Python for data visualization

← All tutorials

Visualization · Practical guide

Python for data visualization

Build clear scientific figures with deliberate subplot layouts, readable axes, and compact legends.

PythonMatplotlibFigure design

Start learning

What you’ll learn

Choose a layout for single and multipanel figures.
Control axes, ticks, grids, and shared labels.
Place and customize legends.

Before you begin Python basics

Python and Matplotlib.
The layout examples use empty axes; no dataset is needed.
The legend snippet continues an existing plot with ax and legend handles.

Create axes with subplots

Start with a single axis or a regular grid. Shared figure labels are useful when several panels use the same variables.

Parameters & layout notes

1.1 plt.subplots
 fig,ax = plt.subplots(nrows, ncols, figsize=(width, height), dpi, sharex=False, sharey=False, **kwargs)
 Parameters:
------------------------------
  - fig: figure
  - ax: axes or array of Axes.
  - dpi: the resolution of figure.
  - nrows/ncols: number of rows/columns of the subplot grid.
  - sharex/sharey: bool, default: False.
                   Controls sharing of properties among x (sharex) or y (sharey) axes:
  - figsize: the size of the created figure (Width, height) in inches (float, float).
  - **kwargs: other parameters such as squeeze, width_ratios, height_ratios, subplot_kw, 
              gridspec_kw, etc. that are passed to the pyplot.figure call

import matplotlib.pyplot as plt
# generate a single axes figure (output 1).
fig, ax = plt.subplots(1,1, figsize=(2, 2), dpi=100)
# generate a two rows and two columns axes fugure (output 2). 
# ax[0] (1st row, 1st coloumn), ax[1] (1st row, 2nd coloumn) (1st row, 1st coloumn), ax[2], ax[3] represent each axes.
fig, ax = plt.subplots(2,2, figsize=(3, 3), dpi=100)
# initialize the axes.
config = {"font.family":'Helvetica'}
plt.subplots_adjust(wspace = 0.1,hspace = 0.1)
plt.rcParams.update(config)

# generate a two rows and two columns axes fugure. 
# ax1 (1st row, 1st coloumn), ax2 (1st row, 2nd coloumn) (1st row, 1st coloumn), ax3, ax4 represent each axes.
fig, ([ax1,ax2],[ax3,ax4]) = plt.subplots(2,2, figsize=(3, 3), dpi=100)
# all the ases share the same x and y label (output 3).
fig.text(0.5, 0, 'x', ha='center')
fig.text(0, 0.5, 'y', va='center',rotation='vertical')

Build panels with figure

Create the figure first, then add each panel in a loop. This is useful when panel creation is part of an iterative workflow.

Default figure background · Click to enlarge

Custom figure background · Click to enlarge

Parameters & layout notes

1.2 plt.figure
 fig = plt.figure(figsize=None, dpi=None, facecolor=None, edgecolor=None, frameon=True, **kwargs)
 Parameters:
------------------------------
  - fig: figure
  - dpi: the resolution of figure.
  - figsize: the size of the created figure (Width, height) in inches (float, float).
  - dpi:the resolution of the figure in dots-per-inch.
  - facecolorcolor: the background color.
  - edgecolorcolor: the border color.
  - frameonbool, default: True. If False, suppress drawing the figure frame.
  - **kwargs: additional keyword arguments are passed to the Figure constructor.
------------------------------
  Useful for iteratively figure plotting.

import matplotlib.pyplot as plt
# plot a 2*2 axes figure with 100 dpi (output 1).
fig = plt.figure(figsize=(3,3), dpi=100)
config = {"font.family":'Helvetica'}
plt.subplots_adjust(wspace =0.4,hspace =0.3)
plt.rcParams.update(config)
for i in range(4):
    ax = fig.add_subplot(2,2,i+1)

# plot a 2*2 axes figure with 100 dpi, yellow face and edgecolor (output 2).
fig = plt.figure(figsize=(3,3), dpi=100, facecolor="y", edgecolor="y")
config = {"font.family":'Helvetica'}
plt.subplots_adjust(wspace =0.4,hspace =0.3)
plt.rcParams.update(config)
for i in range(4):
    ax = fig.add_subplot(2,2,i+1)

Arrange panels with GridSpec

Use a grid to make panels span different numbers of rows or columns. The example arranges four panels across a flexible grid.

