Update app.py

app.py

@@ -1,10 +1,9 @@
 import gradio as gr
-from gradio_leaderboard import Leaderboard, ColumnFilter, SelectColumns
 import pandas as pd
 from apscheduler.schedulers.background import BackgroundScheduler
-from huggingface_hub import snapshot_download
-
-from src.about import (
+# Removed Hugging Face Hub imports as they are not needed for the simplified leaderboard
+# from huggingface_hub import snapshot_download, HfApi
+from src.about import ( # Assuming these still exist and are relevant for other tabs
     CITATION_BUTTON_LABEL,
     CITATION_BUTTON_TEXT,
     EVALUATION_QUEUE_TEXT,
@@ -12,112 +11,124 @@ from src.about import (
     LLM_BENCHMARKS_TEXT,
     TITLE,
 )
-from src.display.css_html_js import custom_css
-from src.display.utils import (
-    BENCHMARK_COLS,
-    COLS,
-    EVAL_COLS,
-    EVAL_TYPES,
-    AutoEvalColumn,
-    ModelType,
-    fields,
-    WeightType,
-    Precision
-)
-from src.envs import API, EVAL_REQUESTS_PATH, EVAL_RESULTS_PATH, QUEUE_REPO, REPO_ID, RESULTS_REPO, TOKEN
-from src.populate import get_evaluation_queue_df, get_leaderboard_df
-from src.submission.submit import add_new_eval
-
-
+from src.display.css_html_js import custom_css # Keep custom CSS
+# Removed utils imports related to the old leaderboard
+# from src.display.utils import (...)
+from src.envs import REPO_ID # Keep if needed for restart_space or other functions
+# Removed constants related to old data paths and repos if not needed elsewhere
+# from src.envs import API, EVAL_REQUESTS_PATH, EVAL_RESULTS_PATH, QUEUE_REPO, REPO_ID, RESULTS_REPO, TOKEN
+# Removed old data processing functions
+# from src.populate import get_evaluation_queue_df, get_leaderboard_df
+from src.submission.submit import add_new_eval # Keep submission logic
+
+# --- New Elo Leaderboard Configuration ---
+INITIAL_MODELS = [
+    "gpt-4o-mini", "gpt-4o", "gemini-2.0-flash", "deepseek-v3",
+    "gemini-2.0-pro", "o3-mini", "deepseek-r1", "gemini-2.5-pro"
+]
+CATEGORIES = ["MLE-Lite", "Tabular", "NLP", "CV"]
+DEFAULT_ELO = 1200
+
+# Placeholder data structure for Elo scores per category
+# *** MODIFY THE SCORES HERE AS NEEDED ***
+elo_data = {
+    category: pd.DataFrame({
+        "Model": INITIAL_MODELS,
+        "Elo Score": [DEFAULT_ELO] * len(INITIAL_MODELS)
+    }) for category in CATEGORIES
+}
+# Example: How to set specific scores for a category
+# elo_data["NLP"] = pd.DataFrame({
+#     "Model": INITIAL_MODELS,
+#     "Elo Score": [1300, 1450, 1250, 1350, 1400, 1150, 1320, 1500] # Example scores
+# })
+
+# --- Helper function to update leaderboard ---
+def update_leaderboard(category):
+    """Returns the DataFrame for the selected category."""
+    df = elo_data.get(category)
+    if df is None:
+        # Return default if category not found (shouldn't happen with radio)
+        return elo_data[CATEGORIES[0]]
+    return df
+
+# --- Mock/Placeholder functions/data for other tabs ---
+# Since we removed the snapshot download, the original queue fetching will fail.
+# Provide empty DataFrames or mock data if you want the queue display to work without the original data source.
+# This is a placeholder - replace with actual data loading if needed for the submission tab.
+print("Warning: Evaluation queue data fetching is disabled/mocked due to leaderboard changes.")
