How can I plot st error bars with seaborn relplot?

Question

I am studying different variables. I want to plot the results with the stadnard error.

I use the filter function because depending on what I want to analyse, I am interested in just plotting mineral, or just plotting one material...etc. I mention this because it is important for the error bars. With seaborn it is not possible to plot the error bars (I used the raw data and I introduced in the seaborn function cd='', but it does not work. Therefore, I have calculated the mean and st error in excel and I plot that directly. The table is the result of the average and the st error that I use in the script. If I add ci in the seaborn, does not do anything. Therefore I want to add it externally in a second line. But I have tried with ax.errorbar(), I cant either plot the st error.

import os
import io
import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
from matplotlib.ticker import FormatStrFormatter
Results = pd.read_excel('results.xlsx',sheet_name='Sheet1',usecols="A:J")
df=pd.DataFrame(Results)

RR_filtered=Results[(Results['Mineral ']=='IC60') | (Results['Mineral ']=='MinFree')]
R_filtered=RR_filtered[(Results['Material']=='A')]

palette = ["#fdae61","#abd9e9"]
sns.set_palette(palette)
ax1=sns.relplot( data=R_filtered,x="Impeller speed (rpm)", y="Result",col="Media size ",hue="Mineral content (g/g fibre)",
palette=palette,size="Media size ",sizes=(50, 200))

R2_filtered=RR_filtered[(Results['Material']=='B')]
ax2=sns.relplot( data=R2_filtered,x="Impeller speed (rpm)", y="Result",col="Media size ",hue="Mineral content (g/g fibre)",
palette=palette,size="Media size ",sizes=(50, 200))
plt.show()

data as image

Media size  Material    Impeller speed (rpm)    Energy input (kWh/t)    Mineral     Mineral content (g/g fibre) Result  ster
1.7 A   400 3000    IC60    4   3.42980002276166    0.21806853183829
1.7 A   650 3000    IC60    4   5.6349292302978 0.63877270588513
1.7 A   900 3000    IC60    4   6.1386616444364 0.150420705145224
1.7 A   1150    3000    IC60    4   5.02677117937851    1.05459146256349
1.7 A   1400    3000    IC60    4   3.0654271029038 0.917937247698497
3   A   400 3000    IC60    4   8.06973541574516    2.07869756201064
3   A   650 3000    IC60    4   4.69110601906018    1.21725878149246
3   A   900 3000    IC60    4   10.2119514553564    1.80680816945106
3   A   1150    3000    IC60    4   7.3271067522139 0.438931805677489
3   A   1400    3000    IC60    4   4.86901883487513    2.04826541508181
1.7 A   400 3000    MinFree 0   1.30614274245145    0.341512517371074
1.7 A   650 3000    MinFree 0   0.80632268273782    0.311762840996982
1.7 A   900 3000    MinFree 0   1.35958635068886    0.360649049944933
1.7 A   1150    3000    MinFree 0   1.38784671261469    0.00524838126778526
