逻辑回归实践[1]

逻辑回归实践[1]

1数据处理

1 导入数据： https://pan.baidu.com/s/1wO9qJRjnrm8uhaSP67K0lw
说明：这份数据集是金融数据（非原始数据，已经处理过了），我们要做的是预测贷款用户是否会逾期。表格中 “status” 是结果标签：0 表示未逾期，1 表示逾期

• 分析——原始大小为（4754,90）

  • 将特征缺失太多的样本删除data.dropna(thesh=50)——（4476,90）
  • 将数据分为data_num和data_obj——利用data.info()
  • 对data_num和data_obj分别分析——data.describe()

    data_obj.drop(['trade_no','bank_card_no','source','trade_no','id_name'], axis=1, inplace=True)
    data_num.drop(['custid', 'student_feature', 'Unnamed: 0'], axis=1, inplace=True)
    

  • 填充和转换
  • 保存数据——（4476,89）

• 遇到问题分析

  1. 读中文乱码（加入encoding=‘gbk’）
  2. 删除样本后，保留的原来的索引，利用pd.contact(), 注意index是否相同

• 代码

  import pandas as pd
  from sklearn.preprocessing import LabelBinarizer,Imputer
  
  '''==================
  1.读取数据
  '''
  data = pd.read_csv('data.csv',encoding='gbk')     #excel和csv的中文存错格式是GBK
  
  '''
  1_1.删除确实样本和重复样本
  '''
  data_del = data.dropna(thresh=50)
  data_del.drop_duplicates(inplace=True)   #移除重复项
  
  '''
  1_2.分析，将数据分为data_num和data_obj
  '''
  object_column = ['trade_no','bank_card_no','reg_preference_for_trad',  'source','id_name', 'latest_query_time', 'loans_latest_time']
  data_obj = data_del[object_column]
  data_num = data_del.drop(object_column,axis=1)
  '''
  1_3.删除一些特征
  '''
  data_obj.drop(['trade_no','bank_card_no','source','trade_no','id_name'], axis=1, inplace=True)
  data_num.drop(['custid', 'student_feature', 'Unnamed: 0'], axis=1, inplace=True)
  
  '''
  1_4.对于数值型数据，采取均值填充
  '''
  imputer =  Imputer(strategy='mean')
  num = imputer.fit_transform(data_num)
  data_num = pd.DataFrame(num, index=data_num.index,columns=data_num.columns)
  
  '''
  1_5.对于非数值数据，采取前值填充
  '''
  data_obj.fillna(method='ffill',inplace=True)
  
  '''======================================================
  1_6.将reg_preference_for_trad转为数值型特征        one-hot编码
  '''
  encoder = LabelBinarizer()
  reg_preference_1hot = encoder.fit_transform(data_obj['reg_preference_for_trad'])
  
  data_obj.drop(['reg_preference_for_trad'], axis=1, inplace=True)
  reg_preference_df = pd.DataFrame(reg_preference_1hot, index=data_obj.index,columns=encoder.classes_)
  data_obj = pd.concat([data_obj, reg_preference_df], axis=1)
  
  '''
  1_7.将latest_query_time和loans_time 两种日期型数据拆分为月份和星期两个特征
  '''
  data_obj['latest_query_time'] = pd.to_datetime(data_obj['latest_query_time'])
  data_obj['latest_query_time_month'] = data_obj['latest_query_time'].dt.month
  data_obj['latest_query_time_weekday'] = data_obj['latest_query_time'].dt.weekday
  data_obj['loans_latest_time'] = pd.to_datetime(data_obj['loans_latest_time'])
  data_obj['loans_latest_time_month'] = data_obj['loans_latest_time'].dt.month
  data_obj['loans_latest_time_weekday'] = data_obj['loans_latest_time'].dt.weekday
  
  data_obj = data_obj.drop(['latest_query_time', 'loans_latest_time'], axis=1)
  
  '''==========
  1_7.合并data_num,和data_obj
  '''
  data_processed = pd.concat([data_num, data_obj], axis=1)
  
  '''===========
  1_8.保存处理后的数据
  '''
  data_processed.to_csv('data_processed.csv', index=False)
  
  
  

2. 逻辑回归实践

• 在将数据标准化后，准确率提升了很多

  import pandas as pd
  from sklearn.model_selection import train_test_split
  from sklearn.linear_model import LogisticRegression
  from sklearn.metrics import f1_score
  from sklearn.preprocessing import StandardScaler
  
  '''------------------------------------------
  1 读取数据
  '''
  data = pd.read_csv('data_processed.csv',encoding='gbk') 
  
  '''-------------------------------------------
  1.1 划分训练集何验证集
  '''
  train, test = train_test_split(data, test_size=0.1, random_state=666)
  
  '''----------------------------------------
  1.2 获取标签
  '''
  y_train = train.status
  train.drop(['status'], axis=1, inplace=True)
  
  y_test = test.status
  test.drop(['status'], axis=1, inplace=True)
  
  '''
  1.3 数据标准化
  '''
  scaler = StandardScaler()
  train = pd.DataFrame(scaler.fit_transform(train),index=train.index, columns=test.columns)
  test = pd.DataFrame(scaler.fit_transform(test),index=test.index, columns=test.columns)
  
  '''----------------------------------------
  1.4 训练模型
  '''
  model = LogisticRegression(C=1, dual=True)
  model.fit(train, y_train)
  
  '''--------------------------------------
  1.5 预测模型
  '''
  y_test_pre = model.predict(test)
  
  '''----------------------------------------
  1.6 评分
  '''
  score_vail = f1_score(y_test, y_test_pre, average='macro')
  
  print('-------------------------------')
  print('验证分数：{}'.format(score_vail))
  

结果：