Upload folder using huggingface_hub

convert_to_parquet.py

@@ -0,0 +1,44 @@
+#!/usr/bin/env python3
+"""Convert JSONL datasets to Hugging Face parquet format."""
+
+import json
+import pandas as pd
+from pathlib import Path
+
+
+def jsonl_to_parquet(jsonl_path: str, parquet_path: str) -> int:
+    """Convert a JSONL file to parquet format."""
+    records = []
+    with open(jsonl_path, 'r') as f:
+        for line in f:
+            line = line.strip()
+            if line:
+                records.append(json.loads(line))
+
+    df = pd.DataFrame(records)
+    df.to_parquet(parquet_path, index=False)
+    return len(df)
+
+
+def main():
+    base_dir = Path(__file__).parent
+
+    # Convert lite split
+    lite_input = base_dir / "dataset.jsonl"
+    lite_output = base_dir / "lite.parquet"
+    lite_count = jsonl_to_parquet(lite_input, lite_output)
+    print(f"Converted {lite_count} records to {lite_output}")
+
+    # Convert full split
+    full_input = base_dir / "dataset_full.jsonl"
+    full_output = base_dir / "full.parquet"
+    full_count = jsonl_to_parquet(full_input, full_output)
+    print(f"Converted {full_count} records to {full_output}")
+
+    print("\nDone! Hugging Face dataset splits:")
+    print(f"  - lite: {lite_output} ({lite_count} tasks)")
+    print(f"  - full: {full_output} ({full_count} tasks)")
+
+
+if __name__ == "__main__":
+    main()

full.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d521242cc66c6659d11b25e4e114c7731c8ffaef97bc3550f9f49c77a6c53f59
+size 143528

lite.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7666ab2611caf0205e87958f29f74a771779d4900dc678cff4b5e76ec0c7bb54
+size 19066