Files changed (1) hide show
  1. app.py +15 -1
app.py CHANGED
@@ -67,7 +67,21 @@ class DecodeRequest(BaseModel):
67
  class DecodeResponse(BaseModel):
68
  text: str
69
 
70
-
 
 
 
 
 
 
 
 
 
 
 
 
 
 
71
  # -----------------------------
72
  # Embedding Endpoint (text)
73
  # -----------------------------
 
67
  class DecodeResponse(BaseModel):
68
  text: str
69
 
70
+ class BatchTokenizeRequest(BaseModel):
71
+ texts: List[str]
72
+
73
+ class BatchTokenizeResponse(BaseModel):
74
+ input_ids: List[List[int]]
75
+
76
+ @app.post("/batch_tokenize", response_model=BatchTokenizeResponse)
77
+ def batch_tokenize(req: BatchTokenizeRequest):
78
+ encs = tokenizer(
79
+ req.texts,
80
+ add_special_tokens=False,
81
+ return_attention_mask=False,
82
+ return_token_type_ids=False,
83
+ )
84
+ return {"input_ids": encs["input_ids"]}
85
  # -----------------------------
86
  # Embedding Endpoint (text)
87
  # -----------------------------