Update README.md

README.md

@@ -10,7 +10,7 @@ metrics:
 ## Model description
 This biLSTMhard model is a Bidirectional Long-Short-Term Memory (BiLSTM) architecture trained from scratch on the Hotel Arabic Reviews Dataset (HARD) dataset with one bidirectional LSTM layer with 150 hidden states. 
 It achieves the following results on the evaluation test set: 
- - Accuracy: 80%
+ - Accuracy: 0.8013
 
 
 ## BibTeX Citation:
@@ -41,6 +41,8 @@ The following hyperparameters were used during training:
 | :-- | :-- |
 train_batch_size:| 128
 num_epochs: |15
+dropout: |0.3
+dropout: |0.3
 
 
  ## Model Plot