从seaborn barplot中删除图例但保留自定义颜色

Question

I'm trying to create a barplot with custom colors based on one of the column in the df. However I don't want legend to display. As soon as I try to remove legend color is back to default.

Here is my code.

b = sns.barplot(x="game_week", y="shots", data=data, hue="color")
b.tick_params(labelsize=5)
b.legend_.remove()
b.set(ylim=(0, np.nanmax(df[cols])))

Any ideas? Is it even possible with sns.barplot?

Answer 1

You can set the color palette with seaborn.color_palette like this:

palette = sns.color_palette('hls', len(data.color.unique())

g = sns.barplot(
    x="game_week", 
    y="shots", 
    data=data, 
    hue="color",
    palette=palette
)
g.legend_.remove()

By setting the palette this way you will get a number of colors corresponding to the number of categories in the color column of your dataframe that is also colorblind friendly !