add annotations

Caltech-101.py

@@ -14,10 +14,13 @@
 """Caltech 101 loading script"""
 
 
+from __future__ import annotations
+
 from pathlib import Path
 
 import datasets
 import numpy as np
+import scipy.io
 from datasets.tasks import ImageClassification
 
 _CITATION = """\
@@ -147,6 +150,15 @@ _NAMES = [
     "wrench",
     "yin_yang",
 ]
+# For some reason, the category names in "101_ObjectCategories" and
+# "Annotations" do not always match. This is a manual map between the
+# two. Defaults to using same name, since most names are fine.
+_ANNOTATION_NAMES_MAP = {
+    "Faces": "Faces_2",
+    "Faces_easy": "Faces_3",
+    "Motorbikes": "Motorbikes_16",
+    "airplanes": "Airplanes_Side_2",
+}
 
 _TRAIN_POINTS_PER_CLASS = 30
 
@@ -173,36 +185,56 @@ class Caltech101(datasets.GeneratorBasedBuilder):
     ]
 
     def _info(self):
-        return datasets.DatasetInfo(
-            description=_DESCRIPTION,
-            features=datasets.Features(
+        if self.config.name == self._BUILDER_CONFIG_WITHOUT_BACKGROUND.name:
+            features = datasets.Features(
                 {
                     "image": datasets.Image(),
                     "label": datasets.features.ClassLabel(names=_NAMES),
+                    "annotation": {
+                        "obj_contour": datasets.features.Array2D(
+                            shape=(2, None), dtype="float64"
+                        ),
+                        "box_coord": datasets.features.Array2D(
+                            shape=(1, 4), dtype="int64"
+                        ),
+                    },
                 }
-            ),
-            supervised_keys=("image", "label"),
+            )
+        else:
+            features = datasets.Features(
+                {
+                    "image": datasets.Image(),
+                    "label": datasets.features.ClassLabel(names=_NAMES),
+                }
+            )
+        return datasets.DatasetInfo(
+            description=_DESCRIPTION,
+            features=features,
             homepage=_HOMEPAGE,
             license=_LICENSE,
             citation=_CITATION,
-            task_templates=ImageClassification(
-                image_column="image", label_column="label"
-            ),
         )
 
     def _split_generators(self, dl_manager):
         data_root_dir = dl_manager.download_and_extract(_DATA_URL)
-        compress_folder_path = [
+        img_folder_compress_path = [
             file
             for file in dl_manager.iter_files(data_root_dir)
             if Path(file).name == "101_ObjectCategories.tar.gz"
         ][0]
-        data_dir = dl_manager.extract(compress_folder_path)
+        annotations_folder_compress_path = [
+            file
+            for file in dl_manager.iter_files(data_root_dir)
+            if Path(file).name == "Annotations.tar"
+        ][0]
+        img_dir = dl_manager.extract(img_folder_compress_path)
+        annotation_dir = dl_manager.extract(annotations_folder_compress_path)
         return [
             datasets.SplitGenerator(
                 name=datasets.Split.TRAIN,
                 gen_kwargs={
-                    "filepath": data_dir,
+                    "img_dir": Path(img_dir) / "101_ObjectCategories",
+                    "annotation_dir": Path(annotation_dir) / "Annotations",
                     "split": "train",
                     "config_name": self.config.name,
                 },
@@ -210,58 +242,76 @@ class Caltech101(datasets.GeneratorBasedBuilder):
             datasets.SplitGenerator(
                 name=datasets.Split.TEST,
                 gen_kwargs={
-                    "filepath": data_dir,
+                    "img_dir": Path(img_dir) / "101_ObjectCategories",
+                    "annotation_dir": Path(annotation_dir) / "Annotations",
                     "split": "test",
                     "config_name": self.config.name,
                 },
             ),
         ]
 
-    def _generate_examples(self, filepath, split, config_name):
+    def _generate_examples(self, img_dir, annotation_dir, split, config_name):
         # Same stratagy as the one proposed in TF datasets: 30 random examples from each class are added to the train
         # split, and the remainder are added to the test split.
         # Source: https://github.com/tensorflow/datasets/blob/1106d587f97c4fca68c5b593dc7dc48c790ffa8c/tensorflow_datasets/image_classification/caltech.py#L88-L140
 
         is_train_split = split == "train"
-        data_dir = Path(filepath) / "101_ObjectCategories"
+
         # Sets random seed so the random partitioning of files is the same when
         # called for the train and test splits.
         numpy_original_state = np.random.get_state()
         np.random.seed(1234)
 
-        for class_dir in data_dir.iterdir():
-            # print(class_dir)
-            fnames = [
-                image_path
+        for class_dir in img_dir.iterdir():
+            class_name = class_dir.name
+            index_codes = [
+                image_path.name.split("_")[1][: -len(".jpg")]
                 for image_path in class_dir.iterdir()
                 if image_path.name.endswith(".jpg")
             ]
             # _TRAIN_POINTS_PER_CLASS datapoints are sampled for the train split,
             # the others constitute the test split.
-            if _TRAIN_POINTS_PER_CLASS > len(fnames):
+            if _TRAIN_POINTS_PER_CLASS > len(index_codes):
                 raise ValueError(
-                    "Fewer than {} ({}) points in class {}".format(
-                        _TRAIN_POINTS_PER_CLASS, len(fnames), class_dir.name
-                    )
+                    f"Fewer than {_TRAIN_POINTS_PER_CLASS} ({len(index_codes)}) points in class {class_dir.name}"
                 )
-            train_fnames = np.random.choice(
-                fnames, _TRAIN_POINTS_PER_CLASS, replace=False
+
+            train_indices = np.random.choice(
+                index_codes, _TRAIN_POINTS_PER_CLASS, replace=False
             )
-            test_fnames = set(fnames).difference(train_fnames)
-            fnames_to_emit = train_fnames if is_train_split else test_fnames
+            test_indices = set(index_codes).difference(train_indices)
+
+            indices_to_emit = train_indices if is_train_split else test_indices
 
             if (
-                class_dir.name == "BACKGROUND_Google"
+                class_name == "BACKGROUND_Google"
                 and config_name == self._BUILDER_CONFIG_WITHOUT_BACKGROUND.name
             ):
                 print("skip BACKGROUND_Google")
                 continue
 
-            for image_file in fnames_to_emit:
+            for indice in indices_to_emit:
                 record = {
-                    "image": str(image_file),
+                    "image": str(class_dir / f"image_{indice}.jpg"),
                     "label": class_dir.name.lower(),
                 }
-                yield "%s/%s" % (class_dir.name.lower(), image_file), record
+                if config_name == self._BUILDER_CONFIG_WITHOUT_BACKGROUND.name:
+                    if class_name in _ANNOTATION_NAMES_MAP:
+                        annotations_class_name = _ANNOTATION_NAMES_MAP[class_name]
+                    else:
+                        annotations_class_name = class_name
+                    data = scipy.io.loadmat(
+                        str(
+                            annotation_dir
+                            / annotations_class_name
+                            / f"annotation_{indice}.mat"
+                        )
+                    )
+                    # raise ValueError(data["obj_contour"].dtype, data["box_coord"])
+                    record["annotation"] = {
+                        "obj_contour": data["obj_contour"],
+                        "box_coord": data["box_coord"],
+                    }
+                yield f"{class_dir.name.lower()}/{f'image_{indice}.jpg'}", record
         # Resets the seeds to their previous states.
         np.random.set_state(numpy_original_state)