Plotly:如何在甘特图上标记条形?

2022-01-21 00:00:00 python plotly visualization gantt-chart

问题描述

有什么方法可以在 Plotly Gantt 图中创建条形图上的任务标签和 y 轴上的资源标签?

plotly.figure_factory.create_gantt的

解决方案

您可以根据设置在 fig['layout']['annotations'] 中添加结构为字典列表的注释就像

代码:

情节的结构与上面的源代码相似,但我已将其设置为在离线 Jupyter Notebook 中与 iplot() 一起使用.

# 导入从 plotly.offline 导入 download_plotlyjs, init_notebook_mode, plot, iplot从 IPython.core.display 导入显示,HTML将 plotly.figure_factory 导入为 ff导入 plotly.graph_objs# 设置display(HTML("<style>.container { width:50% !important; } .widget-select > select {background-color: gainboro;}</style>"))init_notebook_mode(连接=真)#%qtconsole --style vim# 日期开始A = '2009-01-01'StartB = '2009-03-05'StartC = '2009-02-20'FinishA='2009-02-28'FinishB='2009-04-15'FinishC='2009-05-30'标签日期A='2009-01-25'标签日期B='2009-03-20'LabelDateC='2009-04-01'# 样本数据df = [dict(Task="Task A", Start=StartA, Finish=FinishA),dict(Task="Task B", Start=StartB, Finish=FinishB),dict(Task="Task C", Start=StartC, Finish=FinishC)]# 数字图 = ff.create_gantt(df)# 添加注释annots = [dict(x=LabelDateA,y=0,text="任务标签 A", showarrow=False, font=dict(color='white')),dict(x=LabelDateB,y=1,text="任务标签 B", showarrow=False, font=dict(color='White')),dict(x=LabelDateC,y=2,text="任务标签 C", showarrow=False, font=dict(color='White'))]# 绘制图形fig['layout']['annotations'] = annotsiplot(无花果)

可能的改进:

如您所见,我已经硬编码了标签所在的日期.您可以轻松地计算中间日期开始和结束日期.但是为什么不简单地使用 (x=LabelDateC,y=2,align="center",text="Task label C", showarrow=False, font 中的 align=center 调整标签=dict(color='White'))?这不起作用,因为标签似乎附加到日期本身,而不是条的结构或大小.

Is there any way to create in Plotly Gantt chart with Task labels on bars and Resource labels on y-axis?

There is no such example in documentation of plotly.figure_factory.create_gantt. An abstract example of desirable chart is presented below:

解决方案

You can add annotations structured as a list of dictionaries in fig['layout']['annotations'] based on a setup like the first example from your link.

Plot:

Code:

The structure of the plot is similar to the source above, but I've set it up to work with iplot() in an off-line Jupyter Notebook.

# imports
from plotly.offline import download_plotlyjs, init_notebook_mode, plot, iplot
from IPython.core.display import display, HTML
import plotly.figure_factory as ff
import plotly.graph_objs as go

# setup
display(HTML("<style>.container { width:50% !important; } .widget-select > select {background-color: gainsboro;}</style>"))
init_notebook_mode(connected=True)

#%qtconsole --style vim

# dates
StartA = '2009-01-01'
StartB = '2009-03-05'
StartC = '2009-02-20'

FinishA='2009-02-28'
FinishB='2009-04-15'
FinishC='2009-05-30'

LabelDateA='2009-01-25'
LabelDateB='2009-03-20'
LabelDateC='2009-04-01'

# sample data
df = [dict(Task="Task A", Start=StartA, Finish=FinishA),
      dict(Task="Task B", Start=StartB, Finish=FinishB),
      dict(Task="Task C", Start=StartC, Finish=FinishC)]

# figure
fig = ff.create_gantt(df)

# add annotations
annots =  [dict(x=LabelDateA,y=0,text="Task label A", showarrow=False, font=dict(color='white')),
           dict(x=LabelDateB,y=1,text="Task label B", showarrow=False, font=dict(color='White')),
           dict(x=LabelDateC,y=2,text="Task label C", showarrow=False, font=dict(color='White'))]

# plot figure
fig['layout']['annotations'] = annots
iplot(fig)

Possible improvements:

As you can see, I've hard-coded the dates where the labels are positioned. You could just as easily calculate the middle date between the start and end dates. But why not simply adjust the labels using align=center in (x=LabelDateC,y=2,align="center",text="Task label C", showarrow=False, font=dict(color='White'))? That won't work because the label seems to be attached to the dates itself and not the structure or size of the bar.

相关文章