Simplify sentiment model initialization by using a direct model reference

model/sentiment.py

@@ -3,9 +3,6 @@ from typing import Any
 
 class Sentiment:
     def __init__(self, line: str) -> (list | list[Any] | Any | None):
-        model = AutoModelForSequenceClassification.from_pretrained("models/feel-it-italian-sentiment", local_files_only=True)
-        tokenizer = AutoTokenizer.from_pretrained("models/feel-it-italian-sentiment", local_files_only=True)
-        
-        self.pipe = pipeline("text-classification", model=model, tokenizer=tokenizer)
+        self.pipe = pipeline("text-classification", model="MilaNLProc/feel-it-italian-sentiment")
         self.result = self.pipe(line)