Convert dataset to Parquet

Convert dataset to Parquet.

README.md

@@ -32,18 +32,28 @@ dataset_info:
           '0': negative
           '1': positive
           '2': no_impact
+          '3': mixed
   splits:
   - name: train
-    num_bytes: 48555
+    num_bytes: 48551
     num_examples: 892
   - name: validation
-    num_bytes: 5788
+    num_bytes: 5784
     num_examples: 105
   - name: test
-    num_bytes: 5588
+    num_bytes: 5584
     num_examples: 104
-  download_size: 49870
-  dataset_size: 59931
+  download_size: 48150
+  dataset_size: 59919
+configs:
+- config_name: default
+  data_files:
+  - split: train
+    path: data/train-*
+  - split: validation
+    path: data/validation-*
+  - split: test
+    path: data/test-*
 train-eval-index:
 - config: default
   task: text-classification

data/test-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d641bfc593a0a35b8af376f76fab573953476eb8affc91d1ae07c16d8479ac0
+size 6161

data/train-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ca6372cf331a5db4de59d7a4de11710315c84866a868e65ae05eb4964465737
+size 35646

data/validation-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97408b0166f4a456289d2d8ab8429bcac5c47b921d14dc92ecdad8b59732abb3
+size 6343