add CeLLaTe-V2 dataset train-val split

CeLLaTe-V2 dataset split into training and validaiton sets using 80:20 ratio

README.md

@@ -0,0 +1,24 @@
+---
+dataset_info:
+  features:
+  - name: tokens
+    list: string
+  - name: tags
+    list: string
+  splits:
+  - name: train
+    num_bytes: 343017.22799575824
+    num_examples: 754
+  - name: validation
+    num_bytes: 85981.77200424178
+    num_examples: 189
+  download_size: 102901
+  dataset_size: 428999.0
+configs:
+- config_name: default
+  data_files:
+  - split: train
+    path: data/train-*
+  - split: validation
+    path: data/validation-*
+---

data/train-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9aaea45b1cf1def7bb92c6c9ae27a8576256fba21d75a90905960c34e01136f4
+size 77963

data/validation-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f17b75603f98ea330f1ae4d8d4a568382501957f9eca5ecb46e6880dde39cfb
+size 24938