Update app.py

app.py

@@ -1,114 +1,64 @@
-import os
-import logging
 import gradio as gr
-import openai
-import multiprocessing
+from huggingface_hub import InferenceClient
 
-from process_run import process_run
+"""
+For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
+"""
+client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
 
-# Configure logging
-logging.basicConfig(
-    level=logging.INFO,
-    format='%(asctime)s - %(levelname)s - %(message)s',
-    handlers=[
-        logging.StreamHandler(),
-    ]
-)
-logger = logging.getLogger(__name__)
-logger.setLevel('INFO')
-
-# Set your OpenAI API key
-openai.api_key = os.getenv("OPENAI_API_KEY")
 
+def respond(
+    message,
+    history: list[tuple[str, str]],
+    system_message,
+    max_tokens,
+    temperature,
+    top_p,
+):
+    messages = [{"role": "system", "content": system_message}]
 
-# Example instructions to display
-EXAMPLES = [
-    "When did the solar system form? Find on wikipedia.",
-    "Find the rating of Monopoly (1935) on boardgamegeek.com",
-]
+    for val in history:
+        if val[0]:
+            messages.append({"role": "user", "content": val[0]})
+        if val[1]:
+            messages.append({"role": "assistant", "content": val[1]})
 
-URL_EXAMPLES = [
-    "about:blank",
-    "https://www.wikipedia.org",
-    "https://www.boardgamegeek.com"
-]
+    messages.append({"role": "user", "content": message})
 
-def main():
-    logger.info("Starting BrowserGym web agent")
-    
-    with gr.Blocks(title="WebShephered Demo") as demo:
-        # Add CSS for outlined groups
-        gr.Markdown("# WebShephered Demo")
-        with gr.Row():
-            with gr.Column(scale=2):
-                with gr.Column():
-                    instruction = gr.Textbox(
-                        label="Instruction",
-                        placeholder="Enter your instruction here",
-                        lines=2,
-                    )
-                    gr.Examples(
-                        examples=[[e] for e in EXAMPLES],
-                        inputs=instruction,
-                        cache_examples=False,
-                    )
-                
-                gr.Markdown("\n\n")
-                
-                with gr.Column():
-                    start_url = gr.Textbox(
-                        label="Starting URL",
-                        placeholder="URL to start the browser at",
-                        value="about:blank"
-                    )
-                    gr.Examples(
-                        examples=URL_EXAMPLES,
-                        inputs=start_url,
-                        cache_examples=False,
-                    )
+    response = ""
 
-                gr.Markdown("\n\n")
+    for message in client.chat_completion(
+        messages,
+        max_tokens=max_tokens,
+        stream=True,
+        temperature=temperature,
+        top_p=top_p,
+    ):
+        token = message.choices[0].delta.content
 
-                model_name = gr.Dropdown(
-                    label="Agent Model",
-                    choices=["gpt-4o"],
-                    value="gpt-4o"
-                )
-                run_btn = gr.Button("Run Demo")
+        response += token
+        yield response
 
-                gr.Markdown("---")
-                
-                with gr.Column():
-                    gr.Markdown("## Current State")
-                    state_view = gr.Markdown()
-                    browser_view = gr.Image(label="Browser View")
 
-                    gr.Markdown("### Task Checklist from WebShephered")
-                    checklist_view = gr.Markdown()
-                    
-                    gr.Markdown("### Action Selection in current step")
-                    with gr.Row() as rm_row:
-                        rm_cards_container = gr.HTML()
-            with gr.Column(scale=2):
-                gr.Markdown("## Trajectory")
-                trajectory_container = gr.HTML() # Placeholder for our custom trajectory component
-
-        
-            
-        run_btn.click(
-            fn=process_run,
-            inputs=[instruction, model_name, start_url],
-            outputs=[state_view, browser_view, checklist_view, rm_cards_container, trajectory_container],
-            api_name="run_agent",
-            concurrency_limit=32,
-            show_progress=True
-        )
+"""
+For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
+"""
+demo = gr.ChatInterface(
+    respond,
+    additional_inputs=[
+        gr.Textbox(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="Temperature"),
+        gr.Slider(
+            minimum=0.1,
+            maximum=1.0,
+            value=0.95,
+            step=0.05,
+            label="Top-p (nucleus sampling)",
+        ),
+    ],
+)
 
-    logger.info("Launching Gradio interface")
-    # Set max_threads to allow multiple concurrent requests
-    demo.launch(share=True, max_threads=32)
 
 if __name__ == "__main__":
-    # Add support for multiprocessing on Windows
-    multiprocessing.freeze_support()
-    main()
+    demo.launch()