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

@@ -5,8 +5,6 @@ language_creators:
 - machine-generated
 language:
 - it
-language_bcp47:
-- it-IT
 license:
 - unknown
 multilinguality:
@@ -22,6 +20,8 @@ task_ids:
 - extractive-qa
 paperswithcode_id: squad-it
 pretty_name: SQuAD-it
+language_bcp47:
+- it-IT
 dataset_info:
   features:
   - name: id
@@ -38,13 +38,20 @@ dataset_info:
       dtype: int32
   splits:
   - name: train
-    num_bytes: 50864824
+    num_bytes: 50864680
     num_examples: 54159
   - name: test
-    num_bytes: 7858336
+    num_bytes: 7858312
     num_examples: 7609
-  download_size: 8776531
-  dataset_size: 58723160
+  download_size: 13797580
+  dataset_size: 58722992
+configs:
+- config_name: default
+  data_files:
+  - split: train
+    path: data/train-*
+  - split: test
+    path: data/test-*
 ---
 
 # Dataset Card for "squad_it"

data/test-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72eb99d7e396079f5f97d2a4a0322afd9f186539318d22c522456fc966802f5d
+size 1624373

data/train-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61afb4156ad32e349e33e4d82bb97bca8f2195863bf070c2fdd6d00c4b1c76bd
+size 12173207