Update README.md

README.md

@@ -1,4 +1,20 @@
-BoolQ Validation dataset Evaluation: <br/>
+## BoolQ Validation dataset Evaluation: <br/>
 support => 3270 <br/>
 accuracy => 0.73 <br/>
-macro f1 => 0.71
+macro f1 => 0.71
+
+## SQuAD Validation dataset Evaluation: <br/>
+eval_HasAns_exact      = 55.9885<br/>
+eval_HasAns_f1         = 70.4997<br/>
+eval_HasAns_total      =    5928<br/>
+eval_NoAns_exact       = 20.5719<br/>
+eval_NoAns_f1          = 20.5719<br/>
+eval_NoAns_total       =    5945<br/>
+eval_best_exact        = 50.0969<br/>
+eval_best_exact_thresh =     0.0<br/>
+eval_best_f1           =    50.1<br/>
+eval_best_f1_thresh    =     0.0<br/>
+eval_exact             = 38.2549<br/>
+eval_f1                =    45.5<br/>
+eval_samples           =   12165<br/>
+eval_total             =   11873