Alejandro Moreo Fernandez
|
4c74ff02a3
|
svmperf in one-vs-all bugfix
|
2023-02-11 10:47:27 +01:00 |
Alejandro Moreo Fernandez
|
7b2d3cb7f1
|
example using svmperf
|
2023-02-11 10:08:31 +01:00 |
Alejandro Moreo Fernandez
|
952cf5e767
|
fixing bugs in one-vs-all
|
2023-02-10 19:02:17 +01:00 |
Alejandro Moreo Fernandez
|
33a21db52c
|
more examples, one-vs-all fixed
|
2023-02-09 19:43:24 +01:00 |
Alejandro Moreo Fernandez
|
9584e5152e
|
more examples, one-vs-all fixed
|
2023-02-09 19:39:39 +01:00 |
Alejandro Moreo Fernandez
|
e28abfc362
|
more examples, one-vs-all fixed
|
2023-02-09 19:39:16 +01:00 |
Alejandro Moreo Fernandez
|
2485117f05
|
adding documentation and adding one new example
|
2023-02-08 19:06:53 +01:00 |
Alejandro Moreo Fernandez
|
ceb88792c5
|
added DistributionMatching method, a generic model for distribution matching for multiclass quantification problems that takes the divergence and number of bins as hyperparameters
|
2023-01-31 15:08:58 +01:00 |
Alejandro Moreo Fernandez
|
f9a199d859
|
fixing hyperparameters with prefixes, and replacing learner with classifier in aggregative quantifiers
|
2023-01-27 18:13:23 +01:00 |
Alejandro Moreo Fernandez
|
adf799c8ec
|
recalibration
|
2023-01-24 09:48:21 +01:00 |
Alejandro Moreo Fernandez
|
3c48841480
|
Merge branch 'protocols' of github.com:HLT-ISTI/QuaPy into protocols
|
2023-01-18 19:46:31 +01:00 |
Alejandro Moreo Fernandez
|
09abcfc935
|
adding calibration methods from the abstension package to quapy
|
2023-01-18 19:46:19 +01:00 |
Alejandro Moreo Fernandez
|
3541f559e3
|
Merge pull request #16 from pglez82/protocols
Protocols
|
2023-01-18 16:55:51 +01:00 |
Pablo González
|
38aa42e4c5
|
fixing a bug
|
2023-01-18 16:44:56 +01:00 |
Pablo González
|
17cc1a3a5b
|
Merge branch 'HLT-ISTI:protocols' into protocols
|
2023-01-18 16:14:40 +01:00 |
Pablo González
|
8da4b4c5f3
|
placing the legend
|
2023-01-18 16:12:38 +01:00 |
Pablo González
|
7ed7c9b2e9
|
changing the logaritmic scale
|
2023-01-18 16:05:40 +01:00 |
Alejandro Moreo Fernandez
|
1d4fa40f3e
|
Merge branch 'protocols' of github.com:HLT-ISTI/QuaPy into protocols
|
2023-01-18 15:19:57 +01:00 |
Alejandro Moreo Fernandez
|
850f0e25db
|
Merge pull request #15 from pglez82/protocols
Protocols
|
2023-01-18 15:19:26 +01:00 |
Pablo González
|
f10a3139d9
|
changes to plots again
|
2023-01-18 14:53:46 +01:00 |
Pablo González
|
50d886bffe
|
testing log scale
|
2023-01-18 13:06:38 +01:00 |
Alejandro Moreo Fernandez
|
6e910075ab
|
adding calibration methods from abstension package
|
2023-01-17 13:53:48 +01:00 |
Pablo González
|
c888346fcf
|
solving a bug in show_legend
|
2023-01-17 11:03:52 +01:00 |
Pablo González
|
7bcf8b24e9
|
fixing bug
|
2023-01-16 17:17:02 +01:00 |
Pablo González
|
948f63fade
|
updating plot to center it better
|
2023-01-16 17:00:24 +01:00 |
Alejandro Moreo Fernandez
|
8b0b9f522a
|
some bugfixes, unittest and minor changes
|
2023-01-16 13:51:29 +01:00 |
Alejandro Moreo Fernandez
|
bb7a77c7c0
|
missing param in documentation of some protocols
|