1.7 A   1400    3000    MinFree 0   1.12365621425779    0.561737044169193
3   A   400 3000    MinFree 0   4.61104587078813    0.147526557483362
3   A   650 3000    MinFree 0   4.40934493149759    0.985706944001226
3   A   900 3000    MinFree 0   5.06333415444978    0.00165055503033251
3   A   1150    3000    MinFree 0   3.85940865344646    0.731238210429852
3   A   1400    3000    MinFree 0   3.75572328102963    0.275897272330075
3   A   400 3000    GIC 4   6.05239906571977    0.0646300937591957
3   A   650 3000    GIC 4   7.9023202316634 0.458062146361444
3   A   900 3000    GIC 4   6.97774277141699    0.171777036954104
3   A   1150    3000    GIC 4   11.0705742735252    1.3960974547215
3   A   1400    3000    GIC 4   9.37948091546579    0.0650589433632627
1.7 A   869 3000    IC60    4   2.39416757908564    0.394947207603093
3   A   859 3000    IC60    4   10.2373958352881    1.55162686552938
1.7 A   885 3000    BHX 4   87.7569689333017    10.2502550323564
3   A   918 3000    BHX 4   104.135074642339    4.77467275433362
1.7 B   400 3000    MinFree 0   1.87573877068556    0.34648345153664
1.7 B   650 3000    MinFree 0   1.99555403904079    0.482200923313764
1.7 B   900 3000    MinFree 0   2.54989484285768    0.398071770532481
1.7 B   1150    3000    MinFree 0   3.67636872311402    0.662270521850053
1.7 B   1400    3000    MinFree 0   3.5664978541551 0.164453275639932
3   B   400 3000    MinFree 0   2.62948341485392    0.0209463845730038
3   B   650 3000    MinFree 0   3.0066638279753 0.305024483713006
3   B   900 3000    MinFree 0   2.79255446831386    0.472851866083359
3   B   1150    3000    MinFree 0   5.64970870330824    0.251859240942665
3   B   1400    3000    MinFree 0   7.40595580787647    0.629256778750272
1.7 B   400 3000    IC60    4   0.38040036521839    0.231869270120922
1.7 B   650 3000    IC60    4   0.515922221163329   0.434661621954815
1.7 B   900 3000    IC60    4   3.06358032815653    0.959408177590503
1.7 B   1150    3000    IC60    4   4.04800689693192    0.255594912271896
1.7 B   1400    3000    IC60    4   3.69967975589305    0.469944383688801
3   B   400 3000    IC60    4   1.35706340378197    0.134829945730943
3   B   650 3000    IC60    4   1.91317966458018    1.77106692180411
3   B   900 3000    IC60    4   0.874227487043329   0.493348110823194
3   B   1150    3000    IC60    4   2.71732337235447    0.0703901684702626
3   B   1400    3000    IC60    4   4.96743231003956    0.45853815499614
3   B   400 3000    GIC 4   0.325743752029247   0.325743752029247
3   B   650 3000    GIC 4   3.12776074994155    0.452049425276085
3   B   900 3000    GIC 4   3.25564762321322    0.319567445434468
3   B   1150    3000    GIC 4   5.99730462724499    1.03439035936441
3   B   1400    3000    GIC 4   7.51312624370307    0.38399627585515

