Update app.py

app.py

@@ -1,56 +1,41 @@
-import random
 import os
-import datetime
-import sys
-from txagent import TxAgent
-import spaces
+import random
 import gradio as gr
+from datetime import datetime
+
+from txagent import TxAgent
 
-# Set environment variables
+# ==== Environment Setup ====
 current_dir = os.path.dirname(os.path.abspath(__file__))
 os.environ["MKL_THREADING_LAYER"] = "GNU"
 os.environ["TOKENIZERS_PARALLELISM"] = "false"
 HF_TOKEN = os.environ.get("HF_TOKEN", None)
 
+# ==== UI Content ====
 DESCRIPTION = '''
 <div>
 <h1 style="text-align: center;">TxAgent: An AI Agent for Therapeutic Reasoning Across a Universe of Tools </h1>
 </div>
 '''
-INTRO = """
-Precision therapeutics require multimodal adaptive models that provide personalized treatment recommendations...
-"""
-LICENSE = """
-DISCLAIMER: THIS WEBSITE DOES NOT PROVIDE MEDICAL ADVICE...
-"""
+INTRO = "Precision therapeutics require multimodal adaptive models..."
+LICENSE = "DISCLAIMER: THIS WEBSITE DOES NOT PROVIDE MEDICAL ADVICE..."
 
-PLACEHOLDER = """
-<div style="padding: 30px; text-align: center; display: flex; flex-direction: column; align-items: center;">
+PLACEHOLDER = '''
+<div style="padding: 30px; text-align: center;">
    <h1 style="font-size: 28px; margin-bottom: 2px; opacity: 0.55;">TxAgent</h1>
-   <p style="font-size: 18px; margin-bottom: 2px; opacity: 0.65;">Tips before using TxAgent:</p>
-   <p style="font-size: 18px; margin-bottom: 2px; opacity: 0.55;">Please click clear🗑️
- (top-right) to remove previous context before submitting a new question.</p>
-   <p style="font-size: 18px; margin-bottom: 2px; opacity: 0.55;">Click retry🔄 (below message) to get multiple versions of the answer.</p>
+   <p style="font-size: 18px;">Click clear 🗑️ before asking a new question.</p>
+   <p style="font-size: 18px;">Click retry 🔄 to see another answer.</p>
 </div>
-"""
+'''
 
 css = """
-h1 {
-  text-align: center;
-  display: block;
-}
+h1 { text-align: center; }
 #duplicate-button {
   margin: auto;
   color: white;
   background: #1565c0;
   border-radius: 100vh;
 }
-.small-button button {
-    font-size: 12px !important;
-    padding: 4px 8px !important;
-    height: 6px !important;
-    width: 4px !important;
-}
 .gradio-accordion {
     margin-top: 0px !important;
     margin-bottom: 0px !important;
@@ -62,167 +47,110 @@ chat_css = """
 .gr-button svg { width: 32px !important; height: 32px !important; }
 """
 
-model_name = 'mims-harvard/TxAgent-T1-Llama-3.1-8B'
-rag_model_name = 'mims-harvard/ToolRAG-T1-GTE-Qwen2-1.5B'
-
-question_examples = [
-    ['Given a 50-year-old patient experiencing severe acute pain and considering the use of the newly approved medication, Journavx, how should the dosage be adjusted considering the presence of moderate hepatic impairment?'],
-    ['Given a 50-year-old patient experiencing severe acute pain and considering the use of the newly approved medication, Journavx, how should the dosage be adjusted considering the presence of severe hepatic impairment?'],
-    ['A 30-year-old patient is taking Prozac to treat their depression. They were recently diagnosed with WHIM syndrome and require a treatment for that condition as well. Is Xolremdi suitable for this patient, considering contraindications?'],
-]
-
+# ==== Model Settings ====
+model_name = "mims-harvard/TxAgent-T1-Llama-3.1-8B"
+rag_model_name = "mims-harvard/ToolRAG-T1-GTE-Qwen2-1.5B"
 new_tool_files = {
-    'new_tool': os.path.join(current_dir, 'data', 'new_tool.json'),
+    "new_tool": os.path.join(current_dir, "data", "new_tool.json")
 }
 
