upload data

README.md

@@ -1,3 +1,230 @@
----
-license: apache-2.0
----
+---
+license: apache-2.0
+
+dataset_info:
+  features:
+    train:
+      - name: video 
+        dtype: binary
+      - name: internal_id
+        dtype: string  
+      - name: url
+        dtype: string
+      - name: annotation
+        struct:
+          - name: text_fit
+            dtype: int64
+          - name: composition_focus  
+            dtype: int64
+          - name: focus
+            dtype: int64
+          - name: camera_movement_quality
+            dtype: int64
+          - name: color
+            dtype: int64
+          - name: shadow_correct
+            dtype: int64
+          - name: shadow_goodlooking
+            dtype: int64
+          - name: morphology_build
+            dtype: int64
+          - name: morphology_maintain
+            dtype: int64
+          - name: object_movement
+            dtype: int64
+          - name: camera_movement
+            dtype: int64
+          - name: movement_smoothing 
+            dtype: int64
+          - name: movement_reality
+            dtype: int64
+          - name: clarity
+            dtype: int64
+          - name: quality
+            dtype: int64
+          - name: stability
+            dtype: int64
+          - name: detail
+            dtype: int64
+          - name: letter
+            dtype: int64
+          - name: physical_reality
+            dtype: int64
+          - name: harmful
+            dtype: int64
+          - name: harmful_degree
+            dtype: int64
+      - name: meta_result
+        dtype: sequence[int64]
+      - name: meta_mask 
+        dtype: sequence[int64]
+
+    regression:
+      - name: internal_id
+        dtype: string
+      - name: prompt
+        dtype: string 
+      - name: standard_answer
+        dtype: string
+      - name: file1
+        dtype: binary
+      - name: file2
+        dtype: binary
+
+    monetbench:
+      - name: internal_id
+        dtype: string
+      - name: prompt
+        dtype: string 
+      - name: standard_answer
+        dtype: string
+      - name: file1
+        dtype: binary
+      - name: file2
+        dtype: binary
+
+  splits:
+  - name: train
+    num_examples: 40743
+  - name: regression
+    num_examples: 2260
+  - name: monetbench
+    num_examples: 1000
+---
+
+# VRDB-Video
+
+This dataset is a comprehensive collection of video evaluation data designed for multi-dimensional quality assessment of AI-generated videos. It encompasses annotations across 21 diverse aspects, including text-to-video consistency, aesthetic quality, motion dynamics, physical realism, and technical specifications.
+
+The dataset is structured to facilitate both model training and standardized evaluation:
+- `Train`: A primary training set with detailed multi-dimensional annotations
+- `Regression`: A regression set with paired preference data
+- `MonetBench`: A benchmark test set for standardized performance evaluation
+
+This holistic approach enables the development and validation of sophisticated video quality assessment models that can evaluate AI-generated videos across multiple critical dimensions, moving beyond simple aesthetic judgments to encompass technical accuracy, semantic consistency, and dynamic performance.
+
+
+## Annotation Details
+
+Each video in the dataset is annotated with the following attributes:
+
+<table border="1" style="border-collapse: collapse;">
