Colormaps¶

38 colormaps are registered with Matplotlib automatically on import rwthplots. Every map is also available in a reversed variant by appending _r (e.g. loading_RWTH_r), following the standard Matplotlib convention. All are accessible via plt.set_cmap(), plt.get_cmap(), or the rwth_cmap() factory.

Colormaps overview

Choosing the right colormap¶

Data type	Recommended map
Qualitative / categorical	`extended_RWTH_discrete` with `lut=N`
Ordered / ranked categories	`continuous_RWTH_discrete`
Symmetric diverging (e.g. anomalies)	`divergent_RWTH`, `divergent_bm_RWTH`, `divergent_gy_RWTH`
Strictly positive sequential	`viridis_RWTH`, `thermal_RWTH`
Single-hue gradient	any `<colour>_RWTH` (e.g. `blue_RWTH`)
Voltage deviation	`voltage_RWTH`
Line / transformer loading	`loading_RWTH`

Perceptual uniformity

viridis_RWTH is designed for perceptual uniformity — lightness increases monotonically from violet to yellow, making it suitable for print, screen, and greyscale reproduction. For non-uniform maps (e.g. thermal_RWTH) the lightness path is non-monotonic, which can mislead magnitude judgements.

Using colormaps¶

import matplotlib.pyplot as plt
from rwthplots.cmap import rwth_cmap

# Standard Matplotlib interface (registered on import)
plt.set_cmap("divergent_RWTH")
ax.imshow(data, cmap="loading_RWTH")

# Factory — returns a LinearSegmentedColormap object
cmap = rwth_cmap("extended_RWTH_discrete", lut=6)

# List all registered names
print(rwth_cmap())

Discrete maps¶

These maps assign one distinct colour per data category with no interpolation between steps.

`extended_RWTH_discrete`¶

The full RWTH corporate design palette arranged for maximum pairwise contrast. Use lut=N to request exactly N colours (1–65):

cmap = rwth_cmap("extended_RWTH_discrete", lut=5)

Colour order is chosen so that each successive colour is as perceptually distinct as possible from the preceding ones — the same greedy strategy used by pick_colors().

`continuous_RWTH_discrete`¶

Samples the full 65-colour palette in order, starting from blue and sweeping continuously through hue. Useful when you want to encode a large number of ordered categories.

Diverging maps¶

Diverging maps have a neutral midpoint (usually white or a light colour) and two saturated extremes. Use them when data has a meaningful centre value — temperature anomalies, deviations from a target, positive/negative quantities.

Always pair diverging maps with a centred normalisation:

import matplotlib.colors as mcolors

# Symmetric around 0
norm = mcolors.TwoSlopeNorm(vcenter=0.0, vmin=-2.0, vmax=2.0)
ax.contourf(lon, lat, anomaly, cmap="divergent_RWTH", norm=norm)

Name	Extremes	Mid
`divergent_RWTH`	Blue ↔ Red	Green
`divergent_bm_RWTH`	Blue ↔ Magenta	White
`divergent_gy_RWTH`	Green ↔ Yellow	White
`voltage_RWTH`	Red ↔ Red	Green

Sequential maps¶

Sequential maps encode magnitude on a single perceptual axis from low to high. They work best for data with a natural zero or minimum.

Name	Stops	Notes
`viridis_RWTH`	Violet → Turquoise → May green → Yellow	Perceptually uniform
`thermal_RWTH`	Black → Bordeaux → Red → Orange → Yellow → White	Blackbody / incandescence style
`loading_RWTH`	Blue → White → Yellow → Red → Bordeaux	0 % to overloaded
`blue_RWTH`	Blue 100 % → Blue 10 %	Single-hue tint gradient
(+ 12 single-colour gradients)	One per remaining RWTH base colour

Power-system colormaps¶

Two colormaps are designed specifically for power-system analysis and grid visualisation. They follow the RWTH colour semantics that engineers at IAEW are familiar with from simulation tools.

Voltage deviationLine loading

import matplotlib.colors as mcolors
norm = mcolors.TwoSlopeNorm(vcenter=1.0, vmin=0.88, vmax=1.12)
im = ax.imshow(voltages, cmap=rwth_cmap("voltage_RWTH"), norm=norm)
cbar = fig.colorbar(im, ax=ax)
cbar.set_label("Voltage (pu)")

voltage_RWTH — symmetric red → orange → green → orange → red. Green at the nominal voltage (1.0 pu), red at both over- and under-voltage extremes. Always use with TwoSlopeNorm(vcenter=1.0).

cmap = rwth_cmap("loading_RWTH")
colors = [cmap(min(v, 1.0)) for v in line_loadings]
bars = ax.bar(labels, line_loadings * 100, color=colors)
ax.axhline(100, color="#CC071E", linewidth=1, linestyle="--")

loading_RWTH — blue → white → yellow → red → bordeaux. Blue for unloaded lines, bordeaux for overloaded. Map values to [0, 1] where 1.0 = 100 % loading; clip above 1.0 to bordeaux.

Colour sets (qualitative)¶

Named palettes for discrete data with up to 13 categories. Access colours by name for readable code:

from rwthplots.cmap import rwth_cset

# Full-intensity palette
cset = rwth_cset("rwth_100")
ax.plot(x, y1, color=cset.blue)
ax.plot(x, y2, color=cset.orange)
ax.plot(x, y3, color=cset.green)

# Lighter tints for filled areas, shading, or secondary lines
cset_50 = rwth_cset("rwth_50")
ax.fill_between(x, y1, alpha=0.4, color=cset_50.blue)

# Iterate over all 13 colours
for i, color in enumerate(cset):
    ax.plot(x, data[i], color=color)

Available tint levels:

Name	Lightness
`rwth_100`	100 % (full saturation)
`rwth_75`	75 %
`rwth_50`	50 %
`rwth_25`	25 %
`rwth_10`	10 % (near-white tints)