[R] read txt file - date - no space
PIKAL Petr
petr@p|k@| @end|ng |rom prechez@@cz
Thu Aug 2 09:56:54 CEST 2018
Well,
you followed my advice only partly. Did you get rid of your silly -999 values before averaging? Probably not.
Did you tried aggregating by slightly longer construction
aggregate(test[,-1], list(format(test$date, "%Y-%m-%d")), mean)
which keeps difference in month and year? Probably not.
We do not have your data, we do not know what exactly you want to do so it is really difficult to give you a help.
If I calculate correctly there are 24 hour in one day and you have data for 18 years which gives me approximately 158000 distinct values.
I can get either 18 values (averaging years) or aproximately 6600 values (averaging days).
So my advice is:
Read your data to R
Change date column to POSIX but store it in different column
Change NA values from -999 to real NA values
Check dimension of your data ?dim
Check structure of your data ?str
Check if all dates are changed to POSIX correctly, are some of them NA?
Aggregate your values (not by lubridate function day) and store them in another object
Cheers
Petr
From: Diego Avesani <diego.avesani using gmail.com>
Sent: Thursday, August 2, 2018 9:31 AM
To: jim holtman <jholtman using gmail.com>; PIKAL Petr <petr.pikal using precheza.cz>
Cc: R mailing list <r-help using r-project.org>
Subject: Re: [R] read txt file - date - no space
Dear all,
I have found and error in the date conversion. Now it looks like:
MyData <- read.csv(file="obs_prec.csv",header=TRUE, sep=",")
# change date to real
MyData$date<-as.POSIXct(MyData$date, format="%m/%d/%Y %H:%M")
After that I apply the PIKAL's suggestions:
aggregate(MyData[,-1], list(day(MyData$date)), mean)
And this is the final results:
1 -82.43636 -46.12437 -319.2710
2 2 -82.06105 -45.74184 -319.2696
3 3 -82.05527 -45.52650 -319.2416
4 4 -82.03535 -47.59191 -319.2275
5 5 -77.44928 -50.05953 -320.5798
...
31 -86.10234 -47.06247 -340.0968
However, it is not correct.
This because I have not made myself clear about my purpose. As I told you some days ago, I have a *.csv file with hourly data from 10/21/1998 to 12/31/2016. I would like to compute the daily means. Basically, I would like to have the mean of the hourly date for each day from 10/21/1998 to 12/31/2016 and not 31 values.
Really really thanks again,
Diego
Diego
On 2 August 2018 at 08:55, Diego Avesani <diego.avesani using gmail.com<mailto:diego.avesani using gmail.com>> wrote:
Dear
I have check the one of the line that gives me problem. I mean, which give NA after R processing. I think that is similar to the others:
10/12/1998 10:00,0,0,0
10/12/1998 11:00,0,0,0
10/12/1998 12:00,0,0,0
10/12/1998 13:00,0,0,0
10/12/1998 14:00,0,0,0
10/12/1998 15:00,0,0,0
10/12/1998 16:00,0,0,0
10/12/1998 17:00,0,0,0
@jim: It seems that you suggestion is focus on reading data from the terminal. It is possible to apply it to a *.csv file?
@Pikal: Could it be that there are some date conversion error?
Thanks again,
Diego
Diego
On 1 August 2018 at 17:01, jim holtman <jholtman using gmail.com<mailto:jholtman using gmail.com>> wrote:
Try this:
> library(lubridate)
> library(tidyverse)
> input <- read.csv(text = "date,str1,str2,str3
+ 10/1/1998 0:00,0.6,0,0
+ 10/1/1998 1:00,0.2,0.2,0.2
+ 10/1/1998 2:00,0.6,0.2,0.4
+ 10/1/1998 3:00,0,0,0.6
+ 10/1/1998 4:00,0,0,0
+ 10/1/1998 5:00,0,0,0
+ 10/1/1998 6:00,0,0,0
+ 10/1/1998 7:00,0.2,0,0", as.is<http://as.is> = TRUE)
> # convert the date and add the "day" so summarize
> input <- input %>%
+ mutate(date = mdy_hm(date),
+ day = floor_date(date, unit = 'day')
+ )
>
> by_day <- input %>%
+ group_by(day) %>%
+ summarise(m_s1 = mean(str1),
+ m_s2 = mean(str2),
+ m_s3 = mean(str3)
+ )
>
> by_day
# A tibble: 1 x 4
day m_s1 m_s2 m_s3
<dttm> <dbl> <dbl> <dbl>
1 1998-10-01 00:00:00 0.200 0.0500 0.150
Jim Holtman
Data Munger Guru
What is the problem that you are trying to solve?
