colorbar

Add a colorbar to an image in milliseconds (faster than plt.imshow+plt.colorbar) 💨

colorbar achieves its speed by avoiding matplotlib and instead using numpy, pillow, cmap and pycairo!

🛠️ Install via: pip install colorbar (+ if pycairo causes trouble try conda install anaconda::pycairo)

Usage 💡

import numpy as np
from colorbar import CBar

shape = (400, 800)
im = np.linspace(0, 1, shape[0] * shape[1]).reshape(*shape)  # dummy image
cbar = CBar(cmap='gray',    # see https://cmap-docs.readthedocs.io/en/latest/catalog/
            vmin=im.min(),  # pixels <= vmin become first color of cmap
            vmax=im.max())  # pixels >= vmax become last color of cmap
im_cbar = cbar.draw(im)  # draw colorbar!
im_cbar.show()

draw() returns a PIL.Image that is shown via show() and saved via save()

im_cbar.save('cbar.png')

Use cbar.save() instead of im=cbar.draw()+im.save() to save the image as a vector graphic

...
cbar.save('cbar.pdf', im)  # .pdf, .svg, .ps or .eps

To customize the colorbar, draw() and save() take the arguments:

• vertical: If True, colorbar is vertical. Default: True
• pad: Added pixel rows on the colorbar-side of the image. Default: 0
• pad_color: Color-string of added pixels. Default: 'k'(=black)
• x: Position of the colorbar center relative to image in width-direction. Default: .9
• y: Position of the colorbar center relative to image in length-direction. Default: .5
• width: Size of the colorbar relative to image. Default: .05
• length: Size of the colorbar relative to image. Default: .8
• label: Label above the colorbar. Default: None
• ticks: List (e.g. [0, 1]) or dict (e.g. {0: 'low', 1: 'high'}) of ticks. Default: None
• fontsize: Fontsize of labels and ticks. Default: 20
• linecolor: Color of the lines and the font used to write labels and ticks. Default: w(=white)
• linewidth: Width of ticklines and outlines. Default: 2
• tick_len: Length of ticklines. Default: 2

Finally, if you want draw() to return a numpy array use asarray=True

im_cbar = cbar.draw(im, asarray=True)

to e.g. plot it with matplotlib via plt.imshow(im_cbar).