Update app.py

app.py

@@ -6,7 +6,7 @@ import sys
 from dotenv import loaddotenv, dotenvvalues
 loaddotenv()
 
-# initialize the client
+# 수정 전 코드 (penAI)
 # client = OpenAI(
 #   baseurl="https://api-inference.huggingface.co/v1",
 #   apikey=os.environ.get('APIY')  # Replace with your token
@@ -14,7 +14,7 @@ loaddotenv()
 
 # Create supported models
 modellinks = {
-    "Meta-Llama-3-8B-Instruct": "meta-llama/Meta-Llama-3-8B-Instruct"
+    "Meta-Llama-3-8B-Instruct": "meta-llama/Meta-Llama-3-8B-Instruct",
 }
 
 #andom dog images for error message
@@ -33,7 +33,6 @@ randomdog = ["0f476473-2d8b-415e-b944-483768418a95.jpg",
               "bfb9e165-c643-4993-9b3a-7e73571672a6.jpg"]
 
 
-
 def resetconversation():
     '''
     Resets Conversation
@@ -81,11 +80,11 @@ repoid = modellinks[selectedmodel]
 
 
 st.subheader(f'ypeGP.net - {selectedmodel}')
-# st.title(f'ChatBot Using {selectedmodel}')
+st.title(f'ChatBot Using {selectedmodel}')
 
 # Set a default model
 if selectedmodel not in st.sessionstate:
-    st.sessionstate[selectedmodel] = modellinks[selectedmodel] 
+    st.sessionstate[selectedmodel] = modellinks[selectedmodel]
 
 # Initialize chat history
 if "messages" not in st.sessionstate:
@@ -106,11 +105,11 @@ if prompt := st.chatinput(f"Hi I'm {selectedmodel}, ask me a question"):
         st.markdown(prompt)
     # Add user message to chat history
     st.sessionstate.messages.append({"role": "user", "content": prompt})
-    
 
     # Display assistant response in chat message container
     with st.chatmessage("assistant"):
         try:
+            # 수정 전 코드 (penAI)
             # stream = client.chat.completions.create(
             #     model=modellinks[selectedmodel],
             #     messages=[
@@ -122,16 +121,83 @@ if prompt := st.chatinput(f"Hi I'm {selectedmodel}, ask me a question"):
             #     maxtokens=3000,
             # )
 
-            # 수정 전 코드 (penAI)
-            # response = st.writestream(stream)
-
-            # 수정 후 코드 (Hugging Face)
-            tokenizer = Autookenizer.frompretrained(repoid)
-            model = AutoModelorCausalLM.frompretrained(repoid)
-
-            inputids = tokenizer(prompt, returntensors="pt").inputids
-            output = model.generate(inputids, maxlength=1000, temperature=tempvalues)
-            response = tokenizer.decode(output[0], skipspecialtokens=rue)
+            # 수정 후 코드 (gradio & InferenceClient)
+            import gradio as gr
+            from huggingfacehub import InferenceClient
+
+            """
+            For more information on `huggingfacehub` Inference API support, please check the docs: https://huggingface.co/docs/huggingfacehub/v0.22.2/en/guides/inference
+            """
+            client = InferenceClient(repoid)
+
+            def respond(
+                message,
+                history: list[tuple[str, str]],
+                systemmessage,
+                maxtokens,
+                temperature,
+                topp,
+            ):
+                messages = [{"role": "system", "content": systemmessage}]
+
+                for val in history:
+                    if val[0]:
+                        messages.append({"role": "user", "content": val[0]})
+                    if val[1]:
+                        messages.append({"role": "assistant", "content": val[1]})
+
+                messages.append({"role": "user", "content": message})
+
+                response = ""
+
+                for message in client.chatcompletion(
+                    messages,
+                    maxtokens=maxtokens,
+                    stream=rue,
+                    temperature=temperature,
+                    topp=topp,
+                ):
+                    token = message.choices[0].delta.content
+
+                    response += token
+                    yield response
+
+            """
+            For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
+            """
+            demo = gr.ChatInterface(
+                respond,
+                additionalinputs=[
+                    gr.extbox(
+                        value="You are a friendly Chatbot.", label="System message"
+                    ),
+                    gr.Slider(
+                        minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"
+                    ),
+                    gr.Slider(
+                        minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="emperature"
+                    ),
+                    gr.Slider(
+                        minimum=0.1,
+                        maximum=1.0,
+                        value=0.95,
+                        step=0.05,
+                        label="op-p (nucleus sampling)",
+                    ),
+                ],
+            )
+
+            response = ""
+
+            for message in demo(
+                prompt,
+                st.sessionstate.messages[1:],
+                "You are a friendly Chatbot.",
+                512,
+                0.7,
+                0.95,
+            ):
+                response += message
 
         except Exception as e:
             # st.empty()