最终选用了python+Matplotlib。Matplotlib是著名Python的标配画图包,其绘图函数的名字基本上与 Matlab 的绘图函数差不多。优点是曲线精致,软件开源免费,支持Latex公式插入,且许多时候只需要一行或几行代码就能搞定。
然后小编经过了几天的摸索,找了几个不错的python代码模板,供大家简单修改就能快速上手使用。建议使用Wing Personal 作为PythonIDE,生成的图片能上下左右进行调整:
NO.1
# -*- coding: utf-8 -*- import numpy as np import matplotlib.pyplot as plt plt.rcParams['font.sans-serif']=['Arial']#如果要显示中文字体,则在此处设为:SimHei plt.rcParams['axes.unicode_minus']=False#显示负号x = np.array([3,5,7,9,11,13,15,17,19,21]) A = np.array([0.9708, 0.6429, 1, 0.8333, 0.8841, 0.5867, 0.9352, 0.8000, 0.9359, 0.9405]) B= np.array([0.9708, 0.6558, 1, 0.8095, 0.8913, 0.5950, 0.9352, 0.8000, 0.9359, 0.9419]) C=np.array([0.9657, 0.6688, 0.9855, 0.7881, 0.8667, 0.5952, 0.9361, 0.7848, 0.9244, 0.9221]) D=np.array([0.9664, 0.6701, 0.9884, 0.7929, 0.8790, 0.6072, 0.9352, 0.7920, 0.9170, 0.9254])#label在图示(legend)中显示。若为数学公式,则最好在字符串前后添加"$"符号 #color:b:blue、g:green、r:red、c:cyan、m:magenta、y:yellow、k:black、w:white、、、 #线型:- -- -. : , #marker:. , o v < * + 1 plt.figure(figsize=(10,5)) plt.grid(linestyle = "--") #设置背景网格线为虚线 ax = plt.gca() ax.spines['top'].set_visible(False) #去掉上边框 ax.spines['right'].set_visible(False) #去掉右边框plt.plot(x,A,color="black",label="A algorithm",linewidth=1.5) plt.plot(x,B,"k--",label="B algorithm",linewidth=1.5) plt.plot(x,C,color="red",label="C algorithm",linewidth=1.5) plt.plot(x,D,"r--",label="D algorithm",linewidth=1.5)group_labels=['dataset1','dataset2','dataset3','dataset4','dataset5',' dataset6','dataset7','dataset8','dataset9','dataset10'] #x轴刻度的标识 plt.xticks(x,group_labels,fontsize=12,fontweight='bold') #默认字体大小为10 plt.yticks(fontsize=12,fontweight='bold') plt.title("example",fontsize=12,fontweight='bold') #默认字体大小为12 plt.xlabel("Data sets",fontsize=13,fontweight='bold') plt.ylabel("Accuracy",fontsize=13,fontweight='bold') plt.xlim(3,21) #设置x轴的范围 #plt.ylim(0.5,1)#plt.legend() #显示各曲线的图例 plt.legend(loc=0, numpoints=1) leg = plt.gca().get_legend() ltext = leg.get_texts() plt.setp(ltext, fontsize=12,fontweight='bold') #设置图例字体的大小和粗细plt.savefig('D:\\filename.png') #建议保存为svg格式,再用inkscape转为矢量图emf后插入word中 plt.show()
效果图:
NO.2
# coding=utf-8import numpy as np import matplotlib.pyplot as pltplt.rcParams['font.sans-serif'] = ['Arial'] # 如果要显示中文字体,则在此处设为:SimHei plt.rcParams['axes.unicode_minus'] = False # 显示负号x = np.array([1, 2, 3, 4, 5, 6]) VGG_supervised = np.array([2.9749694, 3.9357018, 4.7440844, 6.482254, 8.720203, 13.687582]) VGG_unsupervised = np.array([2.1044724, 2.9757383, 3.7754183, 5.686206, 8.367847, 14.144531]) ourNetwork = np.array([2.0205495, 2.6509762, 3.1876223, 4.380781, 6.004548, 9.9298])# label在图示(legend)中显示。