Update handler.py

handler.py

@@ -3,16 +3,13 @@ from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
 import torch
 from handler import EndpointHandler
 
-device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
-class EndpointHandler:
-    def __init__(self, path="Deepakvictor/tan-ta"):
-        self.tokenizer = AutoTokenizer.from_pretrained(path).to(device)
-        self.model = AutoModelForSeq2SeqLM.from_pretrained(path).to(device))
+class EndpointHandler():
+    def __init__(self, path=""):
+        self.tokenizer = AutoTokenizer.from_pretrained(path)
+        self.model = AutoModelForSeq2SeqLM.from_pretrained(path)
 
     def __call__(self, data: str) -> str:
         inp = self.tokenizer(data, return_tensors="pt")
-        for q in inp:
-            inp[q] = inp[q].to(device)
         with torch.inference_mode():
             out= self.model.generate(**inp)
         final_output = self.tokenizer.batch_decode(out,skip_special_tokens=True)