How to remove legend in lmplot function of seaborn?

Question

In seaborn lmplot function we have hue argument that will automatically draw a colorbar on the right side. How can I disable/remove the colorbar?

Here are a sample code that will generate a colorbar

df = pd.DataFrame({'x':[1,2,3,1,2,3,1,2,3],'y':[1,2,3,2,3,4,3,4,5],'color':['a','a','a','b','b','b','c','c','c']})
facets = sns.lmplot(data=df, x='x', y='y', hue='color', fit_reg=False)
facets.fig.show()

I tried to call the colorbar object and remove it. In the traditional matplotlib way we can use

collection = ax.scatter(x='a', y='b', data=df)
colorbar = fig.colorbar(collection)

to call it, but I can't see how it works for lmplot. Another alternative way is in sns.heatmap function, where it provides a cbar argument

sns.heatmap(crosstab, cbar=False)

But in lmplot there is not cbar option. What is the best way to do it in lmplot?

Answer 1

A colorbar in Matplotlib/Seaborn is actually like a legend but for continuous data (what you get in a heatmap). In this case, since you have discrete labels (a = blue, b = orange, c = green), you need to access the legend . You can read about the distinction more in the Python Data Science Handbook

Most plot types in Seaborn now already support passing a legend argument into the plot method, such as:

facets = sns.lmplot(data=df, x='x', y='y', hue='color', fit_reg=False, legend=False)