Update app.py

app.py

@@ -10,14 +10,8 @@ openai.api_key = os.environ.get("OPENAI_API_KEY")
 
 BASE_SYSTEM_MESSAGE = """"""
 
-def make_prediction(chat, prompt, max_tokens=None, temperature=None, top_p=None, top_k=None, repetition_penalty=None):
-    if chat == "Chatbot1":
-        openai.api_base = os.environ.get("OPENAI_API_BASE")
-        completion = openai.Completion.create(model="wizardcoder-python-34b-v1.0.Q5_K_M.gguf", prompt=prompt, max_tokens=max_tokens, temperature=temperature, top_p=top_p, top_k=top_k, repetition_penalty=repetition_penalty, stream=True, stop=["</s>", "<|im_end|>"])
-    elif chat == "Chatbot2":
-        openai.api_base = os.environ.get("OPENAI_API_BASE2")
-        completion = openai.Completion.create(model="wizardcoder-python-34b-v1.0.Q5_K_M.gguf", prompt=prompt, max_tokens=max_tokens, temperature=temperature, top_p=top_p, top_k=top_k, repetition_penalty=repetition_penalty, stream=True, stop=["</s>", "<|im_end|>"])
-
+def make_prediction(prompt, max_tokens=None, temperature=None, top_p=None, top_k=None, repetition_penalty=None):
+    completion = openai.Completion.create(model="/workspace/text-generation-webui/models/wizardcoder-python-34b-v1.0.Q5_K_M.gguf", prompt=prompt, max_tokens=max_tokens, temperature=temperature, top_p=top_p, top_k=top_k, repetition_penalty=repetition_penalty, stream=True, stop=["</s>", "<|im_end|>"])
     for chunk in completion:
         yield chunk["choices"][0]["text"]
 
@@ -35,7 +29,7 @@ def user(message, history):
     return "", history
 
 
-def chat(chatbot, history, system_message, max_tokens, temperature, top_p, top_k, repetition_penalty):
+def chat(history, system_message, max_tokens, temperature, top_p, top_k, repetition_penalty):
     history = history or []
 
     messages = BASE_SYSTEM_MESSAGE + system_message.strip() + "\n" + \
@@ -47,7 +41,6 @@ def chat(chatbot, history, system_message, max_tokens, temperature, top_p, top_k
     messages = messages.rstrip()
 
     prediction = make_prediction(
-        chatbot,
         messages,
         max_tokens=max_tokens,
         temperature=temperature,
@@ -74,8 +67,7 @@ CSS ="""
 .contain { display: flex; flex-direction: column; }
 .gradio-container { height: 100vh !important; }
 #component-0 { height: 100%; }
-#chatbot1 { flex-grow: 1; overflow: auto; resize: vertical; }
-#chatbot2 { flex-grow: 1; overflow: auto; resize: vertical; }
+#chatbot { flex-grow: 1; overflow: auto; resize: vertical; }
 """
 
 #with gr.Blocks() as demo:
@@ -84,15 +76,13 @@ with gr.Blocks(css=CSS) as demo:
         with gr.Column():
             gr.Markdown(f"""
                     ## This demo is an unquantized GPU chatbot of [WizardCoder-Python-34B-V1.0-GGUF](https://huggingface.co/TheBloke/WizardCoder-Python-34B-V1.0-GGUF)
+                    Brought to you by your friends at Alignment Lab AI, garage-bAInd, Open Access AI Collective, and OpenChat!
                     """)
     with gr.Row():
         gr.Markdown("# 🔍 WizardCoder-Python-34B-V1.0-GGUF Playground Space! 🔎")
     with gr.Row():
-        with gr.Column():
-            #chatbot = gr.Chatbot().style(height=500)
-            chatbot1 = gr.Chatbot(label="Chatbot1", elem_id="chatbot1")
-        with gr.Column():
-            chatbot2 = gr.Chatbot(label="Chatbot2", elem_id="chatbot2")
+        #chatbot = gr.Chatbot().style(height=500)
+        chatbot = gr.Chatbot(elem_id="chatbot")
     with gr.Row():
         message = gr.Textbox(
             label="What do you want to chat about?",
@@ -117,15 +107,12 @@ with gr.Blocks(css=CSS) as demo:
 
     chat_history_state = gr.State()
     clear.click(clear_chat, inputs=[chat_history_state, message], outputs=[chat_history_state, message], queue=False)
-    clear.click(lambda: None, None, chatbot1, queue=False)
-    clear.click(lambda: None, None, chatbot2, queue=False)
+    clear.click(lambda: None, None, chatbot, queue=False)
 
     submit_click_event = submit.click(
         fn=user, inputs=[message, chat_history_state], outputs=[message, chat_history_state], queue=True
     ).then(
-        fn=chat, inputs=["Chatbot1", chat_history_state, system_msg, max_tokens, temperature, top_p, top_k, repetition_penalty], outputs=['chatbot1', chat_history_state, message], queue=True
-    ).then(
-        fn=chat, inputs=["Chatbot2", chat_history_state, system_msg, max_tokens, temperature, top_p, top_k, repetition_penalty], outputs=['chatbot2', chat_history_state, message], queue=True
+        fn=chat, inputs=[chat_history_state, system_msg, max_tokens, temperature, top_p, top_k, repetition_penalty], outputs=[chatbot, chat_history_state, message], queue=True
     )
     stop.click(fn=None, inputs=None, outputs=None, cancels=[submit_click_event], queue=False)