GridSearchCV 与 RandomizedSearchCV 用法注意随机搜索可以自定义打分 or loss 函数

在一些情况下，sklearn中没有现成的评价函数，sklearn是允许我们自己的定义的，但需要注意格式，接下来给个例子

import numpy as np
from sklearn.metrics import make_scorer
 
def logloss(act, pred):
    epsilon = 1e-15
    pred = sp.maximum(epsilon, pred)
    pred = sp.minimum(1-epsilon, pred)
    ll = sum(act*sp.log(pred) + sp.subtract(1, act)*sp.log(sp.subtract(1, pred)))
    ll = ll * -1.0/len(act)
    return ll
 
#这里的greater_is_better参数决定了自定义的评价指标是越大越好还是越小越好
loss  = make_scorer(logloss, greater_is_better=False)
score = make_scorer(logloss, greater_is_better=True)

https://blog.csdn.net/juezhanangle/article/details/80051256