1d89301089more uci datasets, plots improved (higher fonts), and evaluation script that shows numerical results in command lineAlejandro Moreo Fernandez2021-01-27 22:49:54 +0100
e609c262b4parallel functionality added to quapy in order to allow for multiprocess parallelization (and not threading) handling quapy's environment variablesAlejandro Moreo Fernandez2021-01-27 09:54:41 +0100
03cf73aff6refactor: methods requiring a val_split can now declare a default value in the __init__ method that will be used in case the fit method is called without specifying the val_split, which now is by default None in the fit, i.e., by default takes the value of the init, that is generally set to 0.4; some uci datasets added; ensembles can now be optimized for quantification, and can be trained on samples of smaller sizeAlejandro Moreo Fernandez2021-01-22 18:01:51 +0100
1ba0748b59experimental method ave-pool, not working due to the fact that onevsall is aggregative and ave-pool is notAlejandro Moreo Fernandez2021-01-20 17:03:12 +0100
b30c40b7a0some refactor made in order to accomodate OneVsAll to operate with aggregative probabilistic quantifiers; launching OneVsAll(HDy)Alejandro Moreo Fernandez2021-01-18 16:52:19 +0100
865dafaefcsetting a timeout for model_selection combinations in order to prevent some combinations to stuck the model selectionAlejandro Moreo Fernandez2021-01-15 17:42:19 +0100
326a8ab803added Ensemble methods (methods ALL, ACC, Ptr, DS from Pérez-Gallego et al 2017 and 2019) and some UCI ML datasets used in those articles (only 5 datasets out of 32 they used)Alejandro Moreo Fernandez2021-01-06 14:58:29 +0100
649d412389dataset fetch for polarity reviews (hp, kindle, imdb) and twitter sentiment (11 datasets) addedAlejandro Moreo Fernandez2020-12-14 18:36:19 +0100
c8a1a70c8arefactoring aggregative methods as methods that not only implement 'classify' and 'quantify', but that also implement 'aggregate' and that, by default, have a default implementation of 'quantify' as a pipeline of 'classify' and 'aggregate'; this helps speeding up evaluations A LOT, since the documents can be pre-classified and the samples are carried out across pre-classified values (labels, or posterior probabilities), and thus only aggregate is called many times within the artificial sampling protocolAlejandro Moreo Fernandez2020-12-11 19:28:17 +0100
9bc3a9f28aevaluation by artificial prevalence sampling added. New methods added. New util functions added to quapy.functional and quapy.utilsAlejandro Moreo Fernandez2020-12-10 19:04:33 +0100
9c8d29156caggregative methods adapted. Explicit loss minimization methods (SVMQ, SVMKLD, ...) added and with support to binary or single-label. HDy addedAlejandro Moreo Fernandez2020-12-04 19:32:08 +0100
Andrea Esuli
Alejandro Moreo Fernandez