Loading ...
Sorry, an error occurred while loading the content.

plotting temps over 1 week

Expand Messages
  • dylan
    hi- ploticus is a great piece of software! however, after spending some time with the prefabs and some custom scripts, i am not able to graph 1 week s worth of
    Message 1 of 7 , Dec 8, 2002
    • 0 Attachment
      hi-

      ploticus is a great piece of software! however, after spending some time
      with the prefabs and some custom scripts, i am not able to graph 1 week's
      worth of temp data. daily graphs are no problem, however i need to display
      the hourly temperature over 1 one week time period..

      any ideas?

      thanks in advance,

      dylan
    • Stephen C. Grubb
      Dylan, the following works for me: sample data: datetime temp 12/08/2002.00:00 22 12/08/2002.06:00 24 12/08/2002.12:00 31 12/08/2002.18:00 30 12/09/2002.00:00
      Message 2 of 7 , Dec 9, 2002
      • 0 Attachment
        Dylan,

        the following works for me:

        sample data:

        datetime temp
        12/08/2002.00:00 22
        12/08/2002.06:00 24
        12/08/2002.12:00 31
        12/08/2002.18:00 30
        12/09/2002.00:00 25
        12/09/2002.06:00 25
        12/09/2002.12:00 26
        12/09/2002.18:00 27
        12/10/2002.00:00 30
        12/10/2002.06:00 38
        12/10/2002.12:00 39
        12/10/2002.18:00 35
        12/11/2002.00:00 32
        12/11/2002.06:00 35
        12/11/2002.12:00 40
        12/11/2002.18:00 42
        12/12/2002.00:00 40
        12/12/2002.06:00 39
        12/12/2002.12:00 30
        12/12/2002.18:00 27
        12/13/2002.00:00 19
        12/13/2002.06:00 18
        12/13/2002.12:00 20
        12/13/2002.18:00 19
        12/14/2002.00:00 14
        12/14/2002.06:00 12
        12/14/2002.12:00 18
        12/14/2002.18:00 23

        command:
        pl -prefab chron data=weektemp.dat header=yes x=1 y=2 datefmt=mm/dd/yyyy
        xinc="1 day" mode=line unittype=datetime


        Note that the date & time come from one field.

        When looking into this I did notice a problem.. chron has a parameter
        called 'timefld' and the docs claim that this can be used to accomodate
        separate date and time fields. However this only works when "tab" is
        being done. Perhaps this threw you off.

        -Steve


        On Sun, 8 Dec 2002, dylan wrote:

        > hi-
        >
        > ploticus is a great piece of software! however, after spending some time
        > with the prefabs and some custom scripts, i am not able to graph 1 week's
        > worth of temp data. daily graphs are no problem, however i need to display
        > the hourly temperature over 1 one week time period..
        >
        > any ideas?
        >
        > thanks in advance,
        >
        > dylan
        >
        >
        >
        > To unsubscribe from this group, send an email to:
        > ploticus-unsubscribe@yahoogroups.com
        >
        >
        >
        > Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
        >
        >
        >


        Stephen C. Grubb scg@...
        Scientific Software Engineer, The Jackson Laboratory
        600 Main Street Bar Harbor, Maine 04609 USA
      • dylan
        Thanks! that did the trick for one data set.... now i just need this to work for 2 data sets: air temp and soil temp. i tried specifying a y2 value, but i
        Message 3 of 7 , Dec 9, 2002
        • 0 Attachment
          Thanks!
          that did the trick for one data set.... now i just need this to work for 2
          data sets: air temp and soil temp.

          i tried specifying a y2 value, but i think that the prefab is ignoring it?
          would this require one to make a script file?

          thanks

          dylan

          on 02.12.9 4:31 AM, Stephen C. Grubb at scg@... wrote:

