NLL Poisson Scorer¶
Poisson distribution negative log-likelihood Scorer.
The anomaly score is the negative log likelihood of the actual series values under a Poisson distribution estimated from the stochastic prediction.
- class darts.ad.scorers.nll_poisson_scorer.PoissonNLLScorer(window=1)[source]¶
Bases:
darts.ad.scorers.scorers.NLLScorer
Attributes
Whether the scorer expects a probabilistic prediction for its first input.
Methods
eval_accuracy_from_prediction
(...[, metric])Computes the anomaly score between actual_series and pred_series, and returns the score of an agnostic threshold metric.
score_from_prediction
(actual_series, pred_series)Computes the anomaly score on the two (sequence of) series.
show_anomalies_from_prediction
(...[, ...])Plot the results of the scorer.
- eval_accuracy_from_prediction(actual_anomalies, actual_series, pred_series, metric='AUC_ROC')¶
Computes the anomaly score between actual_series and pred_series, and returns the score of an agnostic threshold metric.
- Parameters
actual_anomalies (
Union
[TimeSeries
,Sequence
[TimeSeries
]]) – The (sequence of) ground truth of the anomalies (1 if it is an anomaly and 0 if not)actual_series (
Union
[TimeSeries
,Sequence
[TimeSeries
]]) – The (sequence of) actual series.pred_series (
Union
[TimeSeries
,Sequence
[TimeSeries
]]) – The (sequence of) predicted series.metric (
str
) – Optionally, metric function to use. Must be one of “AUC_ROC” and “AUC_PR”. Default: “AUC_ROC”
- Returns
- Score of an agnostic threshold metric for the computed anomaly score
float
if actual_series and actual_series are univariate series (dimension=1).Sequence[float]
if actual_series and actual_series are multivariate series (dimension>1),
returns one value per dimension, or * if actual_series and actual_series are sequences of univariate series, returns one value per series
Sequence[Sequence[float]]]
if actual_series and actual_series are sequences
of multivariate series. Outer Sequence is over the sequence input and the inner Sequence is over the dimensions of each element in the sequence input.
- Return type
Union[float, Sequence[float], Sequence[Sequence[float]]]
- property is_probabilistic: bool¶
Whether the scorer expects a probabilistic prediction for its first input.
- Return type
bool
- score_from_prediction(actual_series, pred_series)¶
Computes the anomaly score on the two (sequence of) series.
If a pair of sequences is given, they must contain the same number of series. The scorer will score each pair of series independently and return an anomaly score for each pair.
- Parameters
actual_series (
Union
[TimeSeries
,Sequence
[TimeSeries
]]) – The (sequence of) actual series.pred_series (
Union
[TimeSeries
,Sequence
[TimeSeries
]]) – The (sequence of) predicted series.
- Returns
(Sequence of) anomaly score time series
- Return type
Union[TimeSeries, Sequence[TimeSeries]]
- show_anomalies_from_prediction(actual_series, pred_series, scorer_name=None, actual_anomalies=None, title=None, metric=None)¶
Plot the results of the scorer.
Computes the anomaly score on the two series. And plots the results.
- The plot will be composed of the following:
the actual_series and the pred_series.
the anomaly score of the scorer.
the actual anomalies, if given.
- It is possible to:
add a title to the figure with the parameter title
give personalized name to the scorer with scorer_name
show the results of a metric for the anomaly score (AUC_ROC or AUC_PR), if the actual anomalies is provided.
- Parameters
actual_series (
TimeSeries
) – The actual series to visualize anomalies from.pred_series (
TimeSeries
) – The predicted series of actual_series.actual_anomalies (
Optional
[TimeSeries
]) – The ground truth of the anomalies (1 if it is an anomaly and 0 if not)scorer_name (
Optional
[str
]) – Name of the scorer.title (
Optional
[str
]) – Title of the figuremetric (
Optional
[str
]) – Optionally, Scoring function to use. Must be one of “AUC_ROC” and “AUC_PR”. Default: “AUC_ROC”