TypeError：__init__()在桑坦德客户事务数据库中为参数'scoring‘获取了多个值

Question

我正试图在Santander Customer Transaction database上运行LOFOImportance，我遇到了以下错误：

import pandas as pd
from sklearn.model_selection import KFold
from lofo.lofo_importance import LOFOImportance
from sklearn.metrics import roc_auc_score

df_Train.sort_values("target", inplace=True)

cv = KFold(n_splits=4, shuffle=False, random_state=42)
target = "target"
features = [col for col in df_Train.columns if col != target]

lofo = LOFOImportance(df_Train, features, target, cv=cv, scoring = 'roc_auc')
importance_df = lofo.get_importance()

有没有人遇到过同样的问题？

Answer 1

如果你看一下LOFOImportance.__init__的signature，你会发现第二个位置参数是scoring

def __init__(self, dataset, scoring, model=None, fit_params=None, cv=4, n_jobs=None):

因此你的代码

lofo = LOFOImportance(df_Train, features, target, cv=cv, scoring = 'roc_auc')

为scoring提供两个差异值(如错误所述)：一个作为位置参数features，另一个作为关键字参数，即字符串roc_auc。