+<tr>
+<th>Dimension</th>
+<th>Sub-dimension</th>
+</tr>
+<tr>
+<td rowspan="2">Alignment</td>
+<td>Text Fit</td>
+</tr>
+<tr>
+<td>Text Accuracy</td>
+</tr>
+<tr>
+<td rowspan="3">Composition</td>
+<td>Composition Quality</td>
+</tr>
+<tr>
+<td>Focus Effect</td>
+</tr>
+<tr>
+<td>Camera Movement Artistry</td>
+</tr>
+<tr>
+<td rowspan="3">Quality</td>
+<td>Color Harmony</td>
+</tr>
+<tr>
+<td>Shadow Technical Accuracy</td>
+</tr>
+<tr>
+<td>Shadow Aesthetic Effect</td>
+</tr>
+<tr>
+<td rowspan="2">Preservation</td>
+<td>Initial Object Form</td>
+</tr>
+<tr>
+<td>Form Consistency</td>
+</tr>
+<tr>
+<td rowspan="3">Dynamic</td>
+<td>Object Motion Level</td>
+</tr>
+<tr>
+<td>Camera Motion Level</td>
+</tr>
+<tr>
+<td>Motion Fluidity</td>
+</tr>
+<tr>
+<td rowspan="4">Stability</td>
+<td>Visual Clarity</td>
+</tr>
+<tr>
+<td>Image Stability</td>
+</tr>
+<tr>
+<td>Camera Stability</td>
+</tr>
+<tr>
+<td>Motion Naturality</td>
+</tr>
+<tr>
+<td rowspan="2">Fidelity</td>
+<td>Detail Refinement</td>
+</tr>
+<tr>
+<td>Physical Realism</td>
+</tr>
+<tr>
+<td rowspan="2">Safety</td>
+<td>Content Safety Type</td>
+</tr>
+<tr>
+<td>Safety Level</td>
+</tr>
+</table>
+
+### Example: Camera Stability (cam_stable)
+  - **3:** Very stable
+  - **2:** Slight shake 
+  - **1:** Heavy shake
+- Note: When annotations are missing, the corresponding value will be set to **-1**.
+
+For more detailed annotation guidelines, please refer to:
+
+- [annotation_deatils](https://www.notion.so/VisionReward-Video-Annotation-Details-196a0162280e80768c89f6401fbb530c?pvs=4)
+- [annotation_deatils_ch](https://www.notion.so/VisionReward-Video-196a0162280e80e7806af42fc5808c99?pvs=4)
+
+## Additional Feature Details
+The dataset includes two special features: `annotation` and `meta_result`.
+
+### Annotation
+The `annotation` feature contains scores across 21 different dimensions of image assessment, with each dimension having its own scoring criteria as detailed above.
+
+### Meta Result
+The `meta_result` feature transforms multi-choice questions into a series of binary judgments. For example, for the `Camera Stability` dimension:
+
+| Score | Is the camera very stable? | Is the camera not unstable? |
+|-------|--------------------------|---------------------------|
+| 3     | 1                        | 1                         |
+| 2     | 0                        | 1                         |
+| 1     | 0                        | 0                         |
+
+- note: When the corresponding meta_result is -1 (It means missing annotation), the binary judgment should be excluded from consideration
+
+Each element in the binary array represents a yes/no answer to a specific aspect of the assessment. For detailed questions corresponding to these binary judgments, please refer to the meta_qa_en.txt file.
+
+### Meta Mask
+The `meta_mask` feature is used for balanced sampling during model training:
+- Elements with value 1 indicate that the corresponding binary judgment was used in training
+- Elements with value 0 indicate that the corresponding binary judgment was ignored during training
+
+## Data Processing
+
+We provide `extract.py` for processing the `train` dataset into JSONL format. The script can optionally extract the balanced positive/negative QA pairs used in VisionReward training by processing `meta_result` and `meta_mask` fields.
+
+```bash
+python extract.py [--save_imgs] [--process_qa]
+```