          >
          > Dylan,
          >
          > the following works for me:
          >
          > sample data:
          >
          > datetime temp
          > 12/08/2002.00:00 22
          > 12/08/2002.06:00 24
          > 12/08/2002.12:00 31
          > 12/08/2002.18:00 30
          > 12/09/2002.00:00 25
          > 12/09/2002.06:00 25
          > 12/09/2002.12:00 26
          > 12/09/2002.18:00 27
          > 12/10/2002.00:00 30
          > 12/10/2002.06:00 38
          > 12/10/2002.12:00 39
          > 12/10/2002.18:00 35
          > 12/11/2002.00:00 32
          > 12/11/2002.06:00 35
          > 12/11/2002.12:00 40
          > 12/11/2002.18:00 42
          > 12/12/2002.00:00 40
          > 12/12/2002.06:00 39
          > 12/12/2002.12:00 30
          > 12/12/2002.18:00 27
          > 12/13/2002.00:00 19
          > 12/13/2002.06:00 18
          > 12/13/2002.12:00 20
          > 12/13/2002.18:00 19
          > 12/14/2002.00:00 14
          > 12/14/2002.06:00 12
          > 12/14/2002.12:00 18
          > 12/14/2002.18:00 23
          >
          > command:
          > pl -prefab chron data=weektemp.dat header=yes x=1 y=2 datefmt=mm/dd/yyyy
          > xinc="1 day" mode=line unittype=datetime
          >
          >
          > Note that the date & time come from one field.
          >
          > When looking into this I did notice a problem.. chron has a parameter
          > called 'timefld' and the docs claim that this can be used to accomodate
          > separate date and time fields. However this only works when "tab" is
          > being done. Perhaps this threw you off.
          >
          > -Steve
          >
          >
          > On Sun, 8 Dec 2002, dylan wrote:
          >
          >> hi-
          >>
          >> ploticus is a great piece of software! however, after spending some time
          >> with the prefabs and some custom scripts, i am not able to graph 1 week's
          >> worth of temp data. daily graphs are no problem, however i need to display
          >> the hourly temperature over 1 one week time period..
          >>
          >> any ideas?
          >>
          >> thanks in advance,
          >>
          >> dylan
          >>
          >>
          >>
          >> To unsubscribe from this group, send an email to:
          >> ploticus-unsubscribe@yahoogroups.com
          >>
          >>
          >>
          >> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
          >>
          >>
          >>
          >
          >
          > Stephen C. Grubb scg@...
          > Scientific Software Engineer, The Jackson Laboratory
          > 600 Main Street Bar Harbor, Maine 04609 USA
          >
          >
          >
          > To unsubscribe from this group, send an email to:
          > ploticus-unsubscribe@yahoogroups.com
          >
          >
          >
          > Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
          >
          >
        • Stephen C. Grubb
          Dylan, You re correct.. the chron prefab only handles one curve or set of bars, so a script would be the way to go.. here s one that should be close to what
          Message 4 of 7 , Dec 10, 2002
          • 0 Attachment
            Dylan,

            You're correct.. the chron prefab only handles one curve or set of bars,
            so a script would be the way to go.. here's one that should be close to
            what you want....

            #proc getdata
            file: twotemps.dat
            fieldnameheader: yes

            #proc areadef
            rectangle: 1 1 6 3
            xscaletype: datetime mm/dd/yyyy
            xautorange: datafield=datetime
            yrange: 0 40
            // yautorange: datafields=airtemp,soiltemp
            xaxis.stubs: inc 1 day
            xaxis.stubformat: MMMdd
            yaxis.stubs: inc

            #proc lineplot
            linedetails: color=blue
            xfield: datetime
            yfield: airtemp
            legendlabel: air temp

            #proc lineplot
            linedetails: color=gray(0.3)
            xfield: datetime
            yfield: soiltemp
            legendlabel: soil temp

