如何使用Bokeh（或其他库）在python中制作轮廓图？

Question

I am interested in making a contour plot in Bokeh. I have not been able to find anything on the net so far.

As a reminder, this is contour plot:

Any help would be appreciated. I would also welcome suggestions to other libraries but with prerequisite of allowing for interactive/animated plots, and not only rendering static output (images).

Answer 1

You can also use matplotlibs contour plot for calculation of the contour data and then plot the contours using bokehs multi_line. I'm also plotting text labels (which are unfortunately a bit ugly).

import numpy as np
import matplotlib.pyplot as plt
from bokeh.models import ColumnDataSource
from bokeh.io import output_file
from bokeh.plotting import gridplot,figure, show

def get_contour_data(X, Y, Z):
    cs = plt.contour(X, Y, Z)
    xs = []
    ys = []
    xt = []
    yt = []
    col = []
    text = []
    isolevelid = 0
    for isolevel in cs.collections:
        isocol = isolevel.get_color()[0]
        thecol = 3 * [None]
        theiso = str(cs.get_array()[isolevelid])
        isolevelid += 1
        for i in range(3):
            thecol[i] = int(255 * isocol[i])
        thecol = '#%02x%02x%02x' % (thecol[0], thecol[1], thecol[2])

        for path in isolevel.get_paths():
            v = path.vertices
            x = v[:, 0]
            y = v[:, 1]
            xs.append(x.tolist())
            ys.append(y.tolist())
            xt.append(x[len(x) / 2])
            yt.append(y[len(y) / 2])
            text.append(theiso)
            col.append(thecol)

    source = ColumnDataSource(data={'xs': xs, 'ys': ys, 'line_color': col,'xt':xt,'yt':yt,'text':text})
    return source


output_file("contour.html")

N = 400
x = np.linspace(-1, 1, N)
y = np.linspace(-1, 1, N)
X, Y = np.meshgrid(x, y)
Z = X**2 + Y**2

source = get_contour_data(X,Y,Z)
plot = figure(plot_width=400,plot_height=400,x_range=[-1,1], y_range=[-1,1])
plot.multi_line(xs='xs', ys='ys', line_color='line_color', source=source)
plot.text(x='xt',y='yt',text='text',source=source,text_baseline='middle',text_align='center')
show(plot)

this is the output:

Answer 2

I wrote a filled contour tool based on matplotlib output:

https://github.com/asher-pembroke/bokeh-tools

Answer 3

Plotly has supported contour plots in Python for the past year:

https://plot.ly/python/contour-plots/

https://plot.ly/pandas/contour-plots/

Density plots (contour plots with adjacent histograms):

https://plot.ly/python/2d-density-plots/

https://plot.ly/pandas/2d-density-plots/

The Plotly Python library is on pip, 100% free and open-source.

Answer 4

By choosing a palette with 10 values, like in the example you provide, one can use image (see Bokeh image example ) in bokeh to simulate a contour plot. The black contour lines and the numbers are missing, but the boundaries between the colors are actually the contour lines. Also, as far as I know Bokeh does not provide a color bar, but you can code it as another image (UPDATE: Latests versions of Bokeh do provide colorbar. ) :

from bokeh.io import output_file
from bokeh.plotting import gridplot,figure, show
from bokeh.models import ColumnDataSource,FixedTicker
import numpy as np
from matplotlib import cm,colors

output_file("contour.html")
cmap = cm.get_cmap("jet") #choose any matplotlib colormap here
num_slabs = 10 # number of color steps
jet_10 = [colors.rgb2hex(m) for m in cmap(np.arange(0,cmap.N,cmap.N/(num_slabs-1)))]
vmin = 0
vmax = 1550
N = 200
x = np.linspace(0, 10, N)
y = np.linspace(0, 10, N)
xx, yy = np.meshgrid(x, y)
d = vmax * (1. + np.sin(xx)*np.cos(yy))

source = ColumnDataSource(data={'d': [d], 'xx': [x], 'yy': [y]})
p = figure(plot_width=400,plot_height=400,x_range=[0, 10], y_range=[0, 10],min_border_right=10)
p.image(image="d", x=[0], y=[0], dw=[10], dh=[10], palette=jet_10,source=source)

# The following code is for the colorbar:
pcb = figure(plot_width=80,plot_height=400,x_range=[0, 1], y_range=[0, vmax],min_border_right=10)
pcb.image(image=[np.linspace(vmin,vmax,100).reshape(100,1)],x=[0],y=[0],dw=[1],dh=[vmax-vmin], palette=jet_10)
pcb.xaxis.major_label_text_color = None
pcb.xaxis.major_tick_line_color = None
pcb.xaxis.minor_tick_line_color = None
pcb.yaxis[0].ticker=FixedTicker(ticks=np.linspace(vmin,vmax,num_slabs+1)) # 11 ticks
pgrid = gridplot([[p,pcb]])  # this places the colorbar next to the image
show(pgrid)

The output will look like:

Answer 5

Thanks Pablo your answer really helped me.

Using your example I made a similar plot that adds the contour lines by using the skimage library. I also turned on the hover tool as in the Bokeh image example that you referenced. Hopefully someone else comes along and adds the line labels.

Note that the following code is meant to run in Jupyter Notebook.

from skimage import measure
import numpy as np

from bokeh.plotting import figure, show, output_notebook
from bokeh.models import (ColorBar,
                          FixedTicker, 
                          LinearColorMapper, 
                          PrintfTickFormatter)

output_notebook()

N = 500
x = np.linspace(0, 10, N)
y = np.linspace(0, 10, N)
xx, yy = np.meshgrid(x, y)
d = np.sin(xx)*np.cos(yy)

mapper = LinearColorMapper(palette='Spectral11', low=-1, high=1)

p = figure(x_range=(0, 10), y_range=(0, 10),
           tooltips=[("x", "$x"), ("y", "$y"), ("value", "@image")])

# must give a vector of image data for image parameter
p.image(image=[d], x=0, y=0, dw=10, dh=10, palette='Spectral11')

levels = np.linspace(-1, 1, 12)
color_bar = ColorBar(color_mapper=mapper, 
                     major_label_text_font_size="8pt",
                     ticker=FixedTicker(ticks=levels),
                     formatter=PrintfTickFormatter(format='%.2f'),
                     label_standoff=6, 
                     border_line_color=None, 
                     location=(0, 0))

p.add_layout(color_bar, 'right')

for level in levels:
  contours = measure.find_contours(d, level)
  for contour in contours:
    x = contour[:,1]/50
    y = contour[:,0]/50
    p.line(x, y, color='grey', line_width=2)

show(p)