[ncl-talk] Bar Graph/Histogram plot question
Barry Lynn
barry.h.lynn at gmail.com
Wed May 3 11:57:04 MDT 2017
Hi Mary:
Thank you for all of those helpful pointers.
Barry
On Wed, May 3, 2017 at 8:46 PM, Mary Haley <haley at ucar.edu> wrote:
> Hi Barry,
>
> I think by making the image 7500x7500 this is almost making it too large,
> and hence the perimeter starts to fade.
>
> You might try a width/height of 2500 instead of 7500.
>
> To increase the thickness of any line in NCL, usually there's a
> yyXXXXThicknessF resource associated with it.
>
> In the case of the tickmark border and the tickmarks themselves, you can
> set:
>
> res at tmBorderThicknessF = 5.0
> res at tmXBMajorThicknessF = 5.0
> res at tmYLMajorThicknessF = 5.0
> res at tmXBMinorThicknessF = 5.0
> res at tmYLMinorThicknessF = 5.0
>
> The default is 1.0, so the above will make things 5x as thick. Depending
> on how big the resolution of your image is, you might need to go higher.
>
> To increase the density and/or thickness of the fill pattern lines, you
> can use "gs" resources, because these bars are filled in by using calls to
> gsn_add_polygon internally.
>
> Please see example histo_6a.ncl, which I just added. It shows how to
> change various elements of a histogram plot:
>
> http://www.ncl.ucar.edu/Applications/histo.shtml#ex6a
>
> As for the hist_text.ncl code, I couldn't reproduce your problem. I ran
> your script and it produced a plot with text.
>
> However, I did get a bunch of warnings like this:
>
> warning:Argument 0 of the current function or procedure was coerced to the
> appropriate type and thus will not change if the function or procedure
> modifies its value
>
> which is occurring because you are calling "floattointeger" on a value
> that is already an integer. If you remove the "floattointeger", then the
> warnings go away.
>
> Also, you should be able to get rid of your do loops by taking advantage
> of functions like "num", "where", and "ispan". For example, this loop:
>
>
> do i_int = 0,dims_obs_x-1
> if (obs_x(i_int).eq.0)then
> print("here in loop")
> sum_obs = sum_obs + 1
> end if
> if (mod_x(i_int).eq.0)then
> sum_mod = sum_mod + 1
> end if
> end do
>
> can be rewritten as two lines:
>
> sum_obs = sum_obs + num(obs_x.eq.0)
> sum_mod = sum_mod + num(mod_x.eq.0)
>
> --Mary
>
>
> On Wed, Apr 26, 2017 at 7:21 AM, Barry Lynn <barry.h.lynn at gmail.com>
> wrote:
>
>> Hi:
>>
>> I was unable to read the pdf file that I generated using my original code
>> and these modifications -- although I can read the png.
>>
>> This is a separate problem which can be left to later.
>>
>> To work around this, I made pngs.
>>
>> However, when making a png, the secondary bars become lighter (attached),
>> and I can't see the perimeter. When making maps, I know how to change the
>> perimeter, but not here.
>>
>> I tried to set the fill patterns, but the program didn't recognize them.
>>
>> So, I am wondering:
>>
>> a) How to darken the perimeter.
>>
>> b) How to change and darken the fill pattern (of the second overlaid
>> bars).
>>
>> c) And, I tried to add text, but the program just hangs up.
>>
>> Thank you
>>
>> P.S. I used the code on this page to add text.
>>
>> https://www.ncl.ucar.edu/Applications/Scripts/histo_16.ncl
>>
>> On Tue, Apr 25, 2017 at 9:22 PM, Barry Lynn <barry.h.lynn at gmail.com>
>> wrote:
>>
>>> That is one fancy bit of programming...
>>>
>>> Thank you -- wonderful.
>>>
>>> On Tue, Apr 25, 2017 at 9:09 PM, Dennis Shea <shea at ucar.edu> wrote:
>>>
>>>> Re: Compares two arrays. Both arrays are combined into a single array
>>>> with the first dimension equal to 2
>>>>
>>>> Let x and y be 1D but of different lengths:
>>>>
>>>> nx = dimsizes(x)
>>>> ny = dimsizes(y)
>>>> lxy = max( (/ nx,ny /) )
>>>>
>>>> xy = new( (/2,lxy/), typeof(x), getVarFillValue(x))
>>>>
>>>> xy(0,0:nx-1) = (/ x /) ; (/... /) no meta data
>>>> xy(1,0:ny-1) = (/ y /)
>>>>
>>>> HTH
>>>>
>>>>
>>>> On Tue, Apr 25, 2017 at 11:14 AM, Barry Lynn <barry.h.lynn at gmail.com>
>>>> wrote:
>>>>
>>>>> Hi:
>>>>>
>>>>> I have been using example 6.
