改变 100% 堆叠条形图中的颜色 python

Question

我正在尝试更改 plotly 中堆叠的 100% 条形图的颜色。

我有以下数据框和代码：

import pandas as pd
import plotly
import plotly.express as px
from plotly.offline import plot

df = pd.DataFrame([["A", "Jan", "20%"],
                   ["A", "Fev", "10%"],
                   ["A", "Mar", "15%"],
                   ["A", "Apr", "10%"],
                   ["B", "Jan", "15%"],
                   ["B", "Fev", "15%"],
                   ["B", "Mar", "20%"],
                   ["B", "Apr", "10%"],
                   ["C", "Jan", "25%"],
                   ["C", "Fev", "25%"],
                   ["C", "Mar", "20%"],
                   ["C", "Apr", "30%"],
                   ["D", "Jan", "40%"],
                   ["D", "Fev", "50%"],
                   ["D", "Mar", "45%"],
                   ["D", "Apr", "50%"]],
                  columns=["type", "month", "percentage"])

colors = plotly.colors.qualitative.Prism

fig=px.bar(df, x='month', y='percentage',
            color=df['type'], barmode ='stack', 
            text = df['percentage'])


fig.update_traces(textfont_size=12, marker_color = colors)

fig.update_layout(title = {'text': "Title",
                           'x':0.5, 'xanchor': 'center'},
                  title_font_size=30,
                  legend=dict(yanchor="bottom", y=0.0, 
                              xanchor="right", x=1.2),
                  legend_font_size=16, 
                  xaxis_title = 'Months', 
                  yaxis_title ='%',
                  xaxis_tickangle=-45,
                  width = 1000, height = 600)
fig.show()

但我得到以下信息：

更改整个条形的颜色，而不是更改条形每个部分的颜色。

我要的是把原来的蓝色、红色、绿色、紫色改成：

（这就是我在 fig.update_traces 中没有 marker_color = colors 时得到的结果，试图根据需要更改颜色。）

我怎样才能正确地做到这一点？

Answer 1

在 update_traces 中设置 color_discrete_sequence argument of px.bar 而不是 marker_color:

colors = plotly.colors.qualitative.Prism

fig = px.bar(df, x='month', y='percentage',
             color=df['type'], barmode='stack',
             text=df['percentage'].astype(str) + '%',  # Add Percent Sign
             color_discrete_sequence=colors)

# Don't forget to remove from update_traces
fig.update_traces(textfont_size=12)

数据和导入（删除了字符串百分比，因为它们绘制不正确）：

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

df = pd.DataFrame([["A", "Jan", 20],
                   ["A", "Fev", 10],
                   ["A", "Mar", 15],
                   ["A", "Apr", 10],
                   ["B", "Jan", 15],
                   ["B", "Fev", 15],
                   ["B", "Mar", 20],
                   ["B", "Apr", 10],
                   ["C", "Jan", 25],
                   ["C", "Fev", 25],
                   ["C", "Mar", 20],
                   ["C", "Apr", 30],
                   ["D", "Jan", 40],
                   ["D", "Fev", 50],
                   ["D", "Mar", 45],
                   ["D", "Apr", 50]],
                  columns=["type", "month", "percentage"])

改变 100% 堆叠条形图中的颜色 python

Change colors in 100% stacked barchart plotly python

python

bar-chart

pandas

plotly