如何在python的一列中插入2个图

Posted 2023-02-18

【中文标题】如何在python的一列中插入2个图

【英文标题】：How to insert 2 graphs in one column on python

【发布时间】：2021-12-04 07:14:20

【问题描述】：

现在我正在学习 python。我的第一个问题是“如何在 python 的一列中插入 2 个图表”。请帮帮我

我的代码是：使用情节。

fig0.update_traces(mode='markers', marker_line_width=1, marker_size=5)
fig0.update_layout(title='Confirmed case by gender',yaxis_zeroline=False, xaxis_zeroline=False)
fig0.show()

fig1 = px.line(timegender, x='date', y='deceased', color='sex', markers = True)
fig1.update_traces(mode='markers', marker_line_width=1, marker_size=5)
fig1.update_layout(title='Deceased case by gender',yaxis_zeroline=False, xaxis_zeroline=False)
fig1.show()

【问题讨论】：

两张图一栏？很难解释这一点 - 1. 一个活动仪表板，即 dash 2. 使用同一轴的多个跟踪，其中列是 xaxis 上的类别，即一个图中的多个跟踪 3. 多组轴，即子地块

【参考方案1】：

您显然在按性别划分 COVID 病例。已获取数据来证明这一点 绘图的关键是结构化数据框

使 sex 列包含 male 或 female 使 measure 列包含 cases 或 deaths

数据是针对多个国家/地区的，因此已过滤到仅 USA 用于绘图 然后生成图表的行或列很简单

import requests
import io
import pandas as pd
import plotly.express as px

# fmt: off
# sex disaggregation data
dfsda = pd.read_csv(io.StringIO(requests.get("https://globalhealth5050.org/?_covid-data=datasettable&_extype=csv").text))
# case / death data
dfall = pd.read_csv(io.StringIO(
    requests.get("https://raw.githubusercontent.com/owid/covid-19-data/master/public/data/owid-covid-data.csv").text))
dfall["date"] = pd.to_datetime(dfall["date"])
dfcc = pd.read_html("https://www.iban.com/country-codes")[0]


# data prep - join three data sources
df = dfsda.merge(dfcc, left_on="Country code", right_on="Alpha-2 code").merge(
    dfall, left_on="Alpha-3 code", right_on="iso_code"
)

# calc split between male & female
df2 = (
    df.loc[:,['Cases (% female)', 'Deaths (% female)', 'Alpha-2 code', 'Alpha-3 code', 'location', 'date', 'total_cases', 'total_deaths']]
    .dropna()
    .assign(
        **"Cases (% female)": lambda d: d["Cases (% female)"].str.strip("%").astype(float) / 100,
            "Deaths (% female)": lambda d: d["Deaths (% female)"].str.strip("%").astype(float) / 100,
            "cases_male": lambda d: d["total_cases"] * (1 - d["Cases (% female)"]),
            "cases_female": lambda d: d["total_cases"] * (d["Cases (% female)"]),
            "deaths_male": lambda d: d["total_deaths"] * (1 - d["Deaths (% female)"]),
            "deaths_female": lambda d: d["total_deaths"] * (d["Deaths (% female)"]),
        
    )
)
# fmt: on

# structure for plotting
df3 = (
    pd.wide_to_long(
        df2,
        i=["Alpha-3 code", "location", "date"],
        stubnames=["cases", "deaths"],
        j="sex",
        sep="_",
        suffix=r"\w+",
    )
    .loc[:, ["cases", "deaths"]]
    .stack()
    .reset_index()
    .rename(columns="level_4": "measure", 0: "value")
)

# finally plot
px.line(
    df3.loc[df3["Alpha-3 code"].eq("USA")],
    x="date",
    y="value",
    color="sex",
    markers=True,
    facet_col="measure",
).update_yaxes(matches=None).show()
px.line(
    df3.loc[df3["Alpha-3 code"].eq("USA")],
    x="date",
    y="value",
    color="sex",
    markers=True,
    facet_row="measure",
).update_yaxes(matches=None).show()