DiceScore(average="weighted") produces scores bigger than 1.0

[nkaenzig]

🐛 Bug

Using average=weighted aggregation in the DiceScore produces scores bigger than 1.0, and therefore outside the expected value range of [0,1] for dice.

To Reproduce

The following code sample returns a score of 2.0:

Code sample

import torch
from torchmetrics.segmentation import DiceScore

N_SAMPLES = 4
N_CLASSES = 3

target = torch.full((N_SAMPLES, N_CLASSES, 128, 128), 0, dtype=torch.int8)
preds = torch.full((N_SAMPLES, N_CLASSES, 128, 128), 0, dtype=torch.int8)

target[0, 0], preds[0, 0] = 1, 1

dice = DiceScore(num_classes=3, average='weighted', include_background=False)

print(dice(preds, target))

Expected behavior

Environment

• TorchMetrics version (if build from source, add commit SHA): 1.6.0
• Python & PyTorch Version (e.g., 1.0): Python 3.11.10 & torch==2.3.0
• Any other relevant information such as OS (e.g., Linux): macOS 15.1.1 (24B91)