Update README.md

README.md

@@ -7,6 +7,14 @@ tags:
 model-index:
 - name: roberta-fa-zwnj-base-finetuned-PQuAD-3epochs
   results: []
+datasets:
+- Z-Jafari/PQuAD
+language:
+- fa
+metrics:
+- f1
+- exact_match
+pipeline_tag: question-answering
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -17,6 +25,19 @@ should probably proofread and complete it, then remove this comment. -->
 This model is a fine-tuned version of [HooshvareLab/roberta-fa-zwnj-base](https://huggingface.co/HooshvareLab/roberta-fa-zwnj-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
 - Loss: 1.0544
+- {'exact': 69.28267933016745,
+ 'f1': 82.1337789035037,
+ 'total': 8002,
+ 'HasAns_exact': 63.370565045992116,
+ 'HasAns_f1': 80.26190847336343,
+ 'HasAns_total': 6088,
+ 'NoAns_exact': 88.08777429467085,
+ 'NoAns_f1': 88.08777429467085,
+ 'NoAns_total': 1914,
+ 'best_exact': 69.28267933016745,
+ 'best_exact_thresh': 0.0,
+ 'best_f1': 82.13377890350401,
+ 'best_f1_thresh': 0.0}
 
 ## Model description
 
@@ -58,4 +79,4 @@ The following hyperparameters were used during training:
 - Transformers 4.57.3
 - Pytorch 2.9.0+cu126
 - Datasets 4.0.0
-- Tokenizers 0.22.1
+- Tokenizers 0.22.1