-agent = TxAgent(model_name,
-                rag_model_name,
-                tool_files_dict=new_tool_files,
-                force_finish=True,
-                enable_checker=True,
-                step_rag_num=10,
-                seed=100,
-                additional_default_tools=['DirectResponse', 'RequireClarification'])
-agent.init_model()
-
-def update_model_parameters(enable_finish, enable_rag, enable_summary,
-                            init_rag_num, step_rag_num, skip_last_k,
-                            summary_mode, summary_skip_last_k, summary_context_length, force_finish, seed):
-    return agent.update_parameters(
-        enable_finish=enable_finish,
-        enable_rag=enable_rag,
-        enable_summary=enable_summary,
-        init_rag_num=init_rag_num,
-        step_rag_num=step_rag_num,
-        skip_last_k=skip_last_k,
-        summary_mode=summary_mode,
-        summary_skip_last_k=summary_skip_last_k,
-        summary_context_length=summary_context_length,
-        force_finish=force_finish,
-        seed=seed,
-    )
-
-def update_seed():
-    seed = random.randint(0, 10000)
-    return agent.update_parameters(seed=seed)
-
-def handle_retry(history, retry_data: gr.RetryData, temperature, max_new_tokens, max_tokens, multi_agent, conversation, max_round):
-    update_seed()
-    new_history = history[:retry_data.index]
-    previous_prompt = history[retry_data.index]['content']
-    yield from agent.run_gradio_chat(new_history + [{"role": "user", "content": previous_prompt}], temperature, max_new_tokens, max_tokens, multi_agent, conversation, max_round)
-
-PASSWORD = "mypassword"
-
-def check_password(input_password):
-    if input_password == PASSWORD:
-        return gr.update(visible=True), ""
-    else:
-        return gr.update(visible=False), "Incorrect password, try again!"
-
+question_examples = [
+    ["Given a 50-year-old patient experiencing severe acute pain and considering the use of the newly approved medication, Journavx, how should the dosage be adjusted considering moderate hepatic impairment?"],
+    ["A 30-year-old patient is on Prozac for depression and now diagnosed with WHIM syndrome. Is Xolremdi suitable?"]
+]
 
+# ====== Main Application Entrypoint ======
 if __name__ == "__main__":
-    conversation_state = gr.State([])
-
-    chatbot = gr.Chatbot(
-        height=800, placeholder=PLACEHOLDER, label='TxAgent',
-        type="messages", show_copy_button=True
+    # === Initialize the model (inside __main__) ===
+    agent = TxAgent(
+        model_name,
+        rag_model_name,
+        tool_files_dict=new_tool_files,
+        force_finish=True,
+        enable_checker=True,
+        step_rag_num=10,
+        seed=100,
+        additional_default_tools=["DirectResponse", "RequireClarification"]
     )
+    agent.init_model()
+
+    # === Gradio interface logic ===
+    def handle_chat(message, history, temperature, max_new_tokens, max_tokens, multi_agent, conversation, max_round):
+        return agent.run_gradio_chat(message, history, temperature, max_new_tokens, max_tokens, multi_agent, conversation, max_round)
+
+    def update_model_parameters(enable_finish, enable_rag, enable_summary,
+                                init_rag_num, step_rag_num, skip_last_k,
+                                summary_mode, summary_skip_last_k, summary_context_length,
+                                force_finish, seed):
+        return agent.update_parameters(
+            enable_finish=enable_finish,
+            enable_rag=enable_rag,
+            enable_summary=enable_summary,
+            init_rag_num=init_rag_num,
+            step_rag_num=step_rag_num,
+            skip_last_k=skip_last_k,
+            summary_mode=summary_mode,
+            summary_skip_last_k=summary_skip_last_k,
+            summary_context_length=summary_context_length,
+            force_finish=force_finish,
+            seed=seed
+        )
+
+    def update_seed():
+        seed = random.randint(0, 10000)
+        return agent.update_parameters(seed=seed)
+
+    def handle_retry(history, retry_data: gr.RetryData, temperature, max_new_tokens, max_tokens, multi_agent, conversation, max_round):
+        update_seed()
+        new_history = history[:retry_data.index]
+        previous_prompt = history[retry_data.index]["content"]
+        yield from agent.run_gradio_chat(
+            new_history + [{"role": "user", "content": previous_prompt}],
+            temperature, max_new_tokens, max_tokens, multi_agent, conversation, max_round
+        )
 
+    # ===== Build Gradio Interface =====
     with gr.Blocks(css=css) as demo:
         gr.Markdown(DESCRIPTION)
         gr.Markdown(INTRO)
 
-        default_temperature = 0.3
-        default_max_new_tokens = 1024
-        default_max_tokens = 81920
-        default_max_round = 30
-
-        temperature_state = gr.State(value=default_temperature)
-        max_new_tokens_state = gr.State(value=default_max_new_tokens)
-        max_tokens_state = gr.State(value=default_max_tokens)
-        max_round_state = gr.State(value=default_max_round)
+        temperature = gr.Slider(0, 1, step=0.1, value=0.3, label="Temperature")
+        max_new_tokens = gr.Slider(128, 4096, step=1, value=1024, label="Max New Tokens")
+        max_tokens = gr.Slider(128, 32000, step=1, value=8192, label="Max Total Tokens")
+        max_round = gr.Slider(1, 50, step=1, value=30, label="Max Rounds")
+        multi_agent = gr.Checkbox(label="Enable Multi-agent Reasoning", value=False)
+        conversation_state = gr.State([])
+
+        chatbot = gr.Chatbot(
+            label="TxAgent",
+            placeholder=PLACEHOLDER,
+            height=700,
+            type="messages",
+            show_copy_button=True
+        )
 
