Skip to main content

model for loan risk controll,i.e. score card

Project description

bbbrisk-bbb风控包

本代码由bbbdata为评分卡专门开发的小贷风控包 评分卡教程和本代码的使用说明请参考:https://www.bbbdata.com

Install and Upgrade · 安装与升级

Pip

pip install bbbrisk # to install
pip install -U bbbrisk # to upgrade

Key features · 主要功能

  • 变量分箱与分析 例如变量自动分箱,如下:
bin_set  = bins.autoBin(x, y,enum_var=['city','marital'])    # 自动分箱,如果有枚举变量,必须指出哪些是枚举变量
bin_stat = bins.bin_stat(x,y,bin_set)                        # 统计各个变量的分箱情况

对单个变量使用自动分箱,如下:

bin_set  = bins.merge.chi2(x,y,bin_num=5,init_bin_num=10)    # 使用卡方分箱对变量进行分箱
bin_stat = bins.Bins(bin_set).binStat(x,y)                   # 统计分箱结果

对单个变量手动分箱,如下:

bin_set  = [['-',0],[0,0.37],[0.37,1.2],[1.2,2],[2,'+']]     # 手动设置变量的分箱
bin_stat = bins.Bins(bin_set).binStat(x,y)                   # 统计分箱结果

统计结果中就会打印出变量在各个分箱的badrate、iv值等

  • 构建评分卡 设置好分箱后,就可以将数据构建构建评分卡,如下:
model,card = model.scoreCard(x,y,bin_set)                    # 构建评分卡
card.featureScore                                            # 评分卡-特征得分表
card.baseScore                                               # 评分卡-基础
model.w                                                      # 逻辑回归模型的变量权重
model.b                                                      # 逻辑回归模型的阈值
  • 打印相关报告 例如打印阈值表、分数分布图等,如下:
thd_tb    = report.get_threshold_tb(score,y,bin_step=10)     # 阈值表
report.draw_score_disb(score,y,bin_step=10)                  # 分数分布

Documents · 文档

更新日志

  • v1.0.0: 添加数据
  • v1.0.0: 初步上传

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

bbbrisk-1.0.1.tar.gz (669.0 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

bbbrisk-1.0.1-py3-none-any.whl (708.9 kB view details)

Uploaded Python 3

File details

Details for the file bbbrisk-1.0.1.tar.gz.

File metadata

  • Download URL: bbbrisk-1.0.1.tar.gz
  • Upload date:
  • Size: 669.0 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.0.1 CPython/3.12.7

File hashes

Hashes for bbbrisk-1.0.1.tar.gz
Algorithm Hash digest
SHA256 2caa3111d0753f8e69848d18c325afb569857cfb19a0151f09031e87ba5bc27e
MD5 a328ed8d568878b1a4e14ca8d5eb314c
BLAKE2b-256 abfd9548a891eeb923ba0b7e4fba5c9d7b853b8d85e3c6af0bccfec77ed4087d

See more details on using hashes here.

File details

Details for the file bbbrisk-1.0.1-py3-none-any.whl.

File metadata

  • Download URL: bbbrisk-1.0.1-py3-none-any.whl
  • Upload date:
  • Size: 708.9 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.0.1 CPython/3.12.7

File hashes

Hashes for bbbrisk-1.0.1-py3-none-any.whl
Algorithm Hash digest
SHA256 47b9923ad1a5d9683c77e66ba2a5ccbab43005c38aad6bca0d59701f760bf82b
MD5 a1162b3d0f235bcc4b173a43c04f6ae2
BLAKE2b-256 cc76375cf0f6dc15a501dc85f1aa3379b10052b48405f37bd5cb524f541e4de6

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page