Add application file

app.py

@@ -212,7 +212,7 @@ def QA_model():
     model = HuggingFacePipeline.from_model_id(model_id="daryl149/llama-2-7b-chat-hf",
         task="text-generation",
         model_kwargs={
-                    "torch_dtype" : torch.float16,
+                    "torch_dtype" : torch.float32,
                     "low_cpu_mem_usage" : True,
                     "temperature": 0.2,
                     "max_length": 2048,