Alejandro Moreo Fernandez
ad11b86168
adding environment variables for N_JOBS, and adding a default classifier (sklearn's logistic regression) for when the classifier is not specified in aggregative quantifiers
2024-05-30 10:53:53 +02:00
Alejandro Moreo Fernandez
9ad36ef008
cleaning examples and adding basic example
2024-05-29 14:24:03 +02:00
Alejandro Moreo Fernandez
acfb02c51f
bugfix import and missing certifi in setup
2024-05-17 18:04:22 +02:00
Alejandro Moreo Fernandez
4db21b6945
smoothing the prevalences in kld error function, bugfix
2024-05-08 11:31:56 +02:00
Alejandro Moreo Fernandez
817aab1d99
Merge branch 'devel' of github.com:HLT-ISTI/QuaPy into devel
2024-04-30 09:55:50 +02:00
Alejandro Moreo Fernandez
7f39f4df66
file kept open in utils pickled resource, fixed
2024-04-30 09:55:28 +02:00
Alejandro Moreo Fernandez
8517338765
Merge branch 'devel' into ucimulti
2024-04-29 17:58:06 +02:00
19524f9aa8
ucimulti datasets removed, cleaning
2024-04-29 17:36:13 +02:00
498fd8b050
datasets removed from ucimulti
2024-04-24 17:23:01 +02:00
Alejandro Moreo Fernandez
f1462897ef
Merge pull request #32 from mirkobunse/devel
...
Integrate composable methods from qunfold
2024-04-24 15:26:32 +02:00
f74b048e2d
uci_multi dataset removed
2024-04-24 15:20:14 +02:00
ecfc175622
datasets removed, debug output added
2024-04-23 16:30:17 +02:00
Alejandro Moreo Fernandez
bf33c134fc
Update _kdey.py
...
fix in KDEy: makes the method robust to cases in which the number of positives for any class is smaller than the number k of folds. In such cases, the kde for that class is created from the uniform prevalence vector
2024-04-19 14:23:35 +02:00
da99f78c0c
Merge remote-tracking branch 'fork-origin/devel' into devel
2024-04-18 10:09:17 +02:00
2000c33372
Composable methods integrated from qunfold, which is an extra dependency for quapy.method.composable
2024-04-18 10:08:49 +02:00
Alejandro Moreo Fernandez
e6f380dc5f
update changelog
2024-04-18 09:38:33 +02:00
a64620c377
Dataset.reduce() allows to fix the random_state to have reproducible unit tests. This is required to ensure that the expected hyper-parameters are always chosen, independent of randomness
2024-04-17 14:46:37 +02:00
72b43bd2f8
Omit large datasets (LeQua, IFCB) during CI to avoid overful memory of GitHub Actions runners
2024-04-17 13:46:59 +02:00
69b8327fe9
Remove an erroneous import in the unit tests and add extra test dependencies.
2024-04-17 11:44:23 +02:00
Alejandro Moreo Fernandez
db6ff4ab9e
refactored unittests
2024-04-16 17:46:58 +02:00
Alejandro Moreo Fernandez
561b672200
updated unit tests
2024-04-16 15:12:22 +02:00
Alejandro Moreo Fernandez
99bc8508ac
Merge branch 'devel' of gitea-s2i2s.isti.cnr.it:moreo/QuaPy into devel
2024-04-15 18:00:56 +02:00
Alejandro Moreo Fernandez
9207114cfa
improving unit tests
2024-04-15 18:00:38 +02:00
Alejandro Moreo Fernandez
e0b80167b9
added max_train_instances to fetch_UCIMulticlassLabelledCollection
2024-04-12 18:24:12 +02:00
Alejandro Moreo Fernandez
4abec6629b
integrating more uci-multiclass datasets
2024-04-12 18:08:00 +02:00
Alejandro Moreo Fernandez
3095d7092c
Merge branch 'ucimulti' of github.com:lorenzovolpi/QuaPy into lorenzovolpi-ucimulti
2024-04-12 13:36:38 +02:00
Alejandro Moreo Fernandez
b53d417240
merged
2024-04-12 13:35:13 +02:00
f69fca32b4
Added UCI multiclass datasets; added filter for min instances per class to UCI multiclass datasets
2024-04-11 20:08:52 +02:00
Alejandro Moreo Fernandez
a04723a976
switching
2024-03-20 17:31:07 +01:00
Alejandro Moreo Fernandez
aa894a3472
merging PR; I have taken this opportunity to refactor some issues I didnt like, including the normalization of prevalence vectors, and improving the documentation here and there
2024-03-19 15:01:42 +01:00
Alejandro Moreo Fernandez
36ac6db27d
fixing doc
2024-03-18 23:39:55 +01:00
5cdd158fcc
Add invariant ratio estimators.
2024-03-15 18:14:42 +01:00
d34b086a76
Refactor solving routine
2024-03-15 17:58:23 +01:00
4dd66b1921
Add projection onto the probability simplex
2024-03-15 17:15:16 +01:00
Alejandro Moreo Fernandez
3921b8368e
merging BayesianCC implemented by Pawel Czyz
2024-03-15 16:24:45 +01:00
2cc4908326
Sketch of the Bayesian quantification
2024-03-15 14:01:24 +01:00
3705264529
Fix a typo.
2024-03-14 10:39:26 +01:00
Alejandro Moreo Fernandez
75af15ae4a
force all samples be with replacement in base.LabelledCollection, irrespective of the sample size requested
2024-02-28 08:46:54 +01:00
Alejandro Moreo Fernandez
b3ccf71edb
Merge branch 'devel' of github.com:HLT-ISTI/QuaPy into devel
2024-02-23 16:30:11 +01:00
Alejandro Moreo Fernandez
320b3eac38
small fixes in kdey (now should work with string labels) and EMQ (in case some training prior prob was 0, it broke)
2024-02-23 16:29:53 +01:00
Alejandro Moreo Fernandez
9542eaee61
doing some benchmarking
2024-02-22 15:10:45 +01:00
Alejandro Moreo Fernandez
d50a86daf4
sketching readme system by Lu and King, Hopings and King
2024-02-16 17:34:10 +01:00
Alejandro Moreo Fernandez
40cb8f78fe
pytests before release
2024-02-14 12:27:19 +01:00
Alejandro Moreo Fernandez
7705c92c8c
fixing ifcb and documenting
2024-02-12 12:39:18 +01:00
Alejandro Moreo Fernandez
a8230827e2
testing IFCB dataset
2024-02-08 14:33:22 +01:00
Alejandro Moreo Fernandez
3c28a75b8c
merging conflicts I didn see
2024-02-07 18:51:06 +01:00
Alejandro Moreo Fernandez
a97978b85d
merged
2024-02-07 18:45:42 +01:00
Alejandro Moreo Fernandez
fcc3f8a0d9
fixing sphinx doc
2024-02-07 18:31:34 +01:00
Alejandro Moreo Fernandez
2f2e48d86a
passing pytests
2024-01-29 09:43:29 +01:00
Alejandro Moreo Fernandez
e6dcfbced1
adding M.Bunse's reference for the solver='minimize' option of ACC, PACC
2024-01-25 18:03:35 +01:00
Alejandro Moreo Fernandez
Lorenzo Volpi
Mirko Bunse
Paweł Czyż