prod-mirror

app.py

@@ -1,4 +1,5 @@
 import os
+import time
 import logging
 import gradio as gr
 import pandas as pd
@@ -48,6 +49,8 @@ from src.submission.submit import add_new_eval
 from src.tools.collections import update_collections
 from src.tools.plots import create_metric_plot_obj, create_plot_df, create_scores_df
 
+# Configure logging
+logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
 
 # Start ephemeral Spaces on PRs (see config in README.md)
 enable_space_ci()
@@ -56,13 +59,23 @@ enable_space_ci()
 def restart_space():
     API.restart_space(repo_id=REPO_ID, token=H4_TOKEN)
 
-
+def time_diff_wrapper(func):
+    def wrapper(*args, **kwargs):
+        start_time = time.time()
+        result = func(*args, **kwargs)
+        end_time = time.time()
+        diff = end_time - start_time
+        logging.info(f"Time taken for {func.__name__}: {diff} seconds")
+        return result
+    return wrapper
+
+@time_diff_wrapper
 def download_dataset(repo_id, local_dir, repo_type="dataset", max_attempts=3):
     """Attempt to download dataset with retries."""
     attempt = 0
     while attempt < max_attempts:
         try:
-            print(f"Downloading {repo_id} to {local_dir}")
+            logging.info(f"Downloading {repo_id} to {local_dir}")
             snapshot_download(
                 repo_id=repo_id,
                 local_dir=local_dir,
@@ -84,7 +97,10 @@ def init_space(full_init: bool = True):
     if full_init:
         # These downloads only occur on full initialization
         download_dataset(QUEUE_REPO, EVAL_REQUESTS_PATH)
+        download_dataset(QUEUE_REPO, EVAL_REQUESTS_PATH)
         download_dataset(DYNAMIC_INFO_REPO, DYNAMIC_INFO_PATH)
+        download_dataset(DYNAMIC_INFO_REPO, DYNAMIC_INFO_PATH)
+        download_dataset(RESULTS_REPO, EVAL_RESULTS_PATH)
         download_dataset(RESULTS_REPO, EVAL_RESULTS_PATH)
 
     # Always retrieve the leaderboard DataFrame

src/leaderboard/filter_models.py

@@ -138,7 +138,6 @@ def flag_models(leaderboard_data: list[dict]):
         else:
             flag_key = model_data[AutoEvalColumn.fullname.name]
 
-        print(f"model check: {flag_key}")
         if flag_key in FLAGGED_MODELS:
             print(f"Flagged model: {flag_key}")
             issue_num = FLAGGED_MODELS[flag_key].split("/")[-1]