来自带有逗号分隔分类值的熊猫的 Swarmplot答案

【问题标题】：Swarmplot from pandas with comma separated categorical values来自带有逗号分隔分类值的熊猫的 Swarmplot
【发布时间】：2021-11-08 12:15:27
【问题描述】：

我正在使用一个数据集，该数据集在其列中包含逗号分隔的分类响应，并试图绘制一个类似于下面的散点图。我认为正确的方法可能是使用 seaborn swarmplot，例如：ax = sns.swarmplot(x="id", y="family", data=dftest) 但结果与我想要的相差甚远。我正在寻找一些关于如何重建类似情节的指示？

样本数据：

id	family	city	nation	world
1	next week, next few years	next week, next few years	lifetime	children's lifetime
2	next few years, lifetime	children's lifetime	next few years	next week, next few years, children's lifetime
3	next week, next few years	lifetime	children's lifetime	children's lifetime
4	next week, next few years	next few years	next week, next few years, lifetime,	next week, next few years, children's lifetime
5	next few years, lifetime	children's lifetime	children's lifetime	lifetime

figure source: Meadows, D., Meadows, D., Randers, J. and Behrens, W., 1972. The Limits to growth. A report for the club of Rome's project on the predicament of mankind. 1st ed. New York: Universe Books.

【问题讨论】：

标签： python pandas matplotlib seaborn

【解决方案1】：

据我所知，图像中呈现的预期输出很难用 python 的图形库实现，所以我试图模仿 seaborn 的 swarmplot 的预期输出。我希望这篇文章能从经验丰富的工程师那里得到一些很好的答案。

import pandas as pd
import numpy as np
import io

data = '''
id family city nation world
1 "next week,next few years" "next week,next few years" lifetime "children's lifetime"
2 "next few years,lifetime" "children's lifetime" "next few years" "next week,next few years,children's lifetime"
3 "next week,next few years" lifetime "children's lifetime" "children's lifetime"
4 "next week,next few years" "next few years" "next week,next few years,lifetime," "next week,next few years,children's lifetime"
5 "next few years,lifetime" "children's lifetime" "children's lifetime" lifetime
'''
df = pd.read_csv(io.StringIO(data), delim_whitespace=True)
dfs = pd.concat([
    df['family'].str.split(',', expand=True).rename(columns={0:'family',1:'family'}),
    df['city'].str.split(',', expand=True).rename(columns={0:'city',1:'city'}),
    df['nation'].str.split(',', expand=True).rename(columns={0:'nation',1:'nation',2:'nation'}),
    df['world'].str.split(',', expand=True).rename(columns={0:'world',1:'world',2:'world'})
], axis=1)

dfs = dfs.unstack().to_frame(name='TIME').reset_index().rename(columns={'level_0':'SPACE'})
dfs['TIME'].replace('', np.NaN, inplace=True)
dfs.dropna(subset=['TIME'], inplace=True)

import seaborn as sns

ax = sns.swarmplot(x="TIME", y="SPACE", data=dfs, size=15)
ax.grid(which='minor', axis='both')
print(ax.get_ylim())
ax.set_ylim(-0.5, 3.5)

for y_pos in [0.5, 1.5, 2.5]:
    ax.axhline(y=y_pos, color='k', lw=0.8)

for x_pos in [0.5, 1.5, 2.5]:
    ax.axvline(x=x_pos, color='k', lw=0.8)

【讨论】：