在2D点图上使用鼠标悬停在numpy数组中查找行/位置

Question

I have a 2D dot plot which I generated using matplotlib with 750 points on it. I was trying to use mpldatacursor so that I can mouse over the points to determine the label for a datapoint (ie row/position in the numpy array etc).

When I use mpldatacursor it works well - but I only get x & y co-ords. Any thoughts on how I might find the label/origin of the individual points of interest? I know I could plot labels - but with 750 datapoints it's pretty messy.

thoughts, Sarr.

Answer 1

You could pick a plotted point, passing the information back in data coordinates. Check matplotlib's event handling and picking guide for detailed information on how to achieve this.

Once you have the data coordinates, you can find the corresponding indices using numpy:

import numpy as np

# arr is the array you plot

[...code to plot/pick array...]

# datapt are the data coordinates you get back from the picking event

indices = np.argwhere(arr == datapt)

This approach involves a lot more work than mpldatacursor, but the latter doesn't seem to offer nearly as much flexibility.