dataframe 里面要有 created_at 一列，格式 %Y-%m-%d %H:%M:%S，首先提取出小时，然后分层（组）抽样，保存到 csv 中，话不多说，上代码～

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
# -*- coding: utf-8 -*-
# author:           inspurer(月小水长)
# create_time:      2022/4/2 22:58
# 运行环境           Python3.6+
# github            https://github.com/inspurer
# 微信公众号         月小水长

import pandas as pd

input_file = 'RussiaUkraine1.csv'
output_file = 'RussiaUkraine2.csv'

df = pd.read_csv(input_file)
# 新增一列 hour
df['hour'] = pd.to_datetime(df['created_at']).dt.hour  # 时间
# 抽样比例 1%
res_df = df.groupby(df['hour']).apply(lambda x: x.sample(frac=0.01))
res_df.to_csv(output_file, index=False, encoding='utf-8-sig')