How to set a colormap in Matplotlib

Colormaps turn scalar values into an ordered color scale, which is what makes a heatmap, image, contour, or colored scatter plot readable. In Matplotlib, the colormap belongs to the plotted scalar mappable, so the choice should follow the data shape instead of the figure as a whole.

Pass a registered colormap name or object through the plotting call's cmap argument when creating scalar plots. The current registry is available through matplotlib.colormaps, and Matplotlib image and scatter defaults use rcParams[“image.cmap”] when no cmap is supplied.

Sequential maps such as viridis fit ordered measurements, while diverging maps fit values around a meaningful center. Direct plot colors are different; they set one fixed color for a line, marker, or bar rather than mapping many numeric values across a scale.

1. Choose a colormap family that matches the numeric meaning.

   Use sequential colormaps such as viridis for low-to-high measurements, diverging colormaps such as coolwarm for values around a baseline, and cyclic colormaps such as twilight for values that wrap around. Avoid qualitative maps for ordered scalar magnitudes.

2. Save the colormap example as colormap_set.py.

   colormap_set.py

   from pathlib import Path
    
   import matplotlib
   import matplotlib.pyplot as plt
   import numpy as np
   from matplotlib import colormaps
    
   temperatures = np.array(
       [
           [18, 19, 21, 24, 26, 27, 25],
           [16, 18, 20, 22, 24, 25, 23],
           [21, 22, 24, 27, 29, 30, 28],
           [15, 17, 19, 21, 23, 24, 22],
       ]
   )
    
   stations = ["North", "West", "Central", "South"]
   days = ["Mon", "Tue", "Wed", "Thu", "Fri", "Sat", "Sun"]
   output = Path("colormap-set.png")
    
   cmap = colormaps["viridis"]
    
   fig, ax = plt.subplots(figsize=(6.4, 3.8), layout="constrained")
   image = ax.imshow(temperatures, cmap=cmap, vmin=15, vmax=31)
   ax.set_xticks(range(len(days)), days)
   ax.set_yticks(range(len(stations)), stations)
   ax.set_xlabel("Day")
   ax.set_title("Weekly station temperature")
   colorbar = fig.colorbar(image, ax=ax, label="Temperature (deg C)")
    
   fig.savefig(output, dpi=160)
    
   print(f"matplotlib {matplotlib.__version__}")
   print(f"colormap: {image.get_cmap().name}")
   print(f"colorbar label: {colorbar.ax.get_ylabel()}")
   print(f"saved: {output}")

   colormaps[“viridis”] returns a registered colormap object. Passing cmap=“viridis” directly to imshow() or scatter() is also valid when no reuse is needed.

3. Run the script from the directory where the PNG should be written.

   $ python colormap_set.py
   matplotlib 3.11.0
   colormap: viridis
   colorbar label: Temperature (deg C)
   saved: colormap-set.png

4. Open colormap-set.png and confirm the colorbar matches the plot.

   The printed colormap: viridis and the colorbar label confirm that the saved image is using the selected colormap and scale. If a plotted object already exists, change that object explicitly with image.set_cmap(“magma”); plt.set_cmap() changes the default colormap and the current image if one exists.