Four panels arranged with GridSpec · Click to enlarge

Parameters & layout notes

1.3 plt.figure & GridSpec
 gs = gridspec.GridSpec(nrows, ncols)
 A grid layout to place subplots within a figure.
Parameters:
------------------------------
  - nrows/ncols: the number of rows and columns of the grid (int).
  - other parameters: https://matplotlib.org/stable/api/_as_gen/matplotlib.gridspec.GridSpec.html
  - detailed tutorial: https://matplotlib.org/3.5.0/tutorials/intermediate/gridspec.html
------------------------------
  Useful for plotting irregular axes.
  Useful for iteratively figure plotting.

import matplotlib.pyplot as plt
from matplotlib import gridspec
# plot a 1*4 axes figure with 100 dpi using GridSpec
fig = plt.figure(figsize=(8,2), dpi=100)
gs = gridspec.GridSpec(2, 8)
config = {"font.family":'Helvetica'}
plt.subplots_adjust(wspace =0.7,hspace =0.1)
plt.rcParams.update(config)
# can also use interative loop to generate axes.
ax1 = plt.subplot(gs[0:2, 0:2])
ax2 = plt.subplot(gs[0:2, 2:4])
ax3 = plt.subplot(gs[0:2, 4:6])
ax4 = plt.subplot(gs[0:2, 6:8])

Style axes and ticks

Set labels and limits, format tick values, control spines, and use a light grid to guide the reader without distracting from the data.

Axis labels, ticks, and grid settings · Click to enlarge

Parameters & layout notes

2.1 set axes parameters

import matplotlib.pyplot as plt
import matplotlib.ticker as mtick

fig,ax = plt.subplots(figsize=(6,2), dpi=100)

# x-axis, y-axis label.
ax.set_xlabel('x', fontsize=10, fontweight = 'bold', fontstyle='italic', labelpad = 0.5)
ax.set_ylabel('y', fontsize=10, fontweight = 'bold', fontstyle='italic', labelpad = 0.5)


# tick parameters setting.
"""
ax.tick_params(axis='both', **kwargs)
The parameters include: 
  (1) axis: {'x', 'y', 'both'}, default: 'both'. The axis to which the parameters are applied.
  (2) which: {'major', 'minor', 'both'}, default: 'major'. The group of ticks to which the parameters are applied.
  (3) direction: {'in', 'out', 'inout'}. Puts ticks inside the Axes, outside the Axes, or both.
  (4) length: float. Tick length in points.
  (5) width: float. Tick width in points.
  (6) color: Tick color.
  (7) pad: float. Distance in points between tick and label.
  (8) labelsize: float or str. Tick label font size in points or as a string (e.g., 'large').
  (9) labelcolor: Tick label color.
  (10) colors: Tick color and label color.
  (11) zorder: float. Tick and label zorder.
  (12) bottom, top, left, right: bool. Whether to draw the respective ticks.
  (13) labelbottom, labeltop, labelleft, labelright: bool. Whether to draw the respective tick labels.
  (14) labelrotation: float. Tick label rotation
"""
# Examples
ax.tick_params(labelsize=8)
ax.tick_params(axis = 'x',labelsize=8)
ax.tick_params(axis='both', which='major', direction='in', length=4, width=2,
               color='red', pad = 0.1, labelsize=10, labelrotation=45, labelcolor = "blue", 
               bottom = True, top = False, left = True, right = False,
               labelbottom = True, labeltop = False, labelleft = True, labelright = False)

# set the visiblity of axis
ax.spines[['top', 'right']].set_visible(False)
ax.spines[['bottom','left']].set_linewidth(2)


# set xlim and ylim and display the same number of decimal places.
ax.set_xlim(0,10)
ax.set_ylim(0,10)
ax.xaxis.set_major_formatter(mtick.FormatStrFormatter('%.1f'))
ax.yaxis.set_major_formatter(mtick.FormatStrFormatter('%.1f'))

# set the facecolor and grid.
ax.set_facecolor((0,0,0,0.03))
ax.grid(color='gray', linestyle=':', linewidth=0.3)

Refine the legend

Control placement, columns, spacing, and marker appearance. Run this snippet after creating the plotted series and their legend labels.

# common settings
ax.legend(loc='upper right',fontsize=12, facecolor= 'none',edgecolor = 'none',bbox_to_anchor=(2.0, 0.7), ncol = 2, columnspacing = 0.4)
# remove legend
ax.legend_.remove()
# sometimes we need to re-define the legend as the alpha or other parameter settings in the figure.
handles, labels = ax.get_legend_handles_labels()
k = 0
new_handles = []
for handle in handles:
    new_handle = plt.Line2D([], [], ls = "none", marker='o', color = original_handle_colors[k], markersize = 8, alpha = 1)
    new_handles.append(new_handle)
    k = k+1
legend = ax.legend(handles=new_handles, labels=labels,loc = 'lower right',fontsize=8.5, facecolor= 'none',edgecolor = 'none',bbox_to_anchor=(1.,0))

Keep exploring

Matplotlib tutorials GridSpec reference

Download all examples