← 返回题库
初级

分析income_to_loan特征与违约关系

未完成
初级参考 完整示例代码供参考，建议自己理解后重新输入 
def solve():
    from pyodide.http import open_url
    from io import StringIO
    loans_clean_csv = open_url("https://data.zuihe.com/dbd/riskctrl/state_03/loans_clean.csv").read()
    loans_featured_csv = open_url("https://data.zuihe.com/dbd/riskctrl/state_03/loans_featured.csv").read()
    iv_table_csv = open_url("https://data.zuihe.com/dbd/riskctrl/state_03/iv_table.csv").read()
    import pandas as pd
    from io import StringIO
    df = pd.read_csv(StringIO(loans_featured_csv))
    print(df['income_to_loan'].describe().round(2).to_string())
    df['itl_grp']=pd.qcut(df['income_to_loan'],q=5,labels=['Q1','Q2','Q3','Q4','Q5'],duplicates='drop')
    grp=df.groupby('itl_grp',observed=True).agg(count=('id','count'),default_rate=('isDefault','mean')).reset_index()
    grp['default_rate']=grp['default_rate'].round(4)
    print(grp.to_string(index=False))

示例

输入
solve()
期望输出
count    10000.00
mean         7.20
std          7.33
min          0.00
25%          3.48
50%          5.12
75%          8.08
max        154.85
itl_grp  count  default_rate
     Q1   2000        0.2865
     Q2   2000        0.2315
     Q3   2001        0.1769
     Q4   2005        0.1411
     Q5   1994        0.1179
Python 代码 🔒 登录后使用
🔒

登录后即可练习

注册免费账号，在浏览器中直接运行 Python 代码

登录 免费注册