Functionality rework

app.py

@@ -13,6 +13,7 @@ from models import get_tinyllama, get_qwen2ins1b, GigaChat, response_gigachat, r
 
 INIT_MODELS = dict()
 S3_SESSION = None
+TEST_MD = None
 CURRENT_MODELS = queue.LifoQueue()
 MODEL_LIB = {'TINYLLAMA': get_tinyllama, "QWEN2INS1B": get_qwen2ins1b, "RUBASE": GigaChat.get_giga}
 GEN_LIB = {'TINYLLAMA': response_tinyllama, "QWEN2INS1B": response_qwen2ins1b, "RUBASE": response_gigachat}
@@ -86,7 +87,7 @@ def model_manager(
     CURRENT_MODELS.put(add_model)
     INIT_MODELS[add_model] = model_lib[add_model]()
 
-def tab_arena():
+def tab_online_arena():
     global S3_SESSION, GEN_LIB, MODEL_LIB, INIT_MODELS, CURRENT_MODELS
     with gradio.Row():
         with gradio.Column():
@@ -196,8 +197,29 @@ def tab_leaderboard():
         with gradio.Blocks() as demo:
             gradio.DataFrame(df)
 
-with open("test.md", "r") as f:
-    TEST_MD = f.read()
+def tab_offline_arena(): 
+    # with gradio.Row(): 
+    #     btn_show_history = gradio.Button("🎲  Click here to sample an example + a pair of LLM outputs! ", elem_classes="sample_button")
+    with gradio.Row():
+        with gradio.Column(scale=2):
+            with gradio.Accordion("Choose models to sample from", open=False, elem_classes="accordion-label"):
+                model_options = list(MODEL_LIB.keys())  
+                selected_models = gradio.CheckboxGroup(model_options, info="", value=model_options, show_label=False, elem_id="select-models") 
+                clear_button = gradio.Button("Clear", elem_classes="btn_boderline_gray", scale=1)
+                # clear the selected_models
+                clear_button.click(lambda: {selected_models: {"value": [], "__type__": "update"}}, inputs=[], outputs=[selected_models])
+        with gradio.Column(scale=1):
+            with gradio.Accordion("Choose task types to sample from", open=False, elem_classes="accordion-label"):
+                select_tasks = gradio.CheckboxGroup(['Task 1', "Task 2", "Task 3"], info="", value=['Task 1', "Task 2", "Task 3"], show_label=False, elem_id="select-tasks") 
+                clear_task_button = gradio.Button("Clear", elem_classes="btn_boderline_gray", scale=1)
+                # clear the select_tasks
+                clear_task_button.click(lambda: {select_tasks: {"value": [], "__type__": "update"}}, inputs=[], outputs=[select_tasks])
+            with gradio.Accordion("Choose criteria  to sample from", open=False, elem_classes="accordion-label"):
+                select_tasks = gradio.CheckboxGroup(['Criterion 1', "Criterion 2", "Criterion 3"], info="", value=['Criterion 1', "Criterion 2", "Criterion 3"], show_label=False, elem_id="select-criteria") 
+                clear_task_button = gradio.Button("Clear", elem_classes="btn_boderline_gray", scale=1)
+                # clear the select_tasks
+                clear_task_button.click(lambda: {select_tasks: {"value": [], "__type__": "update"}}, inputs=[], outputs=[select_tasks])
+
 
 def build_demo():
     # global original_dfs, available_models, gpt4t_dfs, haiku_dfs, llama_dfs
@@ -212,10 +234,11 @@ def build_demo():
                 tab_leaderboard()
 
             with gradio.TabItem("🆚 SBS by categories and criteria", elem_id="od-benchmark-tab-table", id=1):
-                gradio.Markdown(TEST_MD, elem_classes="markdown-text-details")
+                # gradio.Markdown(TEST_MD, elem_classes="markdown-text-details")
+                tab_offline_arena()
 
             with gradio.TabItem("🥊 Model arena", elem_id="od-benchmark-tab-table", id=2):
-                tab_arena()
+                tab_online_arena()
                 # _tab_explore()
 
             with gradio.TabItem("💪 About MERA", elem_id="od-benchmark-tab-table", id=3):
@@ -242,6 +265,9 @@ if __name__ == "__main__":
     # data_load(args.result_file)    
     # TYPES = ["number", "markdown", "number"]
 
+    with open("test.md", "r") as f:
+        TEST_MD = f.read()
+
     try:
         session = boto3.session.Session()
         S3_SESSION = session.client(