def run_LDA(X, y, num_components):
lda = sklearn.lda.LDA(n_components=num_components)
lda_array = lda.fit(X, y).transform(X)
return lda, lda_array
def _calculate(self, X, y, categorical):
import sklearn.lda
if len(y.shape) == 1 or y.shape[1] == 1:
kf = sklearn.cross_validation.StratifiedKFold(y, n_folds=10)
else:
kf = sklearn.cross_validation.KFold(y.shape[0], n_folds=10)
accuracy = 0.
try:
for train, test in kf:
lda = sklearn.lda.LDA()
if len(y.shape) == 1 or y.shape[1] == 1:
lda.fit(X[train], y[train])
else:
lda = OneVsRestClassifier(lda)
lda.fit(X[train], y[train])
predictions = lda.predict(X[test])
accuracy += sklearn.metrics.accuracy_score(predictions, y[test])
return accuracy / 10
except scipy.linalg.LinAlgError as e:
self.logger.warning("LDA failed: %s Returned 0 instead!" % e)
return np.NaN
except ValueError as e:
self.logger.warning("LDA failed: %s Returned 0 instead!" % e)
return np.NaN
def _calculate(self, X, y, categorical):
import sklearn.lda
if len(y.shape) == 1 or y.shape[1] == 1:
kf = sklearn.cross_validation.StratifiedKFold(y, n_folds=10)
else:
kf = sklearn.cross_validation.KFold(y.shape[0], n_folds=10)
accuracy = 0.
try:
for train, test in kf:
lda = sklearn.lda.LDA()
if len(y.shape) == 1 or y.shape[1] == 1:
lda.fit(X[train], y[train])
else:
lda = OneVsRestClassifier(lda)
lda.fit(X[train], y[train])
predictions = lda.predict(X[test])
accuracy += sklearn.metrics.accuracy_score(
predictions, y[test])
return accuracy / 10
except scipy.linalg.LinAlgError as e:
self.logger.warning("LDA failed: %s Returned 0 instead!" % e)
return np.NaN
except ValueError as e:
self.logger.warning("LDA failed: %s Returned 0 instead!" % e)
return np.NaN
def _calculate(self, X, y, categorical):
import sklearn.neighbors
kf = sklearn.cross_validation.StratifiedKFold(y, n_folds=10)
accuracy = 0.
for train, test in kf:
lda = sklearn.neighbors.KNeighborsClassifier(1)
lda.fit(X[train], y[train])
predictions = lda.predict(X[test])
accuracy += sklearn.metrics.accuracy_score(predictions, y[test])
return accuracy / 10
def _calculate(self, X, y, categorical):
import sklearn.lda
kf = sklearn.cross_validation.StratifiedKFold(y, n_folds=10)
accuracy = 0.
try:
for train, test in kf:
lda = sklearn.lda.LDA()
lda.fit(X[train], y[train])
predictions = lda.predict(X[test])
accuracy += sklearn.metrics.accuracy_score(predictions, y[test])
return accuracy / 10
except scipy.linalg.LinAlgError as e:
logging.warning("LDA failed: %s Returned 0 instead!" % e)
return 0
except ValueError as e:
logging.warning("LDA failed: %s Returned 0 instead!" % e)
return 0
def run_LDA(X, y, num_components):
# 3
# LDA가 PCA방식으로 차원을 낮추는 것 보다 더 성능이 더 좋다.
# 각 클래스 정보에 맞는 방법을 찾기 때문.(PCA는 이와 관계 없이 일괄적이다.)
X = np.array(X)
y = np.array(y)
lda = sklearn.lda.LDA(n_components=num_components)
lda_array = lda.fit(X, y).transform(X)
return lda, lda_array
def run_LDA(X, y, num_components):
# 2
lda = sklearn.lda.LDA(n_components=num_components)
lda_array = lda.fit(X, y).transform(X)
return lda, lda_array
def run_LDA(X, y, num_components):
# 3
lda = sklearn.lda.LDA(n_components=num_components) # 축소할 차원 set
lda_array = lda.fit(X, y).transform(X) # DF-data 와 클래스 정보까지 고려하여 fitting 시킨다.
return lda, lda_array
