Update README.md

README.md

@@ -56,12 +56,12 @@ __Download links for all Russian sentiment datasets collected by Smetanin can be
 Training were done in this [project](https://github.com/searayeah/vkr-bert) with this parameters:
 
 ```
-tokenizer.max_length: 256
-batch_size: 32
+max_length: 512
+batch_size: 64
 optimizer: adam
 lr: 0.00001
 weight_decay: 0
-epochs: 2
+num_epochs: 5
 ```
 
 Train/validation/test splits are 80%/10%/10%.