Allow only unidic_lite

README.md

@@ -28,6 +28,14 @@ sentence = "LINE株式会社で[MASK]の研究・開発をしている。"
 print(model(**tokenizer(sentence, return_tensors="pt")))
 ```
 
+### Requirements
+
+```txt
+fugashi 
+sentencepiece
+unidic-lite
+```
+
 ## Model architecture
 
 The model architecture is the DitilBERT base model; 6 layers, 768 dimensions of hidden states, 12 attention heads, 66M parameters.

distilbert_japanese_tokenizer.py

@@ -485,22 +485,6 @@ class MecabTokenizer:
                     )
 
                 dic_dir = unidic_lite.DICDIR
-            elif mecab_dic == "unidic":
-                try:
-                    import unidic
-                except ModuleNotFoundError as error:
-                    raise error.__class__(
-                        "The unidic dictionary is not installed. "
-                        "See https://github.com/polm/unidic-py for installation."
-                    )
-
-                dic_dir = unidic.DICDIR
-                if not os.path.isdir(dic_dir):
-                    raise RuntimeError(
-                        "The unidic dictionary itself is not found. "
-                        "See https://github.com/polm/unidic-py for installation."
-                    )
-
             else:
                 raise ValueError("Invalid mecab_dic is specified.")