github python 画图_[挖坑]python画图学习

import numpy as np

import pandas as pd

import matplotlib.pyplot as plt

参考

注意

plt.show()之后会清空画板，所有绘图数据删除，保存前不要执行

在没有图形化界面的环境下，要执行plt.switch_backend('agg')

官方的案例最好了Examples

无图形化的终端画图

import matplotlib as mpl

mpl.use('Agg')

import matplotlib.pyplot as plt

绘图

二维图形绘制

x=np.linspace(0, 2*np.pi, num=100, endpoint=True)

ysin=np.sin(x)

ycos=np.cos(x)

#创建画板,内含1x1

fig, ax = plt.subplots(1,1,sharex=True,sharey=False,figsize=(8,6))

ax.plot(x,ysin,label="sin")

ax.plot(x,ycos,label="cos")

ax.set_xlabel('X')

ax.set_ylabel('Y')

ax.set_title("sin and cos")

ax.legend() #上图例，plt里面的label

显示plt.show(),plt.show()之后会清空画板，所有绘图数据删除，保存前不要执行

保存图形命令

figfile="sincos.png" #支持png,pdf等多种格式

plt.savefig(figfile,dpi=60)

多个画板

fig, [ax1,ax2] = plt.subplots(1,2,sharex=True,sharey=False,figsize=(8,6))

ax1.plot(x,ysin,label="sin")

ax2.plot(x,ycos,label="cos")

ax1.set_title("sin")

ax2.set_title("cos")

ax1.legend() #上图例，plt里面的label

ax2.legend() #上图例，plt里面的label

figfile="sincos.png" #支持png,pdf等多种格式

plt.savefig(figfile,dpi=100)

也可以

fig, ax = plt.subplots(1,2,sharex=True,sharey=False,figsize=(8,6))

ax[0].plot(x,ysin,label="sin")

ax[1].plot(x,ycos,label="cos")

填充图

ysin=np.sin(x)

ycos=np.cos(x)-0.5

fig, ax = plt.subplots(1,1,sharex=True,sharey=False,figsize=(8,6))

ax.fill_between(x, 0, ysin,alpha=0.5) #,, facecolor='blue', alpha=0.5)

ax.fill_between(x, 0, ycos,alpha=0.5) #,, facecolor='red', alpha=0.5)

文本

latex展示

使用r"$Latex语法$"

xyz=[r"$\alpha_{xx}$","yy","zz","average"]

常用代码段

画峰值出x坐标

def peaklabel(plt,xdata,ydata):

scale=1.1

left=ydata[0:-1]-ydata[1:]*scale>0#我在左面时大 right=ydata[1:]-ydata[0:-1]*scale>0#我在右面时大 left=np.append(left,True)

choose=np.append(left[0],left[1:] & right[:] )

#choose=choose & ( ydata > np.average(ydata) ) #choose=choose & ( ydata > np.max(ydata)/5.0 ) x=xdata[choose]

y=ydata[choose]

for xy in zip(x, y):

plt.annotate("%2.3f" % xy[0], xy=xy, xytext=(-20, 10), textcoords='offset points')

周期逐增画上x的位置

deltaeV=1.0 #画图标签间隔,可以理解为带间距N_each_deltaeV=20 #deltaeV内Y方向最多有多少个标签，若超过取余从0计数mine=-1E15

for i in np.arange(gap.size):

#垂直线 plt.vlines(gap[i], ylimit[0], ylimit[1], colors = "g", linestyles = "dashed",color=colors[i])

#plt.annotate(str(round(gap[i],2)), # xytext=(gap[i], 1.0*(ylimit[1]-ylimit[0])/gap.size*(i+1) ), # textcoords='offset points',color=colors[i]) x=gap[i]

if False:

left=gap.size - gapnum

if i < left:

y=(ylimit[1]-ylimit[0])*1.0/(left+2)*(left-i)+ylimit[0]

else:

y=(ylimit[1]-ylimit[0])*1.0/(gapnum+2)*(i-left+1)+ylimit[0]

else:

if(gap[i] - mine > deltaeV ):

mine = gap[i]

position=1

else:

position=position+1

position=position%N_each_deltaeV

y=(ylimit[1]-ylimit[0])*(1.0*position/N_each_deltaeV)#在deltaeV内等间距插入N_each_deltaeV个点 if ( xlimit[0] == None or x > xlimit[0]) and ( xlimit[1] == None or x < xlimit[1] ):

plt.text(x,y,str(round(gap[i],2)),fontsize=15,

color=colors[i],

bbox=dict(boxstyle='square,pad=0.1', fc="white",alpha=1.0,lw=0.0),

verticalalignment="center",horizontalalignment="center")

差值拟合

高斯滤波器

#原始画图方法axs.plot(DBgrid,avTDB[i],label=DBlable[i])

#滤波后画图from scipy.ndimage import gaussian_filter1d

avTDB[i] = gaussian_filter1d(avTDB[i], 2) #参数解释(原始数据,sigma,)axs.plot(DBgrid,avTDB[i],label=DBlable[i])

插值

from scipy.interpolate import interp1d

#获得插值函数的参数f = interp1d(x, y, kind='cubic')#参数(原始数据x,y,插值算法)

#计算插值数据xx = np.linspace(5,10,10000)

yy = f(xx)

插值算法有linear', 'nearest', 'zero', 'slinear', 'quadratic', 'cubic'

linear 线性

cubic 三次

高斯滤波后再插值效果更好