+finished_eval_queue_df = pd.DataFrame(columns=["Model", "Status", "Requested", "Started"])
+running_eval_queue_df = pd.DataFrame(columns=["Model", "Status", "Requested", "Started"])
+pending_eval_queue_df = pd.DataFrame(columns=["Model", "Status", "Requested", "Started"])
+EVAL_COLS = ["Model", "Status", "Requested", "Started"] # Define for the dataframe headers
+EVAL_TYPES = ["str", "str", "str", "str"] # Define for the dataframe types
+
+# --- Keep restart function if relevant ---
+# Assuming HfApi is initialized elsewhere or REPO_ID is sufficient
+# api = HfApi() # Example initialization, adjust as needed
 def restart_space():
-    API.restart_space(repo_id=REPO_ID)
-
-### Space initialisation
-try:
-    print(EVAL_REQUESTS_PATH)
-    snapshot_download(
-        repo_id=QUEUE_REPO, local_dir=EVAL_REQUESTS_PATH, repo_type="dataset", tqdm_class=None, etag_timeout=30, token=TOKEN
-    )
-except Exception:
-    restart_space()
-try:
-    print(EVAL_RESULTS_PATH)
-    snapshot_download(
-        repo_id=RESULTS_REPO, local_dir=EVAL_RESULTS_PATH, repo_type="dataset", tqdm_class=None, etag_timeout=30, token=TOKEN
-    )
-except Exception:
-    restart_space()
-
-
-LEADERBOARD_DF = get_leaderboard_df(EVAL_RESULTS_PATH, EVAL_REQUESTS_PATH, COLS, BENCHMARK_COLS)
-
-(
-    finished_eval_queue_df,
-    running_eval_queue_df,
-    pending_eval_queue_df,
-) = get_evaluation_queue_df(EVAL_REQUESTS_PATH, EVAL_COLS)
-
-def init_leaderboard(dataframe):
-    if dataframe is None or dataframe.empty:
-        raise ValueError("Leaderboard DataFrame is empty or None.")
-    return Leaderboard(
-        value=dataframe,
-        datatype=[c.type for c in fields(AutoEvalColumn)],
-        select_columns=SelectColumns(
-            default_selection=[c.name for c in fields(AutoEvalColumn) if c.displayed_by_default],
-            cant_deselect=[c.name for c in fields(AutoEvalColumn) if c.never_hidden],
-            label="Select Columns to Display:",
-        ),
-        search_columns=[AutoEvalColumn.model.name, AutoEvalColumn.license.name],
-        hide_columns=[c.name for c in fields(AutoEvalColumn) if c.hidden],
-        filter_columns=[
-            ColumnFilter(AutoEvalColumn.model_type.name, type="checkboxgroup", label="Model types"),
-            ColumnFilter(AutoEvalColumn.precision.name, type="checkboxgroup", label="Precision"),
-            ColumnFilter(
-                AutoEvalColumn.params.name,
-                type="slider",
-                min=0.01,
-                max=150,
-                label="Select the number of parameters (B)",
-            ),
-            ColumnFilter(
-                AutoEvalColumn.still_on_hub.name, type="boolean", label="Deleted/incomplete", default=True
-            ),
-        ],
-        bool_checkboxgroup_label="Hide models",
-        interactive=False,
-    )
-
-
+    print(f"Attempting to restart space: {REPO_ID}")
+    # Replace with your actual space restart mechanism if needed
+    # try:
+    #     api.restart_space(repo_id=REPO_ID)
+    #     print("Space restart request sent.")
+    # except Exception as e:
+    #     print(f"Failed to restart space: {e}")
+
+# --- Gradio App Definition ---
 demo = gr.Blocks(css=custom_css)
+
 with demo:
     gr.HTML(TITLE)
     gr.Markdown(INTRODUCTION_TEXT, elem_classes="markdown-text")
 
