Update app.py

app.py

@@ -11,9 +11,12 @@ from transformers import GPT2LMHeadModel
 
 
 model_name = "rinna/japanese-gpt-1b"
-model = GPT2LMHeadModel.from_pretrained(model_name)
+from huggingface_hub import snapshot_download
 
-tokenizer = T5Tokenizer.from_pretrained(model_name)
+download_path = snapshot_download(repo_id=model_name)
+model = GPT2LMHeadModel.from_pretrained(download_path)
+
+tokenizer = T5Tokenizer.from_pretrained(download_path)
 
 model = torch.quantization.quantize_dynamic(
     model, {torch.nn.Linear}, dtype=torch.qint8)