Update app.py

app.py

@@ -1,30 +1,41 @@
-import gradio as gr  
-from llama_cpp import Llama  
+#import gradio as gr  
+#from llama_cpp import Llama  
   
 # Load the Llama model  
-llm = Llama.from_pretrained(  
-    repo_id="GSridhar1982/QA_Llama31_Quantized_GGUF",  
-    filename="QA_llama31_unsloth.Q4_K_M.gguf",  
-)  
+#llm = Llama.from_pretrained(  
+ #   repo_id="GSridhar1982/QA_Llama31_Quantized_GGUF",  
+ #   filename="QA_llama31_unsloth.Q4_K_M.gguf",  
+#)  
   
-def generate_response(user_input):  
+#def generate_response(user_input):  
     # Perform inference  
-    response = llm.create_chat_completion(  
-        messages=[  
-            {  
-                "role": "user",  
-                "content": user_input  
-            }  
-        ]  
-    )  
+#    response = llm.create_chat_completion(  
+#        messages=[  
+#            {  
+#                "role": "user",  
+#                "content": user_input  
+ #           }  
+ #       ]  
+ #   )  
       
     # Extract the model's reply  
-    model_reply = response['choices'][0]['message']['content']  
-    return model_reply  
-  
+  #  model_reply = response['choices'][0]['message']['content']  
+   # return model_reply  
+
+from transformers import pipeline
+
+# Replace with your Hugging Face model space URL
+model_id = "GSridhar1982/AIML_QA_Llama31_FineTuned_UsingLora" 
+
+pipe = pipeline("text2text-generation", model=model_id)
+
+def generate_text(prompt):
+  """Generates text using the loaded Hugging Face model."""
+  return pipe(prompt)[0]['generated_text']
+
 # Create a Gradio interface  
 iface = gr.Interface(  
-    fn=generate_response,  
+    fn=generate_text,  
     inputs="textbox",  
     outputs="text",  
     title="AIML Q&A Chatbot",