Draw a weakness graph in Python and save it in various formats (Raspberry Pi, macOS)


Draw a weakness graph with Japanese and save it in various formats.

--Image (PNG, JPEG, TIFF, SVG, PDF) --GIF animation, MPEG4 video --Real-time drawing

If you want to try it right away, pass the commentary

heart_curve_mini.gif This is an animated GIF output at 50% scale.

What I did (Raspberry Pi, macOS)

  1. Draw a handwritten graph using matplotlib, Python's graph drawing library.
  2. Make Japanese fonts available and make them handwritten.
  3. Write the formula in LaTeX notation. (Japanese fonts cannot be used.)
  4. Draw in real time like an animation.
  5. Save the final result as an image (PNG, JPEG, TIFF, SVG, PDF).
  6. Save the progress as an MPEG4 video.
  7. Save the progress as a GIF animation.

Environment (2021-01-14)

Raspberry Pi: Raspberry Pi 3B, Raspbian Stretch, Python 3.5.3 Raspberry Pi 4B, Raspbian Buster, Python 3.7.3 macOS:Macbook Air 13inch 2017、Mojave、Python 3.7.9

matplotlib 3.0.3 to 3.3.3, numpy 1.17.0 to 1.18.2 ffmpeg 1.4

It's been running the same code for quite some time, so the environment is less version-dependent. Perhaps it will work fairly well in a Windows environment.

Handwritten graph? What is it?

What is xkcd? ?? This is a science or otaku web comic. (→ wikipedia) It looks theoretical and scientific, but it has a black joke flavor. The author is Randall Munroe

The xkcd-like graph is a weak system and looks like this (→ Image Search)

So, this xkcd-like graph drawing function is officially built into the graph drawing library matplotlib. → Quite motivated and weak matplotlib official gallery (I'm just replacing the PNG image with the URL with/xkcd/added.)

If you put this weakness graph in a presentation / class slide, it will give you a nice taste, and if you draw it with animation, you can get more attention.

Explanation of how to draw a weakness system

Pass for those who want to try it immediately

It's very easy if you just want to make it look like xkcd. Just add `plt.xkcd ()`. It adjusts various parameters in the weakening direction.

from matplotlib import pyplot as plt

#After that, weakness system

If you want to make it partially xkcd-like, use `` `with``` blocking.

#Outside the block, it's a serious graph.
with plt.xkcd():
    #If you draw in this block, it is a weakness system.
#Outside the block, it's a serious graph.

Animation in real time

If you want to see what you are drawing in real time, draw in interactive mode with `plt.ion ()`.

plt.ion()     #Start interactive mode
#Draw a graph
plt.ioff()    #End interactive mode
plt.show()    #Leave the window even after you finish drawing.

Save in various formats

Press the button to save the image

The window that pops up when you draw the graph has a save button (an icon that doesn't fit the youth of today). Press this to save as PNG. PythonScreenSnapz014.png

Save various formats programmatically

I was able to save as PNG, JPEG, TIFF, SVG, PDF without any problem.

In the case of macOS, the backend cannot be saved with the default MacOSX, and it becomes a pure white image. I had to choose Agg, a backend without GUI. unsolved. For Raspberry Pi, the default backend PyQt5 is okay. You can save images while displaying the screen.

#You can save multiple types.
    plt.savefig("heart.png ")
    plt.savefig("heart.jpeg ")

# PS,In EPS, the axis etc. has become a straight line.
#If you specify a backend, specify it before importing pyplot.
import matplotlib
from matplotlib import pyplot as plt
#Check the specified backend
print("current backend:", plt.get_backend())

Save as MPEG4 video or GIF animation

The two are similar, switching between writer.

mp4 version

FFMpegWriter = manimation.writers['ffmpeg']
metadata = dict(title='Heart Curve', artist='Matplotlib',
        comment='Movie support!')
writer = FFMpegWriter(fps=15, metadata=metadata)

with writer.saving(fig, "heart_curve.mp4", 100):

GIF animated version

import matplotlib.animation as manimation
from matplotlib.animation import PillowWriter

PillowWriter = manimation.writers['pillow']
fig.set_dpi(50)  #How it works depends on the back end

with writer.saving(fig, "heart_curve.gif", dpi=None):

Show what you are drawing in real time

Every time you draw something, you can pause it as `plt.pause (0.5)`.

