Memo: Lang - Python

Difference between `plt` and `fig,ax`

object-oriented

Inverse the X-axis

1
2
3


import matplotlib.pyplot as plt
fig, ax = plt.subplots()
ax.invert_xaxis()

Add caption below the x-label

1
2
3


fig, ax = plt.subplots()
txt = 'near and far are distance'
fig.text(x=.5, y=.001, s=txt, ha='center')

1
2
3


fig, ax = plt.subplots()
ax.set_xlabel('''z_cam
near and far are distance''')

微分柱形图

Set xticks

xticks is the (fixed) number on the x-axis, while xticklabels can be customized strings.

For plt (Matplotlib.pyplot.xticks()): Matplotlib Set_xticks - Detailed Tutorial - Python Guides

Matplotlib.axes.Axes.set_xticklabels() Matplotlib Set_xticklabels -Python Guides

1
2
3


fig, ax = plt.subplots()
ax.set_xticks([0, np.pi, 2*np.pi, 3*np.pi])
ax.set_xticklabels(labels=['0', r'$\pi$', r'2$\pi$', r'3$\pi$'], fontdict=None, fontsize=6, fontstyle='italic', color='red', verticalalignment='top', horizontalalignment='left', rotation=90, minor=True,

Change the number and text of the showing Ticks

GfG

For plt:

1
2


plt.xticks(ticks=[1,2,3,4], labels=None, **kwargs)
plt.yticks(ticks=[7,13,24,22], labels=None,)

Use locator_param() to change the tightness and number of ticks.

plt.locator_params(axis='both', nbins=4)
Use xlim() to restrict the diplayed area, rather the whoel plot.
1 2

plt.xlim(0,3) plt.locator_params(axis='x', nbins=3)
Use Matplotlib.ticker.MaxNLocator class

Set Number of Ticks in Matplotlib | Delft Stack

1
2
3
4
5
6
7
8


from matplotlib.ticker import MaxNLocator
fig, ax = plt.subplots(1,1)
ax.yaxis.set_major_locator(MaxNLocator(5)) 

# hide specific ticks
for i, tick in enumerate(ax.xaxis.get_ticklabels()):
   if i%2 !=0:
      tick.set_visible(False)

Set y-axis view limits

Matplotlib.axes.Axes.set_ylim() in Python - GeeksforGeeks

1
2
3
4
5
6
7


fig, ax = plt.subplots(facecolor ='# A0F0CC')
x, y = 4*(np.random.rand(2, 100) - .5)
ax.plot(x, y, 'g')
ax.set_ylim(-3, 3)

# 游标
cursor = Cursor(ax, useblit = True, color ='red', linewidth = 2)

Set the 1st x as 1 (not 0)

Python MatplotLib plot x-axis with first x-axis value labeled as 1 (instead of 0)

Ticks

Add extra ticks, labels

Add extra ticks which must be number, not work for string SO

1
2
3


fig, ax= plt.subplots()
extraticks = [2.1, 3, 7.6]
ax.set_xticks(list(ax.get_xticks()) + extraticks)

Use ax.set_xticklabels() for string Create label list - SO

1
2
3
4
5


near, far = 1, 15
ax.set_xticks(list(ax.get_xticks()) + [-near,-far])
# Their labels also have to be appended at the end.
xticklabels= ['-16', '-14', '-12', '-10', '-8', '-6', '-4', '-2','0', '-near', '-far']
ax.set_xticklabels(xticklabels, rotation=90)

Example: Mark the yticks for the highest and lowest points:

1
2
3
4
5
6
7


rmse = np.load("./error_list.npz")['Train_RMSE']
idx = np.arange(len(rmse))

fig, ax = plt.subplots()
ax.plot(idx, rmse)
extraticks = [rmse.min(), rmse.max()]
ax.set_yticks(list(ax.get_yticks()) + extraticks)

Example: 把新 ticklabel 加入并重新排列
- References:
  1. Gemini 2.5P - Forcing Matplotlib xticklabels

Replace xticklabels

References:
1. Gemini 2.5P -

Supports:

(2025-08-22T12:55)

Rebuild the label list

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11


# 3. Define the desired tick locations and get their labels
# The x-values where you want to place ticks
new_locations = [5, 15, 20, 25, 30]

# Get the corresponding labels directly from the DataFrame using a list comprehension
# This assumes the index of the DataFrame corresponds to the x-values of your plot
new_labels = [df[df.columns[0]][loc] for loc in new_locations]

# 4. Set the new ticks and labels on the plot
ax1.set_xticks(new_locations)
ax1.set_xticklabels(new_labels, rotation=90) # Use rotation to prevent labels from overlapping

Set xticks as integer

(Not sure) ax.xaxis.set_major_locator(MaxNLocator(integer=True))

Python：使用f-string保留小数点位数 csdn

用decimal 模块：w3cschool

1
2
3


from decimal import Decimal
a = 12.345
Decimal(a).quantize(Decimal("0.00")) # 使用默认的进位方式（同round）“0.00”表示保留小数点后两位

