Update app.py

app.py

@@ -9,25 +9,29 @@ model = AutoModelForCausalLM.from_pretrained("Ashikan/dut-recipe-generator")
 pipe = pipeline("text-generation", model=model, tokenizer=tokenizer, device=get_device_id())
 
 def perform_model_inference(ingredients_list=None, recipe_name=None):
-    if ingredients_list:
-        for ingredient_index in range(len(ingredients_list)):
-            ingredients_list[ingredient_index] = ingredients_list[ingredient_index].strip()
-
-        # Create a simple prompt without JSON formatting
-        input_text = "Generate a recipe with these ingredients: " + ", ".join(ingredients_list)
-    elif recipe_name:
-        # Simple prompt for recipe name
-        input_text = "Generate ingredients and method for the recipe: " + recipe_name
-    else:
-        return "Invalid input"
-
-    # Limit the length of the input text to avoid long processing times
-    input_text = input_text[:512]  # Truncate if it's too long
-
-    # Use higher temperature for quicker responses
-    output = pipe(input_text, max_length=512, temperature=0.7, do_sample=True, truncation=True)[0]["generated_text"]
-
-    return format_output(output)
+    try:
+        if ingredients_list:
+            for ingredient_index in range(len(ingredients_list)):
+                ingredients_list[ingredient_index] = ingredients_list[ingredient_index].strip()
+
+            # Create a simple prompt for generating a recipe from ingredients
+            input_text = "Generate a recipe with these ingredients: " + ", ".join(ingredients_list)
+        elif recipe_name:
+            # Simple prompt for generating ingredients and method from recipe name
+            input_text = "Generate ingredients and cooking method for the recipe: " + recipe_name
+        else:
+            return "Invalid input"
+
+        # Limit the length of the input text to avoid long processing times
+        input_text = input_text[:512]  # Truncate if it's too long
+
+        # Generate output with a higher temperature for quicker responses
+        output = pipe(input_text, max_length=512, temperature=0.7, do_sample=True, truncation=True)[0]["generated_text"]
+
+        return format_output(output)
+    
+    except Exception as e:
+        return f"Error occurred: {str(e)}"
 
 def chat_function(history, user_input, mode):
     # If mode is "ingredients", process as ingredient list