moreo
/
QuaPy
1
0
Fork 1
Code Issues Pull Requests Releases Wiki Activity

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
This commit is contained in:
Alejandro Moreo Fernandez 2024-04-19 14:23:35 +02:00 committed by GitHub
parent e6f380dc5f
commit bf33c134fc
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 7 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
quapy/method/_kdey.py
View File

@ -62,8 +62,13 @@ class KDEBase:
:param bandwidth: float, the bandwidth of the kernel
:return: a list of KernelDensity objects, each fitted with the corresponding class-specific covariates
"""
return [self.get_kde_function(X[y == cat], bandwidth) for cat in classes]
class_cond_X = []
for cat in classes:
selX = X[y==cat]
if selX.size==0:
selX = [F.uniform_prevalence(len(classes))]
class_cond_X.append(selX)
return [self.get_kde_function(X_cond_yi, bandwidth) for X_cond_yi in class_cond_X]
class KDEyML(AggregativeSoftQuantifier, KDEBase):