more configs

glenda.py

@@ -71,18 +71,16 @@ class GLENDA(datasets.GeneratorBasedBuilder):
             description="Contains images without visible pathology in relation to endometriosis (label = 'No-Pathology') and different endometriosis classes (label is exactly one of: 6.1.1.1_Endo-Peritoneum, 6.1.1.2_Endo-Ovar, 6.1.1.3_Endo-TIE, 6.1.1.4_Endo-Uterus).",
             version=datasets.Version(f"{VERSION}.0"),
         ),
-        # TODO: FIXME: Coming soon
-        # datasets.BuilderConfig(
-        #     name="object_detection",
-        #     description="Contains images without visible pathology in relation to endometriosis and different endometriosis classes with bounding box annotations.",
-        #     version=datasets.Version(f"{VERSION}.0"),
-        # ),
-        # datasets.BuilderConfig(
-        #     name="instance_segmentation",
-        #     # TODO
-        #     description="Contains images without visible pathology in relation to endometriosis and different endometriosis classes with instance segmentation annotations.",
-        #     version=datasets.Version(f"{VERSION}.0"),
-        # ),
+        datasets.BuilderConfig(
+            name="object_detection",
+            description="Contains images without visible pathology in relation to endometriosis and different endometriosis classes with corresponding COCO bounding box annotations.",
+            version=datasets.Version(f"{VERSION}.0"),
+        ),
+        datasets.BuilderConfig(
+            name="instance_segmentation",
+            description="Contains images without visible pathology in relation to endometriosis and different endometriosis classes with COCO instance segmentation annotations.",
+            version=datasets.Version(f"{VERSION}.0"),
+        ),
     ]
 
     def _info(self):
@@ -106,7 +104,6 @@ class GLENDA(datasets.GeneratorBasedBuilder):
         }
 
         task_templates = None
-        supervised_keys = None
 
         if self.config.name in ("binary_classification", "multiclass_classification"):
             class_names = CLASS_NAMES[self.config.name]
@@ -120,6 +117,18 @@ class GLENDA(datasets.GeneratorBasedBuilder):
                     image_column="image", label_column="labels"
                 )
             ]
+        elif self.config.name == "object_detection":
+            features["objects"] = {
+                "area": datasets.Value("int32"),
+                "bbox": datasets.Sequence(feature=datasets.Value("int32")),
+                "category": datasets.Value("string"),
+                "id": datasets.Value("int32"),
+            }
+            supervised_keys = (("image", "objects"),)
+        elif self.config.name == "instance_segmentation":
+            # features["segmentation"] = {
+            # }
+            supervised_keys = (("image", "objects"),)
         else:
             raise NotImplementedError()
 
@@ -152,7 +161,14 @@ class GLENDA(datasets.GeneratorBasedBuilder):
             for category in coco_annotations["categories"]
         }
 
-        image_filepaths, image_metadata, image_labels = [], [], []
+        image_filepaths, image_metadata = [], []
+
+        if self.config.name in ("binary_classification", "multiclass_classification"):
+            label_name = "labels"
+            annotation_list = []
+        elif self.config.name == "object_detection":
+            label_name = "objects"
+            annotation_list = []
 
         for annotation, metadata in zip(
             coco_annotations["annotations"],
@@ -185,9 +201,19 @@ class GLENDA(datasets.GeneratorBasedBuilder):
 
             if self.config.name == "binary_classification":
                 _, positive_label_name = CLASS_NAMES[self.config.name]
-                image_labels.append(positive_label_name)
+                annotation_list.append(positive_label_name)
             elif self.config.name == "multiclass_classification":
-                image_labels.append(category_id2_name[annotation["category_id"]])
+                annotation_list.append(category_id2_name[annotation["category_id"]])
+            elif self.config.name == "object_detection":
+                annotation_list.append({
+                    "area": annotation["area"],
+                    "bbox": annotation["bbox"],
+                    "category": category_id2_name[annotation["category_id"]],
+                    "id": annotation["category_id"],
+                })
+            elif self.config.name == "instance_segmentation":
+                raise ValueError(annotation)
+                raise NotImplementedError()
             else:
                 raise NotImplementedError()
 