Hide ticks

Remove ticks
- References:
  1. GfG - How to Hide Axis Text Ticks or Tick Labels in Matplotlib?
- Supports:
  1. Remove tick labels by setting them to an empty list ^r1-GfG
    1 2
    
    fig, ax1 = plt.subplots(1,1) ax1.set_xticks([])

Set tick bold, color

(2024-08-04)

Set an individual xtick label to be bold:
1

ax.get_xticklabels()[label_ls.index("Input point cloud")].set_weight('bold')
- Ref: How to set the matplotlib axes tick labels fontweight to bold? - SO (Searched by “python matlibplot.pyplot xtick label bold” in DDG)

Set ticklabel color
- References:
  1. Change Single Matplotlib Xtick Color
- Supports:
  
  (2025-08-11T22:40)
  1. ax.get_xticklabels()[2].set_color('red')

Fontsize of yticks

References:

- Asked ChatGPT

Set the fontsize of numbers, not labels for yticks:

1
2
3
4


x = [1, 2, 3, 4]
y = [1e6, 2e6, 2.5e6, 3e6]
fig, ax = plt.subplots()
ax.tick_params(axis='y', labelsize=14)

Set font size for the scale factor (offset text) in the scientific notation

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13


import matplotlib.pyplot as plt
x = [1, 2, 3, 4]
y = [1e6, 2e6, 2.5e6, 3e6]
fig, ax = plt.subplots()
ax.plot(x, y)

# Enable scientific notation on the y-axis
ax.ticklabel_format(axis='y', style='sci', scilimits=(-2, 2))

# Set font size for the scale factor (offset text)
ax.yaxis.get_offset_text().set_fontsize(14)

plt.show()

Margin

to leave more space around the figure to show the ticks of the boundary Geeksforgeeks

Or prevent the markers get clipped by the axes GfG plt xticks

1
2


fig, ax = plt.subplots()
ax.margins(0.5)

Legend

References:

gfg
Docs
How to add legend below subplots in matplotlib? - SO
How to specify legend position in graph coordinates - SO
- Searched by python matplotlib ax.legend specify positional coordinates in DDG

Notes:

Set legend to ax:

1
2


ax.plot([1, 2, 3], label='Inline label')
ax.legend(loc='best', bbox_to_anchor=(0.5, 0., 0.5, 0.5))

Put it below the subplots ^r3-SO:

1
2


ax.legend(loc="upper center", bbox_to_anchor=(0.5, -0.13),
          fancybox=False, shadow=False, ncol=5, fontsize=6)

(2024/12/04)

Specify the accurate posistion of legend ^r4-SO:
1

ax.legend(loc=(0.25, 0.67))
- 1 is for the right-bottom corner of the legend at the top edge of the plot.

Draw a horizontal line

plt : gfg

1

plt.axhline(y=0.5, color='r', linestyle='-')

ax: SO

1

ax.hlines(y=0.2, xmin=0.01, xmax=20, linewidth=2, color='r')

Tight layout

1

plt.tight_layout()

Or: plt.rcParams["savefig.bbox"] = 'tight'

(2024-07-23)

Or:

1
2
3
4
5


fig, ax = plt.subplots()
fig.tight_layout()
ax.set_title("Tweaking points within triangle", fontsize=16)

ax.set(xlim =(0, 2.5), ylim =(-1.25, 0.5)) 

Save as png

gfg

1
2
3
4


ax.set_facecolor('pink') # inner background
plt.savefig('depth_buffer.png', bbox_inches='tight', dpi =100, facecolor="white") # bkg color

plt.savefig('output.jpg', bbox_inches='tight', dpi =40) #is smaller

Draw markers on axes

plt.plot(x,y, zorder=10, clip_on=False)

plotting markers on top of axes

Arrow of axis

matplotlib axis arrow tip

Change fonts

Plot in Ubuntu has ugly fonts (which is like ‘sans-serif’).

plt.rcParams["font.family"] = "cursive", This will change to your computer’s default monospace font. How to change fonts in matplotlib (python)?

散点图

plt.scatter(x,y, c='r', s=4), s can control marker size, docs
marker 是反着画的
matplotlib.pyplot.scatter(x, y, s=None, c=None, marker=None, cmap=None, norm=None, vmin=None, vmax=None, alpha=None, linewidths=None, verts=None, edgecolors=None, hold=None, data=None, **kwargs)
x,y组成了散点的坐标；s为散点的面积；c为散点的颜色（默认为蓝色’b’）；marker为散点的标记；alpha为散点的透明度（0与1之间的数，0为完全透明，1为完全不透明）;linewidths为散点边缘的线宽；如果marker为None，则使用verts的值构建散点标记；edgecolors为散点边缘颜色。 csdn

