refactor: add environment configuration and refactor settings management

.env.example

@@ -0,0 +1 @@
+HF_TOKEN=changethis

.vscode/cspell.json

@@ -1,5 +1,7 @@
 {
     "words": [
+        "changethis",
+        "initialisation",
         "modelcard",
         "sentencepiece"
     ]

.vscode/settings.json

@@ -0,0 +1,11 @@
+{
+    "editor.formatOnSave": true,
+    "[python]": {
+        "editor.formatOnSave": true,
+        "editor.defaultFormatter": "charliermarsh.ruff",
+        "editor.codeActionsOnSave": {
+            "source.fixAll.ruff": "always",
+            "source.organizeImports.ruff": "always"
+        }
+    }
+}

app.py

@@ -1,5 +1,4 @@
 import gradio as gr
-import pandas as pd
 from apscheduler.schedulers.background import BackgroundScheduler
 from gradio_leaderboard import ColumnFilter, Leaderboard, SelectColumns
 from huggingface_hub import snapshot_download
@@ -24,49 +23,54 @@ from src.display.utils import (
     WeightType,
     fields,
 )
-from src.envs import API, EVAL_REQUESTS_PATH, EVAL_RESULTS_PATH, QUEUE_REPO, REPO_ID, RESULTS_REPO, TOKEN
+from src.envs import API, settings
 from src.populate import get_evaluation_queue_df, get_leaderboard_df
 from src.submission.submit import add_new_eval
 
 
 def restart_space():
-    API.restart_space(repo_id=REPO_ID)
+    API.restart_space(repo_id=settings.REPO_ID)
 
 
 # Space initialisation
 try:
-    print(EVAL_REQUESTS_PATH)
+    print(settings.EVAL_REQUESTS_PATH)
     snapshot_download(
-        repo_id=QUEUE_REPO,
-        local_dir=EVAL_REQUESTS_PATH,
+        repo_id=settings.QUEUE_REPO,
+        local_dir=settings.EVAL_REQUESTS_PATH,
         repo_type="dataset",
         tqdm_class=None,
         etag_timeout=30,
-        token=TOKEN,
+        token=settings.TOKEN,
     )
 except Exception:
     restart_space()
 try:
-    print(EVAL_RESULTS_PATH)
+    print(settings.EVAL_RESULTS_PATH)
     snapshot_download(
-        repo_id=RESULTS_REPO,
-        local_dir=EVAL_RESULTS_PATH,
+        repo_id=settings.RESULTS_REPO,
+        local_dir=settings.EVAL_RESULTS_PATH,
         repo_type="dataset",
         tqdm_class=None,
         etag_timeout=30,
-        token=TOKEN,
+        token=settings.TOKEN,
     )
 except Exception:
     restart_space()
 
 
-LEADERBOARD_DF = get_leaderboard_df(EVAL_RESULTS_PATH, EVAL_REQUESTS_PATH, COLS, BENCHMARK_COLS)
+LEADERBOARD_DF = get_leaderboard_df(
+    settings.EVAL_RESULTS_PATH,
+    settings.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)
+) = get_evaluation_queue_df(settings.EVAL_REQUESTS_PATH, EVAL_COLS)
 
 
 def init_leaderboard(dataframe):

pyproject.toml

@@ -21,6 +21,8 @@ dependencies = [
   "transformers",
   "tokenizers>=0.15.0",
   "sentencepiece",
+  "python-dotenv>=1.2.1",
+  "pydantic-settings>=2.11.0",
 ]
 
 [dependency-groups]

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
 
 

src/envs.py

@@ -1,25 +1,67 @@
-import os
+from functools import cached_property
+from pathlib import Path
+from typing import Annotated
 
 from huggingface_hub import HfApi
+from pydantic import Field, computed_field
+from pydantic_settings import BaseSettings, SettingsConfigDict
 
+# ----------------------------------
 # Info to change for your repository
 # ----------------------------------
-TOKEN = os.environ.get("HF_TOKEN")  # A read/write token for your org
 
-OWNER = "demo-leaderboard-backend"  # Change to your org - don't forget to create a results and request dataset, with the correct format!
-# ----------------------------------
 
