Update app.py

app.py

@@ -9,16 +9,10 @@ model_name = "./t5-finetuned-final"
 tokenizer = T5Tokenizer.from_pretrained(model_name)
 model = T5ForConditionalGeneration.from_pretrained(model_name)
 
-device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+# Move model to CPU (explicitly)
+device = torch.device("cpu")
 model.to(device)
 
-if torch.cuda.is_available():
-    model.half()
-    try:
-        model = torch.compile(model)
-    except:
-        pass
-
 def extract_amount(input_text):
     """
     Extracts the amount from the input text using a robust regex.
@@ -67,10 +61,11 @@ def generate_command(input_command):
     prompt = "extract: " + input_command
     input_ids = tokenizer(prompt, return_tensors="pt").input_ids.to(device)
     
+    # Generate output with reduced max_length and beams for faster inference
     output_ids = model.generate(
         input_ids,
-        max_length=64,
-        num_beams=3, #
+        max_length=32,  # Reduced for faster inference
+        num_beams=2,    # Reduced for faster inference
         early_stopping=True
     )