gluonts.ev.metrics module#

class gluonts.ev.metrics.AverageMeanScaledQuantileLoss(quantile_levels: 'Collection[float]')[source]#

Bases: gluonts.ev.metrics.BaseMetricDefinition

static mean(**quantile_losses: numpy.ndarray) → numpy.ndarray[source]#

quantile_levels: Collection[float]#

class gluonts.ev.metrics.BaseMetricDefinition[source]#

Bases: object

add(*others)[source]#

class gluonts.ev.metrics.Coverage(q: 'float')[source]#

Bases: gluonts.ev.metrics.BaseMetricDefinition

q: float#

class gluonts.ev.metrics.DerivedMetric(name: str, metrics: Dict[str, gluonts.ev.metrics.Metric], post_process: Callable)[source]#

Bases: gluonts.ev.metrics.Metric

A Metric that is computed using other metrics.

A derived metric updates multiple, simpler metrics independently and in the end combines their results as defined in post_process.

get() → numpy.ndarray[source]#

metrics: Dict[str, gluonts.ev.metrics.Metric]#

post_process: Callable#

update(data: Mapping[str, numpy.ndarray]) → typing_extensions.Self[source]#: Update metric using a single data instance.

class gluonts.ev.metrics.DirectMetric(name: str, stat: Callable, aggregate: gluonts.ev.aggregations.Aggregation)[source]#

Bases: gluonts.ev.metrics.Metric

A Metric which uses a single function and aggregation strategy.

aggregate: gluonts.ev.aggregations.Aggregation#

get() → numpy.ndarray[source]#

stat: Callable#

update(data: Mapping[str, numpy.ndarray]) → typing_extensions.Self[source]#: Update metric using a single data instance.

class gluonts.ev.metrics.MAE(forecast_type: str = '0.5')[source]#

Bases: gluonts.ev.metrics.BaseMetricDefinition

Mean Absolute Error.

forecast_type: str = '0.5'#

class gluonts.ev.metrics.MAECoverage(quantile_levels: 'Collection[float]')[source]#

Bases: gluonts.ev.metrics.BaseMetricDefinition

static mean(quantile_levels: Collection[float], **coverages: numpy.ndarray) → numpy.ndarray[source]#

quantile_levels: Collection[float]#

class gluonts.ev.metrics.MAPE(forecast_type: str = '0.5')[source]#

Bases: gluonts.ev.metrics.BaseMetricDefinition

Mean Absolute Percentage Error.

forecast_type: str = '0.5'#

class gluonts.ev.metrics.MASE(forecast_type: str = '0.5')[source]#

Bases: gluonts.ev.metrics.BaseMetricDefinition

Mean Absolute Scaled Error.

forecast_type: str = '0.5'#

class gluonts.ev.metrics.MSE(forecast_type: str = 'mean')[source]#

Bases: gluonts.ev.metrics.BaseMetricDefinition

Mean Squared Error.

forecast_type: str = 'mean'#

class gluonts.ev.metrics.MSIS(alpha: float = 0.05)[source]#

Bases: gluonts.ev.metrics.BaseMetricDefinition

Mean Scaled Interval Score.

alpha: float = 0.05#

class gluonts.ev.metrics.MeanAbsoluteLabel[source]#: Bases: gluonts.ev.metrics.BaseMetricDefinition

class gluonts.ev.metrics.MeanScaledQuantileLoss(q: 'float')[source]#

Bases: gluonts.ev.metrics.BaseMetricDefinition

q: float#

class gluonts.ev.metrics.MeanSumQuantileLoss(quantile_levels: 'Collection[float]')[source]#

Bases: gluonts.ev.metrics.BaseMetricDefinition

static mean(**quantile_losses: numpy.ndarray) → numpy.ndarray[source]#

quantile_levels: Collection[float]#

class gluonts.ev.metrics.MeanWeightedSumQuantileLoss(quantile_levels: 'Collection[float]')[source]#

Bases: gluonts.ev.metrics.BaseMetricDefinition

static mean(**quantile_losses: numpy.ndarray) → numpy.ndarray[source]#

quantile_levels: Collection[float]#

class gluonts.ev.metrics.Metric(name: 'str')[source]#

Bases: object

get() → numpy.ndarray[source]#

name: str#

update(data: Mapping[str, numpy.ndarray]) → typing_extensions.Self[source]#: Update metric using a single data instance.

