tracel-ai/burn · error

Outputs and targets must have the same dimensions. Got {:?}

Error message

Outputs and targets must have the same dimensions. Got {:?} and {:?}

What it means

Validation guard in `DiceMetric::update`: the prediction tensor and the target tensor passed to the metric have different shapes, so the per-pixel intersection/union needed for the Dice coefficient cannot be computed element-wise. Typically caused by a mismatch between model output resolution/class layout and label shape.

Source

Thrown at crates/burn-train/src/metric/vision/dice.rs:247

        Self {
            name,
            config,
            ..Default::default()
        }
    }
}

impl<const D: usize> Metric for DiceMetric<D> {
    type Input = DiceInput<D>;

    fn name(&self) -> MetricName {
        self.name.clone()
    }

    fn update(&mut self, item: &Self::Input, _metadata: &MetricMetadata) -> SerializedEntry {
        // Dice coefficient: 2 * (|X ∩ Y|) / (|X| + |Y|)
        if item.outputs.dims() != item.targets.dims() {
            panic!(
                "Outputs and targets must have the same dimensions. Got {:?} and {:?}",
                item.outputs.dims(),
                item.targets.dims()
            );
        }

        let dims = item.outputs.dims();
        let batch_size = dims[0];
        let n_classes = dims[1];

        let mut outputs = item.outputs.clone();
        let mut targets = item.targets.clone();

        if !self.config.include_background && n_classes > 1 {
            // If not including background, we can ignore the first class
            outputs = outputs.slice(s![.., 1..]);
            targets = targets.slice(s![.., 1..]);
        } else if self.config.include_background && n_classes < 2 {

View on GitHub (pinned to d16f7ba2ed)

Solutions

1. Check that model outputs and targets share the same spatial dimensions (resize/crop targets to match)
2. Verify one-hot encoding matches the number of predicted classes
3. Inspect `outputs.dims()` and `targets.dims()` printed in the message to locate the mismatched axis

When it happens

Trigger: Thrown at crates/burn-train/src/metric/vision/dice.rs:247 when the library encounters an invalid state.

Common situations: See trigger scenarios.

AI-assisted analysis of tracel-ai/burn@d16f7ba2ed (2026-09-05). Data as JSON: /api/errors/79608c8bc79c2ec1. Report an issue: GitHub.