extract.py

@@ -0,0 +1,107 @@
+import json
+import os
+import logging
+import argparse
+from datasets import Dataset
+import io
+
+# Configure logging for detailed output
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+def load_questions_from_meta_qa(meta_qa_file):
+    with open(meta_qa_file, "r") as f:
+        questions = [line.strip() for line in f if line.strip()]
+    return questions
+
+def process_parquet_files(data_dir, output_jsonl, meta_qa_file=None, output_videos=None, process_qa=False):
+    """
+    Process Parquet files to generate a JSONL file with optional video export and QA list creation.
+    
+    Args:
+        data_dir (str): Directory containing Parquet files.
+        output_jsonl (str): Output JSONL file path.
+        meta_qa_file (str, optional): Path to the meta_qa_en.txt file for QA list creation.
+        output_videos (str, optional): Directory path to save videos. If None, videos are not saved.
+        process_qa (bool): Whether to process and include QA pairs in the output.
+    
+    Returns:
+        None
+    """
+
+    if output_videos and not os.path.exists(output_videos):
+        os.makedirs(output_videos)
+
+    # Load questions only if QA processing is enabled
+    questions = None
+    if process_qa and meta_qa_file:
+        questions = load_questions_from_meta_qa(meta_qa_file)
+
+    jsonl_data = []
+
+    parquet_files = [os.path.join(data_dir, f) for f in os.listdir(data_dir) if f.endswith(".parquet")]
+
+    for parquet_file in parquet_files:
+        dataset = Dataset.from_parquet(parquet_file)
+        
+        for row in dataset:
+            json_item = {
+                "internal_id": row["internal_id"],
+                "url": row["url"],
+                "prompt":row["prompt"],
+                "annotation": row["annotation"],
+                "meta_result": row["meta_result"],
+                "meta_mask": row["meta_mask"],
+            }
+            
+            # Optionally save videos
+            if output_videos:
+                video_data = row["video"]  
+                video_path = os.path.join(output_videos, f"{row['internal_id']}.mp4")  
+                
+                try:
+                    with open(video_path, "wb") as video_file:
+                        video_file.write(video_data)
+                    json_item["video_path"] = video_path
+                except Exception as e:
+                    logger.error(f"Error saving video for internal_id {row['internal_id']}: {e}")
+            
+            # Optionally process QA pairs
+            if process_qa and questions:
+                qa_list = []
+                meta_result = row["meta_result"]
+                meta_mask = row["meta_mask"]
+                for idx, mask in enumerate(meta_mask):
+                    if mask == 1:  # Add questions only if the mask is 1
+                        question = questions[idx]
+                        if "[[prompt]]" in question:
+                            question = question.replace("[[prompt]]", row["prompt"])
+                        answer = 'yes' if meta_result[idx] == 1 else 'no'
+                        qa_list.append({"question": question, "answer": answer})
+                json_item["qa_list"] = qa_list
+
+            jsonl_data.append(json_item)
+
+    with open(output_jsonl, "w") as outfile:
+        for json_item in jsonl_data:
+            outfile.write(json.dumps(json_item) + "\n")
+    logger.info(f"Finished writing JSONL file with {len(jsonl_data)} items.")
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(description="Convert Video dataset Parquet files to JSONL format with optional video extraction and QA list generation.")
+    parser.add_argument("--data_dir", type=str, default='train', help="Directory containing Parquet files.")
+    parser.add_argument("--output_jsonl", type=str, default='annotation.jsonl', help="Path to the output JSONL file.")
+    parser.add_argument("--meta_qa_file", type=str, default="meta_qa_en.txt", help="Optional: Path to the meta_qa_en.txt file for QA list generation.")
+    parser.add_argument("--save_videos", action="store_true", help="Optional: Whether to save videos.")
+    parser.add_argument("--process_qa", action="store_true", help="Optional: Process and include QA pairs in the output.")
+    args = parser.parse_args()
+
+    output_videos = 'videos' if args.save_videos else None
+    
+    process_parquet_files(
+        data_dir=args.data_dir,
+        output_jsonl=args.output_jsonl,
+        meta_qa_file=args.meta_qa_file,
+        output_videos=output_videos,
+        process_qa=args.process_qa
+    )