Font related

When you install fonts, you need to rebuild the matplotlib font cache.

import matplotlib.font_manager as fm

Code that extracts only the font name and displays it in a list

You will need it when you specify the font family name like this,

plt.rcParams["font.family"] = "Natsume"

import matplotlib.font_manager as fm

font_names=[] for font_path in fm.findSystemFonts(): try: font_name = fm.FontProperties(fname=font_path).get_name() except Exception as e: print(font_path) print("ERROR::: can't get the font name from the file above. ", e) print("********") finally: font_names.append(font_name)

font_names2 = list(set(font_names))

for font_name in font_names2: print(font_name)

### Reference: Handwritten font information

 Original xkcd font [xkcd] [xkcd Script]
 It seems that it was sampled from the handwritten characters of the author of the comic. Of course, only the alphabet.

 Japanese handwritten font [keyword specified by font.family]

 851 MkPOP [851MkPOP]
 851 Handwritten miscellaneous font [851tegakizatsu]
 Natsume Moji / Natsume Moji Suppression [Natsume] [Natsumemozi-o]
 Apricot Moji / Apricot Moji Monospaced [APJapanesefont] [APJapanesefontT]
 Apricot moji performance / Apricot moji 湛 [APJapanesefontK] [APJapanesefontT]
 Apricot Moji Beginning / Apricot Moji Beginning Monospaced [APJapanesefontH] [APJapanesefontHT]
 Ranobe POP [07LightNovelPOP]
 Gagagaga FREE version (katakana) [GAGAGAGA]

### path!
 Aside from the long story.
# Install the font and try it out

 If you don't have handwritten fonts, install "Natsume Moji" for the time being. If you comment out the font specification line, you can try it only with weak lines, but it is less than half the attractiveness, so please install it. Please read the above [Font related](#フォント関係).

 We recommend Kyoko's "Natsume Moji".

 If you want to use your favorite font, use [Code to extract and list only font names](#フォント名だけ抜き出してリスト表示するコード) to find out the font names.

For Raspberry Pi, after installing Rebuild the system font cache. fc-cache -v

Python font cache rebuild import matplotlib.font_manager as fm fm._rebuild()

# Deliverables and codes

 The code can run ipython3 in the terminal and copy and paste all lines in one shot.

 In addition, all are iPython, but REPL such as mu-editor, Jupyter Notebook, etc. is also okay. I wonder if VS Code's Jupyter is crazy. However, note that VS Code's Jupyter environment seems to be slow in real time.

 Of course, you can save it as a .py file and run it.

# Weakness graph


from matplotlib import pyplot as plt import numpy as np


#plt.rcParams["font.family"] = "Gen Shin Gothic P" plt.rcParams["font.family"] = "Natsume" #plt.rcParams["font.family"] = "GenEi LateGo"

plt.plot(np.sin(np.linspace(0, 10))) plt.title('Yeah! You can also use Japanese handwritten fonts') plt.show()

# Weakness pie chart


from matplotlib import pyplot as plt import numpy as np

plt.xkcd() plt.rcParams["font.family"] = "Natsume"

x = np.array([0.2, 0.4, 0.15, 0.25])

#plt.rcParams["font.family"] = "Natsume" plt.rcParams["font.family"] = "851MkPOP" plt.rcParams["font.size"] = 24

labels = ['melon', 'banana', 'Grape', 'Apple'] colors = ['yellowgreen', 'gold', 'darkviolet', 'hotpink'] plt.pie(x, autopct='%d%%', labels=labels, colors=colors) plt.axis('equal') plt.title('Favorite fruit questionnaire results', fontname="Natsume") plt.tight_layout()


# Weakness function


from matplotlib import pyplot as plt from numpy import arange, pi, sin, cos

plt.xkcd() plt.rcParams["font.family"] = "Natsume"

STEP = 0.1 # 0.01 to 0.Speed ​​/ accuracy adjustment with 1

def draw_graph(x, y, title, color): plt.title(title, fontsize=24) plt.gca().set_aspect('equal', adjustable='box') plt.plot(x, y, color=color, linewidth=8) plt.show()

def draw_heart(): intervals = arange(0, 2 * pi, STEP) x = [] y = [] for t in intervals: x.append(16 * sin(t) ** 3) y.append(13 * cos(t) - 5 * cos(2 * t) - 2 * cos(3 * t) - cos(4 * t)) draw_graph(x, y, title='Heart function', color='#FF6597')


# Weakness function with real-time drawing version TeX demo

 This is Jupyter's in-page drawing, it takes a lot of time to draw frame by frame without overwriting each frame. It depends on the machine power, but it may be better not to do it with VS Code. Let's feed iPython from the terminal.