拉长坐标间距

Python设置matplotlib.plot的坐标轴刻度间隔以及刻度范围

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19


import matplotlib.pyplot as plt
from matplotlib.pyplot import MultipleLocator # 用于设置刻度间隔

x_vals = list(range(11))
y_vals = [x**2 for x in x_vals]
plt.plot(x_vals, y_vals, c='green')
plt.title('Squares', fontsize=24)

plt.tick_params(axis='both',which='major',labelsize=14)
plt.xlabel('Numbers',fontsize=14)
plt.ylabel('Squares',fontsize=14)
x_major_locator=MultipleLocator(1) #把x轴的刻度间隔设置为1，并存在变量里
y_major_locator=MultipleLocator(10) #把y轴的刻度间隔设置为10，并存在变量里
ax=plt.gca() #ax为两条坐标轴的实例
ax.xaxis.set_major_locator(x_major_locator) #把x轴的主刻度设置为1的倍数
ax.yaxis.set_major_locator(y_major_locator) #把y轴的主刻度设置为10的倍数
plt.xlim(-0.5,11) #把x轴的刻度范围设置为-0.5到11，因为0.5不满一个刻度间隔，所以数字不会显示出来，但是能看到一点空白
plt.ylim(-5,110) #把y轴的刻度范围设置为-5到110，同理，-5不会标出来，但是能看到一点空白
plt.show()

设置图片尺寸大小

如何指定matplotlib输出图片的尺寸？ - pythonic生物人的回答 - 知乎 https://www.zhihu.com/question/37221233/answer/2250419008

plt.rcParams['figure.figsize'] = (12.0, 8.0)
fig, ax = plt.subplots(figsize=(15,0.5),dpi=300) pool

设置图片分辨率

plt.figure(figsize=(a, b), dpi=dpi)

matplotlib设置分辨率

网格

plt.grid(visible=True,linestyle="--", color='gray', linewidth='1',)

绘制CDF (221211)

Repeat drawing

How to change a matplotlib figure in a different cell in Jupyter Notebook?

1
2


fig.gca().scatter(x3, y3, color='g', linewidth=5)
fig

(2022-12-13)

Get the arguments name

use lib inspect with two .f_back: callers_local_vars = inspect.currentframe().f_back.f_back.f_locals.items(). Getting the name of a variable as a string

Get current color

How to get color of most recent plotted line in Python’s plt

plt.gca().lines[-1].get_color()

use ax Get default line colour cycle

1
2


line = ax.plot(x,y)
ax.plot(x, y+.3, color = line.get_color())

多曲线

Combine two figures

matplotlib: combine different figures and put them in a single subplot sharing a common legend

Multiple subplots

GfG

1
2
3
4
5
6
7


fig, ax = plt.subplots(3, 3)

# draw graph
for i in ax:
    for j in i:
        j.plot(np.random.randint(0, 5, 5), np.random.randint(0, 5, 5))
plt.show()

title of subplots: ax[0][0].set_title("xxx")

hide x ticks for top subplots and y ticks for right plots: How to set a single, main title above all the subplots with Pyplot?

1
2


plt.setp([a.get_xticklabels() for a in axarr[0, :]], visible=False)
plt.setp([a.get_yticklabels() for a in axarr[:, 1]], visible=False)

Set xticks for subplots: How to set xticks in subplots

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12


fig, axes = plt.subplots(nrows=3, ncols=4)

# Set the ticks and ticklabels for all axes
plt.setp(axes, xticks=[0.1, 0.5, 0.9], xticklabels=['a', 'b', 'c'],
        yticks=[1, 2, 3])

# Use the pyplot interface to change just one subplot...
plt.sca(axes[1, 1])
plt.xticks(range(3), ['A', 'Big', 'Cat'], color='red')

fig.tight_layout()
plt.show()

Set the fontsize of numbers of xticks: plt.setp(ax.get_xticklabels(), fontsize=12, fontweight="normal", horizontalalignment="left", rotation=90) set_xticks() needs argument for ‘fontsize’ #12318
set_xticklabels-Docs

多曲线不同尺度

Three lines with different ranges

References:
1. Gemini 2.5P - Plotting Three Curves With Different Scales

Supports:

(2025-08-11T14:56)

ax.twinx() ^r1-Gemini

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41


import matplotlib.pyplot as plt
import numpy as np

# Generate some sample data
x = np.linspace(0, 10, 100)
curve1 = 990 + 410 * np.sin(x) # Range: 990 - 1400
curve2 = 4 + 6 * np.cos(x)      # Range: -2 - 10
curve3 = 9 + 11 * np.sin(x+1)   # Range: -2 - 20

# Create a figure and a primary axes object
fig, ax1 = plt.subplots(figsize=(10, 6))

# Plot the first curve on the primary y-axis
color = 'tab:red'
ax1.set_xlabel('X-axis')
ax1.set_ylabel('Curve 1', color=color)
ax1.plot(x, curve1, color=color)
ax1.tick_params(axis='y', labelcolor=color)

# Create a second y-axis that shares the same x-axis
ax2 = ax1.twinx()
color = 'tab:blue'
ax2.set_ylabel('Curve 2', color=color)
ax2.plot(x, curve2, color=color)
ax2.tick_params(axis='y', labelcolor=color)

# Create a third y-axis
ax3 = ax1.twinx()
color = 'tab:green'
ax3.set_ylabel('Curve 3', color=color)
ax3.plot(x, curve3, color=color)
ax3.tick_params(axis='y', labelcolor=color)


# To prevent the third y-axis from overlapping with the second, we need to move it to the right.
ax3.spines['right'].set_position(('outward', 60))


fig.tight_layout() # otherwise the right y-label is slightly clipped
plt.savefig("three_curves_plot.png")
plt.close()

