Add application file

app.py

@@ -15,7 +15,9 @@ tokenizer.add_tokens(['<sep>', '<pad>', '<mask>', '<predict>', '<FIM_SUFFIX>', '
 
 print("Creat model...")
 model = AutoModelForCausalLM.from_pretrained('IEITYuan/Yuan2-2B-hf', device_map='auto', torch_dtype=torch.bfloat16, trust_remote_code=True)
-
+# using CUDA for an optimal experience
+device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
+model = model.to(device)
 
 # Defining a custom stopping criteria class for the model's text generation.
 class StopOnTokens(StoppingCriteria):

requirements.txt

@@ -1,3 +1,4 @@
 torch>=2.0
 transformers>=4.35.0
-gradio>=4.13.0
+gradio>=4.13.0
+sentencepiece