Update app.py

app.py

@@ -12,21 +12,9 @@ model_name = "google/gemma-2-2b-it"
 lora_model_name="Anlam-Lab/gemma-2-2b-it-anlamlab-SA-Chatgpt4mini"
 
 
-device = "cuda" if torch.cuda.is_available() else "cpu"
-@torch.no_grad()
-def load_model():
-    tokenizer = AutoTokenizer.from_pretrained(model_name)
-    model = AutoModelForCausalLM.from_pretrained(
-        model_name,
-        device_map=device,
-        torch_dtype=torch.float16 if device == "cuda" else torch.float32,
-        low_cpu_mem_usage=True
-    )
-    model = PeftModel.from_pretrained(model, lora_model_name)
-    model.eval() 
-    return model, tokenizer
-
-model, tokenizer = load_model()
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(model_name, device_map="cpu", torch_dtype=torch.float16)
+model = PeftModel.from_pretrained(model, lora_model_name)
 
 def generate_response(text):
     example = f"""<|begin_of_text|><|start_header_id|>system<|end_header_id|>Bir duygu analisti olarak sana verilen metinleri analiz et ve aşağıdaki kategorilerden yalnızca birini seçerek metnin duygu durumunu belirle:Positive,Negative,Neutral<|eot_id|><|start_header_id|>user<|end_header_id|>{text}<|eot_id|><|start_header_id|>assistant<|end_header_id|>"""