Scale y-axis

Log scale

References:
1. Docs

Supports:

plt.yscale

1
2
3
4


x = np.arange(len(Train_RMSE))
fig, ax = plt.subplots()
plt.yscale('log',)
ax.plot(x, Train_RMSE)

Scale

Problems:
1. Make the line only take the middle 50% y-range of the canvas.
References: {{{
1. Gemini 2.5P - Matplotlib Y-Axis Fraction Scaling }}}

Supports:

(2025-08-22T13:57)

Input

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53


# ===
# === Scaling the Line to the Middle 50%
# ===
import matplotlib.pyplot as plt
import matplotlib.ticker as mticker
from fractions import Fraction
import numpy as np

# --- 1. Original Data ---
x = [1, 2, 3, 4, 5]
y = np.array([0.65, 0.72, 0.78, 0.71, 0.68])

# --- 2. Define Canvas and Target Ranges ---
# The full y-range you want to see on the plot
canvas_min, canvas_max = -2, 6

# The middle 50% of the canvas range
canvas_range = canvas_max - canvas_min
target_min = canvas_min + 0.25 * canvas_range
target_max = canvas_max - 0.25 * canvas_range

# --- 3. Scale the Data ---
data_min, data_max = y.min(), y.max()
data_range = data_max - data_min
target_range = target_max - target_min

# Normalize data to a 0-1 scale, then scale to the target range
y_scaled = (((y - data_min) / data_range) * target_range) + target_min

# --- 4. Create an Inverse Formatter for Labels ---
def inverse_scale_formatter(scaled_val, pos):
    """
    Takes a scaled value from the y-axis and converts it back
    to its original data value for display as a label.
    """
    # Inverse normalization from target range back to 0-1
    normalized_val = (scaled_val - target_min) / target_range
    # Inverse scaling from 0-1 back to the original data range
    original_val = normalized_val * data_range + data_min
    # Format as a fraction
    return f'{Fraction(original_val).limit_denominator()}'

# --- 5. Create and Format the Plot ---
fig, ax = plt.subplots(figsize=(8, 6))

# Plot the SCALED data
ax.plot(x, y_scaled, marker='o', color='purple')

# Set the y-axis limits to the full canvas range
ax.set_ylim(canvas_min, canvas_max)

# Apply the special inverse formatter
ax.yaxis.set_major_formatter(mticker.FuncFormatter(inverse_scale_formatter))

Composite 2 imgs

(2023-12-20) Setting alpha:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13


import cv2
import matplotlib.pyplot as plt

img1 = cv2.imread('rot1.png')
pts1 = [(24, 124), (49, 124), (98, 124), (104, 124), (120, 124),
        (18, 146), (37, 146), (65, 146), (102, 146), (133, 146)]
for pt in pts1:
  cv2.circle(img1, pt, 0, (0, 0, 255))
plt.figure(figsize=(20,10))
plt.imshow(img1[:,:, ::-1])

img2 = cv2.imread('rot2.png')
plt.imshow(img2[:,:,::-1], alpha=0.6)

An example

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92


import matplotlib.pyplot as plt
import numpy as np
from matplotlib.ticker import MultipleLocator

epoch = np.array(
    [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30,
     31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58,
     59])
train_loss = np.array(
    [0.27195, 0.07782, 0.05188, 0.03321, 0.03054, 0.02322, 0.01962, 0.01537, 0.01347, 0.01266, 0.00895, 0.01107,
     0.00755, 0.00633, 0.00915, 0.00872, 0.00431, 0.00733, 0.00584, 0.00477, 0.00502, 0.00384, 0.00282, 0.00506,
     0.00307, 0.00325, 0.00588, 0.00277, 0.00445, 0.00318, 0.00181, 0.00532, 0.00218, 0.00379, 0.00190, 0.00201,
     0.00242, 0.00164, 0.00215, 0.00176, 0.00374, 0.00149, 0.00145, 0.00170, 0.00481, 0.00136, 0.00184, 0.00165,
     0.00162, 0.00265, 0.00285, 0.00172, 0.00241, 0.00139, 0.00181, 0.00161, 0.00135, 0.00264, 0.00165, 0.00158])
valid_loss = np.array(
    [0.04518, 0.02028, 0.01517, 0.01853, 0.00928, 0.01604, 0.01256, 0.01068, 0.01229, 0.01066, 0.01251, 0.00756,
     0.00836, 0.00777, 0.00623, 0.00942, 0.00761, 0.00691, 0.00626, 0.00607, 0.00708, 0.00703, 0.01061, 0.00758,
     0.00807, 0.00541, 0.00521, 0.00570, 0.00844, 0.00610, 0.00735, 0.00448, 0.00466, 0.00734, 0.00633, 0.00511,
     0.00737, 0.00380, 0.00410, 0.00604, 0.00685, 0.00546, 0.00618, 0.00411, 0.00543, 0.00572, 0.00631, 0.00821,
     0.00578, 0.00525, 0.00391, 0.00529, 0.00621, 0.00606, 0.00979, 0.00515, 0.00555, 0.00712, 0.00535, 0.00439])