     with gr.Tabs(elem_classes="tab-buttons") as tabs:
         with gr.TabItem("🏅 LLM Benchmark", elem_id="llm-benchmark-tab-table", id=0):
-            leaderboard = init_leaderboard(LEADERBOARD_DF)
+            with gr.Column():
+                gr.Markdown("## Model Elo Rankings") # New title for the section
+                category_selector = gr.Radio(
+                    choices=CATEGORIES,
+                    label="Select Category",
+                    value=CATEGORIES[0], # Default selection
+                    interactive=True,
+                    container=False, # Make radio buttons horizontal if possible with CSS
+                )
+                leaderboard_df_component = gr.Dataframe(
+                    value=update_leaderboard(CATEGORIES[0]), # Initial value
+                    headers=["Model", "Elo Score"],
+                    datatype=["str", "number"],
+                    interactive=False,
+                    row_count=(len(INITIAL_MODELS), "fixed"), # Fixed row count
+                    col_count=(2, "fixed"), # Fixed column count
+                )
+                # Link the radio button change to the update function
+                category_selector.change(
+                    fn=update_leaderboard,
+                    inputs=category_selector,
+                    outputs=leaderboard_df_component
+                )
 
         with gr.TabItem("📝 About", elem_id="llm-benchmark-tab-table", id=2):
             gr.Markdown(LLM_BENCHMARKS_TEXT, elem_classes="markdown-text")
 
         with gr.TabItem("🚀 Submit here! ", elem_id="llm-benchmark-tab-table", id=3):
+            # --- This section remains largely unchanged, but relies on potentially missing data ---
             with gr.Column():
                 with gr.Row():
                     gr.Markdown(EVALUATION_QUEUE_TEXT, elem_classes="markdown-text")
-
                 with gr.Column():
+                     # Displaying queue tables with potentially empty/mock data
                     with gr.Accordion(
-                        f"✅ Finished Evaluations ({len(finished_eval_queue_df)})",
+                        f"✅ Finished Evaluations ({len(finished_eval_queue_df)})", # Length might be 0
                         open=False,
                     ):
                         with gr.Row():
-                            finished_eval_table = gr.components.Dataframe(
+                           finished_eval_table = gr.components.Dataframe(
                                 value=finished_eval_queue_df,
                                 headers=EVAL_COLS,
                                 datatype=EVAL_TYPES,
                                 row_count=5,
-                            )
+                           )
                     with gr.Accordion(
                         f"🔄 Running Evaluation Queue ({len(running_eval_queue_df)})",
                         open=False,
@@ -129,7 +140,6 @@ with demo:
                                 datatype=EVAL_TYPES,
                                 row_count=5,
                             )
-
                     with gr.Accordion(
                         f"⏳ Pending Evaluation Queue ({len(pending_eval_queue_df)})",
                         open=False,
@@ -141,31 +151,35 @@ with demo:
                                 datatype=EVAL_TYPES,
                                 row_count=5,
                             )
+
             with gr.Row():
                 gr.Markdown("# ✉️✨ Submit your model here!", elem_classes="markdown-text")
-
             with gr.Row():
+                 # Submission form - kept as is
                 with gr.Column():
                     model_name_textbox = gr.Textbox(label="Model name")
                     revision_name_textbox = gr.Textbox(label="Revision commit", placeholder="main")
+                    # Using simple strings for dropdowns now, adjust if ModelType/Precision/WeightType classes are still needed
                     model_type = gr.Dropdown(
-                        choices=[t.to_str(" : ") for t in ModelType if t != ModelType.Unknown],
+                        # choices=[t.to_str(" : ") for t in ModelType if t != ModelType.Unknown], # Original
+                        choices=["Type A", "Type B", "Type C"], # Example choices, replace if needed
                         label="Model type",
                         multiselect=False,
                         value=None,
                         interactive=True,
                     )
-
                 with gr.Column():
                     precision = gr.Dropdown(
-                        choices=[i.value.name for i in Precision if i != Precision.Unknown],
+                        # choices=[i.value.name for i in Precision if i != Precision.Unknown], # Original
+                        choices=["float16", "bfloat16", "float32", "int8"], # Example choices
                         label="Precision",
                         multiselect=False,
                         value="float16",
                         interactive=True,
                     )
                     weight_type = gr.Dropdown(
-                        choices=[i.value.name for i in WeightType],
+                        # choices=[i.value.name for i in WeightType], # Original
+                        choices=["Original", "Adapter", "Delta"], # Example choices
                         label="Weights type",
                         multiselect=False,
                         value="Original",
@@ -175,6 +189,8 @@ with demo:
 
             submit_button = gr.Button("Submit Eval")
             submission_result = gr.Markdown()
+
+            # Keep submission logic attached
             submit_button.click(
                 add_new_eval,
                 [
@@ -198,7 +214,11 @@ with demo:
                 show_copy_button=True,
             )
 
-scheduler = BackgroundScheduler()
-scheduler.add_job(restart_space, "interval", seconds=1800)
-scheduler.start()
-demo.queue(default_concurrency_limit=40).launch()
+# --- Keep scheduler if relevant ---
+# scheduler = BackgroundScheduler()
+# scheduler.add_job(restart_space, "interval", seconds=1800) # Restart every 30 mins
+# scheduler.start()
+
+# --- Launch the app ---
+# demo.queue(default_concurrency_limit=40).launch() # Original launch
+demo.launch() # Simpler launch for testing