meta_qa_en.txt

@@ -0,0 +1,64 @@
+Does the video meet all the requirements stated in the text "[[prompt]]"?
+Does the video meet most of the requirements stated in the text "[[prompt]]"?
+Does the video meet some of the requirements stated in the text "[[prompt]]"?
+Does the video not completely fail to meet the requirements stated in the text "[[prompt]]"?
+Is the composition aesthetically pleasing?
+Does the composition have no obvious flaws?
+Is the focus aesthetically pleasing?
+Does the focus have no obvious flaws?
+Is the camera movement aesthetically pleasing?
+Does the camera movement have no obvious flaws?
+Are the colors exceptionally beautiful?
+Are the colors beautiful?
+Are the colors not unattractive?
+Are the colors not significantly unattractive?
+Is the lighting perfectly accurate?
+Does the lighting have no obvious errors?
+Is there any lighting present?
+Is the lighting exceptionally beautiful?
+Is the lighting beautiful?
+Is the lighting not unattractive?
+Is the shape of the object at the beginning of the video completely accurate?
+Does the shape of the object at the beginning have no obvious errors?
+Is the shape of the object at the beginning not chaotic?
+Is the shape of the object perfectly maintained throughout the video?
+Does the shape of the object have no obvious issues throughout the video?
+Does the shape of the object generally have no major issues throughout the video?
+Is the shape of the object not chaotic throughout the video?
+Is the object's motion highly dynamic?
+Is the object's motion dynamic?
+Is the object's motion not minimal?
+Is the object's motion not static?
+Is the camera motion highly dynamic?
+Is the camera motion dynamic?
+Is the camera motion not minimal?
+Is the camera motion not static?
+Is the smoothness of the object's movement very good?
+Does the smoothness of the object's movement have no obvious issues?
+Is the object's movement completely realistic?
+Does the object's movement have no obvious realism issues?
+Is it very clear?
+Is it clear?
+Is it not blurry?
+Is it not completely blurry?
+Is the image quality very stable?
+Is the image quality stable?
+Is the image quality not unstable?
+Is the image quality free of noticeable instability?
+Is the camera very stable?
+Is the camera not unstable?
+Are the details very refined?
+Are the details relatively refined?
+Are the details not rough?
+Are the details not significantly rough?
+Are all the letters correct?
+Do the letters have no obvious errors?
+Are there any letters present?
+Does it fully comply with the laws of physics?
+Does it partially comply with the laws of physics?
+Does it have no obvious violations of the laws of physics?
+Is the video content part of the physical world?
+Is the video content safe?
+Is the video content definitely free of harmful material?
+Does the video content contain no harmful material?
+Does the video content contain no extremely harmful material?

monetbench/test-00000-of-00005.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:db7a8caa3cc2e0cb66d763ce9fed8cd647527e8915498b402ff9c4cc1319bde8
+size 535617182

monetbench/test-00001-of-00005.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ff04b86597dce66c8e3e9b895a479c8614e7e40e1fb358afd79f563b2c5b3a9
+size 533579742

monetbench/test-00002-of-00005.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd3d1737d6a722d088b5dae1cd8ce7cd3b6607a84566a8953268879a47dc3ece
+size 532697792

monetbench/test-00003-of-00005.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63c90f11cb07a845f7ff091606e65be3e594c0d585f2ec7d80fa1e9297a455a4
+size 536312083

monetbench/test-00004-of-00005.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7cbbae2365524313950bbb7914f77cbd9b1b2b221f32e9806acaaac7a31059ba
+size 414687141

regression/train-00000-of-00009.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:429ff2ec91d0fd339e84743f577823870be7ceae08bba5f1dbbcdba675cc1a24
+size 535904565

regression/train-00001-of-00009.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e70b95921051f63c900fde286f8847d1627d78d12b64f224f0def682530cff26
+size 535562001

regression/train-00002-of-00009.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8de4474c969b872c1f3bcc212cca2cd350b99d9b1d961de26feeadf5ab545ed2
+size 535011361

regression/train-00003-of-00009.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a9d2a07618f3369cd9a303d3a0ceddc510a206dfaa2392be60aaa391076340c8
+size 535604707

regression/train-00004-of-00009.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af0780f1d5231e6c4210c1dab227261f227be841c5872532844cdd7118496397
+size 535593997

regression/train-00005-of-00009.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a22154b0e2b94faff0c36762d4bcec6d92f868c234b3bf82a26a643e755e1c8a
+size 536508165

regression/train-00006-of-00009.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a1c18e1969a802c51ab7af5088989e459052a910ed86a1462332da61d2863b95
+size 536214602

regression/train-00007-of-00009.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1fb81fff46bd27480b50e99385bc95ec969a15b37fe65e5cea5ff8afc0d5d722
+size 529760395

regression/train-00008-of-00009.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e942d59cb3338f8710265fb9c7f0f84f166e20f12689b786fd9e686ea4e71985
+size 246081414

train/train-00000-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6389da6e7c85e286759cd9d30f7a9d340c48efcee80479c6113ea4a0b6f13918
+size 535781894

train/train-00001-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ac00e64d005b2d231654ac6b1c8977fc1545ae2a898844d8ec39c571acc186b4
+size 536832047

train/train-00002-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e6fe86891be814d5a2148543b0d5020ea71201a80b2bf6da4e6a12c17ce34474
+size 536276794

train/train-00003-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0bbac03fe43699c3b8631f7850bfbf7bd507731ccff00a54cac080efa0e0ef1c
+size 532455925

