Improved logging

app.py

@@ -48,14 +48,14 @@ def restart_space():
 
 ### Space initialisation
 try:
-    print(EVAL_REQUESTS_PATH)
+    logger.info(f"Downloading evaluation requests from {QUEUE_REPO} to {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)
+    logger.info(f"Downloading evaluation results from {RESULTS_REPO} to {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
     )

src/about.py

@@ -81,4 +81,3 @@ CITATION_BUTTON_TEXT = r"""
     note={Online resource for evaluating LLM security aspects}
 }
 """
-

src/display/utils.py

@@ -1,8 +1,6 @@
 from dataclasses import dataclass, make_dataclass
 from enum import Enum
 
-import pandas as pd
-
 from src.about import Tasks
 
 def fields(raw_class):

src/envs.py

@@ -7,7 +7,7 @@ from huggingface_hub import HfApi
 TOKEN = os.environ.get("HF_TOKEN") # A read/write token for your org
 
 OWNER = "stacklok"
-REPO_ID = "secure-code-leaderboard"
+REPO_ID = "llm_security_leaderboard"
 # ----------------------------------
 
 REPO_ID = f"{OWNER}/{REPO_ID}"

src/leaderboard/read_evals.py

@@ -19,7 +19,7 @@ class EvalResult:
     """
     eval_name: str  # org_model_precision (uid)
     full_model: str  # org/model (path on hub)
-    org: str 
+    org: str
     model: str
     results: dict
     rank : int = 0
@@ -29,7 +29,7 @@ class EvalResult:
     model_type: ModelType = ModelType.Unknown  # Pretrained, fine tuned, ...
     weight_type: WeightType = WeightType.Original  # Original or Adapter
     revision: str = ""  # commit hash, "" if main
-    architecture: str = "Unknown" 
+    architecture: str = "Unknown"
     license: str = "?"
     likes: int = 0
     num_params: int = 0
@@ -116,7 +116,7 @@ class EvalResult:
             self.num_params = request.get("params", 0)
             self.date = request.get("submitted_time", "")
         except Exception:
-            print(f"Could not find request file for {self.org}/{self.model} with precision {self.precision.value.name}")
+            logging.warning(f"Could not find request file for {self.org}/{self.model} with precision {self.precision.value.name}")
 
     def to_dict(self):
         """Converts the Eval Result to a dict compatible with our dataframe display"""
@@ -126,7 +126,7 @@ class EvalResult:
             average = sum([v for v in self.results.values() if v is not None]) / len(Tasks)
         else:
             average = 0.0
-        
+
         data_dict = {
             "eval_name": self.eval_name,  # not a column, just a save name
             AutoEvalColumn.rank.name: self.rank,

utils/create_datasets.py

@@ -1,12 +1,10 @@
-from huggingface_hub import HfApi, create_repo
+from huggingface_hub import HfApi
+from pathlib import Path
 
 # Authenticate with Hugging Face token
 api = HfApi()
 api.create_repo(repo_id="stacklok/requests", repo_type="dataset")
 
-# Example: Push dataset files
-from huggingface_hub import HfApi
-from pathlib import Path
 
 api.upload_folder(
     folder_path=Path("path_to_local_dataset"),