Skip to content

Metrics

CentroidErrorPerForecastDay

Bases: BaseErrorMetricDaily

Euclidean distance (in pixels) between the predicted and target centroids.

The centroid of a (batch, time) frame is its value-weighted center of mass over the spatial dimensions, summed across channels. Frames whose target has (near-)zero total mass have an undefined centroid and are excluded from the average.

MAEPerForecastDay

Bases: BaseErrorMetricDaily

Mean Absolute Error per forecast lead time.

RMSEPerForecastDay

Bases: BaseErrorMetricDaily

Root Mean Squared Error per forecast lead time.

IceNetAccuracy

Bases: Metric

Binary accuracy metric for use at multiple leadtimes.

update(preds, target, sample_weight=None)

Update metric state with a new batch of predictions and targets.

compute()

Compute the final accuracy metric as a percentage at each leadtime.

SIEError

Bases: Metric

Sea Ice Extent error metric (in km^2) for use at multiple lead times.

leadtimes_to_evaluate = leadtimes_to_evaluate instance-attribute

pixel_size = pixel_size instance-attribute

update(preds, target, _sample_weight=None)

Update the SIE accumulators.

Parameters

preds : torch.Tensor Model predictions. target : torch.Tensor Ground truth values. sample_weight : Optional[torch.Tensor] Ignored (present for API compatibility).

compute()

Compute the final Sea Ice Extent error in km².

SeaIceExtentErrorPerForecastDay

Bases: Metric

Sea Ice Extent error (SIEError) metric (in km^2) for use at multiple lead times.

The SIE error is calculated as the absolute difference between the predicted and true sea ice extent for each forecast day. Sea ice presence is defined by having a probability greater than the threshold value.

sum_errors instance-attribute

sample_count instance-attribute

pixel_size = pixel_size instance-attribute

update(preds, target, _sample_weight=None)

Update the SIE accumulators.

Parameters

preds : torch.Tensor Model predictions of shape (B, T, H, W). target : torch.Tensor Ground truth values of shape (B, T, H, W). _sample_weight : Optional[torch.Tensor] Ignored (present for API compatibility).

compute()

Compute the final Sea Ice Extent error in km².