Matplotlib errorbar() Function

Matplotlib 参考文档 Matplotlib Reference Documentation

errorbar()Used to draw line charts with error bars, displaying the uncertainty range of measured values at the data points.

Widely used in scientific experiments, statistical analysis, and engineering data visualization.

Function Definition

pyplot Interface

matplotlib.pyplot.errorbar(x, y, yerr=None, xerr=None, fmt='',
    ecolor=None, elinewidth=None, capsize=None, barsabove=False,
    lolims=False, uplims=False, xlolims=False, xuplims=False,
    errorevery=1, capthick=None, *, **kwargs)

Axes Interface

Axes.errorbar(x, y, yerr=None, xerr=None, fmt='', ecolor=None,
    elinewidth=None, capsize=None, barsabove=False, lolims=False,
    uplims=False, xlolims=False, xuplims=False, errorevery=1,
    capthick=None, *, **kwargs)

Parameter Description

Parameter	Type	Description
x, y	array-like	Data point coordinates
yerr / xerr	float or array-like	Error in the y/x direction. A scalar means the same error for all points; an array means independent errors for each point. A 2D array can specify lower and upper errors as [lo, hi].
fmt	str	Format string for data points, e.g., 'o' (circle), 's' (square), 'o-' (circle + line).
ecolor	color	Color of the error bar lines.
elinewidth	float	Width of the error bar lines.
capsize	float	Length of the error bar cap lines (in points).
capthick	float	Thickness of the error bar cap lines.
barsabove	bool	If True, error bars are drawn above the data points.
lolims / uplims	array-like of bool	Mark the lower/upper limit in the y direction (only one-sided arrows).
xlolims / xuplims	array-like of bool	Mark the lower/upper limit in the x direction.
errorevery	int	Draw error bars every few data points (reduces visual clutter when data is dense).

errorbar() returns aErrorbarContainerobject, containing(plotline, caplines, barlinecols), allowing separate access to the main line and the error bar lines.

Usage Examples

Example 1: Basic Error Bars

Example

import matplotlib.pyplot as plt
import numpy as np

x = np.array([1, 2, 3, 4, 5])
y = np.array([3.5, 5.2, 4.8, 6.1, 7.3])
y_err = np.array([0.3, 0.5, 0.4, 0.6, 0.5]) # Error for each point

fig, ax = plt.subplots(layout='constrained')

# Draw a line chart with error bars
ax.errorbar(x, y, yerr=y_err,
fmt='o-', # Circles + line
color='steelblue',
ecolor='gray', # Error bar color
elinewidth=1.5, # Error bar line width
capsize=5, # Cap line length
capthick=1.5,
markersize=8,
label='Measurement')

ax.set_title('Errorbar Plot with y-errors')
ax.set_xlabel('X')
ax.set_ylabel('Y')
ax.legend()
ax.grid(True, alpha=0.3)
plt.show()

Example 2: Unequal Errors (Different Upper and Lower)

Example

import matplotlib.pyplot as plt
import numpy as np

x = np.arange(5)
y = np.array([10, 12, 9, 15, 13])

# 2D array: first row is lower limit error, second row is upper limit error
y_err = np.array([[0.5, 0.8, 0.6, 1.0, 0.7], # Lower error
[1.5, 1.2, 2.0, 0.8, 1.3]]) # Upper error

fig, ax = plt.subplots(layout='constrained')

ax.errorbar(x, y, yerr=y_err,
fmt='s', # Square marker, no line
color='coral',
ecolor='black',
capsize=6,
markersize=10,
markerfacecolor='white',
markeredgewidth=1.5,
label='Asymmetric Error')

ax.set_title('Asymmetric Error Bars (different upper/lower)')
ax.set_xlabel('X')
ax.set_ylabel('Y')
ax.legend()
ax.grid(True, alpha=0.3)
plt.show()

Example 3: Bidirectional Error Bars (Errors in Both x and y Directions)

Example

import matplotlib.pyplot as plt
import numpy as np

np.random.seed(42)
x = np.arange(8)
y = 2 * x + 1 + np.random.randn(8) * 2
x_err = np.full(8, 0.3) # x-direction error (same for all points)
y_err = np.random.rand(8) * 3 # y-direction error (different for each point)

fig, ax = plt.subplots(figsize=(7, 5), layout='constrained')

ax.errorbar(x, y,
xerr=x_err, # x-direction error
yerr=y_err, # y-direction error
fmt='o',
color='#8e44ad',
ecolor='gray',
capsize=4,
markersize=8,
label='2D Error')

ax.set_title('Error Bars in Both X and Y Directions')
ax.set_xlabel('X')
ax.set_ylabel('Y')
ax.legend()
ax.grid(True, alpha=0.3)
plt.show()

Example 4: Using errorevery to Reduce Visual Clutter

Example

import matplotlib.pyplot as plt
import numpy as np

# Dense data points (100 points)
x = np.linspace(0, 10, 100)
y = np.sin(x) + np.random.randn(100) * 0.1
y_err = np.full(100, 0.15)

fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(12, 4),
layout='constrained')

# Left plot: show error bars for every point (too dense)
ax1.errorbar(x, y, yerr=y_err, fmt='o', markersize=3,
capsize=2, elinewidth=0.5, errorevery=1)
ax1.set_title('errorevery=1 (too dense)')

# Right plot: show an error bar every 8 points
ax2.errorbar(x, y, yerr=y_err, fmt='o', markersize=3,
capsize=2, elinewidth=0.5, errorevery=8)
ax2.set_title('errorevery=8 (cleaner)')

for ax in [ax1, ax2]:
ax.set_xlabel('X')
ax.set_ylabel('Y')
ax.grid(True, alpha=0.3)

plt.show()

Frequently Asked Questions

Different Shapes of yerr?

Scalar: same error for all points.

1D array (N,): symmetric error for each point (upper and lower are the same).

2D array (2, N): first row is lower error, second row is upper error.

What to Do If You Don't Want Lines Between Data Points?

willfmtSet it to a marker-only format, e.g.,'o'、's'、'^', omitting the line style part.

Matplotlib 参考文档 Matplotlib Reference Documentation

Other Extensions