def gradient_boosting():
kfold = model_selection.StratifiedKFold(n_splits=10, random_state=True)
scoreings = []
for train_index, test_index in kfold.split(X, y):
# print("Train:", train_index, "Validation:", test_index)
X_t, X_test = X[train_index], X[test_index]
y_t, y_test = y[train_index], y[test_index]
GSMOTE = EGSmote()
X_train, y_train = GSMOTE.fit_resample(X_t, y_t)
gbc = GradientBoostingClassifier(n_estimators=100,
learning_rate=0.01,
max_depth=3)
gbc.fit(X_train, y_train)
# Predicting the Test set results
y_predict = gbc.predict(X_test)
y_pred = np.where(y_predict > 0.5, 1, 0)
scoreings.append(evaluate2(y_test, y_pred))
scoreings = np.asarray(scoreings)
fscores = scoreings[:, 0]
gmean = scoreings[:, 1]
auc = scoreings[:, 2]
return ["GBC", fscores.mean(), gmean.mean(), auc.mean()]
def logistic_training():
kfold = model_selection.StratifiedKFold(n_splits=10, random_state=True)
scoreings = []
for train_index, test_index in kfold.split(X, y):
# print("Train:", train_index, "Validation:", test_index)
X_t, X_test = X[train_index], X[test_index]
y_t, y_test = y[train_index], y[test_index]
GSMOTE = EGSmote()
X_train, y_train = GSMOTE.fit_resample(X_t, y_t)
regressor = LogisticRegression(max_iter=120)
regressor.fit(X_train, y_train)
# Predicting the Test set results
y_predict = regressor.predict(X_test)
y_pred = np.where(y_predict > 0.5, 1, 0)
scoreings.append(evaluate2(y_test, y_pred))
scoreings = np.asarray(scoreings)
fscores = scoreings[:, 0]
gmean = scoreings[:, 1]
auc = scoreings[:, 2]
return ["LR", fscores.mean(), gmean.mean(), auc.mean()]
def KNN():
# Fitting Simple Linear Regression to the Training set
kfold = model_selection.StratifiedKFold(n_splits=10, random_state=True)
scoreings = []
for train_index, test_index in kfold.split(X, y):
# print("Train:", train_index, "Validation:", test_index)
X_t, X_test = X[train_index], X[test_index]
y_t, y_test = y[train_index], y[test_index]
GSMOTE = EGSmote()
X_train, y_train = GSMOTE.fit_resample(X_t, y_t)
classifier = KNeighborsClassifier(n_neighbors=5,
metric='minkowski',
p=2)
classifier.fit(X_train, y_train)
# Predicting the Test set results
y_pred = classifier.predict(X_test)
scoreings.append(evaluate2(y_test, y_pred))
scoreings = np.asarray(scoreings)
fscores = scoreings[:, 0]
gmean = scoreings[:, 1]
auc = scoreings[:, 2]
return ["KNN", fscores.mean(), gmean.mean(), auc.mean()]
def KNN(X_train, y_train, X_test, y_test):
classifier = KNeighborsClassifier(n_neighbors=5, metric='minkowski')
classifier.fit(X_train, y_train)
# Predicting the Test set results
y_pred = classifier.predict(X_test)
return evaluate2("KNN", y_test, y_pred)
def gradient_boosting(X_train, y_train, X_test, y_test):
gbc = GradientBoostingClassifier(n_estimators=100)
gbc.fit(X_train, y_train)
# Predicting the Test set results
y_predict = gbc.predict(X_test)
y_pred = np.where(y_predict > 0.5, 1, 0)
return evaluate2("GBC", y_test, y_pred)
def decision_tree(X_train, y_train, X_test, y_test):
regressor = DecisionTreeRegressor()
regressor.fit(X_train, y_train)
# Predicting the Test set results
y_predict = regressor.predict(X_test)
y_pred = np.where(y_predict > 0.5, 1, 0)
return evaluate2("DT", y_test, y_pred)
def logistic_training(X_train, y_train, X_test, y_test):
regressor = LogisticRegression()
regressor.fit(X_train, y_train)
# Predicting the Test set results
y_predict = regressor.predict(X_test)
y_pred = np.where(y_predict > 0.5, 1, 0)
return evaluate2("LR", y_test, y_pred)
def logistic_training(X_train, y_train, X_test, y_test):
regressor = LogisticRegression(max_iter=150, C=2, solver='liblinear')
regressor.fit(X_train, y_train)
# Predicting the Test set results
y_predict = regressor.predict(X_test)
y_pred = np.where(y_predict > 0.5, 1, 0)
cm = confusion_matrix(y_test, y_predict)
print(cm)
return evaluate2("LR", y_test, y_pred)
def XGBoost(X_train, y_train, X_test, y_test):
# Fitting X-Gradient boosting
gbc = xgb.XGBClassifier(objective="binary:logistic", random_state=42)
gbc.fit(X_train, y_train)
# Predicting the Test set results
y_predict = gbc.predict(X_test)
y_pred = np.where(y_predict.astype(int) > 0.5, 1, 0)
return evaluate2("XGBoost", y_test, y_pred)
def decision_tree(X_train, y_train, X_test, y_test):
regressor = DecisionTreeClassifier(criterion="entropy",
max_features="auto",
min_samples_leaf=0.00005)
regressor.fit(X_train, y_train)
# Predicting the Test set results
y_predict = regressor.predict(X_test)
y_pred = np.where(y_predict > 0.5, 1, 0)
cm = confusion_matrix(y_test, y_predict)
print(cm)
return evaluate2("DT", y_test, y_pred)
def XGBoost(X_train, y_train, X_test, y_test):
# Fitting X-Gradient boosting
#
gbc = xgb.XGBClassifier(scale_pos_weight=99,
missing=999999,
max_depth=3,
colsample_bytree=0.8)
eval_set = [(X_train, y_train), (X_test, y_test)]
gbc.fit(X_train,
y_train,
eval_metric="auc",
eval_set=eval_set,
verbose=True,
early_stopping_rounds=5)
# Predicting the Test set results
y_predict = gbc.predict(X_test)
y_pred = np.where(y_predict.astype(int) > 0.5, 1, 0)
# cm = confusion_matrix(y_test, y_pred)
# print(cm)
return evaluate2("XGBoost", y_test, y_pred)