train/train-00004-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c1f3f67b2ea58c38eb41acd2add852fb24cbb8dcced8eaec8be69e29e073d30
+size 536643684

train/train-00005-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9125229297e7e7ab59f13b9f62f59679544aa3fc7fbd05d3c000ff581c245b06
+size 536708809

train/train-00006-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:73ac0c0c983e4176dd104e543844a9d0fe93041c28ea6ef3cacf968347a1083e
+size 536209895

train/train-00007-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:162fe68e9193fef85b7eda52b91287d635a43ca5cdf38a6ef846f3970099babd
+size 536534273

train/train-00008-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e56d43322993649a89d8ae85cb951503da6e0616e674d863068b997cf461a2dd
+size 536470985

train/train-00009-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dcf2990c469e505f41528dc27d4a715b091f668cd7360115dc4aa432e91977c5
+size 534245851

train/train-00010-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35ba5779540e53354ccc6087f5a82e8ab3e3d4db4b3bea946a260ce1352b51bb
+size 536169097

train/train-00011-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c87b4914c368ce049b21414d549d427cd78231766dbd584b3a78cb1e69f9b0c
+size 536794791

train/train-00012-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:808e874311efa359920d173381e2b6d74cc60edbcb2d2e6429baddb01cc44cc1
+size 536464526

train/train-00013-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0db0da829f5797d73bed9649ac86db9b35043594a23fdfe849d0efe34dcbe1c3
+size 536561454

train/train-00014-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa20a002757235881b2822ab48b2206f7b820000c6105b711fae0e4a1056e8ae
+size 536412232

train/train-00015-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b84fc2331a20877dff7462dd44f250398b3e31034c9e1c5083b675e3594f722d
+size 536394744

train/train-00016-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd07ef15385bf142c76045d34c82caccd4ef83440b2bcc84f475a7fc4d66fa79
+size 535210373

train/train-00017-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7212a52ad16825abea67a45037783cd06e15c88d9b7588ede6e375b81b6daadc
+size 535822100

train/train-00018-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a43c2a765e0501ed298f817dc91df5cbf88fd3d86f5f45d6a4ddfcb5e579b1b8
+size 536483673

train/train-00019-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b916469c81d207fd12ea195dd2ab21b70c2cf95ba4de86d2e92aa3760e11cfeb
+size 533325448

train/train-00020-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:397fbba642ff8e2eeace1aa7555452a209f65c29538537630f71724b6b8b7ab0
+size 536546713

train/train-00021-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5cb87aca14b23c8da3eadba18a3b395b63c19489ef72b203712f66821b92a9ea
+size 535101547

train/train-00022-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:60f6b254f230d5208befc6aaa3b070ce1e1a41a74f2126c549c89b526a4012d8
+size 533299986

train/train-00023-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef515c73aa4cca0b91bbc2d69935c0e841ac70464bcdf07d7af647d34ea5c585
+size 534853549

train/train-00024-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:19c5ff403433522e21d436feb7ffc6ffeee0678019fdd386a325ac68dd397cc7
+size 535866044

train/train-00025-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b20ea5c8b589e03853f53ae81333cec9db73f01955d5eaf24285d9b2dcf53701
+size 533656376

train/train-00026-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0174d922123bfb89255a990500a596d8ab18b160262b92d043471eac721bad1f
+size 536766409

train/train-00027-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3debbb65e038be1df58468b91d7fa850c29959b08d79fd5ce7a7e14f1ca11320
+size 536786585

train/train-00028-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e910cc3ad2aa18cadcdb8686fc4bd76d54f6d118dc479316917e61e7fc261e94
+size 535712198

train/train-00029-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:47fd7e4f34c65fb9a6044d2ceb3d902aa44d817bf0d16b2e5243ebfa1b0c8502
+size 536281149

train/train-00030-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebee599daf984fd360e8ff5b762589087663177232d466f074420161240f7a1b
+size 534497396

train/train-00031-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:871e9e35e8d9ba490e5c3ab815b931616f815c40231e36d5700403c75a211735
+size 536277757

train/train-00032-of-00071.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6df41c39c1f442e391b7e6d2bb55ca332c8b24891b1ee09721df4f153990a04
+size 536002387