Convert dataset to Parquet

Convert dataset to Parquet.

README.md

@@ -248,16 +248,25 @@ dataset_info:
           '1': 'true'
   splits:
   - name: train
-    num_bytes: 10060799
+    num_bytes: 10060751
     num_examples: 21046
   - name: test
-    num_bytes: 1253810
+    num_bytes: 1253794
     num_examples: 2629
   - name: validation
-    num_bytes: 1266874
+    num_bytes: 1266858
     num_examples: 2573
-  download_size: 7201807
-  dataset_size: 12581483
+  download_size: 3635948
+  dataset_size: 12581403
+configs:
+- config_name: default
+  data_files:
+  - split: train
+    path: data/train-*
+  - split: test
+    path: data/test-*
+  - split: validation
+    path: data/validation-*
 ---
 
 # Dataset Card for Disaster Response Messages 

data/test-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae38ee653cfe31d712b8d8b36114f2106d064855ce76e577b1d2b768bb1a26e0
+size 379323

data/train-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4287f7b70f24b0026565fe1017be00694702188275c422e1fbcb92b93a0e3196
+size 2868843

data/validation-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a3c3f415cdb659113d9a2e956b1ce15cceaebd12d21b12a815d965a3810ea54
+size 387782