python收入

relationship.sort_values(by=['relationship', 'counts'], ascending=False, inplace=True)

# 设置图框比例，并绘图

plt.figure(figsize=(9, 5))

sns.barplot(x='race', y='counts', hue='income', data=race)

plt.show()

plt.figure(figsize=(9, 5))

sns.barplot(x='relationship', y='counts', hue='income', data=relationship)

plt.show()

# 离散变量的重编码

for feature lumns:

if income[feature].dtype == 'object':

income[feature] = pd.Categorical(income[feature]).codes

print(income.head())

# 删除无意义变量

income.drop(['education', 'fnlwgt'], axis=1, inplace=True)

print(income.head())

del_selection import train_test_split

# 数据拆分

X_train, X_test, y_train, y_test = train_test_split(income.loc[:, 'age':'native-country'], income['income'],

train_size=0.75, test_size=0.25, random_state=1234)

print('训练数据集共有%d条观测' % X_train.shape[0])

print('训练数据集共有%d条观测' % X_test.shape[0])

# 导入 K 近邻模型的类

ighbors import KNeighborsClassifier

# 构建 K 近邻模型

kn = KNeighborsClassifier()

kn.fit(X_train, y_train)

print(kn)

# 导入 GBDT 模型的类

semble import GradientBoostingClassifier

# 构建 GBDT 模型

gbdt = GradientBoostingClassifier()

gbdt.fit(X_train, y_train)

print(gbdt)

# K 近邻模型的网格搜索法

# 导入网格搜索法的函数

id_search import GridSearchCV

# 选择不同的参数

k_options = list(range(1, 12))

parameters = {'n_neighbors': k_options}

# 搜索不同的 K 值

grid_kn = GridSearchCV(estimator=KNeighborsClassifier(), param_grid=parameters, cv=10, scoring='accuracy')

grid_kn.fit(X_train, y_train)

# 结果输出

print(id_scores_, grid_kn.best_params_, grid_kn.best_score_)

# GBDT 模型的网格搜索法

# 选择不同的参数

learning_rate_options = [0.01, 0.05, 0.1]

max_depth_options = [3, 5, 7, 9]

n_estimators_options = [100, 300, 500]

parameters = {'learning_rate': learning_rate_options, 'max_depth': max_depth_options,

'n_estimators': n_estimators_options}

grid_gbdt = GridSearchCV(estimator=GradientBoostingClassifier(), param_grid=parameters, cv=10, scoring='accuracy')

grid_gbdt.fit(X_train, y_train)

# 结果输出

print(id_scores_, grid_gbdt.best_params_, grid_gbdt.best_score_)

# 默认的 K 近邻模型

kn_pred = kn.predict(X_test)

sstab(kn_pred, y_test))

# 模型得分

print('模型在训练集上的准确率%f' % kn.score(X_train, y_train))

print('模型在测试集上的准确率%f' % kn.score(X_test, y_test))

# 导入模型评估模块

from sklearn import metrics

# 计算 ROC 曲线的 x 轴和 y 轴数据

fpr, tpr, _ = _curve(y_test, kn.predict_proba(X_test)[:, 1])

# 绘制 ROC 曲线

plt.plot(fpr, tpr, linestyle='solid', color='red')

# 添加阴影

plt.stackplot(fpr, tpr, color='steelblue')

# 绘制参考线

plt.plot([0, 1], [0, 1], linestyle='dashed', color='black')

# 往图中添加文本

plt.show()

# 预测测试集

grid_kn_pred = grid_kn.predict(X_test)

sstab(grid_kn_pred, y_test))

# 模型得分

print('模型在训练集上的准确率%f' % grid_kn.score(X_train, y_train))

print('模型在测试集上的准确率%f' % grid_kn.score(X_test, y_test))

# 绘制 ROC 曲线

fpr, tpr, _ = _curve(y_test, grid_kn.predict_proba(X_test)[:, 1])

plt.plot(fpr, tpr, linestyle='solid', color='red')

plt.stackplot(fpr, tpr, color='steelblue')

plt.plot([0, 1], [0, 1], linestyle='dashed', color='black')

plt.show()

# 默认的 GBDT 模型

gbdt_pred = gbdt.predict(X_test)

sstab(gbdt_pred, y_test))

# 模型得分

print('模型在训练集上的准确率%f' % gbdt.score(X_train, y_train))

print('模型在测试集上的准确率%f' % gbdt.score(X_test, y_test))

# 绘制 ROC 曲线

fpr, tpr, _ = _curve(y_test, gbdt.predict_proba(X_test)[:, 1])

plt.plot(fpr, tpr, linestyle='solid', color='red')

plt.stackplot(fpr, tpr, color='steelblue')

plt.plot([0, 1], [0, 1], linestyle='dashed', color='black')

plt.show()

# 网格搜索的 GBDT 模型

grid_gbdt_pred = grid_gbdt.predict(X_test)

sstab(grid_gbdt_pred, y_test))

# 模型得分

print('模型在训练集上的准确率%f' % grid_gbdt.score(X_train, y_train))

print('模型在测试集上的准确率%f' % grid_gbdt.score(X_test, y_test))

# 绘制 ROC 曲线

fpr, tpr, _ = _curve(y_test, grid_gbdt.predict_proba(X_test)[:, 1])

plt.plot(fpr, tpr, linestyle='solid', color='red')

plt.stackplot(fpr, tpr, color='steelblue')

plt.plot([0, 1], [0, 1], linestyle='dashed', color='black')

plt.show()