# 解决曲线图里面中文显示乱码问题

# plt.rcParams['font.sans-serif'] = ['Hiragino Sans GB']  # 用来正常显示中文标签
#
# plt.rcParams['axes.unicode_minus'] = False
plt.figure(figsize=(8,6))

# 增加曲线说明

A, = plt.plot(epoch, train_loss, lw=1, label='train_loss')
B, = plt.plot(epoch, valid_loss, lw=1, label='valid_loss')
font1 = {'family': 'Times New Roman',
         'weight': 'normal',
         'size': 18, }
# x轴label

plt.xlabel("epoch", labelpad=2, fontsize=18,)

# y轴label

plt.ylabel("loss", labelpad=4, fontsize=18)

# 设置标题

# plt.title("change in loss", x=0.5, y=-0.2, pad=0.3, fontsize=18)
plt.title("change in loss", fontsize=18)
fig=plt.gcf()
fig.set_facecolor('white')


# ------限制显示x,y轴最小-最大值范围（刻度不一定是多少）

plt.xlim(0, 60)

plt.ylim(0, 0.3)

# ------设置x,y轴刻度

x_major_locator = MultipleLocator(10)

y_major_locator = MultipleLocator(0.02)

ax = plt.gca()

# ax为两条坐标轴的实例

ax.xaxis.set_major_locator(x_major_locator)
# ax.xaxis.set_ticks_position('bottom')
# 把x轴的主刻度设置为1的倍数

ax.yaxis.set_major_locator(y_major_locator)

# 把y轴的主刻度设置为10的倍数

# ------显示网格

# plt.grid()

plt.grid(True, linestyle="--", color='gray', linewidth='1', axis='both')

plt.legend(handles=[A, B], prop=font1)
plt.tick_params(labelsize=18)
labels = ax.get_xticklabels() + ax.get_yticklabels()
[label.set_fontname('Time New Roman') for label in labels]

font2 = {'family': 'Time New Roman',
         'weight': 'normal',
         'size': 18}

plt.show()
plt.savefig('saved_figure.jpg')

3D interactive plot

(2024-03-23)

Make 3D interactive Matplotlib plot in Jupyter Notebook - GfG

3D Scatter plot:

Code

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32


# for creating a responsive plot 
%matplotlib widget 

# importing required libraries 
from mpl_toolkits.mplot3d import Axes3D 
import matplotlib.pyplot as plt 

# creating random dataset 
xs = [14, 24, 43, 47, 54, 66, 74, 89, 12, 
      44, 1, 2, 3, 4, 5, 9, 8, 7, 6, 5] 

ys = [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 6, 3,
      5, 2, 4, 1, 8, 7, 0, 5] 

zs = [9, 6, 3, 5, 2, 4, 1, 8, 7, 0, 1, 2,
      3, 4, 5, 6, 7, 8, 9, 0] 

# creating figure 
fig = plt.figure() 
ax = fig.add_subplot(111, projection='3d') 

# creating the plot 
plot_geeks = ax.scatter(xs, ys, zs, color='green') 

# setting title and labels 
ax.set_title("3D plot") 
ax.set_xlabel('x-axis') 
ax.set_ylabel('y-axis') 
ax.set_zlabel('z-axis') 

# displaying the plot 
plt.show()

3D Bar plot:

Code

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28


%matplotlib widget 
from mpl_toolkits.mplot3d import Axes3D 
import matplotlib.pyplot as plt 
import numpy as np 

# creating random dataset 
xs = [2, 3, 4, 5, 1, 6, 2, 1, 7, 2] 
ys = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10] 
zs = np.zeros(10) 
dx = np.ones(10) 
dy = np.ones(10) 
dz = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10] 

# creating figure 
figg = plt.figure() 
ax = figg.add_subplot(111, projection='3d') 

# creating the plot 
plot_geeks = ax.bar3d(xs, ys, zs, dx,  
                      dy, dz, color='green') 

# setting title and labels 
ax.set_title("3D bar plot") 
ax.set_xlabel('x-axis') 
ax.set_ylabel('y-axis') 
ax.set_zlabel('z-axis') 

plt.show()

Camera Poses

(2024-03-23)

Generate by ChatGPT with prompt: “How to plot multiple camera poses in a single figure?”

Code

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56


%matplotlib widget
import numpy as np
import matplotlib.pyplot as plt

def plot_camera_poses(poses, axis_length=0.1):
    # Create a new figure
    fig = plt.figure()
    ax = fig.add_subplot(111, projection='3d')

    # Plot each camera pose
    for pose in poses:
        # Extract camera position and orientation
        cam_position = pose['position']
        cam_orientation = pose['rotation']

        # Plot camera position
        ax.scatter(cam_position[0], cam_position[1], cam_position[2], c='r', marker='o')

        # Plot camera axes
        axes_endpoints = cam_position + axis_length * cam_orientation
        ax.plot3D([cam_position[0], axes_endpoints[0,0]], 
                  [cam_position[1], axes_endpoints[0,1]], 
                  [cam_position[2], axes_endpoints[0,2]], 'r')
        ax.plot3D([cam_position[0], axes_endpoints[1,0]], 
                  [cam_position[1], axes_endpoints[1,1]], 
                  [cam_position[2], axes_endpoints[1,2]], 'g')
        ax.plot3D([cam_position[0], axes_endpoints[2,0]], 
                  [cam_position[1], axes_endpoints[2,1]], 
                  [cam_position[2], axes_endpoints[2,2]], 'b')

