def func_main(MAX_SAMPLES):
#Defining the results array which ´will contain execution time and non-monotonicity score
resultArr = np.zeros((2, ))
#Reading the dataset
df = pd.read_csv('Datasets/CPU.csv')
#Applying monotonicity constraints
fileMon = open('monFeature.txt', 'w')
fileMon.write('MinMainMem \nMaxMainMem \nCachMem \nMinChan \nMaxChan')
fileMon.close()
data = df.values
X = data[:, :-1]
Y = data[:, -1]
model = AdaBoostClassifier(n_estimators=100, random_state=0)
#Fitting the model with the dataset
model = model.fit(X, Y)
#Computing time
start_time = time.time()
#Calling the random testing approach to test weak group monotonicity
cexPair, failedAtt = artGen.funcMainRanTest(model, df, 4, MAX_SAMPLES)
execTime = (time.time() - start_time)
return cexPair, failedAtt, execTime
def func_main(MAX_SAMPLES):
#Defining the results array which ´will contain execution time and non-monotonicity score
resultArr = np.zeros((2, ))
#Reading the dataset
df = pd.read_csv('Datasets/Automobile.csv')
#Applying monotonicity constraints
fileMon = open('monFeature.txt', 'w')
fileMon.write('SymbolingRisk \nWheelBase \nLength \nWidth \nHeight \nCurbWeight \nNumCylinders \nEngineSize \nHorsePow \nPeak-rpm ')
fileMon.close()
data = df.values
X = data[:, :-1]
Y = data[:, -1]
model = svm.LinearSVC(penalty='l2', loss='squared_hinge', dual=True, tol=0.0001, C=1.0, multi_class='ovr', fit_intercept=True, intercept_scaling=1, class_weight=None,
verbose=0, random_state=None, max_iter=1000)
#Fitting the model with the dataset
model = model.fit(X, Y)
#Computing time
start_time = time.time()
#Calling the random testing approach to test weak group monotonicity
cexPair, failedAtt = artGen.funcMainRanTest(model, df, 4, MAX_SAMPLES)
execTime = (time.time() - start_time)
return cexPair, failedAtt, execTime
def func_main(MAX_SAMPLES):
#Defining the results array which ´will contain execution time and non-monotonicity score
resultArr = np.zeros((2, ))
#Reading the dataset
df = pd.read_csv('Datasets/HousingData.csv')
#Applying monotonicity constraints
fileMon = open('monFeature.txt', 'w')
fileMon.write('RM \nRAD \nTAX')
fileMon.close()
data = df.values
X = data[:, :-1]
Y = data[:, -1]
model = GaussianNB()
#Fitting the model with the dataset
model = model.fit(X, Y)
#Computing time
start_time = time.time()
#Calling the random testing approach to test weak group monotonicity
cexPair, failedAtt = artGen.funcMainRanTest(model, df, 4, MAX_SAMPLES)
execTime = (time.time() - start_time)
return cexPair, failedAtt, execTime
def func_main(MAX_SAMPLES):
#Defining the results array which ´will contain execution time and non-monotonicity score
resultArr = np.zeros((2, ))
#Reading the dataset
df = pd.read_csv('Datasets/AutoMPG.csv')
#Applying monotonicity constraints
fileMon = open('monFeature.txt', 'w')
fileMon.write(
'Cylinders \nDisplacement \nHorsePower \nWeight \nAcceleration')
fileMon.close()
data = df.values
X = data[:, :-1]
Y = data[:, -1]
model = GradientBoostingClassifier(n_estimators=100,
learning_rate=1.0,
max_depth=1,
random_state=0)
#Fitting the model with the dataset
model = model.fit(X, Y)
#Computing time
start_time = time.time()
#Calling the random testing approach to test weak group monotonicity
cexPair, failedAtt = artGen.funcMainRanTest(model, df, 4, MAX_SAMPLES)
execTime = (time.time() - start_time)
return cexPair, failedAtt, execTime
def func_main(MAX_SAMPLES):
#Defining the results array which ´will contain execution time and non-monotonicity score
resultArr = np.zeros((2, ))
#Reading the dataset
df = pd.read_csv('Datasets/CPU.csv')
#Applying monotonicity constraints
