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

merged

This commit is contained in:
Alejandro Moreo Fernandez 2020-12-10 19:08:22 +01:00
parent 9bc3a9f28a 2361186a01
commit e55caf82fd
6 changed files with 14 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
TODO.txt
View File

@ -6,3 +6,6 @@ Add prediction - artificial sampling
Add readers for typical datasets used in Quantification Add readers for typical datasets used in Quantification
Add NAE, NRAE Add NAE, NRAE
Add "measures for evaluating ordinal"? Add "measures for evaluating ordinal"?
Document methods with paper references
The parallel training in svmperf seems not to work

1
quapy/error.py
View File

@ -2,6 +2,7 @@ from sklearn.metrics import f1_score
import numpy as np import numpy as np
SAMPLE_SIZE = None SAMPLE_SIZE = None

2
quapy/functional.py
View File

@ -77,6 +77,7 @@ def normalize_prevalence(prevalences):
return np.ones_like(prevalences) / prevalences.size return np.ones_like(prevalences) / prevalences.size
def num_prevalence_combinations(nclasses:int, nprevpoints:int, nrepeats:int): def num_prevalence_combinations(nclasses:int, nprevpoints:int, nrepeats:int):
""" """
Computes the number of prevalence combinations in the nclasses-dimensional simplex if nprevpoints equally distant Computes the number of prevalence combinations in the nclasses-dimensional simplex if nprevpoints equally distant
@ -118,3 +119,4 @@ def get_nprevpoints_approximation(nclasses, nrepeats, combinations_budget):
return nprevpoints-1 return nprevpoints-1
else: else:
nprevpoints+=1 nprevpoints+=1

4
quapy/method/aggregative.py
View File

@ -337,6 +337,10 @@ class OneVsAll(AggregativeQuantifier):
delayed(self._delayed_binary_predict)(c, self.class_method, X) for c in self.classes delayed(self._delayed_binary_predict)(c, self.class_method, X) for c in self.classes
) )
) )
<<<<<<< HEAD
=======
print('one vs all: ', prevalences)
>>>>>>> 2361186a01c53e744f4291e2e2299700216ff139
return F.normalize_prevalence(prevalences) return F.normalize_prevalence(prevalences)
@property @property

2
quapy/utils/util.py
View File

@ -5,6 +5,7 @@ import contextlib
import numpy as np import numpy as np
def get_parallel_slices(n_tasks, n_jobs=-1): def get_parallel_slices(n_tasks, n_jobs=-1):
if n_jobs == -1: if n_jobs == -1:
n_jobs = multiprocessing.cpu_count() n_jobs = multiprocessing.cpu_count()
@ -32,4 +33,3 @@ def temp_seed(seed):
finally: finally:
np.random.set_state(state) np.random.set_state(state)

3
test.py
View File

@ -3,6 +3,7 @@ from sklearn.svm import LinearSVC
import quapy as qp import quapy as qp
import quapy.functional as F import quapy.functional as F
SAMPLE_SIZE=500 SAMPLE_SIZE=500
binary = False binary = False
@ -28,6 +29,7 @@ model = qp.method.aggregative.ClassifyAndCount(learner)
# model = qp.method.aggregative.ProbabilisticClassifyAndCount(learner) # model = qp.method.aggregative.ProbabilisticClassifyAndCount(learner)
# model = qp.method.aggregative.ProbabilisticAdjustedClassifyAndCount(learner) # model = qp.method.aggregative.ProbabilisticAdjustedClassifyAndCount(learner)
# model = qp.method.aggregative.ExpectationMaximizationQuantifier(learner) # model = qp.method.aggregative.ExpectationMaximizationQuantifier(learner)
model.fit(dataset.training) model.fit(dataset.training)
# estimating class prevalences # estimating class prevalences
@ -51,3 +53,4 @@ print(f'Evaluation according to the artificial sampling protocol ({len(true_prev
for error in qp.error.QUANTIFICATION_ERROR: for error in qp.error.QUANTIFICATION_ERROR:
score = error(true_prev, estim_prev) score = error(true_prev, estim_prev)
print(f'{error.__name__}={score:.5f}') print(f'{error.__name__}={score:.5f}')