Update main.py

main.py

@@ -16,14 +16,14 @@ def average_pool(last_hidden_states: Tensor,
 
 # text-ada replacement
 embeddingTokenizer = AutoTokenizer.from_pretrained(
-    './multilingual-e5-base')
-embeddingModel = AutoModel.from_pretrained('./multilingual-e5-base')
+    './models/multilingual-e5-base')
+embeddingModel = AutoModel.from_pretrained('./models/multilingual-e5-base')
 
 # chatGpt replacement
 inferenceTokenizer = AutoTokenizer.from_pretrained(
-    "./ct2fast-flan-alpaca-xl")
+    "./models/ct2fast-flan-alpaca-xl")
 inferenceTranslator = Translator(
-    "./ct2fast-flan-alpaca-xl", compute_type="int8", device="cpu")
+    "./models/ct2fast-flan-alpaca-xl", compute_type="int8", device="cpu")
 
 
 class EmbeddingRequest(BaseModel):
@@ -99,7 +99,7 @@ async def tokens_count(request: TokensCountRequest):
         inferenceTokenizer.encode(input_text))
 
     # create response
-    response = {
+    return {
         'tokens': tokens,
         'total': len(tokens)
     }