easy-ml/feature/feature_utils.py

# -*- coding:utf-8 -*-
"""
@author: yq
@time: 2023/12/28
@desc:  特征工具类
"""
import numpy as np
import pandas as pd
import scorecardpy as sc
import toad as td
from sklearn.preprocessing import KBinsDiscretizer
from statsmodels.stats.outliers_influence import variance_inflation_factor as vif

from entitys import DataSplitEntity
from enums import BinsStrategyEnum


def f_get_bins(data: DataSplitEntity, feat: str, strategy: str = 'quantile', nbins: int = 10) -> pd.DataFrame:
    # 等频分箱
    if strategy == BinsStrategyEnum.QUANTILE.value:
        kbin_encoder = KBinsDiscretizer(n_bins=nbins, encode='ordinal', strategy='quantile')
        feature_binned = kbin_encoder.fit_transform(data[feat])
        return feature_binned.astype(int).astype(str)
    # 等宽分箱
    if strategy == BinsStrategyEnum.WIDTH.value:
        bin_width = (data.train_data()[feat].max() - data.train_data()[feat].min()) / nbins
        return pd.cut(data.train_data()[feat], bins=nbins, labels=[f'Bin_{i}' for i in range(1, nbins + 1)])
    # 使用toad分箱
    '''
    c = td.transfrom.Combiner()
    # method参数需要根据toad指定的几种方法名称选择
    c.fit(data, y = 'target', method = strategy, min_samples=None, n_bins = nbins, empty_separate = False)
    # 返回toad分箱combiner，用于训练集和测试集的分箱
    # 可使用c.export()[feature]查看某一特征的分箱临界值
    return c
    '''


# 此函数入参应为scorecardpy进行woebin函数转换后的dataframe
def f_get_bins_display(bins_info: pd.DataFrame) -> pd.DataFrame:
    df_list = []
    for col, bin_data in bins_info.items():
        tmp_df = pd.DataFrame(bin_data)
        df_list.append(tmp_df)
    result_df = pd.concat(df_list, ignore_index=True)
    total_bad = result_df['bad'].sum()
    total_cnt = result_df['count'].sum()
    # 整体的坏样本率
    br_overall = total_bad / total_cnt
    result_df['lift'] = result_df['badprob'] / br_overall
    result_df = \
        result_df.sort_values(['total_iv', 'variable'], ascending=False).set_index(['variable', 'total_iv', 'bin']) \
            [['count_distr', 'count', 'good', 'bad', 'badprob', 'lift', 'bin_iv', 'woe']]
    return result_df.style.format(subset=['count', 'good', 'bad'], precision=0).format(
        subset=['count_distr', 'bad', 'lift',
                'badprob', 'woe', 'bin_iv'], precision=4).bar(subset=['badprob', 'bin_iv', 'lift'],
                                                              color=['#d65f58', '#5fbb7a'])


# 此函数筛除变量分箱不单调或非U型的变量
def f_bins_filter(bins: pd.DataFrame, cols: list) -> list:
    result_cols = []
    # 遍历原始变量列表
    for tmp_col in cols:
        tmp_br = bins[tmp_col]['bad_prob'].values.tolist()
        tmp_len = len(tmp_br)
        if tmp_len <= 2:
            result_cols.append(tmp_col)
        else:
            tmp_judge = f_judge_monto(tmp_br)
            # f_judge_monto 函数返回1表示list单调，0表示非单调
            if tmp_judge:
                result_cols.append(tmp_col)
    return result_cols


# 此函数判断list的单调性，允许至多N次符号变化
def f_judge_monto(bd_list: list, pos_neg_cnt: int = 1) -> int:
    start_tr = bd_list[1] - bd_list[0]
    tmp_len = len(bd_list)
    pos_neg_flag = 0
    for i in range(2, tmp_len):
        tmp_tr = bd_list[i] - bd_list[i - 1]
        # 后一位bad_rate减前一位bad_rate，保证bad_rate的单调性
        # 记录符号变化, 允许 最多一次符号变化，即U型分布
        if (tmp_tr >= 0 and start_tr >= 0) or (tmp_tr <= 0 and start_tr <= 0):
            # 满足趋势保持，查看下一位
            continue
        else:
            # 记录一次符号变化
            start_tr = tmp_tr
            pos_neg_flag += 1
            if pos_neg_flag > pos_neg_cnt:
                return False
    # 记录满足趋势要求的变量
    if pos_neg_flag <= pos_neg_cnt:
        return True
    return False


