Closed
Description
Bug report
Bug summary
When modifying a colormap with cm.set_bad( color='0.9' ) and saving the image as pdf, the bad colors appear as green instead of gray. This happens for some data sets (I cannot reproduce it with synthetic data sets, it happens with some of my datasets), with some dpi values (for the attached example, it happens with dpi of 1200, but not with dpi of 1000), with some grayscale values (for the attached example, it doesn't happen with color values of '0.79' or darker), and when saving as pdf (it does not happen when displaying the figure with plt.show()).
This is the dataset used for the example code: Data.csv
Code for reproduction
# Paste your code here
#
#
import numpy as np
import pandas as pd
from copy import copy
from matplotlib import cm
from matplotlib import pyplot as plt
fig_size_px = (1280, 960)
fig_dpi = 200
fig_size_inch = tuple( fi/fig_dpi for fi in fig_size_px )
n = 1730
m = 1731
data = pd.read_csv( 'Data.csv', index_col=0, engine='c', dtype=np.float64 ).values
Mask = np.uint8( np.reshape( data[ :, 0 ], ( m, n ) ) )
data = np.reshape( data[ :, 1 ], ( m, n ) )
cmap_jet = copy( cm.jet )
cmap_jet.set_bad( color='0.9' )
# cmap_jet.set_bad( color='0.78' )
# cmap_jet.set_bad( color=(0.9, 0.9, 0.9) )
# cmap_jet.set_bad( color='#d8dcd6' )
fig4 = plt.figure( 4, figsize=fig_size_inch, dpi=fig_dpi, facecolor='w', clear=True, tight_layout=True )
fig4.clf()
ax4 = fig4.add_subplot(111, aspect='equal')
ax4.cla()
data_masked = np.ma.MaskedArray( data, Mask==0 )
h4 = ax4.imshow( data_masked, origin='lower', cmap=cmap_jet, interpolation='Antialiased' )
ax4.get_xaxis().set_visible( False )
ax4.get_yaxis().set_visible( False )
fig4.savefig( 'Plot_Bad.pdf', dpi=fig_dpi*6, format='pdf', transparent=True )
fig4.savefig( 'Plot_Good.pdf', dpi=fig_dpi*5, format='pdf', transparent=True )
plt.show()Actual outcome
# If applicable, paste the console output here
#
#
Expected outcome
Matplotlib version
- Operating system: Gentoo Linux
- Matplotlib version (
import matplotlib; print(matplotlib.__version__)): 3.4.2 - Matplotlib backend (
print(matplotlib.get_backend())): GTK3Agg - Python version: 3.9.5
- Jupyter version (if applicable):
- Other libraries:
From the distribution repositories