Create README.md

This model is a fine-tuned version from arabertv2 on xtreme dataset on Arabic POS tagging which achieved:
validation loss: 0.139
f1-score: 95%
Accuracy: 96%

README.md

@@ -0,0 +1,12 @@
+---
+datasets:
+- xtreme
+language:
+- ar
+metrics:
+- name: f1
+  value: 95
+- name: accuracy
+  value: 96
+pipeline_tag: token-classification
+---