    # Set plot limits and labels
    ax.set_xlabel('X')
    ax.set_ylabel('Y')
    ax.set_zlabel('Z')

    # Show plot
    plt.show()

# Example camera poses (positions and rotation matrices)
poses = [
    {
        'position': np.array([0, 0, 0]),
        'rotation': np.array([[1, 0, 0],
                              [0, 1, 0],
                              [0, 0, 1]])
    },
    {
        'position': np.array([1, 1, 1]),
        'rotation': np.array([[0, -1, 0],
                              [1, 0, 0],
                              [0, 0, 1]])
    }
]

# Plot camera poses
plot_camera_poses(poses)

3D arrow

(2024-03-23)

Putting arrowheads on vectors in a 3d plot - SO

Code

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35


%matplotlib widget
import numpy as np
import matplotlib.pyplot as plt
from matplotlib.patches import FancyArrowPatch
from mpl_toolkits.mplot3d import proj3d

class Arrow3D(FancyArrowPatch):
    def __init__(self, xs, ys, zs, *args, **kwargs):
        super().__init__((0,0), (0,0), *args, **kwargs)
        self._verts3d = xs, ys, zs

    def do_3d_projection(self, renderer=None):
        xs3d, ys3d, zs3d = self._verts3d
        xs, ys, zs = proj3d.proj_transform(xs3d, ys3d, zs3d, self.axes.M)
        self.set_positions((xs[0],ys[0]),(xs[1],ys[1]))

        return np.min(zs)

arrow_prop_dict = dict(mutation_scale=10, arrowstyle='-|>', color='k', shrinkA=0, shrinkB=0)

fig = plt.figure()
ax = fig.add_subplot(111, projection='3d') 
    
a = Arrow3D([0, 1], 
            [0, 0], 
            [0, 0], **arrow_prop_dict)
ax.add_artist(a)
a = Arrow3D([0, 0], 
            [0, 1], 
            [0, 0], **arrow_prop_dict)
ax.add_artist(a)
a = Arrow3D([0, 0], 
            [0, 0], 
            [0, 1], **arrow_prop_dict)
ax.add_artist(a)

Convex polygon on 2D

(2024-03-26)

How to fill an area within a polygon in Python using matplotlib? - SO

Example from Scipy:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16


# Convex hull of a random set of points:
from scipy.spatial import ConvexHull, convex_hull_plot_2d
import numpy as np
rng = np.random.default_rng()
points = rng.random((30, 2))   # 30 random points in 2-D
hull = ConvexHull(points)

# Plot it:
import matplotlib.pyplot as plt
plt.plot(points[:,0], points[:,1], 'o')
for simplex in hull.simplices:
    plt.plot(points[simplex, 0], points[simplex, 1], 'k-')

plt.plot(points[hull.vertices,0], points[hull.vertices,1], 'r--', lw=2)
plt.plot(points[hull.vertices[0],0], points[hull.vertices[0],1], 'ro')
plt.show()

Fill 3D Polygon

(2024-03-26)

ax.fill is used for 2D, and doesn’t work for 3D.

ax.add_collection3d cannot follow the counter-clockwise order of convex hull to fill the polygon. For example, when drawing a rectangle, it will fill 2 triangles:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14


%matplotlib widget
import numpy as np
import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D
from mpl_toolkits.mplot3d.art3d import Poly3DCollection

fig = plt.figure()
ax = fig.add_subplot(111, projection='3d') 
verts = np.array([[ [0,0,0], 
                    [1,0,0], 
                    [0,1,0], 
                    [1,1,0] ]]) # (1,4,3)
rec = Poly3DCollection(verts, alpha=0.5, facecolors='cyan', edgecolors='k')
ax.add_collection3d(rec)

The Poly3DCollection code is generated by ChatGPT with prompt: “How to fill a 3D triangle with matlabplot”

So, I draw 2 triangles to form a quadrilateral:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19


%matplotlib widget
import numpy as np
import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D
from mpl_toolkits.mplot3d.art3d import Poly3DCollection

fig = plt.figure()
ax = fig.add_subplot(111, projection='3d') 
verts = np.array([[ [0,0,0], 
                    [1,0,0], 
                    [0,1,0],]])
tri = Poly3DCollection(verts, alpha=0.5, facecolors='cyan',)
ax.add_collection3d(tri)

verts = np.array([[ [1,0,0], 
                    [0,1,0],
                    [1,1,0] ]])
tri = Poly3DCollection(verts, alpha=0.5, facecolors='cyan',)
ax.add_collection3d(tri)

Plotting 3D Polygons - SO

Project 3D point onto plane

(2024-03-26)

How to project a point onto a plane in 3D? - SO

The distance from 3D point to the plane: dot product for 3D point and plane normal vector.
3D point minus the distance, resulting in the projection.

