Upgrade TorchMetrics version

anomalib/utils/metrics/__init__.py

@@ -47,15 +47,15 @@ def metric_collection_from_names(metric_names: List[str], prefix: Optional[str])
         AnomalibMetricCollection: Collection of metrics.
     """
     metrics_module = importlib.import_module("anomalib.utils.metrics")
-    metrics = AnomalibMetricCollection([], prefix=prefix, compute_groups=False)
+    metrics = AnomalibMetricCollection([], prefix=prefix)
     for metric_name in metric_names:
         if hasattr(metrics_module, metric_name):
             metric_cls = getattr(metrics_module, metric_name)
-            metrics.add_metrics(metric_cls(compute_on_step=False))
+            metrics.add_metrics(metric_cls())
         elif hasattr(torchmetrics, metric_name):
             try:
                 metric_cls = getattr(torchmetrics, metric_name)
-                metrics.add_metrics(metric_cls(compute_on_step=False))
+                metrics.add_metrics(metric_cls())
             except TypeError:
                 warnings.warn(f"Incorrect constructor arguments for {metric_name} metric from TorchMetrics package.")
         else:

anomalib/utils/metrics/adaptive_threshold.py

@@ -13,7 +13,7 @@ class AdaptiveThreshold(Metric):
     def __init__(self, default_value: float = 0.5, **kwargs):
         super().__init__(**kwargs)
 
-        self.precision_recall_curve = PrecisionRecallCurve(num_classes=1, compute_on_step=False)
+        self.precision_recall_curve = PrecisionRecallCurve(num_classes=1)
         self.add_state("value", default=torch.tensor(default_value), persistent=True)  # pylint: disable=not-callable
         self.value = torch.tensor(default_value)  # pylint: disable=not-callable
 

anomalib/utils/metrics/optimal_f1.py

@@ -13,7 +13,7 @@ class OptimalF1(Metric):
     def __init__(self, num_classes: int, **kwargs):
         super().__init__(**kwargs)
 
-        self.precision_recall_curve = PrecisionRecallCurve(num_classes=num_classes, compute_on_step=False)
+        self.precision_recall_curve = PrecisionRecallCurve(num_classes=num_classes)
 
         self.threshold: torch.Tensor
 

requirements/base.txt

@@ -5,7 +5,7 @@ omegaconf>=2.1.1
 opencv-python>=4.5.3.56
 pandas>=1.1.0
 pytorch-lightning>=1.6.0
-torchmetrics==0.8.0
+torchmetrics==0.9.0
 torchvision>=0.9.1
 torchtext>=0.9.1
 wandb==0.12.17