Python Pandas error bars are not plotted and how to customize the index

Arnold Published at Dev

arnold

I need your help on plotting error bars using Pandas in Python. I have read the Pandas documentation, and did some trials and errors, but got no satisfying result.

Here is my code:

'''
usage : (python) rc-joint-plot-error-bar.py
'''

from __future__ import print_function
import pandas as pd
import matplotlib.pyplot as plt

filename = 'rc-plot-error-bar.csv'

df = pd.read_csv(filename, low_memory = False)

headers = ['Specimen', 'CA_Native_Mean', 'CA_Implant_Mean', 'CP_Native_Mean',
    'CP_Implant_Mean', 'CA_Native_Error', 'CA_Implant_Error', 'CP_Native_Error',
    'CP_Implant_Error']
    
for header in headers :
    df[header] = pd.to_numeric(df[header], errors = 'coerce')
    
CA_means = df[['CA_Native_Mean','CA_Implant_Mean']]
CA_errors = df[['CA_Native_Error','CA_Implant_Error']]

CP_means = df[['CP_Native_Mean', 'CP_Implant_Mean']]
CP_errors = df[['CP_Native_Error', 'CP_Implant_Error']]

CA_means.plot.bar(yerr=CA_errors)
CP_means.plot.bar(yerr=CP_errors)

plt.show()

Here is what my dataframe looks like:

   Specimen  CA_Native_Mean  CA_Implant_Mean  CP_Native_Mean  CP_Implant_Mean  \
0         1               1         0.738366               1         1.087530
1         2               1         0.750548               1         1.208398
2         3               1         0.700343               1         1.394535
3         4               1         0.912814               1         1.324024
4         5               1         1.782425               1         1.296495
5         6               1         0.415147               1         0.479259
6         7               1         0.934014               1         1.084714
7         8               1         0.526591               1         0.873022
8         9               1         1.409730               1         2.051518
9        10               1         1.745822               1         2.134407

   CA_Native_Error  CA_Implant_Error  CP_Native_Error  CP_Implant_Error
0                0          0.096543                0          0.283576
1                0          0.076927                0          0.281199
2                0          0.362881                0          0.481450
3                0          0.400091                0          0.512375
4                0          2.732206                0          1.240796
5                0          0.169731                0          0.130892
6                0          0.355951                0          0.272396
7                0          0.258266                0          0.396502
8                0          0.360461                0          0.451923
9                0          0.667345                0          0.404856

If I ran the code, I got the following figures:

My questions are:

Could you please let me know how to make the error bars appear in the figures?
How to change the index (the values of x-axis) from 0-9 into 1-10?

Big thanks!

Regards, Arnold A.

Sergey Bushmanov

You're almost there!

For your error bars to show up in the plot, the column names in yerr should match those of the data in the bar plot. Try renaming CA_errors.
For changing x-labels, try ax.set_xticklabels(df.Specimen);


_, ax= plt.subplots() 
CA_means = df[['CA_Native_Mean','CA_Implant_Mean']] 
CA_errors = df[['CA_Native_Error','CA_Implant_Error']].\ 
                rename(columns={'CA_Native_Error':'CA_Native_Mean', 
                                'CA_Implant_Error':'CA_Implant_Mean'}) 
CA_means.plot.bar(yerr=CA_errors, ax=ax) 
ax.set_xticklabels(df.Specimen);