一、比赛说明
比赛地址：/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 pdtrain = 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)import xgboost as xgbfrom sklearn.model_selection import train_test_splitparam = {'boosting_type':'gbdt','objective' : 'binary:logistic', #'eval_metric' : 'auc','eta' : 0.01,'max_depth' : 15,'colsample_bytree':0.8,'subsample': 0.9,'subsample_freq': 8,'alpha': 0.6,'lambda': 0,}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)train_data = xgb.DMatrix(X_train, label=y_train)valid_data = xgb.DMatrix(X_valid, label=y_valid)test_data = xgb.DMatrix(test)model = xgb.train(param, train_data, evals=[(train_data, 'train'), (valid_data, 'valid')], num_boost_round = 10000, early_stopping_rounds=200, verbose_eval=25)predict = model.predict(test_data)test['Attrition'] = predictprint(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_xgb.csv')
三、分数
0.83709
