recallTimeSeriesWrapper
Python template · sinapsis_darts_forecasting.templates.time_series_metrics
Overview
Dynamic Template for Darts metrics computation.
This template wraps any metric from the `darts.metrics` module and makes it available as a Sinapsis template. The wrapped metric is applied to the predictions and ground truth of each time series packet in the input container, and the result is stored in the `generic_data` field of the time series packet.
IMPORTANT NOTE: Validation mode should be enabled during model fitting and prediction (e.g., in the `TimeSeriesModel` template), to ensure that the metric is computed on the correct portion of the time series (i.e., the forecast horizon). If validation mode is not enabled, forecasted values and ground truth values may not be properly aligned, which can lead to incorrect metric calculations.
recall — inherited documentation
Recall Score [1]_.
For the true series :math:`y` and predicted series :math:`\hat{y}` of length :math:`T`, it is computed per component/column as:
.. math:: \frac{TP}{TP + FN}
Where :math:`TP` are the true positives, and :math:`FN` are the false negatives.
If :math:`\hat{y}_t` are stochastic (contains several samples), it takes the label with the highest count from each time step and component. If :math:`\hat{y}_t` represent the predict class label probabilities, it takes the label with the highest probability from each time step and component.
Parameters ---------- actual_series The (sequence of) actual series. pred_series The (sequence of) predicted series. intersect For time series that are overlapping in time without having the same time index, setting `True` will consider the values only over their common time interval (intersection in time). labels Optionally, the labels and their order to compute the metric on. If `None`, will use all unique values from the actual and predicted series. component_reduction Optionally, a function to aggregate the metrics over the component/column axis. It must reduce a `np.ndarray` of shape `(t, c)` to a `np.ndarray` of shape `(t,)`. The function takes as input a ``np.ndarray`` and a parameter named `axis`, and returns the reduced array. The `axis` receives value `1` corresponding to the component axis. If `None`, will return a metric per component. series_reduction Optionally, a function to aggregate the metrics over multiple series. It must reduce a `np.ndarray` of shape `(s, t, c)` to a `np.ndarray` of shape `(t, c)` The function takes as input a ``np.ndarray`` and a parameter named `axis`, and returns the reduced array. The `axis` receives value `0` corresponding to the series axis. For example with `np.nanmean`, will return the average over all series metrics. If `None`, will return a metric per component. label_reduction The method to reduce the label-specific metrics. Can be one of:
- ``None``: no reduction, returns a metric per label. - ``'micro'``: computes the metric globally by counting the total true positives, false negatives and false positives. - ``'macro'``: computes the metric for each label, and returns the unweighted mean. - ``'weighted'``: computes the metric for each label, and returns the weighted mean by support (the number of true instances for each label). n_jobs The number of jobs to run in parallel. Parallel jobs are created only when a ``Sequence[TimeSeries]`` is passed as input, parallelising operations regarding different ``TimeSeries``. Defaults to `1` (sequential). Setting the parameter to `-1` means using all the available processors. verbose Optionally, whether to print operations progress.
Returns ------- float A single metric score (when `label_reduction != None`) for:
- a single univariate series. - a single multivariate series with `component_reduction`. - a sequence (list) of uni/multivariate series with `series_reduction` and `component_reduction`. np.ndarray A numpy array of metric scores. The array has shape (n components * n labels,) without component reduction, and shape (n labels,) with component reduction. `n labels` is the number of labels when `label_reduction=None`, and `1` otherwise. For:
- the same input arguments that result in the `float` return case from above but with `label_reduction=None`. - a single multivariate series and at least `component_reduction=None`. - a sequence of uni/multivariate series including `series_reduction` and `component_reduction=None`. list[float] Same as for type `float` but for a sequence of series. list[np.ndarray] Same as for type `np.ndarray` but for a sequence of series.
References ---------- .. [1] https://en.wikipedia.org/wiki/Precision_and_recall
Capabilities
Capabilities: not setInstallation
Install the published package using pip or uv.
pip install sinapsis-darts-forecasting==1.0.0uv pip install sinapsis-darts-forecasting==1.0.0uv add sinapsis-darts-forecasting==1.0.0Configuration
Effective fields from the inspector schema, including nested Pydantic model fields.
Required
Default: null
Optional
Default: null
Optional
Default: 1
Optional
Default: false
Optional
Default: true
Optional
Default: "macro"
Field used to set the template instance name and agent name. Defaults to None. recall intersect <class 'bool'>: default: True labels typing.Union[int, list[int], numpy.ndarray, NoneType]: default: None label_reduction… Read full field description
Field used to set the template instance name and agent name. Defaults to None. recall intersect <class 'bool'>: default: True labels typing.Union[int, list[int], numpy.ndarray, NoneType]: default: None label_reduction typing.Union[str, NoneType, darts.metrics.utils._LabelReduction]: default: _LabelReduction.MACRO n_jobs <class 'int'>: default: 1 verbose <class 'bool'>: default: False
Optional
Default: null
Configuration example
Generated from documented defaults. Replace any placeholders and complete required nested settings before running.
templates:
- template_name: "recallTimeSeriesWrapper"
class_name: "recallTimeSeriesWrapper"
attributes:
recall:
labels: null
n_jobs: 1
verbose: false
intersect: true
label_reduction: "macro"
Inheritance
sinapsis_darts_forecasting.templates.time_series_metrics.TimeSeriesMetricsEnvironment variables
Environment definitions belong to their canonical Python modules.