- 0.5.0 (2015-07-07)
- added CMAP (Co-occurrence Map)
- new representation of Element class (and created new class ElementDict instead of ElementPool)
- updated SPADEm's method sorted_elements (and renamed to "grouped")
- improved scheme to pick up "master" elements (special elements that will be merged with others)
- performance enhancement
- 0.4.2 (2015-06-14)
- fixed incorrect logic (in Element's method get_equivalence_relation_diff)
- performance enhancements (one among others is the removed recursion in SPADEm's method enumerate_frequent_sequences)
- 0.4.1 (2015-06-04)
- performance enhancement
- 0.4.0 (2015-05-29)
- current implementation of SPADE will produce only maximal sequential patterns (no sub-sequences of frequent sequences)
- name of master class is changed from SPADE to SPADEm (where "m" means "maximal")
- removed option "--maximal" (it is applied by default)
- 0.3.2 (2015-05-27)
- fixed algorithm implementation (fixed incorrect logic and added corresponding methods, e.g. methods to deal with equivalence classes)
- improved memory usage (for long sequences, with length more than 10^3, it is recommended to request maximal frequent sequences only)
- option "--k" is changed to the option "--length" (maximum length of frequent sequences)
- added option "--maximal" to get only maximal frequent sequences (no sub-sequences of frequent sequences)
- added option "--sort" to sort output sequences (base on sequence length)
- 0.2.1 (2015-04-27)
- light enhancements
- 0.2.0 (2015-04-20)
- added option to get top-n longest sequential patterns (option "--top n")
- 0.1.2
- enhanced calculating of difference between sequences
- improved procedure for processing of 1- and 2-sequences
- 0.1.1 (2015-03-31)
- doc-strings updates
- moved sequence length checking procedure out of loop in method enumerate_frequent_sequences
- 0.1.0 (2015-03-24)
- first release