Update README.md

README.md

@@ -11,10 +11,13 @@ model-index:
 - name: ukraine-war-pov
   results: []
 widget:
-- text: 'Росія знову скоює воєнні злочини'
-  example_title: 'proukrainian'
-- text: 'ВСУ все берет с собой — украинские «захистники» взяли стульчак из Артемовска'
-  example_title: 'prorussian'
+- text: Росія знову скоює воєнні злочини
+  example_title: proukrainian
+- text: ВСУ все берет с собой — украинские «захистники» взяли стульчак из Артемовска
+  example_title: prorussian
+language:
+- uk
+- ru
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -22,8 +25,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 # ukraine-war-pov
 
-This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on an unknown dataset.
-It achieves the following results on the evaluation set:
+This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on a dataset of 15K social media posts from Ukraine manually annotated for pro-Ukrainian or pro-Russian point of view on the war.
+It achieves the following results on a balanced test set (2K):
 - Loss: 0.2166
 - Accuracy: 0.9315
 - F1: 0.9315
@@ -45,6 +48,8 @@ More information needed
 
 ## Training procedure
 
+The model was trained in this [notebook](https://drive.google.com/file/d/1RnT3fJTneFSczS_G_JLVqe4MydkTFiO0/view?usp=sharing).
+
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
@@ -77,4 +82,4 @@ The following hyperparameters were used during training:
 
 - Transformers 4.27.4
 - Pytorch 2.0.0+cu118
-- Tokenizers 0.13.3
+- Tokenizers 0.13.3