LAMBO Tokenization skeleton.

combo/main.py

@@ -86,8 +86,8 @@ flags.DEFINE_integer(name="batch_size", default=1,
 flags.DEFINE_boolean(name="silent", default=True,
                      help="Silent prediction to file (without printing to console).")
 flags.DEFINE_enum(name="predictor_name", default="combo-spacy",
-                  enum_values=["combo", "combo-spacy", "lambo"],
-                  help="Use predictor with whitespace or spacy tokenizer.")
+                  enum_values=["combo", "combo-spacy", "combo-lambo"],
+                  help="Use predictor with whitespace, spacy or LAMBO tokenizer.")
 
 
 def run(_):

combo/predict.py

@@ -10,15 +10,15 @@ from allennlp.data import tokenizers
 from allennlp.predictors import predictor
 from overrides import overrides
 
-from combo import data
 from combo.data import sentence2conllu, tokens2conllu, conllu2sentence
-from combo.utils import download, graph
+from combo.utils import download, graph, lambo
 
 logger = logging.getLogger(__name__)
 
 
 @predictor.Predictor.register("combo")
 @predictor.Predictor.register("combo-spacy", constructor="with_spacy_tokenizer")
+@predictor.Predictor.register("combo-lambo", constructor="with_lambo_tokenizer")
 class COMBO(predictor.Predictor):
 
     def __init__(self,
@@ -230,6 +230,11 @@ class COMBO(predictor.Predictor):
     def with_spacy_tokenizer(cls, model: models.Model,
                              dataset_reader: allen_data.DatasetReader):
         return cls(model, dataset_reader, tokenizers.SpacyTokenizer())
+    
+    @classmethod
+    def with_lambo_tokenizer(cls, model: models.Model,
+                             dataset_reader: allen_data.DatasetReader):
+        return cls(model, dataset_reader, lambo.LamboTokenizer())
 
     @classmethod
     def from_pretrained(cls, path: str, tokenizer=tokenizers.SpacyTokenizer(),

combo/utils/lambo.py

+from typing import List
+
+from allennlp.data.tokenizers.tokenizer import Tokenizer
+from allennlp.data.tokenizers.token_class import Token
+
+class LamboTokenizer(Tokenizer):
+
+    def __init__(self, language: str = "??",) -> None:
+        self.language = language
+
+    def tokenize(self, text: str) -> List[Token]:
+        #TODO
+        return None
\ No newline at end of file