如何解决训练集和测试集的分布差距过大问题

StratifiedKFold

可以借用sklearn中的StratifiedKFold来来实现K折交叉验证，同时根据标签中不同类别占比来进行拆分数据的，从而解决样本不均衡问题。

#!/usr/bin/python3
# -*- coding:utf-8 -*-
"""
@author: xcd
@file: StratifiedKFold-test.py
@time: 2021/1/26 10:14
@desc: 
"""

import numpy as np
from sklearn.model_selection import KFold, StratifiedKFold

X = np.array([
    [1, 2, 3, 4],
    [11, 12, 13, 14],
    [21, 22, 23, 24],
    [31, 32, 33, 34],
    [41, 42, 43, 44],
    [51, 52, 53, 54],
    [61, 62, 63, 64],
    [71, 72, 73, 74]
])

y = np.array([1, 1, 1, 1, 1, 1, 0, 0])

sfolder = StratifiedKFold(n_splits=4, random_state=0, shuffle=True)
folder = KFold(n_splits=4, random_state=0, shuffle=False)

for train, test in sfolder.split(X, y):
    print(train, test)

print("-------------------------------")
for train, test in folder.split(X, y):
    print(train, test)

for fold, (train_idx, val_idx) in enumerate(sfolder.split(X, y)):
    train_set, val_set = X[train_idx], X[val_idx]

跟KFold有明显的对比，StratifiedKFold用法类似Kfold，但是他是分层采样，确保训练集，测试集中各类别样本的比例与原始数据集中相同。

###
Parameters

n_splits : int, default=3
Number of folds. Must be at least 2.

shuffle : boolean, optional
Whether to shuffle each stratification of the data before splitting into batches.

random_state :
int, RandomState instance or None, optional, default=None

If int, random_state is the seed used by the random number generatorIf RandomState instance, random_state is the random number generator;

If None, the random number generator is the RandomState instance used

by `np.random`. Used when ``shuffle`` == True.
###