@@ -214,6 +240,7 @@ class GLENDA(datasets.GeneratorBasedBuilder):
                 string=str(image_filename_with_parent_folder),
                 pattern=NO_PATHOLOGY_IMAGE_METADATA_REGEX,
             )
+
             try:
                 metadata.update(
                     {
@@ -223,14 +250,20 @@ class GLENDA(datasets.GeneratorBasedBuilder):
                 )
             except AttributeError:
                 if match is None:
-                    print("Could not get metadata for: ", image_filename_with_parent_folder)
+                    print(
+                        "Could not get metadata for: ",
+                        image_filename_with_parent_folder,
+                    )
                     raise ValueError(image_filename_with_parent_folder)
-                metadata.update({
-                    "video_id": None,
-                    "frame_id": None,
-                    "from_seconds": None,
-                    "to_seconds": None,
-                })
+
+                metadata.update(
+                    {
+                        "video_id": None,
+                        "frame_id": None,
+                        "from_seconds": None,
+                        "to_seconds": None,
+                    }
+                )
             # NOTE: Only defined for `endometriosis` images
             metadata["case_id"] = None
             image_metadata.append(metadata)
@@ -240,27 +273,64 @@ class GLENDA(datasets.GeneratorBasedBuilder):
                 "multiclass_classification",
             ):
                 negative_class_label_name, *_ = CLASS_NAMES[self.config.name]
-                image_labels.append(negative_class_label_name)
+                annotation_list.append(negative_class_label_name)
+            elif self.config.name == "object_detection":
+                negative_category_id = 0
+                negative_class_label_name, *_ = CLASS_NAMES["multiclass_classification"]
+                annotation_list.append({
+                    "area": None,
+                    "bbox": [],
+                    "category": negative_class_label_name,
+                    "id": negative_category_id,
+                })
+            elif self.config.name == "instance_segmentation":
+                raise NotImplementedError()
+            else:
+                raise NotImplementedError()
+
 
         return [
             datasets.SplitGenerator(
                 name=datasets.Split.TRAIN,
                 gen_kwargs={
                     "image_filepaths": image_filepaths,
-                    "labels": image_labels,
                     "metadata": image_metadata,
+                    label_name: annotation_list,
                 },
             )
         ]
 
-    def _generate_examples(self, image_filepaths, labels, metadata):
-        for example_id, (image_filepath, label, image_metadata) in enumerate(
-            zip(image_filepaths, labels, metadata)
-        ):
-            with open(image_filepath, "rb") as image_file:
-                yield example_id, {
-                    "image": {"path": str(image_filepath), "bytes": image_file.read()},
-                    "labels": label,
-                    "metadata": image_metadata,
-                }
-datasets.load_dataset(__file__, name="binary_classification")
+    def _generate_examples(self, **kwargs):
+        if self.config.name in ("binary_classification", "multiclass_classification"):
+            for example_id, (image_filepath, label, image_metadata) in enumerate(
+                zip(
+                    kwargs["image_filepaths"],
+                    kwargs["labels"],
+                    kwargs["metadata"]
+                )
+            ):
+                with open(image_filepath, "rb") as image_file:
+                    yield example_id, {
+                        "image": {"path": str(image_filepath), "bytes": image_file.read()},
+                        "labels": label,
+                        "metadata": image_metadata,
+                    }
+        elif self.config.name == "object_detection":
+            for example_id, (image_filepath, objects, image_metadata) in enumerate(
+                zip(
+                    kwargs["image_filepaths"],
+                    kwargs["objects"],
+                    kwargs["metadata"]
+                )
+            ):
+                with open(image_filepath, "rb") as image_file:
+                    yield example_id, {
+                        "image": {"path": str(image_filepath), "bytes": image_file.read()},
+                        "objects": objects,
+                        "metadata": image_metadata,
+                    }
+        else:
+            raise NotImplementedError()
+
+
+datasets.load_dataset(__file__, name="instance_segmentation")