update_all(stream: Iterator[Mapping[str, numpy.ndarray]]) → typing_extensions.Self[source]#: Update metric using a stream of data instances.

class gluonts.ev.metrics.MetricCollection(metrics: 'List[Metric]')[source]#

Bases: object

get() → Dict[str, numpy.ndarray][source]#

metrics: List[gluonts.ev.metrics.Metric]#

update(data: Mapping[str, numpy.ndarray]) → typing_extensions.Self[source]#: Update metrics using a single data instance.

update_all(stream: Iterator[Mapping[str, numpy.ndarray]]) → typing_extensions.Self[source]#: Update metrics using a stream of data instances.

class gluonts.ev.metrics.MetricDefinition(*args, **kwargs)[source]#: Bases: typing_extensions.Protocol

class gluonts.ev.metrics.MetricDefinitionCollection(metrics: 'List[BaseMetricDefinition]')[source]#

Bases: gluonts.ev.metrics.BaseMetricDefinition

metrics: List[gluonts.ev.metrics.BaseMetricDefinition]#

class gluonts.ev.metrics.ND(forecast_type: str = '0.5')[source]#

Bases: gluonts.ev.metrics.BaseMetricDefinition

Normalized Deviation.

forecast_type: str = '0.5'#

static normalized_deviation(sum_absolute_error: numpy.ndarray, sum_absolute_label: numpy.ndarray) → numpy.ndarray[source]#

class gluonts.ev.metrics.NRMSE(forecast_type: str = 'mean')[source]#

Bases: gluonts.ev.metrics.BaseMetricDefinition

RMSE, normalized by the mean absolute label.

forecast_type: str = 'mean'#

static normalize_root_mean_squared_error(root_mean_squared_error: numpy.ndarray, mean_absolute_label: numpy.ndarray) → numpy.ndarray[source]#

class gluonts.ev.metrics.OWA(forecast_type: str = '0.5')[source]#

Bases: gluonts.ev.metrics.BaseMetricDefinition

Overall Weighted Average.

static calculate_OWA(smape: numpy.ndarray, smape_naive2: numpy.ndarray, mase: numpy.ndarray, mase_naive2: numpy.ndarray) → numpy.ndarray[source]#

forecast_type: str = '0.5'#

class gluonts.ev.metrics.RMSE(forecast_type: str = 'mean')[source]#

Bases: gluonts.ev.metrics.BaseMetricDefinition

Root Mean Squared Error.

forecast_type: str = 'mean'#

static root_mean_squared_error(mean_squared_error: numpy.ndarray) → numpy.ndarray[source]#

class gluonts.ev.metrics.SMAPE(forecast_type: str = '0.5')[source]#

Bases: gluonts.ev.metrics.BaseMetricDefinition

Symmetric Mean Absolute Percentage Error.

forecast_type: str = '0.5'#

class gluonts.ev.metrics.SumAbsoluteError(forecast_type: 'str' = '0.5')[source]#

Bases: gluonts.ev.metrics.BaseMetricDefinition

forecast_type: str = '0.5'#

class gluonts.ev.metrics.SumAbsoluteLabel[source]#: Bases: gluonts.ev.metrics.BaseMetricDefinition

class gluonts.ev.metrics.SumError(forecast_type: 'str' = '0.5')[source]#

Bases: gluonts.ev.metrics.BaseMetricDefinition

forecast_type: str = '0.5'#

class gluonts.ev.metrics.SumNumMaskedTargetValues[source]#: Bases: gluonts.ev.metrics.BaseMetricDefinition

class gluonts.ev.metrics.SumQuantileLoss(q: 'float')[source]#

Bases: gluonts.ev.metrics.BaseMetricDefinition

q: float#

class gluonts.ev.metrics.WeightedSumQuantileLoss(q: 'float')[source]#

Bases: gluonts.ev.metrics.BaseMetricDefinition

q: float#

static weight_sum_quantile_loss(sum_quantile_loss: numpy.ndarray, sum_absolute_label: numpy.ndarray) → numpy.ndarray[source]#