How to group together boxes of different traces for subplots in plotly?

user_4451521 0
Viewed 214

In the documentation about plotly boxes we are told that for Group Box plots we use boxmode so that to group together boxes for each value x

fig.update_layout(
    yaxis_title='normalized moisture',
    boxmode='group' # group together boxes of the different traces for each value of x
)

My question is how do you apply this in case you are using Box subplots? An example would be nice.

In my case I have 5 box subplots and I have put boxmode in the layout but this affects all subplots, even the ones that have only one trace for each value of x, which gives a weird appearance to the plot.

Related to this, if you have 5 box subplots with only one trace for x value, the boxmode option affects the plots anyway

plotly python
1 Answers
  • extending https://plotly.com/python/box-plots/#grouped-box-plots example to do sub plots
  • this does work, however it does offset x-axis in subplots (either as rows or columns)
  • have tried setting various xaxis layout parameters to prevent offsetting without success
  • is this the behaviour you are noting does not work well in sub-plots?
import itertools
import plotly.graph_objects as go
from plotly.subplots import make_subplots
import numpy as np

x = ['day 1', 'day 1', 'day 1', 'day 1', 'day 1', 'day 1',
     'day 2', 'day 2', 'day 2', 'day 2', 'day 2', 'day 2']

fig = make_subplots(rows=1, cols=3)
for r,n in itertools.product(range(3), ["kale","carrots","radishes"]):
    fig.add_trace(go.Box(x=[f"{xx}.{r}" for xx in x], y=np.random.uniform(0,1, len(x)), name=f"{n}-{r}"), row=1, col=r+1)
    
fig.update_layout(boxmode="group")
fig.show()

fig = make_subplots(rows=3, cols=1)
for r,n in itertools.product(range(3), ["kale","carrots","radishes"]):
    fig.add_trace(go.Box(x=[f"{xx}.{r}" for xx in x], y=np.random.uniform(0,1, len(x)), name=f"{n}-{r}"), row=r+1, col=1)
    
fig.update_layout(boxmode="group")

enter image description here

user_94414040
Related