Code

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33


%matplotlib widget
import numpy as np
import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D
from mpl_toolkits.mplot3d.art3d import Poly3DCollection

R = np.array([[ 0.97026268,  0.00747991,  0.24193878], # cam x-axis
              [-0.01474287,  0.99949291,  0.02822342], # cam y-axis
              [-0.24160499, -0.030951  ,  0.96988095]]) #cam z-axis

fig = plt.figure()
ax = fig.add_subplot(111, projection='3d') 

# fill rectangle
verts = np.array([[ [0,0,0], R[0]*2, R[1]*2, ]]) # (1,4,3)
tri = Poly3DCollection(verts, alpha=0.5, facecolors='cyan', )
ax.add_collection3d(tri)
verts = np.array([[ R[0]*2, R[1]*2, (R[0]+R[1])*2 ]]) # (1,4,3)
tri = Poly3DCollection(verts, alpha=0.5, facecolors='cyan', )
ax.add_collection3d(tri)

p = np.array([1,1,1])
ax.scatter(*p, c='r', marker='o')
plane_normal = R[2]

len2plane = R[2]@ p.T

projection = p-len2plane * plane_normal

ax.scatter(*projection, marker='*')
ax.plot3D([p[0], projection[0]], 
          [p[1], projection[1]], 
          [p[2], projection[2]], 'gray')

Set line color & marker

(2024-07-23)

Set the line to orange and triangle marker:

1

edge_1, = axis.plot([0,0], [0,0.5], color='orange', marker='>')

Set marker appear only at the endpoint:
1

seg_2, = ax.plot([0,0], [0,0], color='green', marker='>', markersize='10', markevery=[-1])
- Ref: matplotlib: apply marker only to start point or end point? - SO (Searched by “python matplotlib pyplot set line marker only for one end” in DDG)
Set line end arrow
- Use quiver by setting starting point and direction: Arrow on a line plot (Searched by “python matplotlib line arrow” in DDG)

Python 制作动图

(2024-07-23)

用 Python 制作程序动画，并保存为 GIF

Ask Claude3.5: “How to use Python to create a GIF image showing a point moving within in a triangle?”

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45


import numpy as np
import matplotlib.pyplot as plt
from matplotlib.animation import FuncAnimation
from PIL import Image
import io

# Define the triangle vertices
triangle = np.array([(0, 0), (1, 0), (0.5, np.sqrt(3)/2)])

# Set up the figure and axis
fig, ax = plt.subplots()
ax.set_xlim(-0.1, 1.1)
ax.set_ylim(-0.1, 1)
ax.set_aspect('equal')

# Plot the triangle
ax.plot(np.append(triangle[:, 0], triangle[0, 0]),
        np.append(triangle[:, 1], triangle[0, 1]), 'k-')

# Initialize the point
point, = ax.plot([], [], 'ro')

# Animation function
def animate(frame):
    t = frame / 100  # Adjust speed
    # Calculate point position (example: move in a circular path within the triangle)
    x = 0.5 + 0.3 * np.cos(2 * np.pi * t)
    y = np.sqrt(3)/4 + 0.2 * np.sin(2 * np.pi * t)
    point.set_data(x, y)
    return point,

# Create the animation
anim = FuncAnimation(fig, animate, frames=100, interval=50, blit=True)

# Save the animation as a GIF (Don't work)
# frames = []
# for i in range(100):
#     img_buf = io.BytesIO()
#     plt.savefig(img_buf, format='png')
#     img_buf.seek(0)
#     frames.append(Image.open(img_buf))
# 
# frames[0].save('moving_point.gif', save_all=True, append_images=frames[1:], duration=50, loop=0)

plt.close(fig)

Saving animated plot as .gif with PillowWriter doesn’t work: MovieWriter ffmpeg unavailable; trying to use <class ‘matplotlib.animation.PillowWriter’> instead - SO (Search in DDG)

FuncAnimation

(2024-07-23)

Example: Create Animations with FuncAnimation in Python - PythonAlgos (Searched by “python make func animation” in DDG)

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27


%matplotlib widget
import numpy as np
import matplotlib.pyplot as plt
from matplotlib.animation import FuncAnimation

fig, ax = plt.subplots()

#plot sine function
_range = np.arange(0, 2*np.pi, 0.001)
sine = np.sin(_range)
line = plt.plot(_range, sine)

ax = plt.axis([0, 2*np.pi, -1, 1])
# The comma is a must, for unpackaging the tuple
dot, = plt.plot([0], [np.sin(0)], 'ro')
def func(i):
    dot.set_data(i, np.sin(i))
    return dot,

_animation = FuncAnimation(fig, func, 
      frames=np.arange(0, 2*np.pi, 0.1), 
      interval=10)

_animation.save('sine.mp4', fps=30, 
          extra_args=['-vcodec', 'libx264'])

plt.show()

Saving .mp4 requires: sudo apt install ffmpeg
Other Refs:
- More examples: Animations Using Python: A Comprehensive Guide - Medium
- Example of sine wave: Matplotlib.animation.FuncAnimation class in Python - GfG

Save animation with specifying format (mp4, gif) and dpi: matplotlib.animation.Animation.save - Docs (Found in DDG)