fileMon = open('monFeature.txt', 'w')
fileMon.write('MinMainMem \nMaxMainMem \nCachMem \nMinChan \nMaxChan')
fileMon.close()
data = df.values
X = data[:, :-1]
Y = data[:, -1]
model = RandomForestClassifier(n_estimators=100, criterion='entropy', max_depth=None, min_samples_split=2, min_samples_leaf=1, min_weight_fraction_leaf=0.0, max_features='auto', max_leaf_nodes=None,
min_impurity_decrease=0.0, min_impurity_split=None, bootstrap=True, oob_score=False, n_jobs=1, random_state=None, verbose=0)
#Fitting the model with the dataset
model = model.fit(X, Y)
#Computing time
start_time = time.time()
#Calling the random testing approach to test weak group monotonicity
cexPair, failedAtt = artGen.funcMainRanTest(model, df, 4, MAX_SAMPLES)
execTime = (time.time() - start_time)
return cexPair, failedAtt, execTime
def func_main(MAX_SAMPLES):
#Defining the results array which ´will contain execution time and non-monotonicity score
resultArr = np.zeros((2, ))
#Reading the dataset
df = pd.read_csv('Datasets/AdultMod.csv')
#Applying monotonicity constraints
fileMon = open('monFeature.txt', 'w')
fileMon.write('Age \nEducation \nCapital-gain \nhours-per-week \n')
fileMon.close()
data = df.values
X = data[:, :-1]
Y = data[:, -1]
model = KNeighborsClassifier(n_neighbors=7, weights='uniform')
#Fitting the model with the dataset
model = model.fit(X, Y)
#Computing time
start_time = time.time()
#Calling the random testing approach to test weak group monotonicity
cexPair, failedAtt = artGen.funcMainRanTest(model, df, 4, MAX_SAMPLES)
execTime = (time.time() - start_time)
return cexPair, failedAtt, execTime
def func_main(MAX_SAMPLES):
#Defining the results array which ´will contain execution time and non-monotonicity score
resultArr = np.zeros((2, ))
#Reading the dataset
df = pd.read_csv('Datasets/Automobile.csv')
#Applying monotonicity constraints
fileMon = open('monFeature.txt', 'w')
fileMon.write(
'SymbolingRisk \nWheelBase \nLength \nWidth \nHeight \nCurbWeight \nNumCylinders \nEngineSize \nHorsePow \nPeak-rpm '
)
fileMon.close()
data = df.values
X = data[:, :-1]
Y = data[:, -1]
model = GaussianNB()
#Fitting the model with the dataset
model = model.fit(X, Y)
#Computing time
start_time = time.time()
#Calling the random testing approach to test weak group monotonicity
cexPair, failedAtt = artGen.funcMainRanTest(model, df, 4, MAX_SAMPLES)
execTime = (time.time() - start_time)
return cexPair, failedAtt, execTime
def func_main(MAX_SAMPLES):
#Reading the dataset
df = pd.read_csv('Datasets/AdultMod.csv')
#Applying monotonicity constraints
fileMon = open('monFeature.txt', 'w')
fileMon.write('Age \nEducation \nCapital-gain \nhours-per-week \n')
fileMon.close()
data = df.values
X = data[:, :-1]
Y = data[:, -1]
model = AdaBoostClassifier(n_estimators=100, random_state=0)
#Fitting the model with the dataset
model = model.fit(X, Y)
#Computing time
start_time = time.time()
#Calling the random testing approach to test weak group monotonicity
cexPair, failedAtt = artGen.funcMainRanTest(model, df, 4, MAX_SAMPLES)
execTime = (time.time() - start_time)
return cexPair, failedAtt, execTime
def func_main(MAX_SAMPLES):
#Reading the dataset
df = pd.read_csv('Datasets/Automobile.csv')
#Applying monotonicity constraints
fileMon = open('monFeature.txt', 'w')
fileMon.write('SymbolingRisk \nWheelBase \nLength \nWidth \nHeight \nCurbWeight \nNumCylinders \nEngineSize \nHorsePow \nPeak-rpm ')
fileMon.close()
data = df.values
X = data[:, :-1]
Y = data[:, -1]
model = AdaBoostClassifier(n_estimators=100, random_state=0)
#Fitting the model with the dataset
model = model.fit(X, Y)