-REPO_ID = f"{OWNER}/leaderboard"
-QUEUE_REPO = f"{OWNER}/requests"
-RESULTS_REPO = f"{OWNER}/results"
+class Settings(BaseSettings):
+    model_config = SettingsConfigDict(env_file=".env")
+
+    TOKEN: Annotated[str, Field(..., alias="HF_TOKEN", description="A read/write token for your org")]
+
+    # Change to your org - don't forget to create a results and request dataset, with the correct format!
+    OWNER: Annotated[
+        str,
+        Field("y-playground-backend"),
+    ]
+
+    @computed_field
+    def REPO_ID(self) -> str:
+        return (Path(self.OWNER) / "leaderboard").as_posix()
+
+    @computed_field
+    def QUEUE_REPO(self) -> str:
+        return (Path(self.OWNER) / "requests").as_posix()
+
+    @computed_field
+    def RESULTS_REPO(self) -> str:
+        return (Path(self.OWNER) / "results").as_posix()
+
+    CACHE_PATH: Annotated[
+        str,
+        Field(".", alias="HF_HOME", description="If you setup a cache later, just change `HF_HOME`"),
+    ]
+
+    # Local caches
+
+    @computed_field
+    def EVAL_REQUESTS_PATH(self) -> str:
+        return (Path(self.CACHE_PATH) / "eval-queue").as_posix()
+
+    @computed_field
+    def EVAL_RESULTS_PATH(self) -> str:
+        return (Path(self.CACHE_PATH) / "eval-results").as_posix()
+
+    @computed_field
+    def EVAL_REQUESTS_PATH_BACKEND(self) -> str:
+        return (Path(self.CACHE_PATH) / "eval-queue-bk").as_posix()
+
+    @computed_field
+    def EVAL_RESULTS_PATH_BACKEND(self) -> str:
+        return (Path(self.CACHE_PATH) / "eval-results-bk").as_posix()
 
-# If you setup a cache later, just change HF_HOME
-CACHE_PATH = os.getenv("HF_HOME", ".")
+    @computed_field
+    @cached_property
+    def API(self) -> HfApi:
+        return HfApi(token=self.TOKEN)
 
-# Local caches
-EVAL_REQUESTS_PATH = os.path.join(CACHE_PATH, "eval-queue")
-EVAL_RESULTS_PATH = os.path.join(CACHE_PATH, "eval-results")
-EVAL_REQUESTS_PATH_BACKEND = os.path.join(CACHE_PATH, "eval-queue-bk")
-EVAL_RESULTS_PATH_BACKEND = os.path.join(CACHE_PATH, "eval-results-bk")
 
-API = HfApi(token=TOKEN)
+settings = Settings()
+API = settings.API

src/leaderboard/read_evals.py

@@ -1,6 +1,5 @@
 import glob
 import json
-import math
 import os
 from dataclasses import dataclass
 

src/submission/check_validity.py

@@ -1,8 +1,6 @@
 import json
 import os
-import re
 from collections import defaultdict
-from datetime import datetime, timedelta, timezone
 
 import huggingface_hub
 from huggingface_hub import ModelCard
@@ -43,12 +41,12 @@ def is_model_on_hub(
         )
         if test_tokenizer:
             try:
-                tk = AutoTokenizer.from_pretrained(
+                _tk = AutoTokenizer.from_pretrained(
                     model_name, revision=revision, trust_remote_code=trust_remote_code, token=token
                 )
             except ValueError as e:
                 return (False, f"uses a tokenizer which is not in a transformers release: {e}", None)
-            except Exception as e:
+            except Exception:
                 return (
                     False,
                     "'s tokenizer cannot be loaded. Is your tokenizer class in a stable transformers release, and correctly configured?",
@@ -63,7 +61,7 @@ def is_model_on_hub(
             None,
         )
 
-    except Exception as e:
+    except Exception:
         return False, "was not found on hub!", None
 
 

src/submission/submit.py

@@ -4,7 +4,7 @@ import sys
 from datetime import datetime, timezone
 
 from src.display.formatting import styled_error, styled_message, styled_warning
-from src.envs import API, EVAL_REQUESTS_PATH, QUEUE_REPO, TOKEN
+from src.envs import API, settings
 from src.submission.check_validity import (
     already_submitted_models,
     check_model_card,
@@ -32,7 +32,7 @@ def add_new_eval(
     global REQUESTED_MODELS
     global USERS_TO_SUBMISSION_DATES
     if not REQUESTED_MODELS:
-        REQUESTED_MODELS, USERS_TO_SUBMISSION_DATES = already_submitted_models(EVAL_REQUESTS_PATH)
+        REQUESTED_MODELS, USERS_TO_SUBMISSION_DATES = already_submitted_models(settings.EVAL_REQUESTS_PATH)
 
     user_name = ""
     model_path = model
@@ -53,13 +53,13 @@ def add_new_eval(
     # Is the model on the hub?
     if weight_type in ["Delta", "Adapter"]:
         base_model_on_hub, error, _ = is_model_on_hub(
-            model_name=base_model, revision=revision, token=TOKEN, test_tokenizer=True
+            model_name=base_model, revision=revision, token=settings.TOKEN, test_tokenizer=True
         )
         if not base_model_on_hub:
             return styled_error(f'Base model "{base_model}" {error}')
 
     if not weight_type == "Adapter":
-        model_on_hub, error, _ = is_model_on_hub(model_name=model, revision=revision, token=TOKEN, test_tokenizer=True)
+        model_on_hub, error, _ = is_model_on_hub(model_name=model, revision=revision, token=settings.TOKEN, test_tokenizer=True)
         if not model_on_hub:
             return styled_error(f'Model "{model}" {error}')
 