Tell me what you want to do, not how you want to do it.
On Tue, Jul 31, 2018 at 11:54 PM Diego Avesani <diego.avesani using gmail.com<mailto:diego.avesani using gmail.com>> wrote:
Dear all,
I am sorry, I did a lot of confusion. I am sorry, I have to relax and stat
all again in order to understand.
If I could I would like to start again, without mixing strategy and waiting
for your advice.
I am really appreciate you help, really really.
Here my new file, a *.csv file (buy the way, it is possible to attach it in
the mailing list?)
date,str1,str2,str3
10/1/1998 0:00,0.6,0,0
10/1/1998 1:00,0.2,0.2,0.2
10/1/1998 2:00,0.6,0.2,0.4
10/1/1998 3:00,0,0,0.6
10/1/1998 4:00,0,0,0
10/1/1998 5:00,0,0,0
10/1/1998 6:00,0,0,0
10/1/1998 7:00,0.2,0,0
I read it as:
MyData <- read.csv(file="obs_prec.csv",header=TRUE, sep=",")
at this point I would like to have the daily mean.
What would you suggest?
Really Really thanks,
You are my lifesaver
Thanks
Diego
On 1 August 2018 at 01:01, Jeff Newmiller <jdnewmil using dcn.davis.ca.us<mailto:jdnewmil using dcn.davis.ca.us>> wrote:
> ... and the most common source of NA values in time data is wrong
> timezones. You really need to make sure the timezone that is assumed when
> the character data are converted to POSIXt agrees with the data. In most
> cases the easiest way to insure this is to use
>
> Sys.setenv(TZ="US/Pacific")
>
> or whatever timezone from
>
> OlsonNames()
>
> corresponds with your data. Execute this setenv function before the
> strptime or as.POSIXct() function call.
>
> You can use
>
> MyData[ is.na<http://is.na>(MyData$datetime), ]
>
> to see which records are failing to convert time.
>
> [1] https://github.com/jdnewmil/eci298sp2016/blob/master/QuickHowto1
>
> On July 31, 2018 3:04:05 PM PDT, Jim Lemon <drjimlemon using gmail.com<mailto:drjimlemon using gmail.com>> wrote:
> >Hi Diego,
> >I think the error is due to NA values in your data file. If I extend
> >your example and run it, I get no errors:
> >
> >MyData<-read.table(text="103001930 103001580 103001530
> >1998-10-01 00:00:00 0.6 0 0
> >1998-10-01 01:00:00 0.2 0.2 0.2
> >1998-10-01 02:00:00 0.6 0.2 0.4
> >1998-10-01 03:00:00 0 0 0.6
> >1998-10-01 04:00:00 0 0 0
> >1998-10-01 05:00:00 0 0 0
> >1998-10-01 06:00:00 0 0 0
> >1998-10-01 07:00:00 0.2 0 0
> >1998-10-01 08:00:00 0.6 0 0
> >1998-10-01 09:00:00 0.2 0.2 0.2
> >1998-10-01 10:00:00 0.6 0.2 0.4
