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

@@ -36,10 +36,15 @@ dataset_info:
           '7': fear
   splits:
   - name: train
-    num_bytes: 1541746
+    num_bytes: 1541738
     num_examples: 10065
-  download_size: 1563138
-  dataset_size: 1541746
+  download_size: 862018
+  dataset_size: 1541738
+configs:
+- config_name: default
+  data_files:
+  - split: train
+    path: data/train-*
 ---
 
 # Dataset Card for Emotional Tone in Arabic

data/train-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a500309dc6f4bfccada744edea4d3699312177f3a559c2001ad03c2b236d54c6
+size 862018