extendedMD.pruning¶

extendedMD.pruning.extract_ts_from_pointers(ts, pointers)¶

Parameters:	ts (1d array) – 1-dimensional time-series either resulting from the PCA method or the original 1-dimensional time-series pointers (list of int) – list of indexes related to the subsequence one wishes to extract from ts
Returns:	time-series subsequence
Return type:	1d array

extendedMD.pruning.prune_motifs(ts, sorted_dic_list, r)¶

Parameters:	ts (1d array) – 1-dimensional time-series either resulting from the PCA method or the original 1-dimensional time-series sorted_dic_list (list of dic) – list of motif dictionaries returned from the emd algorithm, ordered by relevance r (float) – maximum distance to the center of the motif
Returns:	list of dictionaries with the most relevant motifs
Return type:	list of dic

extendedMD.pruning.prune_motifs_with_dist(ts, motif_dic_list, r, mdl_bins)¶

Parameters:	ts (1d array) – 1-dimensional time-series either resulting from the PCA method or the original 1-dimensional time-series motif_dic_list (list of dic) – list of motif dictionaries returned from the emd algorithm r (float) – maximum distance to the center of the motif mdl_bins (int) – number of bins to break the MDL cost range
Returns:	list of dictionaries with the most relevant motifs. The list is ordered based on MDL cost and motif’s compactness
Return type:	list of dic

extendedMD.pruning.prune_motifs_with_mdl(ts, motif_dic_list, r)¶

This function returns the most relevant motifs from the original list of motif extracted from the emd algorithm, based on the computed MDL cost and avoiding overlapping motifs

Parameters:	ts (1d array) – 1-dimensional time-series either resulting from the PCA method or the original 1-dimensional time-series motif_dic_list (list of dic) – list of motif dictionaries returned from the emd algorithm r (float) – maximum distance to the center of the motif
Returns:	list of dictionaries with the most relevant motifs. The list is ordered based on the MDL cost
Return type:	list of dic