When observation data, dynamic image than a static image to better reflect the trends and characteristics of the data. This article documents how to use matplotlib dynamic graphics rendering.
The main advantage of the Animation API class
1.matplotlib.animation
First, let's draw a line, a line drawing of the process dynamic:
import numpy as np
import matplotlib.pyplot as plt
from matplotlib.animation import FuncAnimation #导入负责绘制动画的接口
#其中需要输入一个更新数据的函数来为fig提供新的绘图信息
fig, ax = plt.subplots() #生成轴和fig, 可迭代的对象
x, y= [], [] #用于接受后更新的数据
line, = plt.plot([], [], '.-') #绘制线对象，plot返回值类型，要加逗号
#------说明--------#
#核心函数包含两个：
#一个是用于初始化画布的函数init()
#另一个是用于更新数据做动态显示的update()
def init():
#初始化函数用于绘制一块干净的画布，为后续绘图做准备
ax.set_xlim(-5, 15*np.pi) #初始函数，设置绘图范围
ax.set_ylim(-3, 3)
return line
def update(step): #通过帧数来不断更新新的数值
x.append(step)
y.append(np.cos(step/3)+np.sin(step**2)) #计算y
line.set_data(x, y)
return line
#fig 是绘图的画布
#update 为更新绘图的函数，step数值是从frames 传入
#frames 数值是用于动画每一帧的数据
ani = FuncAnimation(fig, update, frames=np.linspace(0, 13*np.pi, 128),
init_func=init,interval=20)
plt.show()
To make the program stops at the number of frames we need instead of continuing to draw, you need to slightly modify the above procedures, limit the number of frames added:
fig, ax = plt.subplots()
x, y= [], []
line, = plt.plot([], [], '.-',color='orange')
nums = 50 #需要的帧数
def init():
ax.set_xlim(-5, 60)
ax.set_ylim(-3, 3)
return line
def update(step):
if len(x)>=nums: #通过控制帧数来避免不断的绘图
return line
x.append(step)
y.append(np.cos(step/3)+np.sin(step**2)) #计算y
line.set_data(x, y)
return line
ani = FuncAnimation(fig, update, frames=nums, #nums输入到frames后会使用range(nums)得到一系列step输入到update中去
init_func=init,interval=20)
plt.show()
2. Draw a traveling wave
Traveling wave can be understood as the continuous phase encode positive cosine wave, then the function will modify the above update:
fig, ax = plt.subplots()
x, y= [], []
line, = plt.plot([], [], '.-',color='green')
nums = 100 #需要的帧数
def init():
ax.set_xlim(-5, 40)
ax.set_ylim(-2, 2)
return line
def update(step):
x = np.linspace(0,10*np.pi,100)
y = np.cos(1.2*x+step) #这里只改变相位
line.set_data(x, y) #设置新的 x，y
return line
ani = FuncAnimation(fig, update, frames=nums, #nums输入到frames后会使用range(nums)得到一系列step输入到update中去
init_func=init,interval=100)
plt.show()
3. A simple little animation
We then try to draw two-dimensional animation, just traveling wave can be turned into a two-dimensional waves to draw:
fig, ax = plt.subplots()
x = np.linspace(-10*np.pi,10*np.pi,100)
y = np.linspace(-10*np.pi,10*np.pi,100)
X,Y = np.meshgrid(x,y)
img = plt.imshow(np.sin(X))
nums = 1000 #需要的帧数
def init():
ax.set_xlim(0, 100)
ax.set_ylim(0, 100)
return img
def update(step):
# z = np.cos((0.5*X*(np.cos(0.01*step)+0.1))+step)+np.sin(0.5*(Y*(np.sin(0.01*step)+0.1))+step)
z = np.cos((0.5*X+np.cos(2*step))+step)+np.sin(0.5*(Y+np.sin(2*step))+step)
img.set_data(z) #设置新的 x，y
return img
ani = FuncAnimation(fig, update, frames=nums, #nums输入到frames后会使用range(nums)得到一系列step输入到update中去
init_func=init,interval=40)
plt.show()
Modifications to the parameters of the more interesting point:
fig, ax = plt.subplots()
x = np.linspace(-10*np.pi,10*np.pi,300)
y = np.linspace(-10*np.pi,10*np.pi,300)
X,Y = np.meshgrid(x,y)
img = plt.imshow(np.sin(X),cmap='ocean')
nums = 1000 #需要的帧数
def init():
ax.set_xlim(0, 100)
ax.set_ylim(0, 100)
return img
def update(step):
z = np.cos((0.5*X*(np.cos(0.01*step)+0.1))+step)+np.sin(0.5*(Y*(np.sin(0.01*step)+0.1))+step)
# z = np.cos((X*(np.cos(0.01*step)+0.2))+step)+np.sin((Y*(np.sin(0.01*step)+0.2))+step)
img.set_data(z) #设置新的 x，y
return img
ani = FuncAnimation(fig, update, frames=nums, #nums输入到frames后会使用range(nums)得到一系列step输入到update中去
init_func=init,interval=40)
plt.show()
Ready to paint waves, but seemed to see the aurora. . .
In addition to the above method, matplotlib is also provided a method for image-based rendering animation list further:
This is used in animation ArtistAnimation, the difference of this function requires that before drawing to calculate all survive drawing object, the drawing corresponds to the result of the calculation of dynamic playback. The above traveling wave may be displayed in such a manner:
fig, ax = plt.subplots()
x = np.linspace(-10*np.pi,10*np.pi,300)
y = np.linspace(-10*np.pi,10*np.pi,300)
X,Y = np.meshgrid(x,y)
nums = 100 #需要的帧数,这里要注意不要爆内存，另外大一点有利于显示完全函数的变换周期
imgs = [] #用于存储所有的img
fig = plt.figure() #只需要一个图，不需要iter了
def make_z(step):
z = np.cos((0.5*X*(np.cos(0.01*step)+0.1))+step)+np.sin(0.5*(Y*(np.sin(0.01*step)+0.1))+step)
return z
for step in range(nums):
img = make_z(step)
im = plt.imshow(img, animated=True)
imgs.append([im]) #这里将多个图像对象叠加在一起，imgs为列表对象
ani = animation.ArtistAnimation(fig, imgs, interval=100, repeat_delay=100)
#设置每帧间隔iterval ms，以及播放完以后循环播放检测
plt.show()
TODO(rjj): another effects learning
ref:
https://matplotlib.org/api/animation_api.html?highlight=animation#module-matplotlib.animation
https://matplotlib.org/gallery/animation/dynamic_image.html
https://zhuanlan.zhihu.com/p/32444081
https://blog.csdn.net/liang890319/article/details/52067251
https://blog.csdn.net/liang890319/article/details/52062905
https://zhuanlan.zhihu.com/p/35818908
gif:https://blog.csdn.net/weixin_33895695/article/details/87990582
imagemagic:https://www.cnblogs.com/congtou5/p/5222769.html
picture from:https://www.easyicon.net/1225581-animal_character_psyduck_screech_yellow_icon.html