> >1998-10-01 11:00:00 0 0 0.6
> >1998-10-01 12:00:00 0 0 0
> >1998-10-01 13:00:00 0 0 0
> >1998-10-01 14:00:00 0 0 0
> >1998-10-01 15:00:00 0.2 0 0
> >1998-10-01 16:00:00 0.6 0 0
> >1998-10-01 17:00:00 0.2 0.2 0.2
> >1998-10-01 18:00:00 0.6 0.2 0.4
> >1998-10-01 19:00:00 0 0 0.6
> >1998-10-01 20:00:00 0 0 0
> >1998-10-01 21:00:00 0 0 0
> >1998-10-01 22:00:00 0 0 0
> >1998-10-01 23:00:00 0.2 0 0
> >1998-10-02 00:00:00 0.6 0 0
> >1998-10-02 01:00:00 0.2 0.2 0.2
> >1998-10-02 02:00:00 0.6 0.2 0.4
> >1998-10-02 03:00:00 0 0 0.6
> >1998-10-02 04:00:00 0 0 0
> >1998-10-02 05:00:00 0 0 0
> >1998-10-02 06:00:00 0 0 0
> >1998-10-02 07:00:00 0.2 0 0
> >1998-10-02 08:00:00 0.6 0 0
> >1998-10-02 09:00:00 0.2 0.2 0.2
> >1998-10-02 10:00:00 0.6 0.2 0.4
> >1998-10-02 11:00:00 0 0 0.6
> >1998-10-02 12:00:00 0 0 0
> >1998-10-02 13:00:00 0 0 0
> >1998-10-02 14:00:00 0 0 0
> >1998-10-02 15:00:00 0.2 0 0
> >1998-10-02 16:00:00 0.6 0 0
> >1998-10-02 17:00:00 0.2 0.2 0.2
> >1998-10-02 18:00:00 0.6 0.2 0.4
> >1998-10-02 19:00:00 0 0 0.6
> >1998-10-02 20:00:00 0 0 0
> >1998-10-02 21:00:00 0 0 0
> >1998-10-02 22:00:00 0 0 0
> >1998-10-02 23:00:00 0.2 0 0",
> >skip=1,stringsAsFactors=FALSE)
> >names(MyData)<-c("date","time","st1","st2","st3")
> >MyData$datetime<-strptime(paste(MyData$date,MyData$time),
> > format="%Y-%m-%d %H:%M:%S")
> >MyData$datetime
> >st1_daily<-by(MyData$st1,MyData$date,mean)
> >st2_daily<-by(MyData$st2,MyData$date,mean)
> >st3_daily<-by(MyData$st3,MyData$date,mean)
> >st1_daily
> >st2_daily
> >st3_daily
> >
> >Try adding na.rm=TRUE to the "by" calls:
> >
> >st1_daily<-by(MyData$st1,MyData$date,mean,na.rm=TRUE)
> >st2_daily<-by(MyData$st2,MyData$date,mean,na.rm=TRUE)
> >st3_daily<-by(MyData$st3,MyData$date,mean,na.rm=TRUE)
> >
> >Jim
> >
> >On Tue, Jul 31, 2018 at 11:11 PM, Diego Avesani
> ><diego.avesani using gmail.com<mailto:diego.avesani using gmail.com>> wrote:
> >> Dear all,
> >>
> >> I have still problem with date.
> >> Could you please tel me how to use POSIXct.
> >> Indeed I have found this command:
> >> timeAverage, but I am not able to convert MyDate to properly date.
> >>
> >> Thank a lot
> >> I hope to no bother you, at least too much
> >>
> >>
> >> Diego
> >>
> >>
> >> On 31 July 2018 at 11:12, Diego Avesani <diego.avesani using gmail.com<mailto:diego.avesani using gmail.com>>
> >wrote:
> >>>
> >>> Dear Jim, Dear all,
> >>>
> >>> thanks a lot.
