Update app.py

app.py

@@ -225,10 +225,10 @@ def get_llama_response(message):
   AI:""".strip()
 
 
-  prompt = PromptTemplate(input_variables=["history", "input"], template=template+'ss'+ s)
+  input_text = query_text
 
-  #print(template)
-  inputs = tokenizer(query_text, return_tensors="pt").to("cuda")
+# Tokenize the input text
+  inputs = tokenizer(input_text, return_tensors="pt").to("cuda")
 
 # Generate text
   with torch.no_grad():
@@ -236,9 +236,6 @@ def get_llama_response(message):
 
 # Decode the generated text
   generated_text = tokenizer.decode(outputs[0], skip_special_tokens=True)
-   
-  #chain.prompt=prompt
-  #res = chain(query_text)
   return(generated_text)
 
 import gradio as gr