[BUG] Size of implant's electrode don't seem to agree with the provided electrode radius

[alex-bene]

Bug Description
Trying to visualize various custom parametrized square implants, it seems like the 'r' parameter does not agree with the radius of the electrode when the implant is visualized using plot_implant_on_axon_map.

To Reproduce
plot_implant_on_axon_map(ProsthesisSystem(ElectrodeGrid((2, 2), r=1000, spacing=(2000), etype=DiskElectrode)))

Expected behavior
In the diagram from the above snippet, there should be 4 electrodes with a radius of 1000 microns touching each other and covering a total area of 4000x4000 microns.

Screenshots
However we get this

Why this is happening
Looking around the code this seems to trackback to this line of code where the markersize is set

pulse2percept/pulse2percept/viz/axon_map.py

Line 161 in 9acf659

markersize=np.sqrt(implant[e].r) * 2)

In the End
In the end, what matter is if it is just a visual thing and you couldn't find how to better visualize it (because I looked into it and I didn't find something very useful except that the units of markersize are points^2 --> area) or if the 'r' in the code snippet above means something different than the radius of the electrode in microns (since this would have impact on the results from the simulations)

[mbeyeler]

Thank you for this detailed report. Yeah, as you pointed out towards the end, it is a pain to get the marker to show up with the correct size because of the auto-scaling... I realize though that it is important to get this right - I will look into it

[alex-bene]

In my case, it just happened to be using the visualization to check for electrode overlaps so it was a bit jarring. In any case, I suppose if it was documented that the visual size of the electrode is not the real one would be a good start.

As for a possible solution, I suppose one could use the circle object from matplotlib.

ax.add_patch(plt.Circle((implant[e].x, implant[e].y), implant[e].r, color='w'))

As for using the 'markersize' option, according to kite is seems like:

Using matplotlib.pyplot.plot() determines the width of points. Using matplotlib.pyplot.scatter() determines the area of points
source

In any case, it seems like:
actual_area = markersize*k in the case of scatter
actual_diameter = markersize*k in the case of plot
where k could be calculated through some measurements or trial and error maybe. (although I strongly believe that the use of plt.Circle is the best)

I will try to make some of the above modifications probably in the next week if I find some time.

[mbeyeler]

A PR would be greatly appreciated! Then we can ship this as part of v0.7, which is imminent (~next couple weeks)

[mbeyeler]

Looking further into this: The use of these plotters should actually be deprecated. This is a documentation fail.

The new preferred way is to use the plot methods of the corresponding object - those correctly use Circle.

For example, plot Argus II:

import pulse2percept as p2p
p2p.implants.ArgusII().plot()

Plot Argus II on axon map:

p2p.models.AxonMapModel().plot()
p2p.implants.ArgusII().plot()

Or use your own implant:

p2p.implants.ElectrodeGrid((2, 2), r=1000, spacing=2000, etype=p2p.implants.DiskElectrode).plot()

On axon map:

p2p.models.AxonMapModel().plot()
p2p.implants.ElectrodeGrid((2, 2), r=1000, spacing=2000, etype=p2p.implants.DiskElectrode).plot()

[alex-bene]

Ohh, I actually didn't test that. It seems you're right. Well as a "plus" I suppose the same functionality could be added to the "plot_implant_on_axon_map" function for consistency or completely remove it.