1

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()

enter image description here

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
Trenton McKinney
  • 56,955
  • 33
  • 144
  • 158
Javier Rodriguez
  • 35
  • 5

1 Answers1

1
  • Tested in python 3.11, pandas 1.5.2, matplotlib 3.6.2, seaborn 0.12.1

Sample DataFrame

import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt

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)

Map plt.errorbar onto sns.relplot

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

# plot the filtered dataframe
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 the errorbars
g.map(plt.errorbar, "Impeller speed (rpm)", "Result", "ster", marker="none", color='r', ls='none')

enter image description here

Specify color for each group of errorbars

  • plt.errorbar only accepts one value for color. In order to match the colors to a palette, the specific data for each facet needs to be selected, and the proper color for that group passed to the color parameter.
    • errorbars that are smaller than the circle can't be seen.
# create a palette dictionary for the unique values in the hue column
palette = dict(zip(filtered['Mineral content (g/g fibre)'].unique(), ["#fdae61", "#abd9e9"]))

# plot the filtered dataframe
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), palette=palette)

# iterate through each facet of the facetgrid
for (material, media), ax in g.axes_dict.items():
    # select the data for the facet
    data = filtered[filtered['Material'].eq(material) & filtered['Media size'].eq(media)]
    # select the data for each hue group
    for group, selected in data.groupby('Mineral content (g/g fibre)'):
        # plot the errorbar with the correct color for each group 
        ax.errorbar(data=selected, x="Impeller speed (rpm)", y="Result", yerr="ster", marker="none", color=palette[group], ls='none')

enter image description here

Trenton McKinney
  • 56,955
  • 33
  • 144
  • 158