replaced cuda() with device

app.py

@@ -218,7 +218,7 @@ def load_model():
     #tokenizer_gpt = AutoTokenizer.from_pretrained("netgvarun2005/GPTVirtualTherapistTokenizer", pad_token='<|pad|>',bos_token='<|startoftext|>',eos_token='<|endoftext|>')
     tokenizer_gpt = AutoTokenizer.from_pretrained("netgvarun2005/GPTTherapistDeepSpeedTokenizer", pad_token='<|pad|>',bos_token='<|startoftext|>',eos_token='<|endoftext|>')
     #model_gpt = AutoModelForCausalLM.from_pretrained("netgvarun2005/GPTVirtualTherapist")
-    model_gpt = AutoModelForCausalLM.from_pretrained("netgvarun2005/GPTTherapistDeepSpeedModel").cuda()
+    model_gpt = AutoModelForCausalLM.from_pretrained("netgvarun2005/GPTTherapistDeepSpeedModel").to(device)
    
     return multiModel,tokenizer,model_gpt,tokenizer_gpt