lorenzo.volpi
/
QuaPy
forked from moreo/QuaPy
1
0
Fork 0
Code Pull Requests Activity

adding the possibility to estimate the training prevalence, instead of using the true training prevalence, as a starting point in emq

This commit is contained in:
Alejandro Moreo Fernandez 2022-12-12 09:34:09 +01:00
parent 643a19228b
commit eb860e9678
2 changed files with 20 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
examples/lequa2022_experiments.py
View File

@ -1,6 +1,8 @@
import numpy as np import numpy as np
from sklearn.calibration import CalibratedClassifierCV
from sklearn.linear_model import LogisticRegression from sklearn.linear_model import LogisticRegression
import quapy as qp import quapy as qp
import quapy.functional as F
from data.datasets import LEQUA2022_SAMPLE_SIZE, fetch_lequa2022 from data.datasets import LEQUA2022_SAMPLE_SIZE, fetch_lequa2022
from evaluation import evaluation_report from evaluation import evaluation_report
from method.aggregative import EMQ from method.aggregative import EMQ
@ -14,7 +16,8 @@ qp.environ['SAMPLE_SIZE'] = LEQUA2022_SAMPLE_SIZE[task]
training, val_generator, test_generator = fetch_lequa2022(task=task) training, val_generator, test_generator = fetch_lequa2022(task=task)
# define the quantifier # define the quantifier
quantifier = EMQ(learner=LogisticRegression()) learner = CalibratedClassifierCV(LogisticRegression())
quantifier = EMQ(learner=learner)
# model selection # model selection
param_grid = {'C': np.logspace(-3, 3, 7), 'class_weight': ['balanced', None]} param_grid = {'C': np.logspace(-3, 3, 7), 'class_weight': ['balanced', None]}
@ -24,6 +27,10 @@ quantifier = model_selection.fit(training)
# evaluation # evaluation
report = evaluation_report(quantifier, protocol=test_generator, error_metrics=['mae', 'mrae', 'mkld'], verbose=True) report = evaluation_report(quantifier, protocol=test_generator, error_metrics=['mae', 'mrae', 'mkld'], verbose=True)
pd.set_option('display.max_columns', None) # printing results
pd.set_option('display.width', 1000) pd.set_option('display.expand_frame_repr', False)
report['estim-prev'] = report['estim-prev'].map(F.strprev)
print(report) print(report)
print('Averaged values:')
print(report.mean())

12
quapy/method/aggregative.py
View File

@ -501,17 +501,25 @@ class EMQ(AggregativeProbabilisticQuantifier):
maximum-likelihood estimation, in a mutually recursive way, until convergence. maximum-likelihood estimation, in a mutually recursive way, until convergence.
:param learner: a sklearn's Estimator that generates a classifier :param learner: a sklearn's Estimator that generates a classifier
:param exact_train_prev: set to True (default) for using, as the initial observation, the true training prevalence;
or set to False for computing the training prevalence as an estimate, akin to PCC, i.e., as the expected
value of the posterior probabilities of the trianing documents as suggested in
`Alexandari et al. paper <http://proceedings.mlr.press/v119/alexandari20a.html>`_:
""" """
MAX_ITER = 1000 MAX_ITER = 1000
EPSILON = 1e-4 EPSILON = 1e-4
def __init__(self, learner: BaseEstimator): def __init__(self, learner: BaseEstimator, exact_train_prev=True):
self.learner = learner self.learner = learner
self.exact_train_prev = exact_train_prev
def fit(self, data: LabelledCollection, fit_learner=True): def fit(self, data: LabelledCollection, fit_learner=True):
self.learner, _ = _training_helper(self.learner, data, fit_learner, ensure_probabilistic=True) self.learner, _ = _training_helper(self.learner, data, fit_learner, ensure_probabilistic=True)
self.train_prevalence = F.prevalence_from_labels(data.labels, self.classes_) if self.exact_train_prev:
self.train_prevalence = F.prevalence_from_labels(data.labels, self.classes_)
else:
self.train_prevalence = PCC(learner=self.learner).fit(data, fit_learner=False).quantify(data.X)
return self return self
def aggregate(self, classif_posteriors, epsilon=EPSILON): def aggregate(self, classif_posteriors, epsilon=EPSILON):