数据清洗

任务2 数据清洗

def preprocessingData(data):
    # 填充缺失值
    data['rentType'][data['rentType'] == '--'] = '未知方式'
    
    # 转换object类型数据
    columns = ['rentType','communityName','houseType', 'houseFloor', 'houseToward', 'houseDecoration',  'region', 'plate']
    
    for feature in columns:
        data[feature] = LabelEncoder().fit_transform(data[feature])

    # 将buildYear列转换为整型数据
    buildYearmean = pd.DataFrame(data[data['buildYear'] != '暂无信息']['buildYear'].mode())
    data.loc[data[data['buildYear'] == '暂无信息'].index, 'buildYear'] = buildYearmean.iloc[0, 0]
    data['buildYear'] = data['buildYear'].astype('int')

    # 处理pv和uv的空值
    data['pv'].fillna(data['pv'].mean(), inplace=True)
    data['uv'].fillna(data['uv'].mean(), inplace=True)
    data['pv'] = data['pv'].astype('int')
    data['uv'] = data['uv'].astype('int')

    # 分割交易时间
    def month(x):
        month = int(x.split('/')[1])
        return month
    def day(x):
        day = int(x.split('/')[2])
        return day
    data['month'] = data['tradeTime'].apply(lambda x: month(x))
    data['day'] = data['tradeTime'].apply(lambda x: day(x))
    
    # 去掉部分特征
    data.drop('city', axis=1, inplace=True)
    data.drop('tradeTime', axis=1, inplace=True)
    data.drop('ID', axis=1, inplace=True)
    return data

data_train = preprocessingData(data_train)