1

README.md

@@ -17,11 +17,15 @@ tags:
 - multimodal
 - benchmark
 configs:
-- config_name: images_metadata
+- config_name: image
   default: true
   data_files:
   - split: train
-    path: card_samples/parquet/images_metadata.parquet
+    path: card_samples/parquet/image.parquet
+- config_name: qa
+  data_files:
+  - split: train
+    path: card_samples/parquet/qa.parquet
 ---
 
 # KnowCP
@@ -40,8 +44,10 @@ It supports recognition-style tasks (region/text extraction) and knowledge/reaso
 
 ## Dataset Viewer Layout
 
-The dataset card uses one simple config: `images_metadata`.
-It displays the first 100 images and their metadata.
+The dataset card has two simple series:
+
+- `image`: first 100 single-image artworks with metadata
+- `qa`: 14 question types, 10 items per type (140 total)
 
 The card data is generated only from core folders:
 
@@ -55,11 +61,13 @@ The card data is generated only from core folders:
 - `images/**`: image assets
 - `kb/knowledge_base.json`: artwork metadata (`title`, `artist`, `dynasty`, `institution`)
 - `annotations/*.json`: seal/inscription/object/technique annotation exports
-- `card_samples/parquet/images_metadata.parquet`: viewer-facing image metadata table
+- `questions/by_type/*.jsonl`: source questions by type
+- `card_samples/parquet/image.parquet`: image series table
+- `card_samples/parquet/qa.parquet`: qa series table
 
 ## Data Fields in Viewer
 
-### `images_metadata`
+### `image`
 
 - `image` (rendered image)
 - `image_id`
@@ -69,6 +77,14 @@ The card data is generated only from core folders:
 - `institution`
 - `material`
 
+### `qa`
+
+- `image` (rendered image)
+- `qid`
+- `type`
+- `question`
+- `ground_truth`
+
 ## Usage
 
 ```python
@@ -76,9 +92,11 @@ from datasets import load_dataset
 
 repo_id = "g41/KnowCP"
 
-images_metadata = load_dataset(repo_id, name="images_metadata")
+image_ds = load_dataset(repo_id, name="image")
+qa_ds = load_dataset(repo_id, name="qa")
 
-print(images_metadata["train"][0]["title"])
+print(image_ds["train"][0]["title"])
+print(qa_ds["train"][0]["qid"])
 ```
 
 To regenerate the dataset-card samples:
@@ -87,7 +105,8 @@ To regenerate the dataset-card samples:
 python for_hf/scripts/build_dataset_card_samples.py \
   --hf-repo hf_repo \
   --out-dir hf_repo/card_samples \
-  --images-limit 100
+  --images-limit 100 \
+  --qa-per-type 10
 ```
 
 ## Evaluation with This Project

card_samples/parquet/{images_metadata.parquet → image.parquet}

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:627042528a9f046200350b1e1d583a5e71e6df9292d650791c74cb94011cd7c6
-size 497425345
+oid sha256:683a17d592310dd81539b62b0ffe5c39823127e7d57ae4b0661d52047d9ab560
+size 528434862

card_samples/parquet/qa.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1389412e2da43aaf5daec04fa136094127a413637dd054a5b9eb7bc8f2e3dee6
+size 339904117

card_samples/summary.json

@@ -1,6 +1,29 @@
 {
-  "images_metadata_rows": 100,
+  "image_rows": 100,
+  "qa_rows": 140,
+  "qa_per_type": 10,
+  "qa_per_type_counts": {
+    "CC": 10,
+    "ER_choice": 10,
+    "ER_fillin": 10,
+    "IR": 10,
+    "ITT_MHQA_choice": 10,
+    "ITT_MHQA_fillin": 10,
+    "MITT_MHQA_choice": 10,
+    "MITT_MHQA_fillin": 10,
+    "PR": 10,
+    "SR": 10,
+    "TR_choice": 10,
+    "TR_fillin": 10,
+    "TTI": 10,
+    "VA": 10
+  },
+  "single_image_only": true,
+  "excluded_image_ids": [
+    "IMG001211"
+  ],
   "images_limit": 100,
-  "parquet_path": "hf_repo/card_samples/parquet/images_metadata.parquet",
+  "image_parquet_path": "hf_repo/card_samples/parquet/image.parquet",
+  "qa_parquet_path": "hf_repo/card_samples/parquet/qa.parquet",
   "out_dir": "hf_repo\\card_samples"
 }