'bbox_to_anchor' 的 4 元素元组参数在 matplotlib 中是什么意思? [英] What does a 4-element tuple argument for 'bbox_to_anchor' mean in matplotlib?

问题描述

在图例指南"的

In the "Legend location" section of the "Legend guide" in the matplotlib website, there's a small script where line 9 is plt.legend(bbox_to_anchor=(0., 1.02, 1., .102), loc=3, ncol=2, mode="expand", borderaxespad=0.). All the tuples I've seen passed to bbox_to_anchor have 2 elements in it, but this one has 4. What does each element mean if the tuple passed has 4 elements?

I was looking at it in the pyplot.legend docs, and it said something about bbox_transform coordinates. So I looked around and found matplotlib.transforms.Bbox with a static from_bounds(x0, y0, width, height).

I was guessing the setup for the 4-tuple parameter was based on this from_bounds. I copied the script to Spyder, did %matplotlib in an Ipython console, and changed some values. It seemed to make sense, but when I tried only changing .102 to something like 0.9, the legend didn't change. I think the tuple is based on from_bounds, I just don't know why changing the last value in the tuple did nothing.

解决方案

You're right, the 4-tuple in plt.legend(bbox_to_anchor=(0., 1.02, 1., .102), loc=3) is set as (x0, y0, width, height) where (x0,y0) are the lower left corner coordinates of the bounding box.

While those parameters set the bounding box for the legend, the legend's actual vertical size is shrunk to the size that is needed to put the elements in. Further its position is determined only in conjunction with the loc parameter. The loc parameter sets the alignment of the legend inside the bounding box, such that for some cases, no difference will by seen when changing the height, compare e.g. plot (2) and (4).

这篇关于'bbox_to_anchor' 的 4 元素元组参数在 matplotlib 中是什么意思?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！