            #proc legend
            location: min+1 min+0.5


            ......and here's a sample data file.....
            datetime airtemp soiltemp
            12/08/2002.00:00 22 26
            12/08/2002.06:00 24 27
            12/08/2002.12:00 31 28
            12/08/2002.18:00 30 28
            12/09/2002.00:00 25 27
            12/09/2002.06:00 25 26
            12/09/2002.12:00 26 26
            12/09/2002.18:00 27 26
            12/10/2002.00:00 30 27
            12/10/2002.06:00 38 29
            12/10/2002.12:00 39 31
            12/10/2002.18:00 35 33
            12/11/2002.00:00 32 32
            12/11/2002.06:00 35 33
            12/11/2002.12:00 40 33
            12/11/2002.18:00 42 35
            12/12/2002.00:00 40 37
            12/12/2002.06:00 39 38
            12/12/2002.12:00 30 36
            12/12/2002.18:00 27 31
            12/13/2002.00:00 19 27
            12/13/2002.06:00 18 25
            12/13/2002.12:00 20 23
            12/13/2002.18:00 19 21
            12/14/2002.00:00 14 18
            12/14/2002.06:00 12 16
            12/14/2002.12:00 18 17
            12/14/2002.18:00 23 19
          • dylan
            Steve- Thanks for the starting point... with your template, and the handbook section on the website i was able to get results that are getting closer to want i
            Message 5 of 7 , Dec 12, 2002
            • 0 Attachment
              Steve-

              Thanks for the starting point... with your template, and the handbook
              section on the website i was able to get results that are getting closer to
              want i need.
              (http://fungus.ucdavis.edu/weather/images/weekly_temps.gif)

              however, i am still having problems getting the legend to show up under the
              graph without overlapping the axis...

              also, i have come upon another problem when trying to plot precipitation
              data from a 5 year precipitation record. However, x axis stubs are only
              readable when i set the 'xaxis.stubs' to 100 years! i can't seem to figure
              out why this is happening...

              thanks in advance...

              data looks like this: (no headers: columns = 1.date 2.mm_of_rain)

              01/01/1997.00:00 88.00
              01/02/1997.00:00 32.00
              01/03/1997.00:00 0.00
              01/04/1997.00:00 0.00
              01/05/1997.00:00 0.00
              01/06/1997.00:00 0.00
              01/07/1997.00:00 0.00
              01/08/1997.00:00 0.00
              01/09/1997.00:00 0.00
              01/10/1997.00:00 0.00
              ...
              11/29/2002.00:00 0.00
              11/30/2002.00:00 0.00
              12/01/2002.00:00 0.00
              12/02/2002.00:00 0.00
              12/03/2002.00:00 0.00
              12/04/2002.00:00 0.00
              12/05/2002.00:00 0.00
              12/06/2002.00:00 0.00
              12/07/2002.00:00 0.00



              here is my script file
              --
              #proc settings
              cpulimit: 25

              #proc getdata
              file: anual_precip.data
              fieldnames: datetime precip

              #proc areadef
              rectangle: 1 1 9 4
              xscaletype: datetime mm/dd/yyyy
              xautorange: datafield=datetime
              xaxis.label: Date
              xaxis.stubs: inc 100 Year
              xaxis.stubformat: YY
              yaxis.label: Precipitation (mm)
              yaxis.stubs: inc 10
              yautorange: datafields=precip


              #proc bars
              locfield: datetime
              lenfield: precip
              thinbarline: color=blue
            • Stephen C. Grubb
              Dylan, I took a look at your http://fungus.ucdavis.edu/weather web site. It looks like you found how to adjust the legend (using the location attribute). On
              Message 6 of 7 , Feb 12, 2003
              • 0 Attachment
                Dylan,

                I took a look at your http://fungus.ucdavis.edu/weather web site.

                It looks like you found how to adjust the legend (using the location
                attribute).

                On the 5 year precip graph, one issue is that the graph probably exceeds
                8" wide. As pointed out on the troubleshooting page (etc.) ploticus
                allocates a default image workspace of 8" x 8".. for larger plots page
                a size must be specified in advance, either using -pagesize w,h on the
                command line, or proc page pagesize. This probably explains why the
                5-year graph is getting cut off on the right side.