2022-12-13 16:57:11 +01:00 |
Alejandro Moreo Fernandez
|
c20d9d5ea4
|
the heuristic exact_train_prev is performed via kFCV, using a new function qp.model_selection.cross_val_predict
|
2022-12-12 17:32:30 +01:00 |
Alejandro Moreo Fernandez
|
eb860e9678
|
adding the possibility to estimate the training prevalence, instead of using the true training prevalence, as a starting point in emq
|
2022-12-12 09:34:09 +01:00 |
Alejandro Moreo Fernandez
|
643a19228b
|
data reader for lequa 2022 competition
|
2022-11-28 12:02:08 +01:00 |
Alejandro Moreo Fernandez
|
fb79a29204
|
todos and change log
|
2022-11-08 16:36:52 +01:00 |
Alejandro Moreo Fernandez
|
eafc82c96a
|
full example of training, model selection, and evaluation using the lequa2022 dataset with the new protocols
|
2022-11-04 15:15:12 +01:00 |
Alejandro Moreo Fernandez
|
6cb9f388e0
|
full example of training, model selection, and evaluation using the lequa2022 dataset with the new protocols
|
2022-11-04 15:06:08 +01:00 |
Alejandro Moreo Fernandez
|
d75b777a13
|
Merge branch 'protocols' of github.com:HLT-ISTI/QuaPy into protocols
|
2022-11-04 15:04:50 +01:00 |
Alejandro Moreo Fernandez
|
f2550fdb82
|
full example of training, model selection, and evaluation using the lequa2022 dataset with the new protocols
|
2022-11-04 15:04:36 +01:00 |
Alejandro Moreo Fernandez
|
a4c33a8e4d
|
import fix
|
2022-10-04 17:44:16 +02:00 |
Alejandro Moreo Fernandez
|
e40c409609
|
bugfix in NeuralClassifierTrainer; it was only configured to work well in binary problems
|
2022-10-04 11:03:08 +02:00 |
Alejandro Moreo Fernandez
|
8e14bbc527
|
Merge branch 'master' of github.com:HLT-ISTI/QuaPy
|
2022-10-04 09:13:34 +02:00 |
Alejandro Moreo Fernandez
|
3af7c70a53
|
restoring the default legend in diag plot
|
2022-10-04 09:12:51 +02:00 |
Alejandro Moreo Fernandez
|
1890e20057
|
Update README.md
|
2022-08-29 12:03:03 +02:00 |
Alejandro Moreo Fernandez
|
543003f914
|
Merge pull request #13 from pglez82/dys_implementation
Dys implementation
|
2022-07-12 13:05:35 +02:00 |
Pablo González
|
a4584b79db
|
changing gridsearchQ to ensure reproducibility
|
2022-07-11 16:27:02 +02:00 |
Pablo González
|
c91961cff5
|
adding to __init__.py
|
2022-07-11 14:10:04 +02:00 |
Pablo González
|
428f10fb2d
|
adding SMM
|
2022-07-11 14:06:14 +02:00 |
Alejandro Moreo Fernandez
|
ecd0ad7ec7
|
unit test for replicability based on qp.util.temp_seed
|
2022-07-11 14:00:25 +02:00 |
Pablo González
|
46e294002f
|
dys implementation
|
2022-07-11 12:21:49 +02:00 |
Alejandro Moreo Fernandez
|
1742b75504
|
Merge pull request #12 from pglez82/protocols
changing app to use prevalence_linspace function with smooth limits
|
2022-06-24 14:44:50 +02:00 |
Pablo González
|
1914b854ea
|
Merge branch 'HLT-ISTI:protocols' into protocols
|
2022-06-24 14:21:29 +02:00 |
Pablo González
|
750814ef2a
|
fixing bug in ACC when using cross validation
|
2022-06-24 14:20:08 +02:00 |
Pablo González
|
02dd2846ff
|
changing app to use prevalence_linspace function with smooth limits
|
2022-06-24 14:05:47 +02:00 |