Plotting error bars using pandas dataframe

Question

I'm sure this is relatively easy, but I can't seem to make it work. I would like to plot this df, with date as the x-axis, gas as the y-axis and std as errorbars using the matplotlib module. I can get it to work using the pandas wrapper, but then I have no idea how to style the errorbars.

Using pandas matplotlib wrapper

I can plot the error bars using the matplotlib pandas wrapper trip.plot(yerr='std', ax=ax, marker ='D') But then i'm not sure how to access the error bars to style them like one could in matplotlib using plt.errorbar()

Using Matplotlib

fig, ax = plt.subplots()
ax.bar(trip.index, trip.gas, yerr=trip.std)

or

plt.errorbar(trip.index, trip.gas, yerr=trip.std)

The above code throws this error TypeError: unsupported operand type(s) for -: 'float' and 'instancemethod'

So basically, what I would like help with is plotting the errorbars using standard matplotlib module rather than the pandas wrapper.

DF ==

        date       gas       std
0 2015-11-02  6.805351  7.447903
1 2015-11-03  4.751319  1.847106
2 2015-11-04  2.835403  0.927300
3 2015-11-05  7.291005  2.250171

tacaswell · Accepted Answer · 2016-02-22 19:58:30Z

23

std is a method on a dataframe, ex df.std().

Use

plt.errorbar(trip.index, trip['gas'], yerr=trip['std'])

or if you have mpl1.5.0+

plt.errorbar(trip.index, 'gas', yerr='std', data=trip)

answered Feb 22, 2016 at 19:58

tacaswell

88.2k23 gold badges222 silver badges203 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Trenton McKinney · Accepted Answer · 2023-06-11 21:12:41Z

As pointed out, the primary issue is using the . notation to access a column name that mirrors a built-in method. While using . is acceptable, it doesn't always work, and may be confused as calling a method instead of a column.
- The clearest way to select a subset of columns is with [] for a single column, or [[]] for multiple columns, as shown in How do I select a subset of a DataFrame?
pandas.DataFrame.plot has the yerr parameter, which accepts the column name as a string.
Also see How to add error bars on a grouped barplot from a pandas column.

import pandas as pd
from datetime import date
import matplotlib.pyplot as plt

# sample dataframe
data = {'date': [date(2015, 11, 2), date(2015, 11, 3), date(2015, 11, 4), date(2015, 11, 5)],
        'gas': [6.805351, 4.751319, 2.835403, 7.291005], 'std': [7.447903, 1.847106, 0.9273, 2.250171]}
trip = pd.DataFrmae(data)

# plot the dataframe with error bars
ax = trip.plot(kind='bar', x='date', y='gas', yerr='std', rot=0)

matplotlib.axes.Axes.bar and matplotlib.pyplot.bar also have the yerr parameter.

fig, ax = plt.subplots()
ax.bar(trip['date'], trip['gas'], yerr=trip['std'])

fig, ax = plt.subplots()
ax.bar('date', 'gas', yerr='std', data=trip)

`trip`

         date       gas       std
0  2015-11-02  6.805351  7.447903
1  2015-11-03  4.751319  1.847106
2  2015-11-04  2.835403  0.927300
3  2015-11-05  7.291005  2.250171

Collectives™ on Stack Overflow

Plotting error bars using pandas dataframe

2 Answers 2

Comments

`trip`

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

trip

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related

`trip`