Add explicit config list

dataset.py

@@ -0,0 +1,134 @@
+import os
+import glob
+from pathlib import Path
+from typing import List
+
+import datasets
+
+_CITATION = """\
+@article{liu2025causal3d,
+  title={CAUSAL3D: A Comprehensive Benchmark for Causal Learning from Visual Data},
+  author={Liu, Disheng and Qiao, Yiran and Liu, Wuche and Lu, Yiren and Zhou, Yunlai and Liang, Tuo and Yin, Yu and Ma, Jing},
+  journal={arXiv preprint arXiv:2503.04852},
+  year={2025}
+}
+"""
+
+_DESCRIPTION = """\
+Causal3D is a benchmark for evaluating causal reasoning in physical and hypothetical visual scenes. 
+It includes both real-world recordings and rendered synthetic scenes demonstrating causal interactions.
+"""
+
+_HOMEPAGE = "https://huggingface.co/datasets/LLDDSS/Causal3D"
+_LICENSE = "CC-BY-4.0"
+
+# Automatically discover configs from folder names
+def _discover_configs(base_dir: str) -> List[datasets.BuilderConfig]:
+    configs = []
+    for root_folder in ["real_scenes", "hypothetical_scenes"]:
+        full_path = os.path.join(base_dir, root_folder)
+        if not os.path.exists(full_path):
+            continue
+        for scene in sorted(os.listdir(full_path)):
+            scene_path = os.path.join(full_path, scene)
+            if os.path.isdir(scene_path):
+                config_name = f"{root_folder}_{scene}"
+                configs.append(
+                    datasets.BuilderConfig(
+                        name=config_name,
+                        version=datasets.Version("1.0.0"),
+                        description=f"{scene} in {root_folder}",
+                    )
+                )
+    return configs
+
+
+def _discover_configs(base_dir: str) -> List[datasets.BuilderConfig]:
+    configs = []
+    for root_folder in ["real_scenes", "hypothetical_scenes"]:
+        full_path = os.path.join(base_dir, root_folder)
+        if not os.path.exists(full_path):
+            continue
+        for scene in sorted(os.listdir(full_path)):
+            scene_path = os.path.join(full_path, scene)
+            if os.path.isdir(scene_path):
+                config_name = f"{root_folder}_{scene}"
+                configs.append(
+                    datasets.BuilderConfig(
+                        name=config_name,
+                        version=datasets.Version("1.0.0"),
+                        description=f"{scene} in {root_folder}",
+                    )
+                )
+    return configs
+
+
+class Causal3D(datasets.GeneratorBasedBuilder):
+    BUILDER_CONFIGS = [
+        datasets.BuilderConfig(
+            name="hypothetical_scenes_Hypothetic_v2_linear",
+            version=datasets.Version("1.0.0"),
+            description="Hypothetic v2 linear scene with images and csv."
+        ),
+        datasets.BuilderConfig(
+            name="real_scenes_Real_Parabola",
+            version=datasets.Version("1.0.0"),
+            description="Real Parabola scene."
+        ),
+        ]
+
+    def _info(self):
+        return datasets.DatasetInfo(
+            description=_DESCRIPTION,
+            features=datasets.Features({
+                "image": datasets.Image(),
+                "file_name": datasets.Value("string"),
+                "metadata": datasets.Value("string"),  # optionally replace with structured fields
+            }),
+            homepage=_HOMEPAGE,
+            license=_LICENSE,
+            citation=_CITATION,
+        )
+
+    def _split_generators(self, dl_manager):
+        parts = self.config.name.split("_", 2)
+        category = parts[0] + "_" + parts[1]  # real_scenes or hypothetical_scenes
+        scene = parts[2]
+        data_dir = os.path.join(category, scene)
+
+        return [
+            datasets.SplitGenerator(
+                name=datasets.Split.TRAIN,
+                gen_kwargs={"data_dir": data_dir},
+            )
+        ]
+
+    def _generate_examples(self, data_dir):
+        # Find the .csv file
+        csv_files = list(Path(data_dir).rglob("*.csv"))
+        if not csv_files:
+            raise FileNotFoundError(f"No CSV metadata found in {data_dir}")
+        csv_path = csv_files[0]
+        df = pd.read_csv(csv_path)
+
+        # Make sure a 'file_name' column exists
+        if "file_name" not in df.columns:
+            raise ValueError(f"'file_name' column not found in {csv_path.name}")
+
+        # Load image paths
+        image_files = {}
+        for ext in ("*.png", "*.jpg", "*.jpeg"):
+            for img_path in Path(data_dir).rglob(ext):
+                relative_path = str(img_path.relative_to(data_dir))
+                image_files[relative_path] = str(img_path)
+
+        # Match CSV rows with image paths
+        for idx, row in df.iterrows():
+            fname = row["file_name"]
+            if fname not in image_files:
+                continue  # skip missing images
+            yield idx, {
+                "image": image_files[fname],
+                "file_name": fname,
+                "metadata": row.to_json(),
+            }