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

@@ -30,10 +30,15 @@ dataset_info:
           '1': negative
   splits:
   - name: train
-    num_bytes: 114670811
+    num_bytes: 114670791
     num_examples: 50000
-  download_size: 31510992
-  dataset_size: 114670811
+  download_size: 56234303
+  dataset_size: 114670791
+configs:
+- config_name: default
+  data_files:
+  - split: train
+    path: data/train-*
 ---
 
 # Dataset Card for ImDB Urdu Reviews

data/train-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df742bf617460342a5ef9691e66be05b4ce9787cc3a12d6875d53b8c24e515a0
+size 56234303