Customize Bar Chart using Plotly-CodePudding

I have a question regarding bar chart color customization. I have a Grouped Bar Chart like the one below.

I want to know how it is possible to customize only the first column of each occurrence.

Part of the code I created to build this chart is below:

# Create the chart
colors = ['royalblue'] * len(df5)

for i in range(0,len(df5),1):

if df5['wip_higher'][i] == 0:
    colors[df5.index[i]] = 'palegreen'
    
elif df5['wip_higher'][i] == 1:
    colors[df5.index[i]] = 'moccasin'
    
elif df5['wip_higher'][i] == 2:
    colors[df5.index[i]] = 'goldenrod'
    
elif df5['wip_higher'][i] == 3:
    colors[df5.index[i]] = 'salmon'

elif df5['wip_higher'][i] == 4:
    colors[df5.index[i]] = 'black'


fig = px.bar(df5,
            x='Issue key',
            y=['time_wip', 'planned_effort'],
            template='plotly_white',
            text_auto=True,
            barmode='group',
            width=1000,
            labels={'y': 'Time in WIP (in Days)',
                'x': 'Issue Key'}
            )

fig.update_layout(legend=dict(orientation="h",
                            yanchor="bottom",
                            y=1.02,
                            xanchor="right",
                            x=1),
                legend_title_text='')

fig.update_traces(marker_color=colors)

fig.show()

I have tried different approaches, but none worked.

I appreciate your help.

Regards, Marcelo

CodePudding user response：

How about creating them as seperate traces and style each individually?

fig = px.Bar()
fig.add_trace(go.Bar(x=df['Issue Key'], y=df['time_wip']...style as needed)
fig.add_trace(go.Bar(x=df['Issue Key'], y=df['planned_effort']... style as needed)
fig.update_layout(barmode='group')

CodePudding user response：

thanks for all the answers, I found in the documentation (which is not very well organized) this option. I tried and I could solve it without changing the rest of my code.