> >>>
> >>> Unfortunately, I get the following error:
> >>>
> >>>
> >>> st1_daily<-by(MyData$st1,MyData$date,mean)
> >>> Error in tapply(seq_len(0L), list(`MyData$date` = c(913L, 914L,
> >925L, :
> >>> arguments must have same length
> >>>
> >>>
> >>> This is particularly strange. indeed, if I apply
> >>>
> >>>
> >>> mean(MyData$str1,na.rm=TRUE)
> >>>
> >>>
> >>> it works
> >>>
> >>>
> >>> Sorry, I have to learn a lot.
> >>> You are really boosting me
> >>>
> >>> Diego
> >>>
> >>>
> >>> On 31 July 2018 at 11:02, Jim Lemon <drjimlemon using gmail.com<mailto:drjimlemon using gmail.com>> wrote:
> >>>>
> >>>> Hi Diego,
> >>>> One way you can get daily means is:
> >>>>
> >>>> st1_daily<-by(MyData$st1,MyData$date,mean)
> >>>> st2_daily<-by(MyData$st2,MyData$date,mean)
> >>>> st3_daily<-by(MyData$st3,MyData$date,mean)
> >>>>
> >>>> Jim
> >>>>
> >>>> On Tue, Jul 31, 2018 at 6:51 PM, Diego Avesani
> ><diego.avesani using gmail.com<mailto:diego.avesani using gmail.com>>
> >>>> wrote:
> >>>> > Dear all,
> >>>> > I have found the error, my fault. Sorry.
> >>>> > There was an extra come in the headers line.
> >>>> > Thanks again.
> >>>> >
> >>>> > If I can I would like to ask you another questions about the
> >imported
> >>>> > data.
> >>>> > I would like to compute the daily average of the different date.
> >>>> > Basically I
> >>>> > have hourly data, I would like to ave the daily mean of them.
> >>>> >
> >>>> > Is there some special commands?
> >>>> >
> >>>> > Thanks a lot.
> >>>> >
> >>>> >
> >>>> > Diego
> >>>> >
> >>>> >
> >>>> > On 31 July 2018 at 10:40, Diego Avesani <diego.avesani using gmail.com<mailto:diego.avesani using gmail.com>>
> >>>> > wrote:
> >>>> >>
> >>>> >> Dear all,
> >>>> >> I move to csv file because originally the date where in csv
> >file.
> >>>> >> In addition, due to the fact that, as you told me, read.csv is a
> >>>> >> special
> >>>> >> case of read.table, I prefer start to learn from the simplest
> >one.
> >>>> >> After that, I will try also the *.txt format.
> >>>> >>
> >>>> >> with read.csv, something strange happened:
> >>>> >>
> >>>> >> This us now the file:
> >>>> >>
> >>>> >> date,st1,st2,st3,
> >>>> >> 10/1/1998 0:00,0.6,0,0
> >>>> >> 10/1/1998 1:00,0.2,0.2,0.2
> >>>> >> 10/1/1998 2:00,0.6,0.2,0.4
> >>>> >> 10/1/1998 3:00,0,0,0.6
> >>>> >> 10/1/1998 4:00,0,0,0
> >>>> >> 10/1/1998 5:00,0,0,0
> >>>> >> 10/1/1998 6:00,0,0,0
> >>>> >> 10/1/1998 7:00,0.2,0,0
> >>>> >> 10/1/1998 8:00,0.6,0.2,0
> >>>> >> 10/1/1998 9:00,0.2,0.4,0.4
> >>>> >> 10/1/1998 10:00,0,0.4,0.2
> >>>> >>
> >>>> >> When I apply:
> >>>> >> MyData <- read.csv(file="obs_prec.csv",header=TRUE, sep=",")
> >>>> >>
> >>>> >> this is the results:
> >>>> >>
> >>>> >> 10/1/1998 0:00 0.6 0.00 0.0 NA
> >>>> >> 2 10/1/1998 1:00 0.2 0.20 0.2 NA
> >>>> >> 3 10/1/1998 2:00 0.6 0.20 0.4 NA
> >>>> >> 4 10/1/1998 3:00 0.0 0.00 0.6 NA
> >>>> >> 5 10/1/1998 4:00 0.0 0.00 0.0 NA
> >>>> >> 6 10/1/1998 5:00 0.0 0.00 0.0 NA
> >>>> >> 7 10/1/1998 6:00 0.0 0.00 0.0 NA
> >>>> >> 8 10/1/1998 7:00 0.2 0.00 0.0 NA
> >>>> >>
> >>>> >> I do not understand why.
