Upload tokenizer

tokenization_phylogpn.py

@@ -1,5 +1,5 @@
-from transformers import PreTrainedTokenizer
 from typing import List, Dict, Optional, Tuple
+from transformers import PreTrainedTokenizer
 
 class PhyloGPNTokenizer(PreTrainedTokenizer):
     model_input_names = ["input_ids"]
@@ -24,7 +24,10 @@ class PhyloGPNTokenizer(PreTrainedTokenizer):
             **kwargs,
         )
 
+        self._receptive_field_size = 1
+
     def _tokenize(self, seq: str) -> List[str]:
+        assert len(seq) >= 481, "Input must be at least 481 bp long"
         return list(seq)
     
     def _convert_token_to_id(self, token: str) -> int: