i am a beginner with pandas and matplotlib but i was able to use them previously with data structures i created in my own. (before i knew i could load to pandas from a file) i am trying to plot data from 2 sources and i am having some trouble. the first source is a log from windows PerfMon, below is a snippet of the data:
"(PDH-CSV 4.0) (Coordinated Universal Time)(0)","\\MainUser\Process(App)\% Processor Time"
"04/03/2020 16:09:58.398"," "
"04/03/2020 16:09:59.399","85.795642489249616"
"04/03/2020 16:10:00.401","99.635659557592177"
"04/03/2020 16:10:01.405","71.470475691927575"
"04/03/2020 16:10:02.394","85.172618028129037"
"04/03/2020 16:10:03.401","96.138262475057346"
"04/03/2020 16:10:04.393","88.060038497795688"
"04/03/2020 16:10:05.402","77.300336739910918"
"04/03/2020 16:10:06.397","92.497815954042423"
And the second data source is a log file from GPU-Z:
Date , Memory Used [MB] , GPU Load [%] , System Memory Used [MB] ,
2020-04-03 16:08:56 , 1409 , 14 , 6565 ,
2020-04-03 16:08:57 , 1410 , 14 , 6569 ,
2020-04-03 16:08:58 , 1410 , 14 , 6569 ,
2020-04-03 16:08:59 , 1412 , 16 , 6573 ,
2020-04-03 16:09:00 , 1412 , 15 , 6571 ,
2020-04-03 16:09:01 , 1412 , 15 , 6572 ,
2020-04-03 16:09:02 , 1410 , 14 , 6572 ,
2020-04-03 16:09:03 , 1410 , 15 , 6572 ,
i decided to tackle the CPU graph first, but after fiddling with the data, learning how to replace empty strings and converting from 'object' dtype to date and float types - i passed the data frame to .plot() and the maximum i got was a straight line with a bunch of errors..
Right now, all i got is this:
import matplotlib.pyplot as plt
import pandas as pd
from datetime import datetime
ts = pd.read_csv(r'CSV\CPU_000001.csv',na_values=' ')
ts[list(ts.columns)[0]] = ts[list(ts.columns)[0]].map(lambda x: datetime.strptime(str(x), '%d/%m/%Y %H:%M:%S.%f'))
ts[list(ts.columns)[1]] = ts[list(ts.columns)[1]].astype('float')
ts.plot()
plt.show()
i think that my problems might be related to the fact that the column headers in the CPU data file are in quotes.. but i don't know how to tackle this.
I know i am missing something but the online tutorials are not helpful, they all use randomly generated data and would really appreciate it if someone could help me with this real world problem.
