Random VQA Sample button for VLM direct chat

fastchat/serve/gradio_block_arena_vision.py

@@ -8,6 +8,8 @@
 """
 
 import os
+import json
+import numpy as np
 
 import gradio as gr
 
@@ -31,14 +33,24 @@
 logger = build_logger("gradio_web_server_multi", "gradio_web_server_multi.log")
 
 
+def get_vqa_sample(example_json: gr.components.Textbox):
+    with open(example_json, "r") as f:
+        vqa_samples = json.load(f)
+    random_sample = np.random.choice(vqa_samples)
+    question, path = random_sample["question"], random_sample["path"]
+    return question, path
+
+
 def clear_history_example(request: gr.Request):
     ip = get_ip(request)
     logger.info(f"clear_history_example. ip: {ip}")
     state = None
     return (state, []) + (disable_btn,) * 5
 
 
-def build_single_vision_language_model_ui(models, add_promotion_links=False):
+def build_single_vision_language_model_ui(
+    models, add_promotion_links=False, random_questions=None
+):
     promotion = (
         """
 | [GitHub](https://github.com/lm-sys/FastChat) | [Dataset](https://github.com/lm-sys/FastChat/blob/main/docs/dataset_release.md) | [Twitter](https://twitter.com/lmsysorg) | [Discord](https://discord.gg/HSWAKCrnFx) |
@@ -140,9 +152,8 @@ def build_single_vision_language_model_ui(models, add_promotion_links=False):
                 flag_btn = gr.Button(value="⚠️  Flag", interactive=False)
                 regenerate_btn = gr.Button(value="🔄  Regenerate", interactive=False)
                 clear_btn = gr.Button(value="🗑️  Clear", interactive=False)
-
-    if add_promotion_links:
-        gr.Markdown(acknowledgment_md, elem_id="ack_markdown")
+                if random_questions:
+                    random_btn = gr.Button(value="🎲  Random", interactive=True)
 
     # Register listeners
     btn_list = [upvote_btn, downvote_btn, flag_btn, regenerate_btn, clear_btn]
@@ -194,4 +205,12 @@ def build_single_vision_language_model_ui(models, add_promotion_links=False):
         [state, chatbot] + btn_list,
     )
 
+    if random_questions:
+        questions_textbox = gr.Textbox(value=random_questions, visible=False)
+        random_btn.click(
+            get_vqa_sample,  # First, get the VQA sample
+            questions_textbox,  # Pass the path to the VQA samples
+            [textbox, imagebox],  # Outputs are textbox and imagebox
+        )
+
     return [state, model_selector]

fastchat/serve/gradio_web_server_multi.py

@@ -131,7 +131,9 @@ def build_demo(models, vl_models, elo_results_file, leaderboard_table_file):
             with gr.Tab("Vision Direct Chat", id=3, visible=args.multimodal):
                 single_vision_language_model_list = (
                     build_single_vision_language_model_ui(
-                        vl_models, add_promotion_links=True
+                        vl_models,
+                        add_promotion_links=True,
+                        random_questions=args.random_questions,
                     )
                 )
 
@@ -202,6 +204,9 @@ def build_demo(models, vl_models, elo_results_file, leaderboard_table_file):
     parser.add_argument(
         "--multimodal", action="store_true", help="Show multi modal tabs."
     )
+    parser.add_argument(
+        "--random-questions", type=str, help="Load random questions from a JSON file"
+    )
     parser.add_argument(
         "--register-api-endpoint-file",
         type=str,