Update convert_czner.py

convert_czner.py

@@ -33,7 +33,7 @@ def whitespace_tokenize_with_offsets(text):
     return tokens, start_tok_offsets, end_tok_offsets
 
 
-def proc_dataset(dataset, max_text_length=420):
+def proc_dataset(dataset, max_text_length=200):
     r = []
     for doc in dataset:
         text = doc["text"]