plotnine中如何将变量映射到scale_color_gradient？

Let's break down what's happening here and resolve that warning for you:

The Root Issue #

Your code maps the bin count to the fill aesthetic in geom_histogram(aes(fill='..count..')), but then you're trying to adjust the color scale with scale_color_gradient(). These are two distinct aesthetic properties in ggplot:

• fill controls the interior color of the histogram bars
• color controls the border color of the bars

Since you haven't mapped any variable to the color aesthetic, Plotnine throws that warning because it can't generate a legend for an aesthetic that isn't being used in your plot.

The Simple Fix #

Just replace scale_color_gradient() with scale_fill_gradient() to match the aesthetic you're actually using for the bar fill:

import pandas as pd
from plotnine import ggplot, aes, geom_histogram, scale_fill_gradient

gg = pd.DataFrame()
gg['x'] = u  # Assuming 'u' is your pre-defined dataset variable
p = ggplot(aes(x='x'), gg) + geom_histogram(aes(fill='..count..'), binwidth=0.1)
p = p + scale_fill_gradient(low='blue', high='white')
print(p)

Extra Context #

• If you wanted the bar border color to also use a gradient based on count, you'd add color='..count..' to the aes() inside geom_histogram, and then you could use both scale_fill_gradient() and scale_color_gradient() (though this might make the histogram look cluttered for most use cases).
• The ..count.. variable is a special computed value in ggplot that returns the number of observations in each bin, which is exactly what you're using to drive your fill gradient here.

内容的提问来源于stack exchange，提问作者bill fite