Update src/main.py

Added commitScheduler

src/main.py

@@ -8,8 +8,9 @@ from typing import Any, Dict, List, Tuple
 import streamlit as st
 import pandas as pd
 from PIL import Image
-from huggingface_hub import HfApi, hf_hub_download
+from huggingface_hub import HfApi, hf_hub_download, CommitScheduler
 from huggingface_hub.utils import HfHubHTTPError
+from pathlib import Path
 
 
 # =========================
@@ -43,6 +44,24 @@ CHALLENGE_TYPES = ["Object Detection", "Instance Segmentation"]
 LEADERBOARD_METRICS = ["bbox_mAP", "bbox_AP50", "segm_mAP", "segm_AP50"]
 DEFAULT_SORT_METRIC = "segm_AP50"
 
+# =========================
+# COMMIT SCHEDULER SETUP
+# =========================
+
+# Local folder for scheduler to watch
+SCHEDULER_FOLDER = Path("submissions_queue")
+SCHEDULER_FOLDER.mkdir(exist_ok=True)
+
+# Initialize CommitScheduler
+scheduler = CommitScheduler(
+    repo_id=DB_REPO_ID,
+    repo_type=DB_REPO_TYPE,
+    folder_path=SCHEDULER_FOLDER,
+    token=SUBMISSIONS_TOKEN,
+    path_in_repo="submissions",  # Where files go in the repo
+    every=5,  # Commit every 5 minutes
+)
+
 
 # =========================
 # HELPERS
@@ -101,25 +120,14 @@ def _validate_submission_json(obj: Any) -> Tuple[bool, str]:
     return True, "OK"
 
 
-def _upload_json(api: HfApi, data: Dict[str, Any] | List[Any], path_in_repo: str) -> None:
-    with tempfile.NamedTemporaryFile(mode="w", suffix=".json", delete=False) as tmp:
-        json.dump(data, tmp, ensure_ascii=False)
-        tmp_path = tmp.name
-
-    try:
-        api.upload_file(
-            path_or_fileobj=tmp_path,
-            path_in_repo=path_in_repo,
-            repo_id=DB_REPO_ID,
-            repo_type=DB_REPO_TYPE,
-            token=SUBMISSIONS_TOKEN,
-            commit_message=f"Add {path_in_repo}",
-        )
-    finally:
-        try:
-            os.remove(tmp_path)
-        except OSError:
-            pass
+def _save_json_to_scheduler(data: Dict[str, Any] | List[Any], filename: str) -> None:
+    """
+    Save JSON to the scheduler folder.
+    The scheduler will automatically commit it to the repo.
+    """
+    file_path = SCHEDULER_FOLDER / filename
+    with open(file_path, "w") as f:
+        json.dump(data, f, ensure_ascii=False, indent=2)
 
 
 def _create_submission_record(
@@ -132,11 +140,10 @@ def _create_submission_record(
     original_filename: str,
 ) -> str:
     """
-    Writes pred/meta/status to the private DB dataset repo.
+    Writes pred/meta/status to the scheduler folder.
     Returns submission_id.
     """
     _require_token()
-    api = HfApi()
 
     submission_id = str(uuid.uuid4())
     ts = int(time.time())
@@ -153,10 +160,10 @@ def _create_submission_record(
 
     status = {"state": "queued", "timestamp": ts}
 
-    base = f"submissions/{submission_id}"
-    _upload_json(api, pred, f"{base}/pred.json")
-    _upload_json(api, meta, f"{base}/meta.json")
-    _upload_json(api, status, f"{base}/status.json")
+    # Save to scheduler folder (will be auto-committed)
+    _save_json_to_scheduler(pred, f"{submission_id}_pred.json")
+    _save_json_to_scheduler(meta, f"{submission_id}_meta.json")
+    _save_json_to_scheduler(status, f"{submission_id}_status.json")
 
     return submission_id