Add temporal training/prediction pipeline, chronological splits, feature selection, and reporting enhancements

scripts/aggregate_report.py

@@ -17,6 +17,16 @@
 HIGHER_IS_BETTER = {"R2_test", "R2_val"}
 
 
+def _status_ready(path: Path) -> bool:
+    if not path.exists():
+        return False
+    try:
+        status = json.loads(path.read_text(encoding="utf-8"))
+    except Exception:
+        return False
+    return bool(status.get("end_time"))
+
+
 def _fig_to_base64(fig) -> str:
     buf = io.BytesIO()
     fig.savefig(buf, format="png", bbox_inches="tight")
@@ -205,9 +215,11 @@ def main() -> None:
     print("[aggregator] waiting for scenario statuses", flush=True)
     deadline = time.time() + 60 * 60
     while time.time() < deadline:
-        if all(p.exists() for p in status_paths):
+        if all(_status_ready(p) for p in status_paths):
             break
         time.sleep(5)
+    else:
+        print("[aggregator] timeout waiting for fresh scenario statuses; proceeding with available files", flush=True)
 
     out_dir = Path("results/final")
     out_dir.mkdir(parents=True, exist_ok=True)

scripts/run_scenario.py

@@ -67,6 +67,7 @@ def main() -> None:
         "split_metadata_path": str(sdir / "model" / "split_metadata.json"),
     }
 
+    (sdir / "status.json").write_text(json.dumps(status, indent=2), encoding="utf-8")
     try:
         print(f"[{scenario}] scenario started", flush=True)
         csv = _resolve_dataset_path(scenario, args.dataset)

tests/test_aggregate_report_features.py

@@ -24,6 +24,7 @@ def test_aggregate_report_handles_feature_and_new_metrics(tmp_path, monkeypatch)
                 "preds_path": str(sdir / "preds.csv"),
                 "dataset_path": "",
                 "split_metadata": {"train_start_index": 0, "train_end_index": 5, "val_start_index": 6, "val_end_index": 7, "test_start_index": 8, "test_end_index": 9, "train_rows": 6, "val_rows": 2, "test_rows": 2, "train_pct": 0.6, "val_pct": 0.2, "test_pct": 0.2},
+                "end_time": "2024-01-01T00:00:00Z",
             }
         ),
         encoding="utf-8",
@@ -38,6 +39,7 @@ def test_aggregate_report_handles_feature_and_new_metrics(tmp_path, monkeypatch)
                 "features": ["dl_cqi", "ul_sinr"],
                 "epochs": 1,
                 "split_metadata": {"train_start_index": 0, "train_end_index": 5, "val_start_index": 6, "val_end_index": 7, "test_start_index": 8, "test_end_index": 9, "train_rows": 6, "val_rows": 2, "test_rows": 2, "train_pct": 0.6, "val_pct": 0.2, "test_pct": 0.2},
+                "end_time": "2024-01-01T00:00:00Z",
             }
         ),
         encoding="utf-8",

tests/test_aggregate_waiting.py

@@ -0,0 +1,14 @@
+from __future__ import annotations
+
+import json
+
+from scripts.aggregate_report import _status_ready
+
+
+def test_status_ready_requires_end_time(tmp_path) -> None:
+    p = tmp_path / "status.json"
+    p.write_text(json.dumps({"success": True}), encoding="utf-8")
+    assert _status_ready(p) is False
+
+    p.write_text(json.dumps({"success": True, "end_time": "2024-01-01T00:00:00Z"}), encoding="utf-8")
+    assert _status_ready(p) is True