Petrônio Cândido
1237f3c2e3
Improvements on pwfts to support GranularFTS; Improvementos on mvfts methods, fuzzy sets and partitioners to support interval and probabilistic forecasting
2019-06-21 11:32:56 -03:00
Petrônio Cândido
48fcf8daca
Improvements on forecast_ahead benchmarks
2019-06-10 13:33:53 -03:00
Petrônio Cândido
87686e5ff0
Bugfixes in pwfts, distributed and composite
2018-12-27 18:46:55 -02:00
Petrônio Cândido
386f28c47d
pySpark support
2018-12-17 21:49:48 -02:00
Petrônio Cândido
6dcdd8f59e
Bugfix on pwfts.forecast_ahead_distribution
2018-12-13 14:46:16 -02:00
Petrônio Cândido
0d033c3ac2
Improvements on cmvfts and pwfts for multivariate forecasting
2018-12-11 22:21:19 -02:00
Petrônio Cândido
c0385e556e
Bugfixes on pwfts
2018-12-11 15:15:40 -02:00
Petrônio Cândido
1773499059
Optimizations and refactorings on hofts and pwfts; fuzzyfy function on Partitioner; Improvements on cmvfts to enable all models to be used with him
2018-12-05 21:17:34 -02:00
Petrônio Cândido
0b63ed118a
Documentation update
2018-10-29 16:48:05 -03:00
Petrônio Cândido
bb983b53dd
Fuzzyfy optimization with binary search. Several minor parameter refactorings
2018-09-18 14:59:03 -03:00
Petrônio Cândido
07bfceae78
Introducing FTS.lags, the lag indexing for HOFTS, IFTS and PWFTS
2018-07-02 18:55:28 -03:00
Petrônio Cândido
aeb8236a7f
Introducing alpha_cut on fuzzyfication processes
2018-06-29 15:35:14 -03:00
Petrônio Cândido
5563af3079
Bugfixes
2018-05-14 00:16:19 -03:00
Petrônio Cândido
ffd97bacfc
Minor code standadization on models; Improvements and bugfixes on benchmarks
2018-05-08 17:59:53 -03:00
Petrônio Cândido
abe9a45a47
k-Nearest Neighbors benchmark method
2018-04-25 11:36:01 -03:00
Petrônio Cândido
6adf4710b6
pwfts.visualize_distributions
2018-04-19 16:21:59 -03:00
Petrônio Cândido
3db6e69119
Improvements on benchmarks
2018-04-13 15:23:54 -03:00
Petrônio Cândido
b65af00526
Bugfixes in pwfts
2018-04-11 14:11:58 -03:00
Petrônio Cândido
f9204b87c9
Bugfixes in pwfts
2018-04-11 01:31:17 -03:00
Petrônio Cândido
50c2b501b1
Refactoring to centralize the apply_transformations and apply_transformations_inverse inside fts.predict method
2018-04-11 01:12:55 -03:00
Petrônio Cândido
581f404b18
Improvements and bugfixes on pwfts.forecast_interval and pwfts.point_expected_value
2018-04-11 00:19:05 -03:00
Petrônio Cândido
24c4fadc08
Bugfixes at pwfts and improvements on benchmarks.plot_*
2018-04-10 12:39:56 -03:00
Petrônio Cândido
1312d96246
Bugfixes at pwfts.forecast_ahead_distribution
2018-04-09 14:52:25 -03:00
Petrônio Cândido
1d7801bdbf
- Benchmark refactoring to expose a unified and simpler interface (the method sliding_window_benchmarks)
...
- Improvements at FTS.fit and FTS.predict
- PWFTS bugfixes and improvements
2018-04-09 11:09:28 -03:00
Petrônio Cândido
3d64c5065e
- Bugfixes due to refactorings
2018-03-05 15:07:02 -03:00
Petrônio Cândido
9718f48b39
- common.Util.distributed_train
...
- Big refactoring to change FTS.sets from list to dict. This refactoring allow to remove references to the fuzzy sets from the FLRG and save memory.
- HOFTS and PWFTS train and forecasting simplification by using the method generate_lhs_flrg
- Small others bugfixes/improvements
2018-03-03 20:07:50 -03:00
Petrônio Cândido
00db6a30ad
- Compacting datasets with bz2
...
- Refactoring generate_flrg and train methods
- Introducing batches and model saving on fit method
2018-03-02 19:20:21 -03:00
Petrônio Cândido
7a2dd7e54c
Notebooks and method bugfixes and refactoring
2018-02-27 18:30:20 -03:00
Petrônio Cândido
37862c661d
- More data files and artificial data generators
2018-02-27 12:56:05 -03:00
Petrônio Cândido
e4838693a6
- Deep refactor on project folders
2018-02-26 13:29:11 -03:00
