Add tqdm progress bar support and disable option for evaluation scripts

evaluate_horizon.py

@@ -19,6 +19,11 @@ import pandas as pd
 import torch
 from torch.utils.data import DataLoader
 
+try:
+    from tqdm import tqdm
+except Exception:  # pragma: no cover
+    tqdm = None
+
 from utils import (
     EvalRecordDataset,
     build_dataset_from_config,
@@ -72,6 +77,11 @@ def parse_args() -> argparse.Namespace:
         nargs="+",
         default=[1, 5, 10, 20, 50],
     )
+    p.add_argument(
+        "--no_tqdm",
+        action="store_true",
+        help="Disable tqdm progress bars",
+    )
     return p.parse_args()
 
 
@@ -102,6 +112,8 @@ def main() -> None:
     args = parse_args()
     seed_everything(args.seed)
 
+    show_progress = (not args.no_tqdm)
+
     run_dir = args.run_dir
     cfg = load_train_config(run_dir)
 
@@ -117,6 +129,7 @@ def main() -> None:
         subset=test_subset,
         age_bins_years=age_bins_years,
         seed=args.seed,
+        show_progress=show_progress,
     )
 
     device = torch.device(args.device)
@@ -139,8 +152,16 @@ def main() -> None:
     print("DISCLAIMER: AUC here is horizon-dependent label AUC (no IPCW / censoring adjustment).")
     print("DISCLAIMER: Brier is unadjusted diagnostic/proxy (no censoring adjustment).")
 
-    scores = predict_cifs(model, head, criterion, loader,
+    scores = predict_cifs(
-                          horizons, device=device)
+        model,
+        head,
+        criterion,
+        loader,
+        horizons,
+        device=device,
+        show_progress=show_progress,
+        progress_desc="Inference (horizons)",
+    )
     # scores shape: (N, K, H)
     if scores.ndim != 3:
         raise ValueError(
@@ -157,7 +178,12 @@ def main() -> None:
     per_cause_rows: List[Dict[str, object]] = []
     workload_rows: List[Dict[str, object]] = []
 
-    for h_idx, tau in enumerate(horizons):
+    horizon_iter = enumerate(horizons)
+    if show_progress and tqdm is not None:
+        horizon_iter = tqdm(horizon_iter, total=len(
+            horizons), desc="Metrics by horizon")
+
+    for h_idx, tau in horizon_iter:
         s_tau = scores[:, :, h_idx]
         y_tau = build_labels_within_tau_flat(
             N, K, evt_rec_idx, evt_cause, evt_dt, tau)

evaluate_next_event.py

@@ -7,6 +7,11 @@ import pandas as pd
 import torch
 from torch.utils.data import DataLoader
 
+try:
+    from tqdm import tqdm  # noqa: F401
+except Exception:  # pragma: no cover
+    tqdm = None
+
 from utils import (
     EvalRecordDataset,
     build_dataset_from_config,
@@ -53,6 +58,11 @@ def parse_args() -> argparse.Namespace:
         default=20,
         help="Minimum positives for per-cause AUC",
     )
+    p.add_argument(
+        "--no_tqdm",
+        action="store_true",
+        help="Disable tqdm progress bars",
+    )
     return p.parse_args()
 
 
@@ -60,6 +70,8 @@ def main() -> None:
     args = parse_args()
     seed_everything(args.seed)
 
+    show_progress = (not args.no_tqdm)
+
     run_dir = args.run_dir
     cfg = load_train_config(run_dir)
 
@@ -72,6 +84,7 @@ def main() -> None:
         subset=test_subset,
         age_bins_years=age_bins_years,
         seed=args.seed,
+        show_progress=show_progress,
     )
 
     device = torch.device(args.device)
@@ -91,7 +104,16 @@ def main() -> None:
     )
 
     tau = float(args.tau_short)
-    scores = predict_cifs(model, head, criterion, loader, [tau], device=device)
+    scores = predict_cifs(
+        model,
+        head,
+        criterion,
+        loader,
+        [tau],
+        device=device,
+        show_progress=show_progress,
+        progress_desc="Inference (next-event)",
+    )
     # scores shape: (N,K,1) for multi-taus; squeeze last
     if scores.ndim == 3:
         scores = scores[:, :, 0]
@@ -167,7 +189,7 @@ def main() -> None:
 
     aucs = auc[np.isfinite(auc)]
 
-    if aucs:
+    if aucs.size > 0:
         metrics_rows.append(
             {"metric": "macro_ovr_auc", "value": float(np.mean(aucs))})
     else:

utils.py

@@ -10,6 +10,11 @@ import numpy as np
 import torch
 from torch.utils.data import DataLoader, Dataset, Subset, random_split
 
+try:
+    from tqdm import tqdm as _tqdm
+except Exception:  # pragma: no cover
+    _tqdm = None
+
 from dataset import HealthDataset
 from losses import (
     DiscreteTimeCIFNLLLoss,
@@ -23,6 +28,12 @@ DAYS_PER_YEAR = 365.25
 N_TECH_TOKENS = 2  # pad=0, DOA=1, diseases start at 2
 
 
+def _progress(iterable, *, enabled: bool, desc: str, total: Optional[int] = None):
+    if enabled and _tqdm is not None:
+        return _tqdm(iterable, desc=desc, total=total)
+    return iterable
+
+
 def make_inference_dataloader_kwargs(
     device: torch.device,
     num_workers: int,
@@ -278,6 +289,7 @@ def build_event_driven_records(
     subset: Subset,
     age_bins_years: Sequence[float],
     seed: int,
+    show_progress: bool = False,
 ) -> List[EvalRecord]:
     if len(age_bins_years) < 2:
         raise ValueError("age_bins must have at least 2 boundaries")
@@ -296,7 +308,12 @@ def build_event_driven_records(
     # Speed: avoid calling dataset.__getitem__ for every patient here.
     # We only need DOA + event times/codes to create evaluation records.
     eps = 1e-6
-    for patient_idx in indices:
+    for patient_idx in _progress(
+        indices,
+        enabled=show_progress,
+        desc="Building eval records",
+        total=len(indices),
+    ):
         patient_id = dataset.patient_ids[patient_idx]
 
         doa_days = float(dataset._doa[patient_idx])
@@ -445,6 +462,8 @@ def predict_cifs(
     loader: DataLoader,
     taus_years: Sequence[float],
     device: torch.device,
+    show_progress: bool = False,
+    progress_desc: str = "Inference",
 ) -> np.ndarray:
     model.eval()
     head.eval()
@@ -453,7 +472,12 @@ def predict_cifs(
 
     all_out: List[np.ndarray] = []
     with torch.no_grad():
-        for batch in loader:
+        for batch in _progress(
+            loader,
+            enabled=show_progress,
+            desc=progress_desc,
+            total=len(loader) if hasattr(loader, "__len__") else None,
+        ):
             event_seq, time_seq, cont, cate, sex, baseline_pos = batch
             event_seq = event_seq.to(device, non_blocking=True)
             time_seq = time_seq.to(device, non_blocking=True)