2

In MATLAB™ one can use cplot.m which can generate colored plot basically looks like 2d plot with 3rd axis (z-axis) value as colorbar. Is there any tool/plotting technique I can use to generate a similar plot in Python or IDL Programming Language?. The previous question on stack overflow dealing with different problem as given in a link.

cplot image (Linear colored line plot

gboffi
  • 22,939
  • 8
  • 54
  • 85
Pavan
  • 31
  • 8
  • What do you mean with "lineared" ?? – gboffi Nov 17 '19 at 21:53
  • The duplicate does not address the IDL part of the question. See http://www.idlcoyote.com/graphics_tips/coloredline.html for an example solution (use `plots` or `cgplots` if you have the Coyote library). – sappjw Nov 18 '19 at 12:58

2 Answers2

1

Matplotlib has not a cplot direct equivalent but you can use a LineCollection.

With this understanding you have to modify the usual boilerplate adding a specific import

In [1]: import numpy as np 
   ...: import matplotlib.pyplot as plt 
   ...: from matplotlib.collections import LineCollection

Now, generate some data (c is the 3rd value associated with the (x, y) point)

In [2]: x = np.linspace(0, 6.3, 64) 
   ...: y = np.sin(x) ; c = np.cos(x)

LineCollection needs a 3D array, i.e. a list of segments, each segment a list of points, each point a list of coordinates, that we build using this recipe

In [3]: points = np.array([x, y]).T.reshape(-1,1,2) 
   ...: segments = np.concatenate([points[:-1], points[1:]], axis=1)

Now we instantiate the LineCollection, specifying the colormap that we want and the line width, and immediately after we tell to our instance that its array (what is mapped to colors) is the array c 

In [4]: lc = LineCollection(segments, cmap='plasma', linewidth=3) 
   ...: lc.set_array(c)

and eventually we plot lc in its own way, call autoscale because it's needed (try not to call it...) and add a colorbar.

In [5]: fig, ax = plt.subplots()                                                          
   ...: ax.add_collection(lc) 
   ...: ax.autoscale() 
   ...: plt.colorbar(lc);

enter image description here

I know, it's a bit clunky but it works.

gboffi
  • 22,939
  • 8
  • 54
  • 85
0

IDL v8 has an easy to use keyword for the PLOT function called VERT_COLORS:

; generate some sample data
x = cos(dindgen(100)/20)
y = sin(dindgen(100)/20)
z = dindgen(100)+100

; plot the data
p = plot(x, y, vert_colors=bytscl(z), rgb_table=39, xrange=[-2,2], yrange=[-2,2], thick=3, /aspect_ratio)
cb = colorbar(range=[min(z), max(z)], target=p)

plot with colorbar

The z data is scaled to a byte index of the colortable number 39. The colorbar needs to know the data range explicitly.

jitter
  • 346
  • 5
  • 12
  • 1
    The colorbar should show me the values on the z-axis. but it showing else h=colorbar(p). any idea on bringing z-axis value on the colorbar. – Pavan Nov 20 '19 at 04:29
  • I updated the example in the post above to include a colorbar. – jitter Nov 24 '19 at 15:11