         chatbot.retry(
             handle_retry,
             chatbot, chatbot,
-            temperature_state, max_new_tokens_state,
-            max_tokens_state,
-            gr.Checkbox(value=False, render=False),
-            conversation_state,
-            max_round_state
+            temperature, max_new_tokens, max_tokens,
+            multi_agent, conversation_state, max_round
         )
 
         gr.ChatInterface(
-            fn=agent.run_gradio_chat,
+            fn=handle_chat,
             chatbot=chatbot,
-            fill_height=True, fill_width=True, stop_btn=True,
-            additional_inputs_accordion=gr.Accordion(label="⚙️ Inference Parameters", open=False, render=False),
             additional_inputs=[
-                temperature_state, max_new_tokens_state, max_tokens_state,
-                gr.Checkbox(label="Activate multi-agent reasoning mode", value=False, render=False),
-                conversation_state,
-                max_round_state,
-                gr.Number(label="Seed", value=100, render=False)
+                temperature, max_new_tokens, max_tokens,
+                multi_agent, conversation_state, max_round
             ],
             examples=question_examples,
-            cache_examples=False,
             css=chat_css,
+            cache_examples=False,
+            fill_height=True,
+            fill_width=True,
+            stop_btn=True
         )
 
-        with gr.Accordion("Settings", open=False):
-            temperature_slider = gr.Slider(0, 1, step=0.1, value=default_temperature, label="Temperature")
-            max_new_tokens_slider = gr.Slider(128, 4096, step=1, value=default_max_new_tokens, label="Max new tokens")
-            max_tokens_slider = gr.Slider(128, 32000, step=1, value=default_max_tokens, label="Max tokens")
-            max_round_slider = gr.Slider(0, 50, step=1, value=default_max_round, label="Max round")
-
-            temperature_slider.change(lambda x: x, inputs=temperature_slider, outputs=temperature_state)
-            max_new_tokens_slider.change(lambda x: x, inputs=max_new_tokens_slider, outputs=max_new_tokens_state)
-            max_tokens_slider.change(lambda x: x, inputs=max_tokens_slider, outputs=max_tokens_state)
-            max_round_slider.change(lambda x: x, inputs=max_round_slider, outputs=max_round_state)
-
-            password_input = gr.Textbox(label="Enter Password for More Settings", type="password")
-            incorrect_message = gr.Textbox(visible=False, interactive=False)
-
-            with gr.Accordion("⚙️ Settings", open=False, visible=False) as protected_accordion:
-                with gr.Row():
-                    with gr.Column(scale=1):
-                        with gr.Accordion("⚙️ Model Loading", open=False):
-                            model_name_input = gr.Textbox(label="Enter model path", value=model_name)
-                            load_model_btn = gr.Button(value="Load Model")
-                            load_model_btn.click(agent.load_models, inputs=model_name_input, outputs=gr.Textbox(label="Status"))
-                    with gr.Column(scale=1):
-                        with gr.Accordion("⚙️ Functional Parameters", open=False):
-                            enable_finish = gr.Checkbox(label="Enable Finish", value=True)
-                            enable_rag = gr.Checkbox(label="Enable RAG", value=True)
-                            enable_summary = gr.Checkbox(label="Enable Summary", value=False)
-                            init_rag_num = gr.Number(label="Initial RAG Num", value=0)
-                            step_rag_num = gr.Number(label="Step RAG Num", value=10)
-                            skip_last_k = gr.Number(label="Skip Last K", value=0)
-                            summary_mode = gr.Textbox(label="Summary Mode", value='step')
-                            summary_skip_last_k = gr.Number(label="Summary Skip Last K", value=0)
-                            summary_context_length = gr.Number(label="Summary Context Length", value=None)
-                            force_finish = gr.Checkbox(label="Force FinalAnswer", value=True)
-                            seed = gr.Number(label="Seed", value=100)
-
-                            submit_btn = gr.Button("Update Parameters")
-                            updated_parameters_output = gr.JSON()
-                            submit_btn.click(
-                                fn=update_model_parameters,
-                                inputs=[
-                                    enable_finish, enable_rag, enable_summary, init_rag_num, step_rag_num, skip_last_k,
-                                    summary_mode, summary_skip_last_k, summary_context_length, force_finish, seed
-                                ],
-                                outputs=updated_parameters_output
-                            )
-
-            submit_button = gr.Button("Submit")
-            submit_button.click(
-                check_password,
-                inputs=password_input,
-                outputs=[protected_accordion, incorrect_message]
-            )
-
         gr.Markdown(LICENSE)
 
-    demo.launch(share=True)
+    demo.launch()