趋势走向

最新推荐文章于 2025-08-03 14:18:09 发布

转载最新推荐文章于 2025-08-03 14:18:09 发布 · 204 阅读

0 ·

CC 4.0 BY-SA版权

原文链接：www.baidu.com

本文介绍了一种使用Python的matplotlib和numpy库进行性能数据可视化的方法，包括CPU使用率和内存消耗的图表绘制。通过读取特定文件中的数据，生成了反映导航和巡航模式下，有无ADAS辅助驾驶系统的CPU及内存使用情况的图表。

#!/usr/bin/env python
# -*- coding: utf-8 -*-

import matplotlib.pyplot as plt
import numpy as np
import sys
import datetime

# 修改文件路径
filePath = "/home/st/Desktop/draw-performance-pgen5/"
file_tag = "Performance"
x_time = 0
x_cpu = 1
x_mem = 2
nowTime=datetime.datetime.now().strftime('%Y-%m-%d')

filename = sys.argv[1]
# x轴显示时间标签
x_lable = []
i = 0
with open(filePath + filename, 'r') as perf_file:
line = perf_file.readline()
while line:
i += 1
x_lable.append(str(line.strip().split(',')[x_time].split(' ')[1]).strip())
line = perf_file.readline()
# cpu图
x = np.arange(i)
z = np.loadtxt(filePath + filename, delimiter=",", usecols=(x_cpu), unpack=True)
#plt.bar(x, z, facecolor='#79b3f2', edgecolor='none')
plt.plot(x, z, color='b', label='mem')
# x轴显示时间标签的个数（30个）
x_show_num = 30
N = i
if N > x_show_num:
x_freq = N / x_show_num
x_show = list(x)[::x_freq]
x_lable_show = x_lable[::x_freq]
else:
x_show = x
x_lable_show = x_lable
# x轴显示时间的倾斜角
plt.xlim(x_lable_show[0], x_lable_show[len(x_lable_show) - 1])
plt.ylim(0,400)
plt.xticks(x_show, x_lable_show, rotation=90)
plt.subplots_adjust(bottom=0.35)
plt.grid(True)
# x轴标题
plt.xlabel('Time/s')
# y轴标题
plt.ylabel('CPU/%')
# title
if filename == 'performance-guide.txt':
plt.title('%s Navigation CPU Without ADAS %s' %(nowTime, file_tag))
cpu_fig_name = '%s-guide-cpu-without-adas.png' % nowTime
elif filename == 'performance-guide-adas.txt':
plt.title('%s Navigation CPU With ADAS %s' %(nowTime, file_tag))
cpu_fig_name = '%s-guide-cpu-with-adas.png' % nowTime
elif filename == 'performance-cruise.txt':
plt.title('%s Cruise CPU Without ADAS %s' %(nowTime, file_tag))
cpu_fig_name = '%s-cruise-cpu-without-adas.png' % nowTime
else:
plt.title('%s Cruise CPU With ADAS %s' %(nowTime, file_tag))
cpu_fig_name = '%s-cruise-cpu-with-adas.png' % nowTime
plt.savefig(cpu_fig_name)
plt.close()

# mem图
y = np.loadtxt(filePath + filename, delimiter=",", usecols=(x_mem), unpack=True)
# 将获取的mem KB转成MB，保留两位小数
for item in range(0, i):
y[item] = round(y[item]/1024.0, 2)
#plt.bar(x, y, facecolor='#79b3f2', label='mem', edgecolor='none')
plt.plot(x, y, color='b', label='mem')
# x轴显示时间的倾斜角
plt.xlim(x_lable_show[0], x_lable_show[len(x_lable_show) - 1])
plt.ylim(0,2000)
plt.xticks(x_show, x_lable_show, rotation=90)
plt.subplots_adjust(bottom=0.35)
plt.grid(True)
# x轴标题
plt.xlabel('Time/s')
# y轴标题
plt.ylabel('Mem/MB')
# title
if filename == 'performance-guide.txt':
plt.title('%s Navigation Memory Without ADAS %s' %(nowTime, file_tag))
mem_fig_name = '%s-guide-mem-without-adas.png' % nowTime
elif filename == 'performance-guide-adas.txt':
plt.title('%s Navigation Memory With ADAS %s' %(nowTime, file_tag))
mem_fig_name = '%s-guide-mem-with-adas.png' % nowTime
elif filename == 'performance-cruise.txt':
plt.title('%s cruise Memory Without ADAS %s' %(nowTime, file_tag))
mem_fig_name = '%s-cruise-mem-without-adas.png' % nowTime
else:
plt.title('%s Cruise Memory With ADAS %s' %(nowTime, file_tag))
mem_fig_name = '%s-cruise-mem-with-adas.png' % nowTime
plt.savefig(mem_fig_name)
plt.close()