Search

query_search

tssearch.search.query_search.time_series_search(dict_distances, query, sequence, tq=None, ts=None, weight=None, output=('number', 1))

Time series search method locates the k-best occurrences of a given query on a more extended sequence based on a distance measurement.

Parameters:

• dict_distances (dict) – Configuration file with distances.
• query (nd-array) – Query time series.
• sequence (nd-array) – Sequence time series.
• tq (nd-array) – Time stamp time series query.
• ts (nd-array) – Time stamp time series sequence.
• weight (nd-array (Default: None)) – query weight values.
• output (tuple) – number of occurrences.

Returns:

distance_results – time instants, optimal alignment path and distance for each occurrence per distance.

Return type:

dict

segmentation

tssearch.search.segmentation.time_series_segmentation(dict_distances, query, sequence, tq=None, ts=None, weight=None)

Time series segmentation locates the time instants between consecutive query repetitions on a more extended and repetitive sequence.

Parameters:

• dict_distances (dict) – Configuration file with distances
• query (nd-array) – Query time series.
• sequence (nd-array) – Sequence time series.
• tq (nd-array) – Time stamp time series query.
• ts (nd-array) – Time stamp time series sequence.
• weight (nd-array (Default: None)) – query weight values

Returns:

segment_results – Segmented time instants for each given distances

Return type:

dict

utils

tssearch.search.search_utils.elastic_search(dict_distances, query, sequence, tq=None, ts=None, weight=None)

Query search for elastic measures

Parameters:

• dict_distances (dict) – Configuration file with distances
• query (nd-array) – Query time series.
• sequence (nd-array) – Sequence time series.
• tq (nd-array) – Time stamp time series query.
• ts (nd-array) – Time stamp time series sequence.
• weight (nd-array (Default: None)) – query weight values

Returns:

• distance (nd-array) – distance value between query and sequence
• ac (nd-array) – accumulated cost matrix

tssearch.search.search_utils.lockstep_search(dict_distances, query, sequence, weight)

Query search for lockstep measures

Parameters:

• dict_distances (dict) – Configuration file with distances
• query (nd-array) – Query time series.
• sequence (nd-array) – Sequence time series.
• weight (nd-array (Default: None)) – query weight values

Returns:

res – distance value between query and sequence

Return type:

nd-array

tssearch.search.search_utils.start_sequences_index(distance, output=('number', 1), overlap=1.0)

Method to retrieve the k-best occurrences from a given vector distance

Parameters:

• distance (nd-array) – distance values
• output (tuple) – number of occurrences
• overlap (float) – minimum distance between occurrences

Returns:

id_s – indexes of k-best occurrences

Return type:

nd-array