
-        annotation_results.append(
-            {
-                "id": id_,
-                "type": "rectanglelabels",
-                "from_name": "label",
-                "to_name": "image",
-                "original_width": image_width,
-                "original_height": image_height,
-                "image_rotation": 0,
-                "value": {
-                    "rotation": 0,
-                    "x": x,
-                    "y": y,
-                    "width": width,
-                    "height": height,
-                    "rectanglelabels": [label_name],
-                },
-            },
-        )
-    return annotation_results
-
-
-def format_annotation_results_from_ultralytics(
-    results: "Results",
-    labels: list[str],
-    label_mapping: dict[str, str] | None = None,
-) -> list[dict]:
-    annotation_results = []
-    orig_height, orig_width = results.orig_shape
-    boxes = results.boxes
-    classes = boxes.cls.tolist()
-    for i, xyxyn in enumerate(boxes.xyxyn):
-        # Boxes found.
-        if len(xyxyn) > 0:
-            xyxyn = xyxyn.tolist()
-            x1 = xyxyn[0] * 100
-            y1 = xyxyn[1] * 100
-            x2 = xyxyn[2] * 100
-            y2 = xyxyn[3] * 100
-            width = x2 - x1
-            height = y2 - y1
-            label_id = int(classes[i])
-            label_name = labels[label_id]
-            if label_mapping:
-                label_name = label_mapping.get(label_name, label_name)
-            annotation_results.append(
-                {
-                    "id": generate_id(),
-                    "type": "rectanglelabels",
-                    "from_name": "label",
-                    "to_name": "image",
-                    "original_width": orig_width,
-                    "original_height": orig_height,
-                    "image_rotation": 0,
-                    "value": {
-                        "rotation": 0,
-                        "x": x1,
-                        "y": y1,
-                        "width": width,
-                        "height": height,
-                        "rectanglelabels": [label_name],
-                    },
-                },
-            )
-    return annotation_results
-
-
-def generate_id(length: int = 10) -> str:
-    return "".join(random.choices(string.ascii_letters + string.digits, k=length))