Commit Graph

11 Commits

Author SHA1 Message Date
Petrônio Cândido de Lima e Silva
47b1cb57c9 A new method for fuzzy frequency count at PFTS was added but in benchmarks the previous method proved to be most efficient. The new method was keep for future investigations. 2017-02-16 11:54:37 -02:00
Petrônio Cândido de Lima e Silva
cb12810e0a - Probability distributions 2017-02-15 18:16:13 -02:00
Petrônio Cândido de Lima e Silva
8787d749ff Probability density function at FTS 2017-02-14 23:01:44 -02:00
Petrônio Cândido de Lima e Silva
279679b3a4 Refactoring to help tasks automotion 2017-02-10 11:09:59 -02:00
Petrônio Cândido de Lima e Silva
26656153f4 Contextual Multi Season FTS 2017-02-09 11:04:48 -02:00
Petrônio Cândido de Lima e Silva
55d3deadfc - DateTimeSeasonalIndexer
-  persist_obj, load_obj, persist_env, load_env
2017-02-08 13:23:41 -02:00
Petrônio Cândido de Lima e Silva
bb42a6be07 - MultiSeasonal FTS
- Seasonal Indexers for Panda DataFrames
- Indexed FLR's
2017-02-04 20:40:27 -02:00
Petrônio Cândido de Lima e Silva
71c71ca9d9 Several bugfixes and optimizations 2017-02-01 14:54:37 -02:00
Petrônio Cândido de Lima e Silva
0cf938c2a6 Cascaded transformations in all fts models 2017-01-29 21:59:50 -02:00
Petrônio Cândido de Lima e Silva
f53eded47e Several bugfixes in models and benchmarks 2017-01-24 10:40:48 -02:00
Petrônio Cândido de Lima e Silva
8f2d2c8bcd Otimização de forecastAheadDistribution para um método greedy mais confiável 2017-01-13 21:42:00 -02:00