Gait Sequence Detection (GSD)#

Algorithms to detect gait sequences within raw IMU data.

Algorithms#

GsdIluz(*[, pre_filter, window_length_s, ...])

Implementation of the GSD algorithm by Iluz et al. (2014) [Rea5df3d1a481-1].

GsdIonescu(*[, min_n_steps, ...])

Implementation of the GSD algorithm developed by Paraschiv-Ionescu et al. (2014) [R58e9f9eaa522-1].

GsdAdaptiveIonescu(*[, min_n_steps, ...])

Implementation of the GSD algorithm by Paraschiv-Ionescu et al. (2019) [1, 2]_ with adaptive threshold.

Pipelines#

pipeline.GsdEmulationPipeline(algo, *[, ...])

Run a GSD algorithm in isolation on a Gait Dataset.

Base Classes#

Base class for GSD detectors.

BaseGsDetector()

Base class for GS-detectors.

Docu-helper#

base_gsd_docfiller(func)

Decorator to fill common parts of the docstring for subclasses of BaseGsdDetector.

Evaluation#

calculate_matched_gsd_performance_metrics(...)

Calculate commonly known performance metrics for based on the matched overlap with the reference.

calculate_unmatched_gsd_performance_metrics(*, ...)

Calculate general performance metrics that don't rely on matching the detected and reference gait sequences.

categorize_intervals_per_sample(*, ...[, ...])

Evaluate detected gait sequence intervals against a reference on a sample-wise level.

categorize_intervals(*, gsd_list_detected, ...)

Evaluate a gait sequence list against a reference sequence-by-sequence with a minimum overlap threshold.

plot_categorized_intervals(...)

Plot the categorized intervals together with the detected and reference intervals.

get_matching_intervals(*, metrics_detected, ...)

Extract the detected and reference gait sequences that are considered as matches sequence-by-sequence.

Evaluation Challenges#

GsdEvaluation(dataset, scoring, *, ...)

Evaluation challenge for Gait Sequence Detection (GSD) algorithms.

GsdEvaluationCV(dataset, cv_iterator, ...)

Evaluation challenge for Gait Sequence Detection (GSD) algorithms using cross-validation.

gsd_evaluation_scorer(pipeline, datapoint)

Evaluate the performance of a GSD algorithm on a single datapoint.