如何在堆积条形图区域外绘制图例

时间:2018-01-13 14:37:46

标签: plot bokeh

你好,

我的Bokeh库有问题。我需要在情节区域外面制作一个传奇盒子。它适用于圆形和区域绘图。我无法找到堆叠条形图的解决方案。通过图例下方的解决方案形成图形的一部分,它是不可接受的(参见代码下的链接)。我想在框中有一个图例。

此致 SR

import numpy as np
from bokeh.layouts import gridplot
from bokeh.models import ColumnDataSource, 
NumeralTickFormatter,LinearAxis, Range1d, Legend
from bokeh.plotting import show, output_notebook, figure 
from bokeh.core.properties import value
from math import pi
output_notebook()
colors=['#47A440','#9FCDC8']
source = ColumnDataSource(F)
mois=F.mois.tolist()

def insert(colonne, position1, position2,valeur1, valeur2):
    li=[]
    li=colonne.tolist()
    li.insert(position1,valeur1)
    li.insert(position2, valeur2)
    return li
Data_mois_area= insert(F.mois,0,12,'Janvier','Décembre')
Data_conso_area= insert(F.Pui_Conso,0,13,0,0)
Data_pv_area= insert(F.Pui_PV,0,13,0,0)
p = figure(x_range=mois, plot_height=500, 
plot_width=800,toolbar_location=None,tools="hover")
xa=range(len(F.mois))
r0=p.patch(x=Data_mois_area, y=Data_conso_area, color="#003BFB")
r1=p.patch(x=Data_mois_area, y=Data_pv_area, color="#FBB000")
p.vbar_stack(['Pui_Autocon','Pui_Réinj'], x='mois', width=0.5, 
color=colors,line_width=1, line_color='black',
         legend=['Autoconsommation','Réinjection'], source=source)
r3=p.circle(x='mois', y='Taux_autocon', size=10, color="red", 
y_range_name="foo", source=source )
r4=p.circle(x='mois', y='Taux_autopro', size=10, color="purple", 
y_range_name="foo", source=source)
p.y_range.start = 0
p.x_range.range_padding = 0.1
p.xgrid.grid_line_color = None
p.axis.minor_tick_line_color = None
p.outline_line_color = None
p.title.text = "Besoins et Consommation"
p.title.align = "center"
p.title.text_color = "black"
p.title.text_font_size = "15px"
p.xaxis.major_label_orientation = pi/4
p.yaxis.major_label_orientation = "horizontal"
p.yaxis.axis_label = "Energie en kWh"
p.yaxis.axis_label_text_font_style = "italic"
p.extra_y_ranges['foo'] = Range1d(0, 1.02)
p.add_layout(LinearAxis(y_range_name="foo"), 'right')
p.legend.orientation = "horizontal"
legend = Legend(items=[
("Consommation",   [r0]),
("Production", [r1]),
("Taux autoconsommation", [r3]),
("Taux autoproduction ",[r4]),    
 ], location=(1, -30))
 p.add_layout(legend, 'right')}
 show(p) 

enter image description here

1 个答案:

答案 0 :(得分:2)

我解决这个问题的方法如下(hbar堆栈示例但解决方案完全相同)

# make a figure
p = figure(y_range=uniquesubjects, x_range=(0,600), plot_height=900, plot_width=900, title="Subjects In Trouble!",
        toolbar_location=None, tools="")

# add stacked horizontal bars
p.hbar_stack(headers[0:-1], y='subject_name', color=colors, height=1, source=source, legend=[value(x) for x in legendvals])

# add some graph options
p.xgrid.grid_line_color = None
p.axis.minor_tick_line_color = None
p.outline_line_color = None
p.legend.location = "top_right"
p.legend.orientation = "vertical"

new_legend = p.legend[0]
p.legend[0].plot = None
p.add_layout(new_legend, 'right')

show(p)

我正常创建一个"内部图例"然后我会复制它,将其从原始位置移除并将其添加为外部布局。这是在以下三行中完成的:

new_legend = p.legend[0]
p.legend[0].plot = None
p.add_layout(new_legend, 'right')