Convert dataset to Parquet

Convert dataset to Parquet.

README.md

@@ -39,10 +39,10 @@ dataset_info:
           '2': contradiction
   splits:
   - name: train
-    num_bytes: 84729207
+    num_bytes: 84728887
     num_examples: 392702
-  download_size: 42113232
-  dataset_size: 84729207
+  download_size: 54693610
+  dataset_size: 84728887
 - config_name: snli
   features:
   - name: premise
@@ -84,6 +84,11 @@ dataset_info:
     num_examples: 5010
   download_size: 42113232
   dataset_size: 1566267
+configs:
+- config_name: multi_nli
+  data_files:
+  - split: train
+    path: multi_nli/train-*
 ---
 
 # Dataset Card for "kor_nli"

multi_nli/train-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8de5f800e4c9d0fb89bd6a36a800f7ec26be2a97f50c32f2aff3c197ca155f66
+size 54693610