skip_first_row=21,
y_column_index=0,
assignedColumnNames=APS_FULL_COLUMNS,
missingSymbol='na',
needImpute=True,
dropOrNot=True)
X_test, y_test = load_data(ROOT_PATH + APS_TEST,
skip_first_row=21,
y_column_index=0,
assignedColumnNames=APS_FULL_COLUMNS,
missingSymbol='na',
needImpute=True,
dropOrNot=True)
y_train = to_binary_numeric(y_train, classNeg="neg")
y_test = to_binary_numeric(y_test, classNeg="neg")
randForestClf = RandomForestClassifier(n_estimators=50,
random_state=2333,
oob_score=True)
randForestClf.fit(X_train, y_train)
y_predict = randForestClf.predict(X_test)
falsePositiveRate, truePositiveRate, thresholds = roc_curve(
y_test, y_predict)
# compute Area Under the Curve (AUC) using the trapezoidal rule
area = auc(falsePositiveRate, truePositiveRate)
plt.plot(falsePositiveRate,
truePositiveRate,
color='red',
#
__author__ = 'Aaron Yang'
__email__ = '*****@*****.**'
__date__ = '10/4/2019 5:30 PM'
import matplotlib.pyplot as plt
import pandas as pd
import seaborn as sns
from ay_hw_4._global import ROOT_PATH, APS_TRAIN, APS_FULL_COLUMNS
from ay_hw_4.util_data import load_data, to_binary_numeric
if __name__ == "__main__":
pd.set_option('display.max_columns', 100)
X_data, y_data = load_data(ROOT_PATH + APS_TRAIN,
skip_first_row=21,
y_column_index=0,
assignedColumnNames=APS_FULL_COLUMNS,
missingSymbol='na',
needImpute=True,
dropOrNot=True)
y_data = to_binary_numeric(y_data, classNeg="neg")
data = pd.concat([y_data, X_data], axis=1)
correlation = data.corr()
fig = plt.figure(figsize=(20, 15))
sns.heatmap(correlation, vmin=-1, vmax=1, cmap=sns.color_palette("Blues"))
plt.show()
# 把dropOrNot打开 将报错， 因为数据中有10列存在NaN
log_tree = Classifier(classname="weka.classifiers.trees.LMT")
eval_smote_test_obj = Evaluation(smote_test_data)
eval_smote_test_obj.crossvalidate_model(classifier=log_tree,
data=smote_test_data,
num_folds=5,
rnd=Random(1))
print("SMOTE Test CV (5-folds) Error = %.2f%%" %
(eval_smote_test_obj.percent_incorrect))
print(eval_smote_test_obj.matrix())
print("=================\"Summary\"====================")
print(eval_smote_test_obj.summary())
log_tree.build_classifier(smote_test_data)
y_predict = eval_smote_test_obj.test_model(log_tree, smote_test_data)
y_test = to_binary_numeric(y_test.head(500), classNeg="neg")
falsePositiveRate, truePositiveRate, thresholds = roc_curve(
y_test, y_predict)
# compute Area Under the Curve (AUC) using the trapezoidal rule
area = auc(falsePositiveRate, truePositiveRate)
plt.plot(falsePositiveRate,
truePositiveRate,
color='red',
label='ROC = ' + str(area))
plt.plot([0, 1], [0, 1], linestyle='dotted')
plt.xlabel('False Positive Rate')
plt.ylabel('True Positive Rate')
plt.title('ROC & AUC (SMOTE Test)')
plt.legend()
log_tree = Classifier(classname="weka.classifiers.trees.LMT")
eval_smote_train_obj = Evaluation(smote_train_data)
eval_smote_train_obj.crossvalidate_model(classifier=log_tree,
data=smote_train_data,
num_folds=5,
rnd=Random(1))
print("SMOTE Train CV (5-folds) Error = %.2f%%" %
(eval_smote_train_obj.percent_incorrect))
print(eval_smote_train_obj.matrix())
print("=================\"Summary\"====================")
print(eval_smote_train_obj.summary())
log_tree.build_classifier(smote_train_data)
y_predict = eval_smote_train_obj.test_model(log_tree, smote_train_data)
y_train_smote = to_binary_numeric(y_train_smote, classNeg="neg")
falsePositiveRate, truePositiveRate, thresholds = roc_curve(y_train_smote,
y_predict,
pos_label=0)
# compute Area Under the Curve (AUC) using the trapezoidal rule
area = auc(falsePositiveRate, truePositiveRate)
plt.plot(falsePositiveRate,
truePositiveRate,
color='red',
label='ROC = ' + str(area))
plt.plot([0, 1], [0, 1], linestyle='dotted')
plt.xlabel('False Positive Rate')
plt.ylabel('True Positive Rate')
plt.title('ROC & AUC (SMOTE Train)')
export_train_data.to_csv(GENERATED_TRAIN_DATA_FILE_PATH, sep=',', index=False)
train_data = convert.load_any_file(filename=GENERATED_TRAIN_DATA_FILE_PATH)
train_data.class_is_first()
# load logistic model tree algorithm
log_tree = Classifier(classname="weka.classifiers.trees.LMT")
eval_train_obj = Evaluation(train_data)
eval_train_obj.crossvalidate_model(classifier=log_tree, data=train_data, num_folds=5, rnd=Random(1))
print("Train CV (10-folds) Error = %.2f%%" % (eval_train_obj.percent_incorrect))
print(eval_train_obj.matrix())
print("=================\"Summary\"====================")
print(eval_train_obj.summary())
log_tree.build_classifier(train_data)
y_predict = eval_train_obj.test_model(log_tree, train_data)
# y_train = np.array(np.where(y_train.head(500).to_numpy() == 'neg', 0, 1))
y_train = to_binary_numeric(y_train.head(500), classNeg="neg")
falsePositiveRate, truePositiveRate, thresholds = roc_curve(y_train, y_predict, pos_label=0)
# compute Area Under the Curve (AUC) using the trapezoidal rule
area = auc(falsePositiveRate, truePositiveRate)
plt.plot(falsePositiveRate, truePositiveRate, color='red', label='ROC = ' + str(area))
plt.plot([0, 1], [0, 1], linestyle='dotted')
plt.xlabel('False Positive Rate')
plt.ylabel('True Positive Rate')
plt.title('ROC & AUC (Train)')
plt.legend()
plt.show()
def count_neg_and_pos(y_data):
y_value = to_binary_numeric(y_data)
num_neg = np.count_nonzero(y_value)
return len(y_value) - num_neg, num_neg