Update app.py

app.py

@@ -10,19 +10,19 @@ prompt_template = 'USER: {0}\nASSISTANT: '
 # Function to generate responses using the GPT-4 model with custom settings
 def generate_response(prompt, settings):
     # Extract settings from the input
-    max_tokens = settings["max_tokens"]
-    temp = settings["temp"]
-    top_k = settings["top_k"]
-    top_p = settings["top_p"]
-    repeat_penalty = settings["repeat_penalty"]
-    repeat_last_n = settings["repeat_last_n"]
-    n_batch = settings["n_batch"]
-    n_predict = settings["n_predict"]
-    streaming = settings["streaming"]
-    
+    max_tokens = settings[0]
+    temp = settings[1]
+    top_k = settings[2]
+    top_p = settings[3]
+    repeat_penalty = settings[4]
+    repeat_last_n = settings[5]
+    n_batch = settings[6]
+    n_predict = settings[7]
+    streaming = settings[8]
+
     # Generate chat history and input prompt
     chat_history_with_prompt = prompt_template.format(prompt)
-    
+
     # Generate response with custom settings
     response = model.generate(
         chat_history_with_prompt,
@@ -36,33 +36,29 @@ def generate_response(prompt, settings):
         n_predict=n_predict,
         streaming=streaming
     )
-    
+
     return response
 
 # Initialize Gradio Interface
-with gr.Blocks() as chatbot_demo:
-    with gr.Tab("Chat"):
-        gr.Interface(
-            fn=generate_response,
-            inputs=[
-                gr.inputs.Textbox(label="Chat Input", placeholder="Start the conversation..."),
-                gr.inputs.Number(default=200, label="Max Tokens"),
-                gr.inputs.Number(default=0.7, label="Temperature"),
-                gr.inputs.Number(default=40, label="Top-k"),
-                gr.inputs.Number(default=0.4, label="Top-p"),
-                gr.inputs.Number(default=1.18, label="Repeat Penalty"),
-                gr.inputs.Number(default=64, label="Repeat Last n"),
-                gr.inputs.Number(default=8, label="Batch Size"),
-                gr.inputs.Textbox(default="Auto", label="Number of Predictions"),
-                gr.inputs.Checkbox(default=False, label="Streaming"),
-            ],
-            outputs=gr.outputs.Textbox(),
-            title="GPT-4 Chatbot",
-            description="Chat with the GPT-4 based chatbot. Configure generation settings and see the chat history for this session.",
-        )
-
-    with gr.Tab("Settings"):
-        gr.Text("Settings tab content")
+interface = gr.Interface(
+    fn=generate_response,
+    inputs=[
+        gr.inputs.Textbox(label="Chat Input", placeholder="Start the conversation..."),
+        gr.inputs.Number(default=200, label="Max Tokens"),
+        gr.inputs.Number(default=0.7, label="Temperature"),
+        gr.inputs.Number(default=40, label="Top-k"),
+        gr.inputs.Number(default=0.4, label="Top-p"),
+        gr.inputs.Number(default=1.18, label="Repeat Penalty"),
+        gr.inputs.Number(default=64, label="Repeat Last n"),
+        gr.inputs.Number(default=8, label="Batch Size"),
+        gr.inputs.Textbox(default="Auto", label="Number of Predictions"),
+        gr.inputs.Checkbox(default=False, label="Streaming"),
+    ],
+    outputs=gr.outputs.Textbox(),
+    title="GPT-4 Chatbot",
+    description="Chat with the GPT-4 based chatbot. Configure generation settings and see the chat history for this session.",
+)
 
 # Launch Gradio Interface
-chatbot_demo.queue(concurrency_count=75).launch(debug=True)
+if __name__ == "__main__":
+    interface.launch(debug=True)