1 #!usr/bin/env python 2 # -*- coding: utf-8 -*- 3 import os 4 import numpy as np 5 import pandas as pd 6 import matplotlib.pyplot as plt 7 8 base_dir = 'd:/Pattern/' 9 input_base_dir = base_dir + 'data_graph_save/' 10 rmse_list = [] 11 xlabel_list = [str(x) for x in range(6, 241, 6)] 12 xlabel = tuple(xlabel_list) 13 print(xlabel) 14 for i in range(6, 241, 6): 15 fileName = input_base_dir + 'output_' + str(i) + '.csv' 16 print(fileName) 17 data = pd.read_csv(fileName) 18 n_data = data.shape[0] 19 data_true = data.iloc[:, 6] 20 data_predict = data.iloc[:, 7] 21 22 rmse_predict = np.sqrt(np.array(np.square(data_predict - data_true)).sum() / n_data) 23 rmse_list.append(rmse_predict) 24 25 rmse_array = np.array(rmse_list).round(1) 26 print(rmse_array) 27 28 plt.figure(figsize=(15, 8), dpi=80) 29 ax = plt.subplot(1, 1, 1) 30 31 # plt.hist(pd.Series(rmse_array), bins=rmse_array.shape) 32 width = 0.5 33 x_bar = np.arange(len(xlabel_list)) 34 rect1 = ax.bar(left=x_bar, height=rmse_array, width=width, color="blue") 35 36 for rec in rect1: 37 x = rec.get_x() 38 height = rec.get_height() 39 ax.text(x-0.2, 1.02*height, str(height)) 40 41 ax.set_xticks(x_bar) 42 ax.set_xticklabels(xlabel) 43 44 plt.title('Rmse Constraction') 45 plt.xlabel('Forecast time') 46 plt.ylabel('Rmse') 47 plt.savefig('Rmse_tmp.png') 48 plt.show()
结果
