kaggle员工离职预测——GBDT

一、比赛说明

比赛地址：https://www.kaggle.com/c/bi-attrition-predict

问题描述
数据包括员工的各种统计信息，以及该员工是否已经离职，统计的信息包括工资、出差、工作环境满意度、工作投入度、是否加班、是否升职、工资提升比例等。
需要通过训练数据得出员工离职预测的结果。

评分标准
AUC

二、代码

import pandas as pd
train = pd.read_csv('train.csv',index_col=0)
test = pd.read_csv('test.csv',index_col=0)

#print(train['Attrition'].value_counts())
# 处理Attrition字段
train['Attrition'] = train['Attrition'].map(lambda x:1 if x == 'Yes' else 0)
from sklearn.preprocessing import LabelEncoder
# 查看数据是否有空值
#print(train.isna().sum())

# 去掉没用的列 员工号码，标准工时（=80）
train = train.drop(['EmployeeNumber', 'StandardHours'], axis = 1)
test = test.drop(['EmployeeNumber', 'StandardHours'], axis = 1)

# 对于分类特征进行特征值编码
attr=['Age','BusinessTravel','Department','Education','EducationField','Gender','JobRole','MaritalStatus','Over18','OverTime']
lbe_list = []
for feature in attr:
    lbe=LabelEncoder()
    train[feature] = lbe.fit_transform(train[feature])
    test[feature] = lbe.transform(test[feature])
    lbe_list.append(lbe)
#train.to_csv('temp.csv')
#print(train)

from sklearn.ensemble import GradientBoostingRegressor
from sklearn.model_selection import train_test_split

X_train, X_valid, y_train, y_valid = train_test_split(train.drop('Attrition',axis=1), train['Attrition'], test_size=0.2, random_state=42)

# 采用回归算法，可以得到更好的AUC结果
model = GradientBoostingRegressor(random_state=10)
model.fit(X_train, y_train)
predict = model.predict(test)
print(predict)
test['Attrition'] = predict
#print(predict)
# 转化为二分类输出
#test['Attrition'] = test['Attrition'].map(lambda x:1 if x>=0.5 else 0)
test[['Attrition']].to_csv('submit_gbdt.csv')
print('submit_gbdt.csv saved')

三、分数

0.84436