How to change the dataframe used for plotting

[tabedzki]

Hi all,
I have a core set of plotting commands I'd like to reduce copying and pasting. My initial though is to update the data used similar to how one can change aes to change the y-axis.

In the below example, print(base + aes(y="f_core")) works well. What is the correct way to replace what data is used? I would prefer avoiding creating a second ggplot() object just for the new data? Is it possible to group the geom_point() and geom_line() so I can reuse them?
Thanks in advance

from plotnine 
# tmp_data is data from somewhere else

Nb_list = [ 1, 5, 10, 50]

tmp_data = allstuff.loc[allstuff.query("Ns <= 100 and Nb <= 100 and Nb in @Nb_list").groupby(["Ns", "Nb", "backbone", "sides"]).chiNs.idxmin()]

base = ( ggplot( 
    data=tmp_data,
    mapping= aes(x="Ns", y="chiNs_corrected", color="factor(Nb)"),) 
    + geom_point()
    + geom_line()
    + facet_grid("sides ~ backbone", labeller="label_both") \
    + theme(figure_size=(5, 4))
)

print(base)
print(base + aes(y="f_core"))
print(base + aes(y="x") + tmp_data.query("backbone == 'excluded'"))

[has2k1]

Create a base plot without data and hide the geoms in a function that takes data and returns a list of layers that plot that data.

def layers(data):
    return [geom_point(data), geom_line(data)]

base = ( ggplot( 
    mapping= aes(x="Ns", y="chiNs_corrected", color="factor(Nb)"),) 
    + facet_grid("sides ~ backbone", labeller="label_both") \
    + theme(figure_size=(5, 4))
)

base + layers(data)
base + aes(y='d') + layers(data)
base + aes(y='e') + layers(data.query('a > b'))

[tabedzki]

Thanks very much!