                As for the stubs on the 5-year graph, you might try:
                stubs: inc 1 year
                stubformat: yy

                to cause a stub to be placed every year. You can use
                ticincrement: 3 months

                to additionally get a tic for every quarter.

                If you're still getting wierd behavior please send me the complete 5-year
                data and I'll play with it here.


                -Steve


                On Thu, 12 Dec 2002, dylan wrote:

                > Steve-
                >
                > Thanks for the starting point... with your template, and the handbook
                > section on the website i was able to get results that are getting closer to
                > want i need.
                > (http://fungus.ucdavis.edu/weather/images/weekly_temps.gif)
                >
                > however, i am still having problems getting the legend to show up under the
                > graph without overlapping the axis...
                >
                > also, i have come upon another problem when trying to plot precipitation
                > data from a 5 year precipitation record. However, x axis stubs are only
                > readable when i set the 'xaxis.stubs' to 100 years! i can't seem to figure
                > out why this is happening...
                >


                Stephen C. Grubb scg@...
                Scientific Software Engineer, The Jackson Laboratory
                600 Main Street Bar Harbor, Maine 04609 USA
              • kevin mulholland
                The other problem that I noticed is that 99 only has 3 quarters, whereas all the other years have 4, is this some sort of new calendar adjustment? ... -- Kevin
                Message 7 of 7 , Feb 12, 2003
                • 0 Attachment
                  The other problem that I noticed is that 99 only has 3 quarters, whereas
                  all the other years have 4, is this some sort of new calendar
                  adjustment?

                  On Wed, 2003-02-12 at 14:57, Stephen C. Grubb wrote:
                  > Dylan,
                  >
                  > I took a look at your http://fungus.ucdavis.edu/weather web site.
                  >
                  > It looks like you found how to adjust the legend (using the location
                  > attribute).
                  >
                  > On the 5 year precip graph, one issue is that the graph probably exceeds
                  > 8" wide. As pointed out on the troubleshooting page (etc.) ploticus
                  > allocates a default image workspace of 8" x 8".. for larger plots page
                  > a size must be specified in advance, either using -pagesize w,h on the
                  > command line, or proc page pagesize. This probably explains why the
                  > 5-year graph is getting cut off on the right side.
                  >
                  > As for the stubs on the 5-year graph, you might try:
                  > stubs: inc 1 year
                  > stubformat: yy
                  >
                  > to cause a stub to be placed every year. You can use
                  > ticincrement: 3 months
                  >
                  > to additionally get a tic for every quarter.
                  >
                  > If you're still getting wierd behavior please send me the complete 5-year
                  > data and I'll play with it here.
                  >
                  >
                  > -Steve
                  >
                  >
                  > On Thu, 12 Dec 2002, dylan wrote:
                  >
                  > > Steve-
                  > >
                  > > Thanks for the starting point... with your template, and the handbook
                  > > section on the website i was able to get results that are getting closer to
                  > > want i need.
                  > > (http://fungus.ucdavis.edu/weather/images/weekly_temps.gif)
                  > >
                  > > however, i am still having problems getting the legend to show up under the
                  > > graph without overlapping the axis...
                  > >
                  > > also, i have come upon another problem when trying to plot precipitation
                  > > data from a 5 year precipitation record. However, x axis stubs are only
                  > > readable when i set the 'xaxis.stubs' to 100 years! i can't seem to figure
                  > > out why this is happening...
                  > >
                  >
                  >
                  > Stephen C. Grubb scg@...
                  > Scientific Software Engineer, The Jackson Laboratory
                  > 600 Main Street Bar Harbor, Maine 04609 USA
                  >
                  >
                  > To unsubscribe from this group, send an email to:
                  > ploticus-unsubscribe@yahoogroups.com
                  >
                  >
                  >
                  > Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
                  --
                  Kevin Mulholland
                  Principal Systems Engineer
                  Landis+Gyr Ltd
                  desk 00 44 1952 602014
                  mobile 00 44 780882 5979
                  --
                Your message has been successfully submitted and would be delivered to recipients shortly.