Convert dataset to Parquet

Convert dataset to Parquet.

This dataset uses `tasks`, which are deprecated and will raise an error after the next release of `datasets`. See: https://github.com/huggingface/datasets/pull/6999

README.md

@@ -34,10 +34,15 @@ dataset_info:
           '2': adv
   splits:
   - name: train
-    num_bytes: 2278043
+    num_bytes: 2278039
     num_examples: 567
-  download_size: 1228804
-  dataset_size: 2278043
+  download_size: 1398139
+  dataset_size: 2278039
+configs:
+- config_name: default
+  data_files:
+  - split: train
+    path: data/train-*
 ---
 
 # Dataset Card for OneStopEnglish corpus

data/train-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f31161ff98617fc970147ac6bcd0a55a37b60fc1625bd78fcfdcbc500f40d6a8
+size 1398139