realtime drawing version

import matplotlib

from matplotlib import pyplot as plt from numpy import arange, pi, sin, cos

plt.xkcd() plt.rcParams["font.family"] = "Natsume" plt.rcParams["font.size"] = 12 plt.rcParams['xtick.direction'] = 'in' plt.rcParams['ytick.direction'] = 'in'

plt.xticks(arange(-20, 21, 10)) plt.yticks(arange(-20, 21, 5))

plt.title('Heart function', fontsize=24) plt.gca().set_aspect('equal', adjustable='box')

STEP = 0.05 # 0.01 to 0.Speed ​​/ accuracy adjustment with 1

def draw_heart(): intervals = arange(0, 2 * pi, STEP) x = [] y = [] plt.ion()

for t in intervals:
    x.append(16 * sin(t) ** 3)
    y.append(13 * cos(t) - 5 * cos(2 * t) - 2 * cos(3 * t) - cos(4 * t))
    plt.plot(x, y, color='#FF6597', linewidth=8)



draw_heart() #plt.close()

plt.text(0.0, 2.0, r"\mathrm{x = 16sin^3\theta}", horizontalalignment='center') plt.pause(0.5) plt.text(0.0, 0.0, r"\mathrm{y = 13cos\theta - 5cos2\theta - 2cos3\theta - cos4\theta}", horizontalalignment='center') plt.pause(0.5) plt.text(0.0, -2.0, r"\mathrm{0\leqq\theta\leqq2\pi}", horizontalalignment='center') plt.pause(0.5)

plt.ioff() plt.show()

 Animation in Jupyter Lab

Recommended Posts

Draw a weakness graph in Python and save it in various formats (Raspberry Pi, macOS)
Temporarily save a Python object and reuse it in another Python
Draw a graph of a quadratic function in Python
Draw graph in python
Draw a heart in Python
How to interactively draw a machine learning pipeline with scikit-learn and save it in HTML
Draw a scatterplot matrix in python
Make a thermometer with Raspberry Pi and make it viewable with a browser Part 4
Draw a watercolor illusion with edge detection in Python3 and openCV3
Draw a CNN diagram in Python
[Python] How to save the installed package and install it in a new environment at once Mac environment
Get GrovePi + sensor value with Raspberry Pi and store it in kintone
RabbitMQ message notification app in Python with Growl ~ with Raspberry Pi and Julius ~
[Raspberry Pi] Add a thermometer and a hygrometer
Draw a heart in Python Part 2 (SymPy)
Easily graph data in shell and Python
Draw a tree in Python 3 using graphviz
Make a thermometer with Raspberry Pi and make it visible on the browser Part 3
Use libsixel to output Sixel in Python and output a Matplotlib graph to the terminal.
Creating a temperature control system with Raspberry Pi and ESP32 (3) Recipient Python file
How to read a serial number file in a loop, process it, and graph it
Use python on Raspberry Pi 3 and turn on the LED when it gets dark!
Until you get daily data for multiple years of Japanese stocks and save it in a single CSV (Python)
Draw a graph with Japanese labels in Jupyter
Draw a "breast curved surface" in a 3D graph (1)
Create a standard normal distribution graph in Python
[Python] Draw a directed graph with Dash Cytoscape
Block and draw images in Minecraft Pi Edition
I wrote a class in Python3 and Java
Let's write a Python program and run it
POST JSON in Python and receive it in PHP
[Python] How to draw a histogram in Matplotlib
Draw a "breast curved surface" in a 3D graph (2)
Draw multiple photos in a graph from multiple folders
Build a Python development environment on Raspberry Pi
A Python script that crawls RSS in Azure Status and posts it to Hipchat
The result of making a map album of Italy honeymoon in Python and sharing it
I created a stacked bar graph with matplotlib in Python and added a data label
[Selenium] Open the link in a new tab and move it [Python / Chrome Driver]
Draw a graph in Julia ... I tried a little analysis
Create a color sensor using a Raspberry Pi and a camera
Receive the form in Python and do various things
[Python] How to draw a line graph with Matplotlib
Calculate and draw bounded (closed) Voronoi diagrams in Python
Until drawing a 3D graph in Python on windows10
Carefully understand the exponential distribution and draw in Python
Carefully understand the Poisson distribution and draw in Python
Ubuntu 20.04 on raspberry pi 4 with OpenCV and use with python
Install pyenv on Raspberry Pi and version control Python
I wrote python3.4 in .envrc with direnv and allowed it, but I got a syntax error
It is blocked by Proxy, a connection error occurs in Python or pip, and it is retried.
Recursively get the Excel list in a specific folder with python and write it to Excel.
It was great to edit the Python file in the Raspberry Pi with Atom's remote function
A story about trying to use cron on a Raspberry Pi and getting stuck in space
An easy way to view the time taken in Python and a smarter way to improve it