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

@@ -258,6 +258,7 @@ language_bcp47:
 tags:
 - language-identification
 dataset_info:
+  config_name: WiLI-2018 dataset
   features:
   - name: sentence
     dtype: string
@@ -500,16 +501,23 @@ dataset_info:
           '232': tuk
           '233': kan
           '234': ltg
-  config_name: WiLI-2018 dataset
   splits:
   - name: train
-    num_bytes: 65408201
+    num_bytes: 65408153
     num_examples: 117500
   - name: test
-    num_bytes: 66491260
+    num_bytes: 66491212
     num_examples: 117500
-  download_size: 130516351
-  dataset_size: 131899461
+  download_size: 91718265
+  dataset_size: 131899365
+configs:
+- config_name: WiLI-2018 dataset
+  data_files:
+  - split: train
+    path: WiLI-2018 dataset/train-*
+  - split: test
+    path: WiLI-2018 dataset/test-*
+  default: true
 ---
 
 # Dataset Card for wili_2018

WiLI-2018 dataset/test-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a1b582dbc8fc71d6baabc9574835d4a5d925b21f9ab2fcea49c7c4e86acc0df
+size 46000315

WiLI-2018 dataset/train-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:816e63cff8d7d3da5d9a2aaab68527f9d28e9efd22dab45ca0a9b9517c52ecea
+size 45717950