ipyvolume icon indicating copy to clipboard operation
ipyvolume copied to clipboard

Issue displaying integer values in scatter

Open Carifio24 opened this issue 7 months ago • 0 comments

It seems that there's currently an issue with displaying integer values (in particular, a numpy array with an integral dtype) in the ipyvolume scatter. The motivation for this is https://github.com/glue-viz/glue-jupyter/issues/449.

You can easily create some data to reproduce this as follows

import ipyvolume as ipv
from numpy.random import randint, random

ints = randint(0, 100, (100,))
ints_as_floats = ints.astype("float64")
floats_1 = random((100,))
floats_2 = random((100,))

If I create a scatter using the quickscatter with the integer array as one of the coordinate values, nothing displays: Screenshot 2024-07-26 at 10 36 37 PM

But doing the same after converting the dtype of the integer array to float64, the scatter behaves as expected: Screenshot 2024-07-26 at 10 36 49 PM

Carifio24 avatar Jul 27 '24 02:07 Carifio24