Barplot coloring using seaborn.color_palette

Question

I have the following code fragment:

import seaborn
import matplotlib.pyplot as plt
plt.bar(df['Col1'], df['Col2'], width = 0.97, color=seaborn.color_palette("coolwarm", df['Col1'].shape[0], 0.999)) 

Now, my bars are colored in the blue-red spectrum (as given by the parameter coolwarm). How can I change the distribution between these two colors and their order, for example, to get 80% of all bars red and only the rest (i.e., 20%) in blue? (Now it is 50-50% ratio)

Answer 1

There's no built in way to do this using seaborn or the matplotlib colormaps, but here's a solution that seems to do the trick

import seaborn as sns
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np

## function that generates a color palette
def gen_palette(n_left, n_right, cmap_name="coolwarm", desat=0.999):
    """
    n_left: number of colors from "left half" of the colormap
    n_right: number of colors from the "right half"
    cmap_name: name of the color map to use (ideally one of the diverging palettes)

    return: palette, list of RGB-triples
    """
    palette_1 = sns.color_palette(palette=cmap_name, 
                                  n_colors=2 * n_left, 
                                  desat=desat)[:n_left]
    palette_2 = sns.color_palette(palette=cmap_name, 
                                  n_colors=2 * n_right, 
                                  desat=desat)[n_right:]
    return palette_1 + palette_2

## generate example data
N = 20
rng = np.random.default_rng(seed=42)
y_vals = 10 * rng.random(N)
df = pd.DataFrame(
    {"Col1": np.arange(N),
     "Col2": y_vals}
)

## build the color palette with the desired blue-red split
n_red = round(0.8 * N)
palette = gen_palette(N - n_red, n_red)

## plot
plt.bar(df['Col1'], df['Col2'], 
        width=0.97, 
        color=palette)