Option for adding errorbar legend

[kinyatoride]

Description

Hi, I think it will be great if we have an option to show error bar in the legend similar to what shade and fade do.

Current behavior

Currently, we cannot show error bar in the legend. Only the line is shown in the left figure.

import numpy as np
import pandas as pd
import proplot as pplt

state = np.random.RandomState(51423)
data = state.rand(20, 8).cumsum(axis=0).cumsum(axis=1)[:, ::-1]
data = data + 20 * state.normal(size=(20, 8)) + 30
data = pd.DataFrame(data, columns=np.arange(0, 16, 2))

fig, axs = pplt.subplots(ncols=2)

ax = axs[0]
h = ax.plot(data, means=True, label='label')
ax.legend(h)

ax = axs[1]
h = ax.plot(data, means=True, shadestd=1, label='label')
ax.legend(h)

Proplot version

3.4.3
0.9.7

[cvanelteren]

You mean like this:

see here

import numpy as np
import pandas as pd
import proplot as pplt

state = np.random.RandomState(51423)
data = state.rand(20, 8).cumsum(axis=0).cumsum(axis=1)[:, ::-1]
data = data + 20 * state.normal(size=(20, 8)) + 30
data = pd.DataFrame(data, columns=np.arange(0, 16, 2))

fig, ax = pplt.subplots()

h = ax.imshow(np.random.rand(10, 10))
ax.colorbar(h, loc="upper right")
pplt.show(block=1)

[kinyatoride]

Not actually... That's a colorbar. The error bar is not shown in the legend in my original left figure.

[cvanelteren]

Ah my bad I read your original question wrong! Use errorbar instead

import numpy as np
import pandas as pd
import proplot as pplt

state = np.random.RandomState(51423)
data = state.rand(20, 8).cumsum(axis=0).cumsum(axis=1)[:, ::-1]
data = data + 20 * state.normal(size=(20, 8)) + 30
data = pd.DataFrame(data, columns=np.arange(0, 16, 2))

fig, ax = pplt.subplots()
x = np.linspace(0, 10)
y = x
yerr = abs(np.random.randn(x.size))
h = ax.errorbar(x, y, yerr=yerr, label="test")
ax.legend()
pplt.show(block=1)

[cvanelteren]

For completeness

[kinyatoride]

Thanks. My suggestion was to include that option with ax.plot(data, means=True).
Currently, it is supported when you show errors with shading or fading options e.g. ax.plot(data, means=True, shadestd=1).

That way, it will be more consistent.

[cvanelteren]

I see that is a bit inconsistent indeed. I can put this on the stack of todos. Should be an easy fix.

[cvanelteren]

Note to-self:

• Extract and pass label to error bar in plot.py @ 1506