update, use fine tuned model

main.py

@@ -11,9 +11,10 @@ from pydantic import BaseModel
 
 app = FastAPI()
 
-# Load your model and tokenizer
-tokenizer = GPT2Tokenizer.from_pretrained("gpt2")
-model = GPT2LMHeadModel.from_pretrained("gpt2")
+# Load your fine-tuned model and tokenizer
+dir = 'models/fine-tuned-gpt2'
+tokenizer = GPT2Tokenizer.from_pretrained(dir)
+model = GPT2LMHeadModel.from_pretrained(dir)
 device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 model.to(device)