Update README.md

README.md

@@ -1,6 +1,14 @@
 vnPara is a Vietnamese dataset for the paraphrase identification task.
 This repository is a reformatted version of the original dataset, adapted for Hugging Face Datasets, with a standard train/test split (80% / 20%).
 
+```
+Overall label counts: Counter({'1': 1576, '0': 1506, '1,': 1})
+train_gen: 2640 labels: Counter({'0': 1506, '1': 1134})
+dev_gen  : 126 labels: Counter({'1': 126})
+test_gen : 316 labels: Counter({'1': 316})
+neg_pool : 1506 labels: Counter({'0': 1506})
+```
+
 ## Original Resource
 The original dataset and implementation can be found at:  
 https://github.com/nlp-team-itus-hcmc/paraphrase-identification-vietnamese-document