1
2
3
4
5


# Save the animation as a mp4 video
anim.save('Tweaking_within_Triangle.mp4', writer = 'ffmpeg', fps = 10)

# Save the animation as a GIF
anim.save('Tweaking_within_Triangle.gif', writer = 'pillow', fps = 10)

Ref: Animations using Matplotlib - saving animations

Omit frame and ticks

(2024-07-23)

Use transparent=True:

1
2


plt.axis('off')
plt.savefig('imgName',bbox_inches='tight',transparent=True, pad_inches=0)

Ref: savefig without frames, axes, only content - SO

Use fig.subplot_adjust():

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19


import matplotlib.pyplot as plt
import numpy as np
pixels = 512
px = 1/plt.rcParams['figure.dpi']  # pixel in inches
fig, ax = plt.subplots(frameon=False)   # Transparent content background
fig.set_size_inches(pixels*px, pixels*px)
ax = plt.Axes(fig, [0., 0., 1., 1.]) # This is needed for fitting content to 512*512 figure frame otherwise output figure will be less than desired size
ax.set_axis_off()
fig.add_axes(ax)

plt.plot(np.sin(np.arange(10)), c = 'black')

fig.subplots_adjust(bottom = 0)
fig.subplots_adjust(top = 0.00001) # This value should be very small so that marginal axes of subplot would not overlay on the main figure
fig.subplots_adjust(right = 1)
fig.subplots_adjust(left = 0)

plt.savefig('no_content_data.png')
plt.close()

Ref: savefig without frames, axes, only content - SO
Similar solution: Save an image (only content, without axes or anything else) to a file using Matloptlib - SO (Searched by “python matplotlib pyplot save figure without title or axis ticks” in DDG)

Annotation

Add Rectangle

(2024-08-03)

Ask Claude3.5

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45


import matplotlib.pyplot as plt
from matplotlib.patches import Rectangle, Patch

# --- Add a box to enclose methods being compared

from matplotlib.patches import Rectangle, Patch
# Find the indices of the ticks we want to highlight
highlight_methods = ["Method3", "Method4", "Method5"]
highlight_indices = [methods.index(method) for method in highlight_methods] # [3,4,5]


# Get all tick positions
tick_positions = ax.get_xticks()

# Get the x-coordinates of the first and last tick to be highlighted
x_start = tick_positions[highlight_indices[0]]
x_end = tick_positions[highlight_indices[-1]]
# Adjust to extend slightly beyond the ticks
x_start -= 0.24
x_end += 0.24

# Get the y-range of the plot
y_min, y_max = ax.get_ylim()

# Create a rectangle
rect = Rectangle((x_start, y_min+0.05), x_end - x_start, y_max - y_min-0.18, 
                 fill=False, edgecolor='magenta', linestyle='--', linewidth=2)

# Add the rectangle to the plot
ax.add_patch(rect)

# Adjust the plot limits to ensure the rectangle is visible
ax.set_xlim(tick_positions[0] - 0.5, tick_positions[-1] + 0.5)

# Create a custom patch for the legend
legend_patch = Patch(facecolor='none', edgecolor='magenta', linestyle='--', label='Comparison methods')

# Get the current handles and labels
handles, labels = ax.get_legend_handles_labels()

# Add the new patch to the handles
handles.append(legend_patch)

# Create the legend with the updated handles
ax.legend(handles=handles, loc='center left', bbox_to_anchor=(1, 0.5), fontsize=14)

Arrow

Arrow and text

References: {{{
1. python - Arrow properties in matplotlib annotate - Stack Overflow
  - Searched by python matplotlib annotation arrow in DDG
2. matplotlib.pyplot.annotate — Matplotlib 3.10.5 documentation }}}

Supports:

(2025-08-22T13:09)

ax.annotate ^r1-SO

1
2
3
4
5
6
7
8
9


ax.annotate('test', xy=(0.9, 0.9),
             xycoords='data',
             xytext=(0, 0),
             textcoords='data',
             arrowprops=dict(arrowstyle= '<|-|>',
                             color='blue',
                             lw=3.5,
                             ls='--')
           )

xycoords: coordinate system, default: data ^r2-Docs

Table of contents

Difference between plt and fig,ax

Inverse the X-axis

Add caption below the x-label

微分柱形图

Set xticks

Change the number and text of the showing Ticks

Set y-axis view limits

Set the 1st x as 1 (not 0)

Ticks

Add extra ticks, labels

Set xticks as integer

Hide ticks

Set tick bold, color

Fontsize of yticks

Margin

Legend

Draw a horizontal line

Tight layout

Save as png

Draw markers on axes

Arrow of axis

Change fonts

散点图

拉长坐标间距

设置图片尺寸大小

设置图片分辨率

网格

绘制CDF (221211)

Repeat drawing

Get the arguments name

Get current color

多曲线

Combine two figures

Multiple subplots

多曲线不同尺度

Scale y-axis

Composite 2 imgs

An example

3D interactive plot

Camera Poses

3D arrow

Convex polygon on 2D

Fill 3D Polygon

Project 3D point onto plane

Set line color & marker

Python 制作动图

FuncAnimation

Omit frame and ticks

Annotation

Add Rectangle

Arrow

Difference between `plt` and `fig,ax`