def train_lda(data):
model = LinearDiscriminantAnalysis()
print('#### Linear Discriminant Analysis ####')
scores = cross_val_score(model,
data[0],
data[1],
cv=5,
scoring='f1_macro',
n_jobs=8)
print('f1 macro %s' % str(scores))
scores = cross_val_score(model,
data[0],
data[1],
cv=5,
scoring='precision_macro',
n_jobs=8)
print('precision macro %s' % str(scores))
scores = cross_val_score(model,
data[0],
data[1],
cv=5,
scoring='recall_macro',
n_jobs=8)
print('recall macro %s' % str(scores))
model.fit(data[0], data[1])
write_model(model.intercept_, model.coef_, 'lda')
def train_regression(data):
model = LogisticRegression(class_weight='balanced',
solver='liblinear',
max_iter=4000,
penalty='l2',
random_state=1)
print('#### Logistic Regression ####')
scores = cross_val_score(model,
data[0],
data[1],
cv=5,
scoring='f1_macro',
n_jobs=8)
print('f1 macro %s' % str(scores))
scores = cross_val_score(model,
data[0],
data[1],
cv=5,
scoring='precision_macro',
n_jobs=8)
print('precision macro %s' % str(scores))
scores = cross_val_score(model,
data[0],
data[1],
cv=5,
scoring='recall_macro',
n_jobs=8)
print('recall macro %s' % str(scores))
model.fit(data[0], data[1])
write_model(model.intercept_, model.coef_, 'regression')