Removing queue code

app.py

@@ -7,7 +7,6 @@ from huggingface_hub import snapshot_download
 from src.about import (
     CITATION_BUTTON_LABEL,
     CITATION_BUTTON_TEXT,
-    EVALUATION_QUEUE_TEXT,
     INTRODUCTION_TEXT,
     LLM_BENCHMARKS_TEXT,
     TITLE,
@@ -24,9 +23,8 @@ from src.display.utils import (
     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.envs import API, EVAL_RESULTS_PATH, REPO_ID, RESULTS_REPO, TOKEN
+from src.populate import get_leaderboard_df
 
 
 def restart_space():
@@ -34,18 +32,6 @@ def restart_space():
 
 
 ### 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(
@@ -60,7 +46,7 @@ except Exception:
     restart_space()
 
 
-LEADERBOARD_DF = get_leaderboard_df(EVAL_RESULTS_PATH, EVAL_REQUESTS_PATH, COLS, BENCHMARK_COLS)
+LEADERBOARD_DF = get_leaderboard_df(EVAL_RESULTS_PATH, COLS, BENCHMARK_COLS)
 
 # (
 #     finished_eval_queue_df,

src/about.py

@@ -13,17 +13,17 @@ class Task:
 # ---------------------------------------------------
 class Tasks(Enum):
     # task_key in the json file, metric_key in the json file, name to display in the leaderboard
-    task0 = Task("MIMIC CDM Appendicitis", "acc", "CDM App")
-    task1 = Task("MIMIC CDM Cholecystitis", "acc", "CDM Cholec")
-    task2 = Task("MIMIC CDM Diverticulitis", "acc", "CDM Divert")
-    task3 = Task("MIMIC CDM Pancreatitis", "acc", "CDM Pancr")
-    task4 = Task("MIMIC CDM Mean", "acc", "CDM Mean")
-
-    task5 = Task("MIMIC CDM FI Appendicitis", "acc", "CDM FI App")
-    task6 = Task("MIMIC CDM FI Cholecystitis", "acc", "CDM FI Cholec")
-    task7 = Task("MIMIC CDM FI Diverticulitis", "acc", "CDM FI Divert")
-    task8 = Task("MIMIC CDM FI Pancreatitis", "acc", "CDM FI Pancr")
-    task9 = Task("MIMIC CDM FI Mean", "acc", "CDM FI Mean")
+    task0 = Task("MIMIC_CDM_Appendicitis", "acc", "CDM App")
+    task1 = Task("MIMIC_CDM_Cholecystitis", "acc", "CDM Cholec")
+    task2 = Task("MIMIC_CDM_Diverticulitis", "acc", "CDM Divert")
+    task3 = Task("MIMIC_CDM_Pancreatitis", "acc", "CDM Pancr")
+    task4 = Task("MIMIC_CDM_Mean", "acc", "CDM Mean")
+
+    task5 = Task("MIMIC_CDM_FI_Appendicitis", "acc", "CDM FI App")
+    task6 = Task("MIMIC_CDM_FI_Cholecystitis", "acc", "CDM FI Cholec")
+    task7 = Task("MIMIC_CDM_FI_Diverticulitis", "acc", "CDM FI Divert")
+    task8 = Task("MIMIC_CDM_FI_Pancreatitis", "acc", "CDM FI Pancr")
+    task9 = Task("MIMIC_CDM_FI_Mean", "acc", "CDM FI Mean")
 
 
 NUM_FEWSHOT = 0  # Change with your few shot

src/leaderboard/read_evals.py

@@ -14,22 +14,22 @@ from src.submission.check_validity import is_model_on_hub
 
 @dataclass
 class EvalResult:
-    """Represents one full evaluation. Built from a combination of the result and request file for a given run.
-    """
-    eval_name: str # org_model_precision (uid)
-    full_model: str # org/model (path on hub)
-    org: str 
+    """Represents one full evaluation. Built from a combination of the result and request file for a given run."""
+
+    eval_name: str  # org_model_precision (uid)
+    full_model: str  # org/model (path on hub)
+    org: str
     model: str
-    revision: str # commit hash, "" if main
+    revision: str  # commit hash, "" if main
     results: dict
     precision: Precision = Precision.Unknown
-    model_type: ModelType = ModelType.Unknown # Pretrained, fine tuned, ...
-    weight_type: WeightType = WeightType.Original # Original or Adapter
-    architecture: str = "Unknown" 
+    model_type: ModelType = ModelType.Unknown  # Pretrained, fine tuned, ...
+    weight_type: WeightType = WeightType.Original  # Original or Adapter
+    architecture: str = "Unknown"
     license: str = "?"
     likes: int = 0
     num_params: int = 0
-    date: str = "" # submission date of request file
+    date: str = ""  # submission date of request file
     still_on_hub: bool = False
 
     @classmethod
@@ -85,10 +85,10 @@ class EvalResult:
             org=org,
             model=model,
             results=results,
-            precision=precision,  
-            revision= config.get("model_sha", ""),
+            precision=precision,
+            revision=config.get("model_sha", ""),
             still_on_hub=still_on_hub,
-            architecture=architecture
+            architecture=architecture,
         )
 
     def update_with_request_file(self, requests_path):
@@ -105,7 +105,9 @@ 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}")
+            print(
+                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"""
@@ -146,15 +148,12 @@ def get_request_file_for_model(requests_path, model_name, precision):
     for tmp_request_file in request_files:
         with open(tmp_request_file, "r") as f:
             req_content = json.load(f)
-            if (
-                req_content["status"] in ["FINISHED"]
-                and req_content["precision"] == precision.split(".")[-1]
-            ):
+            if req_content["status"] in ["FINISHED"] and req_content["precision"] == precision.split(".")[-1]:
                 request_file = tmp_request_file
     return request_file
 
 
-def get_raw_eval_results(results_path: str, requests_path: str) -> list[EvalResult]:
+def get_raw_eval_results(results_path: str) -> list[EvalResult]:
     """From the path of the results folder root, extract all needed info for results"""
     model_result_filepaths = []
 
@@ -176,7 +175,7 @@ def get_raw_eval_results(results_path: str, requests_path: str) -> list[EvalResu
     for model_result_filepath in model_result_filepaths:
         # Creation of result
         eval_result = EvalResult.init_from_json_file(model_result_filepath)
-        eval_result.update_with_request_file(requests_path)
+        # eval_result.update_with_request_file(requests_path)
 
         # Store results of same eval together
         eval_name = eval_result.eval_name
@@ -188,7 +187,7 @@ def get_raw_eval_results(results_path: str, requests_path: str) -> list[EvalResu
     results = []
     for v in eval_results.values():
         try:
-            v.to_dict() # we test if the dict version is complete
+            v.to_dict()  # we test if the dict version is complete
             results.append(v)
         except KeyError:  # not all eval values present
             continue

src/populate.py

@@ -8,9 +8,9 @@ from src.display.utils import AutoEvalColumn, EvalQueueColumn
 from src.leaderboard.read_evals import get_raw_eval_results
 
 
-def get_leaderboard_df(results_path: str, requests_path: str, cols: list, benchmark_cols: list) -> pd.DataFrame:
+def get_leaderboard_df(results_path: str, cols: list, benchmark_cols: list) -> pd.DataFrame:
     """Creates a dataframe from all the individual experiment results"""
-    raw_data = get_raw_eval_results(results_path, requests_path)
+    raw_data = get_raw_eval_results(results_path)
     all_data_json = [v.to_dict() for v in raw_data]
 
     df = pd.DataFrame.from_records(all_data_json)