Update README.md

README.md

@@ -67,7 +67,7 @@ import re
 model = from_pretrained_keras("vrclc/transliteration")
 ```
 
-### Define Tokens:
+### Define Tokens and Input Sequence Length:
 
 ```
 source_tokens = list('abcdefghijklmnopqrstuvwxyz ')
@@ -91,6 +91,9 @@ target_tokens = [
 ]
 target_tokenizer = Tokenizer(char_level=True, filters='')
 target_tokenizer.fit_on_texts(target_tokens)
+
+max_seq_length = model.get_layer("encoder_input").input_shape[0][1]
+
 ```
 
 ### Wrapper script to split input sentences to words before passing to the model