#Computing time
start_time = time.time()
#Calling the random testing approach to test weak group monotonicity
cexPair, failedAtt = artGen.funcMainRanTest(model, df, 4, MAX_SAMPLES)
execTime = (time.time() - start_time)
return cexPair, failedAtt, execTime
def func_main(MAX_SAMPLES):
#Defining the results array which ´will contain execution time and non-monotonicity score
resultArr = np.zeros((2, ))
#Reading the dataset
df = pd.read_csv('Datasets/Mammographic.csv')
#Applying monotonicity constraints
fileMon = open('monFeature.txt', 'w')
fileMon.write('BI-RADS \nAge \nDensity')
fileMon.close()
data = df.values
X = data[:, :-1]
Y = data[:, -1]
model = LogisticRegression(penalty='l2', dual=False, tol=0.0001, C=1.0, fit_intercept=True, intercept_scaling=1, class_weight=None, random_state=None, solver='lbfgs', max_iter=100, verbose=0, warm_start=False, n_jobs=None)
#Fitting the model with the dataset
model = model.fit(X, Y)
#Computing time
start_time = time.time()
#Calling the random testing approach to test weak group monotonicity
cexPair, failedAtt = artGen.funcMainRanTest(model, df, 4, MAX_SAMPLES)
execTime = (time.time() - start_time)
return cexPair, failedAtt, execTime
def func_main(MAX_SAMPLES):
#Defining the results array which ´will contain execution time and non-monotonicity score
resultArr = np.zeros((2, ))
#Reading the dataset
df = pd.read_csv('Datasets/CarEvaluation.csv')
#Applying monotonicity constraints
fileMon = open('monFeature.txt', 'w')
fileMon.write('NumDoors \nNumPersons \nLugBoot \nSafety')
fileMon.close()
data = df.values
X = data[:, :-1]
Y = data[:, -1]
model = MLPClassifier(activation='relu', alpha=1e-05, batch_size='auto', beta_1=0.9, beta_2=0.999, early_stopping=False, epsilon=1e-08, hidden_layer_sizes=(500, 50), learning_rate='adaptive', learning_rate_init=0.001,
max_iter=200, momentum=0.9, nesterovs_momentum=True, power_t=0.5, random_state=1, shuffle=True, solver='lbfgs', tol=0.0001, validation_fraction=0.1, verbose=False, warm_start=False)
#Fitting the model with the dataset
model = model.fit(X, Y)
#Computing time
start_time = time.time()
#Calling the random testing approach to test weak group monotonicity
cexPair, failedAtt = artGen.funcMainRanTest(model, df, 4, MAX_SAMPLES)
execTime = (time.time() - start_time)
return cexPair, failedAtt, execTime
def func_main(constr, MAX_SAMPLES):
#Defining the results array which ´will contain execution time and non-monotonicity score
resultArr = np.zeros((2, ))
#Reading the dataset
df = pd.read_csv('Datasets/AdultMod.csv')
#Applying monotonicity constraints
fileMon = open('monFeature.txt', 'w')
fileMon.write('Age \nEducation \nCapital-gain \nhours-per-week \n')
noOfFe = 4
fileMon.close()
data = df.values
X = data[:, :-1]
Y = data[:, -1]
dfT = pd.read_csv('Datasets/AdultMod.csv')
dfT.drop('Class', axis=1, inplace=True)
feature_names = dfT.columns.values
feature_monotones = [0] * (len(feature_names))
with open('monFeature.txt') as f:
feArr = f.readlines()
feArr = [x.strip() for x in feArr]
#Adding monotonicity constraints
if(constr == 1):
for i in range(noOfFe):
for j in range(dfT.shape[1]):
if(feArr[i] == dfT.columns.values[j]):
feature_monotones[j] = 1
monotone_model = lgb.LGBMClassifier(min_child_samples=5, monotone_constraints=feature_monotones)
model = monotone_model.fit(data[:, :-1].reshape(len(X), len(feature_names)), Y)
#Computing time
start_time = time.time()
#Calling the random testing approach to test weak group monotonicity
cexPair, failedAtt = artGen.funcMainRanTest(model, df, 4, MAX_SAMPLES)
execTime = (time.time() - start_time)
return cexPair, failedAtt, execTime
