Convert dataset to Parquet

Convert dataset to Parquet.

README.md

@@ -19,6 +19,26 @@ task_ids:
 pretty_name: 'PUGG: IR dataset for Polish'
 tags:
 - wikipedia
+dataset_info:
+  config_name: corpus
+  features:
+  - name: _id
+    dtype: string
+  - name: title
+    dtype: string
+  - name: text
+    dtype: string
+  splits:
+  - name: test
+    num_bytes: 287988378
+    num_examples: 309621
+  download_size: 200782749
+  dataset_size: 287988378
+configs:
+- config_name: corpus
+  data_files:
+  - split: test
+    path: corpus/test-*
 ---
 # PUGG: KBQA, MRC, IR Dataset for Polish
 

corpus/test-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4984cd0db8c68b53552992c69f2912c4b452ff51d8783930b4c12a9b0c82af3a
+size 200782749