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

@@ -9,8 +9,6 @@ license:
 - cc-by-4.0
 multilinguality:
 - monolingual
-pretty_name: 'DISFL-QA: A Benchmark Dataset for Understanding Disfluencies in Question
-  Answering'
 size_categories:
 - 10K<n<100K
 source_datasets:
@@ -20,6 +18,8 @@ task_categories:
 task_ids:
 - extractive-qa
 - open-domain-qa
+pretty_name: 'DISFL-QA: A Benchmark Dataset for Understanding Disfluencies in Question
+  Answering'
 dataset_info:
   features:
   - name: squad_v2_id
@@ -40,16 +40,25 @@ dataset_info:
       dtype: int32
   splits:
   - name: train
-    num_bytes: 7712523
+    num_bytes: 7712491
     num_examples: 7182
   - name: test
-    num_bytes: 3865097
+    num_bytes: 3865065
     num_examples: 3643
   - name: validation
-    num_bytes: 1072731
+    num_bytes: 1072699
     num_examples: 1000
-  download_size: 48935038
-  dataset_size: 12650351
+  download_size: 4246350
+  dataset_size: 12650255
+configs:
+- config_name: default
+  data_files:
+  - split: train
+    path: data/train-*
+  - split: test
+    path: data/test-*
+  - split: validation
+    path: data/validation-*
 ---
 
 # Dataset Card for DISFL-QA: A Benchmark Dataset for Understanding Disfluencies in Question Answering

data/test-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3a8768ab77830bb13e44110d29e41ce49a77698d53ef84f889c30d2bc1e82444
+size 590102

data/train-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e4e82c5d2b8b5c03afe4b8cedb2816abcdcb59dedd9edfa009eb95bd49e3d15
+size 3221857

data/validation-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c88a0ae7fcbd0650cd414734948a89ee62ddf2593de5b93efb5c54ab2137b24
+size 434391