def f_get_woe(data: DataSplitEntity, c: td.transform.Combiner, to_drop: list) -> pd.DataFrame:
    transer = td.transform.WOETransformer()
    # 根据训练数据来训练woe转换器，并选择目标变量和排除变量
    train_woe = transer.fit_transform(c.transform(data.train_data()), data.train_data()['target'],
                                      exclude=to_drop + ['target'])
    test_woe = transer.transform(c.transfrom(data.test_data()))
    oot_woe = transer.transform(c.transform(data.val_data()))
    return train_woe, test_woe, oot_woe


def f_get_psi(train_data: DataSplitEntity, oot_data: DataSplitEntity) -> pd.DataFrame:
    # 计算前，先排除掉不需要的cols
    return td.metrics.PSI(train_data, oot_data)


def f_get_corr(data: pd.DataFrame, meth: str = 'spearman') -> pd.DataFrame:
    return data.corr(method=meth)


def f_get_ivf(data: pd.DataFrame) -> pd.DataFrame:
    if len(data.columns.to_list()) <= 1:
        return None
    vif_v = [vif(data.values, data.columns.get_loc(i)) for i in data.columns]
    vif_df = pd.DataFrame()
    vif_df["变量"] = data.columns
    vif_df['vif'] = vif_v
    return vif_df


def f_calcu_model_ks(data, y_column, sort_ascending):
    var_ks = data.groupby('MODEL_SCORE_BIN')[y_column].agg([len, np.sum]).sort_index(ascending=sort_ascending)
    var_ks.columns = ['样本数', '坏样本数']
    var_ks['好样本数'] = var_ks['样本数'] - var_ks['坏样本数']
    var_ks['坏样本比例'] = (var_ks['坏样本数'] / var_ks['样本数']).round(4)
    var_ks['样本数比例'] = (var_ks['样本数'] / var_ks['样本数'].sum()).round(4)
    var_ks['总坏样本数'] = var_ks['坏样本数'].sum()
    var_ks['总好样本数'] = var_ks['好样本数'].sum()
    var_ks['平均坏样本率'] = (var_ks['总坏样本数'] / var_ks['样本数'].sum()).round(4)
    var_ks['累计坏样本数'] = var_ks['坏样本数'].cumsum()
    var_ks['累计好样本数'] = var_ks['好样本数'].cumsum()
    var_ks['累计样本数'] = var_ks['样本数'].cumsum()
    var_ks['累计坏样本比例'] = (var_ks['累计坏样本数'] / var_ks['总坏样本数']).round(4)
    var_ks['累计好样本比例'] = (var_ks['累计好样本数'] / var_ks['总好样本数']).round(4)
    var_ks['KS'] = (var_ks['累计坏样本比例'] - var_ks['累计好样本比例']).round(4)
    var_ks['LIFT'] = ((var_ks['累计坏样本数'] / var_ks['累计样本数']) / var_ks['平均坏样本率']).round(4)
    return var_ks.reset_index()


def f_get_model_score_bin(df, card, bins=None):
    train_score = sc.scorecard_ply(df, card, print_step=0)
    df['score'] = train_score
    if bins is None:
        _, bins = pd.qcut(df['score'], q=10, retbins=True, duplicates="drop")
        bins = list(bins)
        bins[0] = -np.inf
        bins[-1] = np.inf
    score_bins = pd.cut(df['score'], bins=bins)
    df['MODEL_SCORE_BIN'] = score_bins.astype(str).values
    return df, bins

def f_calcu_model_psi(df_train, df_test):
    tmp1 = df_train.groupby('MODEL_SCORE_BIN')['MODEL_SCORE_BIN'].agg(['count']).sort_index(ascending=True)
    tmp1['样本数比例'] = (tmp1['count'] / tmp1['count'].sum()).round(4)
    tmp2 = df_test.groupby('MODEL_SCORE_BIN')['MODEL_SCORE_BIN'].agg(['count']).sort_index(ascending=True)
    tmp2['样本数比例'] = (tmp2['count'] / tmp2['count'].sum()).round(4)
    psi = ((tmp1['样本数比例'] - tmp2['样本数比例']) * np.log(tmp1['样本数比例'] / tmp2['样本数比例'])).round(4)
    psi = psi.reset_index()
    psi = psi.rename(columns={"样本数比例": "psi"})
    psi['训练样本数'] = list(tmp1['count'])
    psi['测试样本数'] = list(tmp2['count'])
    psi['训练样本数比例'] = list(tmp1['样本数比例'])
    psi['测试样本数比例'] = list(tmp2['样本数比例'])
    return psi