Updated readme and fixed feature definition

README.md

@@ -1,3 +1,62 @@
 ---
 license: bsd-3-clause
+dataset_info:
+  config_name: exdark
+  features:
+  - name: img
+    dtype: image
+  - name: target
+    sequence:
+    - name: labels
+      dtype:
+        class_label:
+          names:
+            '0': Dog
+            '1': Motorbike
+            '2': People
+            '3': Cat
+            '4': Chair
+            '5': Table
+            '6': Car
+            '7': Bicycle
+            '8': Bottle
+            '9': Bus
+            '10': Cup
+            '11': Boat
+    - name: bboxes
+      sequence: float32
+  splits:
+  - name: train
+    num_bytes: 1864877
+    num_examples: 7361
+  download_size: 1487935964
+  dataset_size: 1864877
 ---
+
+The Exclusively Dark (ExDARK) dataset is a collection of low-light
+images from very low-light environments to twilight (i.e 10 different
+conditions) with 12 object classes (similar to PASCAL VOC) annotated on both
+image class level and local object bounding boxes.
+
+The object classes are as follows:
+
+- Dog
+- Motorbike
+- People
+- Cat
+- Chair
+- Table
+- Car
+- Bicycle
+- Bottle
+- Bus
+- Cup
+- Boat
+
+For more information about the original Exclusively Dark Image dataset,
+please visit the official dataset page:
+
+[https://github.com/cs-chan/Exclusively-Dark-Image-Dataset](https://github.com/cs-chan/Exclusively-Dark-Image-Dataset)
+
+Please refer to the original dataset source for any additional details,
+citations, or specific usage guidelines provided by the dataset creators.

ex-dark.py

@@ -103,15 +103,16 @@ class ExDark(datasets.GeneratorBasedBuilder):
             features=datasets.Features(
                 {
                     "img": datasets.Image(),
-                    "label": datasets.Sequence(
+                    "target": datasets.Sequence(
                         feature={
-                            "class": datasets.ClassLabel(names=_LABEL_NAMES),
-                            "x": datasets.Value("int32"),
-                            "y": datasets.Value("int32"),
-                            "w": datasets.Value("int32"),
-                            "h": datasets.Value("int32"),
+                            "labels": datasets.features.ClassLabel(
+                                names=_LABEL_NAMES
+                            ),
+                            "bboxes": datasets.features.Sequence(
+                                feature=datasets.Value("float32")
+                            )
                         }
-                    ),
+                    )
                 }
             ),
             homepage=_HOMEPAGE,
@@ -142,12 +143,13 @@ class ExDark(datasets.GeneratorBasedBuilder):
         df["class"] = df["class"].apply(lambda x: classes.tolist().index(x))
 
         for idx, file_name in enumerate(df.file_name.unique()):
+            img_path = os.path.join(data_dir, "ExDark", file_name)
             sample = df[df.file_name == file_name]
             bboxs = sample[["x", "y", "w", "h"]].to_numpy()
             labels = sample["class"].to_numpy()
             yield idx, {
-                "img": file_name,
-                "label": {
+                "img": img_path,
+                "target": {
                     "labels": labels,
                     "bboxes": bboxs,
                 },