def test_cv_validator_mc(self):
"""
It tests cross-validation for a multi-class TSVM-based estimator.
"""
expected_output = [100, 0.0]
eval_m = Validator(self.mc_X, self.mc_y, ('CV', 5), self.mc_clf)
eval_func = eval_m.choose_validator()
acc, std, _ = eval_func({'C1': 1, 'C2': 1})
self.assertEqual([acc, std], expected_output)
def test_tt_validator_mc(self):
"""
It tests train/test split for a multi-class TSVM-based estimator.
"""
expected_output = [100, 0.0]
eval_m = Validator(self.mc_X, self.mc_y, ('t_t_split', 0.3),
self.mc_clf)
eval_func = eval_m.choose_validator()
acc, std, _ = eval_func({'C1': 1, 'C2': 1})
self.assertEqual([acc, std], expected_output)
def test_save_model(self):
"""
It saves a TSVM-based model on the disk for test purpose.
"""
if not isfile(self.save_model_path):
tsvm_model = TSVM()
eval_m = Validator(self.X, self.y, ('CV', 5), tsvm_model)
save_model(eval_m, {'C1': 1, 'C2': 1}, self.save_model_path)
dataset = DataReader(data_path, sep_char, header)
shuffle_data = True
normalize_data = False
dataset.load_data(shuffle_data, normalize_data)
X, y, file_name = dataset.get_data()
# Step 2: Choose a TSVM-based estimator
kernel = 'RBF'
lstsvm_clf = LSTSVM(kernel=kernel)
# Step 3: Select a multi-class approach
ovo_lstsvm = OneVsOneClassifier(lstsvm_clf)
# Step 4: Evaluate the multi-class estimator using train/test split
eval_method = 't_t_split' # Train/Test split
test_set_size = 20 # 20% of samples
val = Validator(X, y, (eval_method, test_set_size), ovo_lstsvm)
eval_func = val.choose_validator()
# Hyper-parameters of the classifier
h_params = {'C1': 2**-2, 'C2': 2**-2, 'gamma': 2**-7}
acc, std, full_report = eval_func(h_params)
print("Accuracy: %.2f" % acc)
print(full_report)
data_path = '../../dataset/australian.csv'
sep_char = ',' # separtor character of the CSV file
header = True # Whether the dataset has header names.
dataset = DataReader(data_path, sep_char, header)
shuffle_data = True
normalize_data = False
dataset.load_data(shuffle_data, normalize_data)
X, y, file_name = dataset.get_data()
# Step 2: Choose a TSVM-based estimator
kernel = 'linear'
tsvm_clf = TSVM(kernel=kernel)
# Step 3: Evaluate the estimator using train/test split
eval_method = 't_t_split' # Train/Test split
test_set_size = 30 # 30% of samples
val = Validator(X, y, (eval_method, test_set_size), tsvm_clf)
eval_func = val.choose_validator()
# Hyper-parameters of the classifier
h_params = {'C1': 2**-3, 'C2': 2**-5}
acc, std, full_report = eval_func(h_params)
print("Accuracy: %.2f" % acc)
print(full_report)
# LIBTwinSVM: A Library for Twin Support Vector Machines
# Developers: Mir, A. and Mahdi Rahbar
# License: GNU General Public License v3.0
"""
A benchmark script for demonstrating the effectiveness of LIBTwinSVM in terms of
prediction accuracy.
"""
from libtsvm.preprocess import DataReader
from libtsvm.estimators import TSVM
from libtsvm.model_selection import Validator, grid_search
data_path = '../dataset/australian.csv'
dataset = DataReader(data_path, ',', True)
dataset.load_data(True, False)
X, y, _ = dataset.get_data()
tsvm_clf = TSVM(kernel='linear')
val = Validator(X, y, ('CV', 5), tsvm_clf)
eval_method = val.choose_validator()
params = {'C1': (-5, 5), 'C2': (-5, 5), 'gamma': None}
best_acc, best_acc_std, opt_params, _ = grid_search(eval_method, params)
print("Best accuracy: %.2f+-%.2f | Optimal parameters: %s" %
(best_acc, best_acc_std, str(opt_params)))
def test_choose_validator(self):
"""
It tests whether the right validator is returned.
"""
output = []
eval_m = Validator(self.bin_X, self.bin_y, ('CV', 5), self.bin_clf)
output.append(eval_m.choose_validator().__name__ \
== Validator.cv_validator.__name__)
eval_m = Validator(self.bin_X, self.bin_y, ('t_t_split', 0.3),
self.bin_clf)
output.append(eval_m.choose_validator().__name__ \
== Validator.tt_validator.__name__)
eval_m = Validator(self.mc_X, self.mc_y, ('CV', 5), self.mc_clf)
output.append(eval_m.choose_validator().__name__ \
== Validator.cv_validator_mc.__name__)
eval_m = Validator(self.mc_X, self.mc_y, ('t_t_split', 0.3),
self.mc_clf)
output.append(eval_m.choose_validator().__name__ \
== Validator.tt_validator_mc.__name__)
self.assertEqual(True, all(output))