若为数学公式,则最好在字符串前后添加"$"符号 # color:b:blue、g:green、r:red、c:cyan、m:magenta、y:yellow、k:black、w:white、、、 # 线型:- -- -. : , # marker:. , o v < * + 1 plt.figure(figsize=(10, 5)) plt.grid(linestyle="--") # 设置背景网格线为虚线 ax = plt.gca() ax.spines['top'].set_visible(False) # 去掉上边框 ax.spines['right'].set_visible(False) # 去掉右边框plt.plot(x, VGG_supervised, marker='o', color="blue", label="VGG-style Supervised Network", linewidth=1.5) plt.plot(x, VGG_unsupervised, marker='o', color="green", label="VGG-style Unsupervised Network", linewidth=1.5) plt.plot(x, ourNetwork, marker='o', color="red", label="ShuffleNet-style Network", linewidth=1.5)group_labels = ['Top 0-5%', 'Top 5-10%', 'Top 10-20%', 'Top 20-50%', 'Top 50-70%', ' Top 70-100%'] # x轴刻度的标识 plt.xticks(x, group_labels, fontsize=12, fontweight='bold') # 默认字体大小为10 plt.yticks(fontsize=12, fontweight='bold') # plt.title("example", fontsize=12, fontweight='bold') # 默认字体大小为12 plt.xlabel("Performance Percentile", fontsize=13, fontweight='bold') plt.ylabel("4pt-Homography RMSE", fontsize=13, fontweight='bold') plt.xlim(0.9, 6.1) # 设置x轴的范围 plt.ylim(1.5, 16)# plt.legend() #显示各曲线的图例 plt.legend(loc=0, numpoints=1) leg = plt.gca().get_legend() ltext = leg.get_texts() plt.setp(ltext, fontsize=12, fontweight='bold') # 设置图例字体的大小和粗细plt.savefig('./filename.svg', format='svg') # 建议保存为svg格式,再用inkscape转为矢量图emf后插入word中 plt.show()
效果图:
NO.3
# coding=utf-8 import matplotlib.pyplot as plt from matplotlib.pyplot import figure import numpy as npfigure(num=None, figsize=(2.8, 1.7), dpi=300) #figsize的2.8和1.7指的是英寸,dpi指定图片分辨率。那么图片就是(2.8*300)*(1.7*300)像素大小 test_mean_1000S_n = [0.7,0.5,0.3,0.8,0.7,0.5,0.3,0.8,0.7,0.5,0.3,0.8,0.7,0.5,0.3,0.8,0.7,0.5,0.3,0.8] test_mean_1000S = [0.9,0.8,0.7,0.6,0.9,0.8,0.7,0.6,0.9,0.8,0.7,0.6,0.9,0.8,0.7,0.6,0.9,0.8,0.7,0.6] plt.plot(test_mean_1000S_n, 'royalblue', label='without threshold') plt.plot(test_mean_1000S, 'darkorange', label='with threshold') #画图,并指定颜色plt.xticks(fontproperties = 'Times New Roman', fontsize=8) plt.yticks(np.arange(0, 1.1, 0.2), fontproperties = 'Times New Roman', fontsize=8) #指定横纵坐标的字体以及字体大小,记住是fontsize不是size。yticks上我还用numpy指定了坐标轴的变化范围。plt.legend(loc='lower right', prop={'family':'Times New Roman', 'size':8}) #图上的legend,记住字体是要用prop以字典形式设置的,而且字的大小是size不是fontsize,这个容易和xticks的命令弄混plt.title('1000 samples', fontdict={'family' : 'Times New Roman', 'size':8}) #指定图上标题的字体及大小plt.xlabel('iterations', fontdict={'family' : 'Times New Roman', 'size':8}) plt.ylabel('accuracy', fontdict={'family' : 'Times New Roman', 'size':8}) #指定横纵坐标描述的字体及大小plt.savefig('./where-you-want-to-save.png', dpi=300, bbox_inches="tight") #保存文件,dpi指定保存文件的分辨率 #bbox_inches="tight" 可以保存图上所有的信息,不会出现横纵坐标轴的描述存掉了的情况plt.show() #记住,如果你要show()的话,一定要先savefig,再show。如果你先show了,存出来的就是一张白纸。
效果图:
最后在放点Matplotlib相关设置供大家参考:
附颜色表
Marker常见参数
任何程序错误,以及技术疑问或需要解答的,请添加