Answer 1

• Tested in python 3.11 , pandas 1.5.2 , matplotlib 3.6.2 , seaborn 0.12.1

data = {'Media size': [1.7, 1.7, 1.7, 1.7, 1.7, 3.0, 3.0, 3.0, 3.0, 3.0, 1.7, 1.7, 1.7, 1.7, 1.7, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 1.7, 3.0, 1.7, 3.0, 1.7, 1.7, 1.7, 1.7, 1.7, 3.0, 3.0, 3.0, 3.0, 3.0, 1.7, 1.7, 1.7, 1.7, 1.7, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0],
        'Material': ['A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B'],
        'Impeller speed (rpm)': [400, 650, 900, 1150, 1400, 400, 650, 900, 1150, 1400, 400, 650, 900, 1150, 1400, 400, 650, 900, 1150, 1400, 400, 650, 900, 1150, 1400, 869, 859, 885, 918, 400, 650, 900, 1150, 1400, 400, 650, 900, 1150, 1400, 400, 650, 900, 1150, 1400, 400, 650, 900, 1150, 1400, 400, 650, 900, 1150, 1400],
        'Energy input (kWh/t)': [3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000, 3000],
        'Mineral': ['IC60', 'IC60', 'IC60', 'IC60', 'IC60', 'IC60', 'IC60', 'IC60', 'IC60', 'IC60', 'MinFree', 'MinFree', 'MinFree', 'MinFree', 'MinFree', 'MinFree', 'MinFree', 'MinFree', 'MinFree', 'MinFree', 'GIC', 'GIC', 'GIC', 'GIC', 'GIC', 'IC60', 'IC60', 'BHX', 'BHX', 'MinFree', 'MinFree', 'MinFree', 'MinFree', 'MinFree', 'MinFree', 'MinFree', 'MinFree', 'MinFree', 'MinFree', 'IC60', 'IC60', 'IC60', 'IC60', 'IC60', 'IC60', 'IC60', 'IC60', 'IC60', 'IC60', 'GIC', 'GIC', 'GIC', 'GIC', 'GIC'],
        'Mineral content (g/g fibre)': [4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 4, 4, 4, 4, 4, 4, 4, 4, 4, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4], 'Result': [3.42980002276166, 5.6349292302978, 6.1386616444364, 5.02677117937851, 3.0654271029038, 8.06973541574516, 4.69110601906018, 10.2119514553564, 7.3271067522139, 4.86901883487513, 1.30614274245145, 0.80632268273782, 1.35958635068886, 1.38784671261469, 1.12365621425779, 4.61104587078813, 4.40934493149759, 5.06333415444978, 3.85940865344646, 3.75572328102963, 6.05239906571977, 7.9023202316634, 6.97774277141699, 11.0705742735252, 9.37948091546579, 2.39416757908564, 10.2373958352881, 87.7569689333017, 104.135074642339, 1.87573877068556, 1.99555403904079, 2.54989484285768, 3.67636872311402, 3.5664978541551, 2.62948341485392, 3.0066638279753, 2.79255446831386, 5.64970870330824, 7.40595580787647, 0.38040036521839, 0.515922221163329, 3.06358032815653, 4.04800689693192, 3.69967975589305, 1.35706340378197, 1.91317966458018, 0.874227487043329, 2.71732337235447, 4.96743231003956, 0.325743752029247, 3.12776074994155, 3.25564762321322, 5.99730462724499, 7.51312624370307],
        'ster': [0.21806853183829, 0.63877270588513, 0.150420705145224, 1.05459146256349, 0.917937247698497, 2.07869756201064, 1.21725878149246, 1.80680816945106, 0.438931805677489, 2.04826541508181, 0.341512517371074, 0.311762840996982, 0.360649049944933, 0.0052483812677852, 0.561737044169193, 0.147526557483362, 0.985706944001226, 0.0016505550303325, 0.731238210429852, 0.275897272330075, 0.0646300937591957, 0.458062146361444, 0.171777036954104, 1.3960974547215, 0.0650589433632627, 0.394947207603093, 1.55162686552938, 10.2502550323564, 4.77467275433362, 0.34648345153664, 0.482200923313764, 0.398071770532481, 0.662270521850053, 0.164453275639932, 0.0209463845730038, 0.305024483713006, 0.472851866083359, 0.251859240942665, 0.629256778750272, 0.231869270120922, 0.434661621954815, 0.959408177590503, 0.255594912271896, 0.469944383688801, 0.134829945730943, 1.77106692180411, 0.493348110823194, 0.0703901684702626, 0.45853815499614, 0.325743752029247, 0.452049425276085, 0.319567445434468, 1.03439035936441, 0.38399627585515]}
df = pd.DataFrame(data)

# filter the dataframe by Mineral
filtered = df[(df['Mineral']=='IC60') | (df['Mineral']=='MinFree')].copy()

# add a mean column for the mean of each group
filtered['mean_'] = filtered.groupby(['Media size', 'Material', 'Impeller speed (rpm)']).Result.transform('mean')

# plot the data
g = sns.relplot(data=filtered, x="Impeller speed (rpm)", y="Result", col="Media size", row='Material', hue="Mineral content (g/g fibre)", size="Media size", sizes=(50, 200))

# add map plt.errorbar to g
g.map(plt.errorbar, "Impeller speed (rpm)", "mean_", "ster", marker="o", color='r', ls='none')

# if desired, rename the y label, which is overwritten as mean_ when adding the errorbar
# g.set_ylabels("Result", clear_inner=False)

• If the error bar should be over each value, pass 'Reslt' to plt.errorbar , and don't add the 'mean_' column

g = sns.relplot(data=filtered, x="Impeller speed (rpm)", y="Result", col="Media size", row='Material', hue="Mineral content (g/g fibre)", size="Media size", sizes=(50, 200))
g.map(plt.errorbar, "Impeller speed (rpm)", "Result", "ster", marker="none", color='r', ls='none')