@@ -104,7 +104,7 @@ def add_new_eval(
         return styled_warning("This model has been already submitted.")
 
     print("Creating eval file")
-    OUT_DIR = f"{EVAL_REQUESTS_PATH}/{user_name}"
+    OUT_DIR = f"{settings.EVAL_REQUESTS_PATH}/{user_name}"
     os.makedirs(OUT_DIR, exist_ok=True)
     out_path = f"{OUT_DIR}/{model_path}_eval_request_False_{precision}_{weight_type}.json"
 
@@ -115,7 +115,7 @@ def add_new_eval(
     API.upload_file(
         path_or_fileobj=out_path,
         path_in_repo=out_path.split("eval-queue/")[1],
-        repo_id=QUEUE_REPO,
+        repo_id=settings.QUEUE_REPO,
         repo_type="dataset",
         commit_message=f"Add {model} to eval queue",
     )

uv.lock

@@ -681,7 +681,9 @@ dependencies = [
     { name = "matplotlib" },
     { name = "numpy" },
     { name = "pandas" },
+    { name = "pydantic-settings" },
     { name = "python-dateutil" },
+    { name = "python-dotenv" },
     { name = "sentencepiece" },
     { name = "tokenizers" },
     { name = "tqdm" },
@@ -705,7 +707,9 @@ requires-dist = [
     { name = "matplotlib" },
     { name = "numpy" },
     { name = "pandas" },
+    { name = "pydantic-settings", specifier = ">=2.11.0" },
     { name = "python-dateutil" },
+    { name = "python-dotenv", specifier = ">=1.2.1" },
     { name = "sentencepiece" },
     { name = "tokenizers", specifier = ">=0.15.0" },
     { name = "tqdm" },
@@ -1024,6 +1028,20 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/ce/91/2ec36480fdb0b783cd9ef6795753c1dea13882f2e68e73bce76ae8c21e6a/pydantic_core-2.33.2-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:a11c8d26a50bfab49002947d3d237abe4d9e4b5bdc8846a63537b6488e197808", size = 2066678, upload-time = "2025-04-23T18:33:12.224Z" },
 ]
 
+[[package]]
+name = "pydantic-settings"
+version = "2.11.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "pydantic" },
+    { name = "python-dotenv" },
+    { name = "typing-inspection" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/20/c5/dbbc27b814c71676593d1c3f718e6cd7d4f00652cefa24b75f7aa3efb25e/pydantic_settings-2.11.0.tar.gz", hash = "sha256:d0e87a1c7d33593beb7194adb8470fc426e95ba02af83a0f23474a04c9a08180", size = 188394, upload-time = "2025-09-24T14:19:11.764Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/83/d6/887a1ff844e64aa823fb4905978d882a633cfe295c32eacad582b78a7d8b/pydantic_settings-2.11.0-py3-none-any.whl", hash = "sha256:fe2cea3413b9530d10f3a5875adffb17ada5c1e1bab0b2885546d7310415207c", size = 48608, upload-time = "2025-09-24T14:19:10.015Z" },
+]
+
 [[package]]
 name = "pydub"
 version = "0.25.1"
@@ -1063,6 +1081,15 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/ec/57/56b9bcc3c9c6a792fcbaf139543cee77261f3651ca9da0c93f5c1221264b/python_dateutil-2.9.0.post0-py2.py3-none-any.whl", hash = "sha256:a8b2bc7bffae282281c8140a97d3aa9c14da0b136dfe83f850eea9a5f7470427", size = 229892, upload-time = "2024-03-01T18:36:18.57Z" },
 ]
 
+[[package]]
+name = "python-dotenv"
+version = "1.2.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/f0/26/19cadc79a718c5edbec86fd4919a6b6d3f681039a2f6d66d14be94e75fb9/python_dotenv-1.2.1.tar.gz", hash = "sha256:42667e897e16ab0d66954af0e60a9caa94f0fd4ecf3aaf6d2d260eec1aa36ad6", size = 44221, upload-time = "2025-10-26T15:12:10.434Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/14/1b/a298b06749107c305e1fe0f814c6c74aea7b2f1e10989cb30f544a1b3253/python_dotenv-1.2.1-py3-none-any.whl", hash = "sha256:b81ee9561e9ca4004139c6cbba3a238c32b03e4894671e181b671e8cb8425d61", size = 21230, upload-time = "2025-10-26T15:12:09.109Z" },
+]
+
 [[package]]
 name = "python-multipart"
 version = "0.0.20"