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

fixing issue regarding fit_learner=False in QuaNetTrainer

This commit is contained in:
Alejandro Moreo Fernandez 2021-06-21 12:55:39 +02:00
parent f33abb5319
commit b4aeaa97b7
3 changed files with 19 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
quapy/data/base.py
View File

@ -130,11 +130,15 @@ class LabelledCollection:
yield self.uniform_sampling_index(sample_size) yield self.uniform_sampling_index(sample_size)
def __add__(self, other): def __add__(self, other):
if issparse(self.instances) and issparse(other.instances): if other is None:
return self
elif issparse(self.instances) and issparse(other.instances):
join_instances = vstack([self.instances, other.instances]) join_instances = vstack([self.instances, other.instances])
elif isinstance(self.instances, list) and isinstance(other.instances, list): elif isinstance(self.instances, list) and isinstance(other.instances, list):
join_instances = self.instances + other.instances join_instances = self.instances + other.instances
elif isinstance(self.instances, np.ndarray) and isinstance(other.instances, np.ndarray): elif isinstance(self.instances, np.ndarray) and isinstance(other.instances, np.ndarray):
print(self.instances.shape)
print(other.instances.shape)
join_instances = np.concatenate([self.instances, other.instances]) join_instances = np.concatenate([self.instances, other.instances])
else: else:
raise NotImplementedError('unsupported operation for collection types') raise NotImplementedError('unsupported operation for collection types')

13
quapy/method/aggregative.py
View File

@ -125,7 +125,7 @@ def training_helper(learner,
if not (0 < val_split < 1): if not (0 < val_split < 1):
raise ValueError(f'train/val split {val_split} out of range, must be in (0,1)') raise ValueError(f'train/val split {val_split} out of range, must be in (0,1)')
train, unused = data.split_stratified(train_prop=1 - val_split) train, unused = data.split_stratified(train_prop=1 - val_split)
elif val_split.__class__.__name__ == LabelledCollection.__name__: # isinstance(val_split, LabelledCollection): elif isinstance(val_split, LabelledCollection):
train = data train = data
unused = val_split unused = val_split
else: else:
@ -144,10 +144,8 @@ def training_helper(learner,
if not hasattr(learner, 'predict_proba'): if not hasattr(learner, 'predict_proba'):
raise AssertionError('error: the learner cannot be calibrated since fit_learner is set to False') raise AssertionError('error: the learner cannot be calibrated since fit_learner is set to False')
unused = None unused = None
if val_split.__class__.__name__ == LabelledCollection.__name__: if isinstance(val_split, LabelledCollection):
unused = val_split unused = val_split
if data is not None:
unused = unused+data
return learner, unused return learner, unused
@ -307,19 +305,22 @@ class PACC(AggregativeProbabilisticQuantifier):
# fit the learner on all data # fit the learner on all data
self.learner, _ = training_helper(self.learner, data, fit_learner, ensure_probabilistic=True, self.learner, _ = training_helper(self.learner, data, fit_learner, ensure_probabilistic=True,
val_split=None) val_split=None)
classes = data.classes_
else: else:
self.learner, val_data = training_helper( self.learner, val_data = training_helper(
self.learner, data, fit_learner, ensure_probabilistic=True, val_split=val_split) self.learner, data, fit_learner, ensure_probabilistic=True, val_split=val_split)
y_ = self.learner.predict_proba(val_data.instances) y_ = self.learner.predict_proba(val_data.instances)
y = val_data.labels y = val_data.labels
classes = val_data.classes_
self.pcc = PCC(self.learner) self.pcc = PCC(self.learner)
# estimate the matrix with entry (i,j) being the estimate of P(yi|yj), that is, the probability that a # estimate the matrix with entry (i,j) being the estimate of P(yi|yj), that is, the probability that a
# document that belongs to yj ends up being classified as belonging to yi # document that belongs to yj ends up being classified as belonging to yi
confusion = np.empty(shape=(data.n_classes, data.n_classes)) n_classes = len(classes)
for i,class_ in enumerate(data.classes_): confusion = np.empty(shape=(n_classes, n_classes))
for i, class_ in enumerate(classes):
confusion[i] = y_[y == class_].mean(axis=0) confusion[i] = y_[y == class_].mean(axis=0)
self.Pte_cond_estim_ = confusion.T self.Pte_cond_estim_ = confusion.T

12
quapy/method/neural.py
View File

@ -91,12 +91,14 @@ class QuaNetTrainer(BaseQuantifier):
train_data.instances = self.learner.transform(train_data.instances) train_data.instances = self.learner.transform(train_data.instances)
self.quantifiers = { self.quantifiers = {
'cc': CC(self.learner).fit(classifier_data, fit_learner=False), 'cc': CC(self.learner).fit(None, fit_learner=False),
'acc': ACC(self.learner).fit(classifier_data, fit_learner=False, val_split=valid_data), 'acc': ACC(self.learner).fit(None, fit_learner=False, val_split=valid_data),
'pcc': PCC(self.learner).fit(classifier_data, fit_learner=False), 'pcc': PCC(self.learner).fit(None, fit_learner=False),
'pacc': PACC(self.learner).fit(classifier_data, fit_learner=False, val_split=valid_data), 'pacc': PACC(self.learner).fit(None, fit_learner=False, val_split=valid_data),
'emq': EMQ(self.learner).fit(classifier_data, fit_learner=False), # 'emq': EMQ(self.learner).fit(classifier_data, fit_learner=False),
} }
if classifier_data is not None:
self.quantifiers['emq'] = EMQ(self.learner).fit(classifier_data, fit_learner=False)
self.status = { self.status = {
'tr-loss': -1, 'tr-loss': -1,