如何在 pandas 中实现内在联系 [英] How to achieve an inner join in pandas

问题描述

我需要有效地执行用Python实现的内部联接.

I need to effectively do an inner join implemented in Python.

我有2个数据集，它们来自不同的来源，但是共享一个公用密钥.

I have 2 data sets which come from separate sources but share a common key.

(为了争辩)让我们说他们看起来像这样:

Lets say (for the sake of argument) that they look like this:

person_likes = [{'person_id': '1', 'food': 'ice_cream', 'pastimes': 'swimming'},
                {'person_id': '2', 'food': 'paella', 'pastimes': 'banjo'}]

person_accounts = [{'person_id': '1', 'blogs': ['swimming digest', 'cooking puddings']},
                   {'person_id': '2', 'blogs': ['learn flamenca']}]

如何最好地合并这两套数据.我有这样的东西:

How best can I join these two sets of data. I have something like this:

joins = []
for like in person_likes:
    for acc in person_accounts:
        if like['person_id'] == acc['person_id']:
            join = {}
            join.update(like)
            join.update(acc)
            joins.append(join)

print(joins)

这看起来很好用(我还没有对它进行广泛的测试)，乍一看看起来我们可以做的最好-但是我想知道是否有一种已知的算法可以实现更高的性能，以及是否还有更多的算法可以惯用还是Pythonic的方式?

This appears to work fine (I haven't tested it extensively), and at first glance looks like the best we can do - but I wonder if there is a know algorithm which is more performant and also if there is a more idiomatic or Pythonic way of doing this?

推荐答案

在这里，熊猫似乎是一个显而易见的答案.

Pandas seems like an obvious answer here.

import pandas as pd
accounts = pd.DataFrame(person_accounts)
likes = pd.DataFrame(person_likes)
pd.merge(accounts, likes, on='person_id')

                             blogs person_id       food  pastimes
# 0  [swimming digest, cooking puddings]         1  ice_cream  swimming
# 1                     [learn flamenca]         2     paella     banjo

这篇关于如何在 pandas 中实现内在联系的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！