def main():
data = dataset_block(load_cancer_dataset(), withOnes=False)
data_train, data_test = split(data)
reg_const = optimize_regularization(data)
theta = svm.train(data_train, c=reg_const)
stats = svm.test(data_test, theta)
print("precision:%6.2f\nrecall:%6.2f\nerror:%6.2f\nf1-score:%6.2f\n" %
(stats.precision(), stats.recall(), stats.error(), stats.f_score()))
print("regularization constant used: %6.2f\n" % reg_const)
def main():
data = dataset_block(load_cancer_dataset(), withOnes=False)
data_train, data_test = split(data)
reg_const = optimize_regularization(data)
theta = svm.train(data_train, c=reg_const)
stats = svm.test(data_test, theta)
print("precision:%6.2f\nrecall:%6.2f\nerror:%6.2f\nf1-score:%6.2f\n" %
(stats.precision(), stats.recall(), stats.error(), stats.f_score()))
print("regularization constant used: %6.2f\n" % reg_const)
def optimize_regularization(data):
reg_best, f1_best = 0, 0
data_train, data_test = split(data)
for d in range(-10, 40):
reg_current = 0.5 ** d
theta = svm.train(data_train, c=reg_current)
stats = svm.test(data_test, theta)
f1_current = stats.f_score()
if f1_best < f1_current:
reg_best, f1_best = reg_current, f1_current
return reg_best
def optimize_regularization(data):
reg_best, f1_best = 0, 0
data_train, data_test = split(data)
for d in range(-10, 40):
reg_current = 0.5**d
theta = svm.train(data_train, c=reg_current)
stats = svm.test(data_test, theta)
f1_current = stats.f_score()
if f1_best < f1_current:
reg_best, f1_best = reg_current, f1_current
return reg_best