> >>>> >> Something wrong with date?
> >>>> >>
> >>>> >> really really thanks,
> >>>> >> I appreciate a lot all your helps.
> >>>> >>
> >>>> >> Diedro
> >>>> >>
> >>>> >>
> >>>> >> Diego
> >>>> >>
> >>>> >>
> >>>> >> On 31 July 2018 at 01:25, MacQueen, Don <macqueen1 using llnl.gov<mailto:macqueen1 using llnl.gov>>
> >wrote:
> >>>> >>>
> >>>> >>> Or, without removing the first line
> >>>> >>> dadf <- read.table("xxx.txt", stringsAsFactors=FALSE, skip=1)
> >>>> >>>
> >>>> >>> Another alternative,
> >>>> >>> dadf$datetime <- as.POSIXct(paste(dadf$V1,dadf$V2))
> >>>> >>> since the dates appear to be in the default format.
> >>>> >>> (I generally prefer to work with datetimes in POSIXct class
> >rather
> >>>> >>> than
> >>>> >>> POSIXlt class)
> >>>> >>>
> >>>> >>> -Don
> >>>> >>>
> >>>> >>> --
> >>>> >>> Don MacQueen
> >>>> >>> Lawrence Livermore National Laboratory
> >>>> >>> 7000 East Ave., L-627
> >>>> >>> Livermore, CA 94550
> >>>> >>> 925-423-1062
> >>>> >>> Lab cell 925-724-7509
> >>>> >>>
> >>>> >>>
> >>>> >>>
> >>>> >>> On 7/30/18, 4:03 PM, "R-help on behalf of Jim Lemon"
> >>>> >>> <r-help-bounces using r-project.org<mailto:r-help-bounces using r-project.org> on behalf of
> >drjimlemon using gmail.com<mailto:drjimlemon using gmail.com>>
> >>>> >>> wrote:
> >>>> >>>
> >>>> >>> Hi Diego,
> >>>> >>> You may have to do some conversion as you have three fields
> >in
> >>>> >>> the
> >>>> >>> first line using the default space separator and five
> >fields in
> >>>> >>> subsequent lines. If the first line doesn't contain any
> >important
> >>>> >>> data
> >>>> >>> you can just delete it or replace it with a meaningful
> >header
> >>>> >>> line
> >>>> >>> with five fields and save the file under another name.
> >>>> >>>
> >>>> >>> It looks as thought you have date-time as two fields. If
> >so, you
> >>>> >>> can
> >>>> >>> just read the first field if you only want the date:
> >>>> >>>
> >>>> >>> # assume you have removed the first line
> >>>> >>> dadf<-read.table("xxx.txt",stringsAsFactors=FALSE
> >>>> >>> dadf$date<-as.Date(dadf$V1,format="%Y-%m-%d")
> >>>> >>>
> >>>> >>> If you want the date/time:
> >>>> >>>
> >>>> >>>
> >dadf$datetime<-strptime(paste(dadf$V1,dadf$V2),format="%Y-%m-%d
> >>>> >>> %H:%M:%S")
> >>>> >>>
> >>>> >>> Jim
> >>>> >>>
> >>>> >>> On Tue, Jul 31, 2018 at 12:29 AM, Diego Avesani
> >>>> >>> <diego.avesani using gmail.com<mailto:diego.avesani using gmail.com>> wrote:
> >>>> >>> > Dear all,
> >>>> >>> >
> >>>> >>> > I am dealing with the reading of a *.txt file.
