Auto download missing model for bert_gen.py (#146)

* auto download missing model

* support openi

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix wrong delete

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* pass pre-commit

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix repeat login

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

text/japanese_bert.py

@@ -1,10 +1,14 @@
-import torch
-from transformers import AutoTokenizer, AutoModelForMaskedLM
 import sys
-from text.japanese import text2sep_kata
-from config import config
 
-tokenizer = AutoTokenizer.from_pretrained("./bert/deberta-v2-large-japanese")
+import torch
+from transformers import AutoModelForMaskedLM, AutoTokenizer
+
+from config import config
+from text.japanese import text2sep_kata
+
+LOCAL_PATH = "./bert/deberta-v2-large-japanese"
+
+tokenizer = AutoTokenizer.from_pretrained(LOCAL_PATH)
 
 models = dict()
 
@@ -27,9 +31,7 @@ def get_bert_feature_with_token(tokens, word2ph, device=config.bert_gen_config.d
     if not device:
         device = "cuda"
     if device not in models.keys():
-        models[device] = AutoModelForMaskedLM.from_pretrained(
-            "./bert/deberta-v2-large-japanese"
-        ).to(device)
+        models[device] = AutoModelForMaskedLM.from_pretrained(LOCAL_PATH).to(device)
     with torch.no_grad():
         inputs = torch.tensor(tokens).to(device).unsqueeze(0)
         token_type_ids = torch.zeros_like(inputs).to(device)