ML：根據不同機器學習模型輸出的預測值+且與真實值相減得到絕對誤差對比+誤差可視化

?

?

?

目錄

輸出結果

實現代碼

---

?

?

?

輸出結果

?

實現代碼

# -*- coding: utf-8 -*-#ML：根據不同機器學習模型輸出的預測值+且與真實值相減得到絕對誤差對比+誤差可視化#Model_comparison_error_Plot()函數：根據不同模型預測值輸出絕對誤差對比且可視化 def Model_comparison_error_Plot(list_str01,list_str02,list_num00, Xlabel,Ylabel,title):#數字列表相減功能：將兩個字符串列表改為數字列表并相減#(1)、依次將兩個字符串列表改為數字列表import numpy as nplist_num01 = list(map(float, list_str01))list_num02 = list(map(float, list_str02))#(2)、列表內數字相減求差#T1、利用array方法list_array01 = np.array(list_num01)list_array02 = np.array(list_num02)list_array00 = np.array(list_num00) #定義標準列表list_subtraction01 = list_array00 - list_array01list_subtraction02 = list_array00 - list_array02print(list_array00)print('list_subtraction01', list_subtraction01)print('list_subtraction02', list_subtraction02)#T2、利用列表的for循環指針對應取出法error01 = []error02 = []for i in range(len(list_num00)):error01.append(round (list_num01[i] - list_num00[i],3) )error02.append(round (list_num02[i] - list_num00[i],3) )print(error01)print(error02)#(3)、繪制error曲線import matplotlib.pyplot as pltx = range(0,len(list_subtraction01))y1 = list_subtraction01y2 = list_subtraction02y_zero = [0 for x in range(0, len(list_subtraction01))]plt.plot(x,y_zero,'r--',label='zero') plt.plot(x,y1,'g',label='STD-DTR') plt.plot(x,y2,'b',label='STD-XGBR')plt.xlabel(Xlabel)plt.ylabel(Ylabel)plt.title(title)plt.legend(loc=1) plt.show()DTR_list = ['67.330', '66.794', '65.319', '65.435', '67.903', '67.743', '63.994', '62.466', '67.581', '67.505', '64.196', '63.726', '66.749', '67.363', '65.962', '65.630', '66.602', '66.956', '63.730', '63.858', '67.370', '66.902', '63.392', '63.408', '64.428', '73.083', '72.952', '73.561', '73.148', '73.258', '72.558'] XGBR_list = ['66.398', '66.308', '66.197', '66.323', '66.388', '66.388', '65.761', '65.074', '65.516', '65.448', '65.534', '65.530', '64.163', '64.097', '62.860', '62.860', '63.771', '63.926', '62.667', '62.902', '62.981', '62.981', '62.987', '62.897', '66.465', '72.292', '71.947', '71.947', '71.947', '71.947', '71.928'] real_list = [64, 68, 64, 68, 65, 65, 63, 63, 66, 65, 65, 65, 64, 65, 61, 62, 64, 63, 66, 60, 66, 62, 64, 61, 71, 75, 73, 73, 73, 73, 73] title = 'Comparisons and Visualization of Absolute Errors in Output of Predicted Values of Different Models' Xlabel = 'working condition' Ylabel = 'Absolute Error Value' Model_comparison_error_Plot(DTR_list,XGBR_list,real_list,Xlabel,Ylabel,title)

?

?

?

?

?

總結

以上是生活随笔為你收集整理的ML：根据不同机器学习模型输出的预测值+且与真实值相减得到绝对误差对比+误差可视化的全部內容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網站內容還不錯，歡迎將生活随笔推薦給好友。