> >>>> >>> > The txt file the following shape:
> >>>> >>> >
> >>>> >>> > 103001930 103001580 103001530
> >>>> >>> > 1998-10-01 00:00:00 0.6 0 0
> >>>> >>> > 1998-10-01 01:00:00 0.2 0.2 0.2
> >>>> >>> > 1998-10-01 02:00:00 0.6 0.2 0.4
> >>>> >>> > 1998-10-01 03:00:00 0 0 0.6
> >>>> >>> > 1998-10-01 04:00:00 0 0 0
> >>>> >>> > 1998-10-01 05:00:00 0 0 0
> >>>> >>> > 1998-10-01 06:00:00 0 0 0
> >>>> >>> > 1998-10-01 07:00:00 0.2 0 0
> >>>> >>> >
> >>>> >>> > If it is possible I have a coupe of questions, which will
> >sound
> >>>> >>> stupid but
> >>>> >>> > they are important to me in order to understand ho R deal
> >with
> >>>> >>> file
> >>>> >>> or date.
> >>>> >>> >
> >>>> >>> > 1) Do I have to convert it to a *csv file?
> >>>> >>> > 2) Can a deal with space and not ","
> >>>> >>> > 3) How can I read date?
> >>>> >>> >
> >>>> >>> > thanks a lot to all of you,
> >>>> >>> > Thanks
> >>>> >>> >
> >>>> >>> >
> >>>> >>> > Diego
> >>>> >>> >
> >>>> >>> > [[alternative HTML version deleted]]
> >>>> >>> >
> >>>> >>> > ______________________________________________
> >>>> >>> > R-help using r-project.org<mailto:R-help using r-project.org> mailing list -- To UNSUBSCRIBE and
> >more,
> >>>> >>> see
> >>>> >>> > https://stat.ethz.ch/mailman/listinfo/r-help
> >>>> >>> > PLEASE do read the posting guide
> >>>> >>> http://www.R-project.org/posting-guide.html
> >>>> >>> > and provide commented, minimal, self-contained,
> >reproducible
> >>>> >>> code.
> >>>> >>>
> >>>> >>> ______________________________________________
> >>>> >>> R-help using r-project.org<mailto:R-help using r-project.org> mailing list -- To UNSUBSCRIBE and
> >more, see
> >>>> >>> https://stat.ethz.ch/mailman/listinfo/r-help
> >>>> >>> PLEASE do read the posting guide
> >>>> >>> http://www.R-project.org/posting-guide.html
> >>>> >>> and provide commented, minimal, self-contained,
> >reproducible
> >>>> >>> code.
> >>>> >>>
> >>>> >>>
> >>>> >>
> >>>> >
> >>>
> >>>
> >>
> >
> >______________________________________________
> >R-help using r-project.org<mailto:R-help using r-project.org> mailing list -- To UNSUBSCRIBE and more, see
> >https://stat.ethz.ch/mailman/listinfo/r-help
> >PLEASE do read the posting guide
> >http://www.R-project.org/posting-guide.html
> >and provide commented, minimal, self-contained, reproducible code.
>
> --
> Sent from my phone. Please excuse my brevity.
>
[[alternative HTML version deleted]]
______________________________________________
R-help using r-project.org<mailto:R-help using r-project.org> mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Osobní údaje: Informace o zpracování a ochraně osobních údajů obchodních partnerů PRECHEZA a.s. jsou zveřejněny na: https://www.precheza.cz/zasady-ochrany-osobnich-udaju/ | Information about processing and protection of business partner’s personal data are available on website: https://www.precheza.cz/en/personal-data-protection-principles/
Důvěrnost: Tento e-mail a jakékoliv k němu připojené dokumenty jsou důvěrné a podléhají tomuto právně závaznému prohláąení o vyloučení odpovědnosti: https://www.precheza.cz/01-dovetek/ | This email and any documents attached to it may be confidential and are subject to the legally binding disclaimer: https://www.precheza.cz/en/01-disclaimer/
[[alternative HTML version deleted]]
More information about the R-help
mailing list