>>>>>
>>>>> However, the page says: "Compares two arrays. Both arrays are
>>>>> combined into a single array with the first dimension equal to 2,"
>>>>>
>>>>> and implies that both arrays are of the same length.
>>>>>
>>>>> In my case, they are of different lengths.
>>>>>
>>>>> Regarding the bar graphs -- I didn't see this particular page when I
>>>>> first looked for bar graphs. However, it won't be helpful if I can't put
>>>>> one behind the other (and slightly off to the side) like you can with
>>>>> res at gsnHistogramCompare = True
>>>>>
>>>>> Barry
>>>>>
>>>>> On Tue, Apr 25, 2017 at 7:06 PM, Mary Haley <haley at ucar.edu> wrote:
>>>>>
>>>>>> Hi Barry,
>>>>>>
>>>>>> Those numbers at the top are coming from this line, which sets the
>>>>>> main title for your plot:
>>>>>>
>>>>>> res at tiMainString = str_join(""+x,",")
>>>>>>
>>>>>> Remove or comment this line.
>>>>>>
>>>>>> In what way do you want to combine the two graphs? Do you want the
>>>>>> bars side-by-side, or stacked?
>>>>>>
>>>>>> As an example, visit this page:
>>>>>>
>>>>>> http://www.ncl.ucar.edu/Applications/bar.shtml
>>>>>>
>>>>>> and look at:
>>>>>>
>>>>>> unique_5.ncl
>>>>>> bar_10.ncl
>>>>>> bar_15.ncl
>>>>>>
>>>>>> etc
>>>>>>
>>>>>> You can also look at the histogram page, see histo_6.ncl:
>>>>>>
>>>>>> http://www.ncl.ucar.edu/Applications/histo.shtml#ex6
>>>>>>
>>>>>> --Mary
>>>>>>
>>>>>>
>>>>>> On Tue, Apr 25, 2017 at 9:46 AM, Barry Lynn <barry.h.lynn at gmail.com>
>>>>>> wrote:
>>>>>>
>>>>>>> Hi:
>>>>>>>
>>>>>>> I followed Mary's comments to imply that the histogram function
>>>>>>> plots the number of elements within each interval. Hence, when my obs
>>>>>>> distribution show, for example 227, I created 227 consecutive elements with
>>>>>>> the value 0, and then proceeded through the distribution.
>>>>>>>
>>>>>>> I would like to do the following:
>>>>>>>
>>>>>>> 1) remove the list of numbers at the top of each graph,
>>>>>>>
>>>>>>> 2) Combine the graphs.
>>>>>>>
>>>>>>> The typical approach doesn't work (bottom of code) because there are
>>>>>>> different numbers of x and y.
>>>>>>>
>>>>>>> Thank you for your suggestions.
>>>>>>>
>>>>>>> Barry
>>>>>>> P.S. The code runs stand alone.
>>>>>>>
>>>>>>> On Tue, Apr 25, 2017 at 12:34 AM, Mary Haley <haley at ucar.edu> wrote:
>>>>>>>
>>>>>>>> Hi Barry,
>>>>>>>>
>>>>>>>> The gsn_histogram function takes whatever "bin" values you give it,
>>>>>>>> and counts the number of values that fall in each bin, and then draws this
>>>>>>>> as a bar.
>>>>>>>>
>>>>>>>> If you don't provide any bins, then gsn_histogram will pick "nice"
>>>>>>>> bins for you.
>>>>>>>>
>>>>>>>> The binning happens such that no values are binned in more than one
>>>>>>>> interval.
>>>>>>>>
>>>>>>>> Try running the attached script.
>>>>>>>>
>>>>>>>> In the first plot, I let gsn_histogram pick the bin intervals.
>>>>>>>>
>>>>>>>> The first bin, which is from 1 to 2, will capture all values >= 1
>>>>>>>> and < 2, the second bin counts all values >=2 and < 3 and so on. The very
>>>>>>>> last bin, 9 to 10, will count all values >= 9 and <= 10.
>>>>>>>>
>>>>>>>> The second plot I specifically set the bin intervals that I want.
>>>>>>>>
>>>>>>>> In the third plot, instead of counting values that fall in a range,
>>>>>>>> I set an array of discrete values to explicitly count.
>>>>>>>>
>>>>>>>> Hope this helps clear things up.
>>>>>>>>
>>>>>>>> --Mary
>>>>>>>>
>>>>>>>>
>>>>>>>> On Sun, Apr 23, 2017 at 2:23 AM, Barry Lynn <barry.h.lynn at gmail.com
>>>>>>>> > wrote:
>>>>>>>>
>>>>>>>>> Hi:
>>>>>>>>>
>>>>>>>>> I have data specified as below.
>>>>>>>>>
>>>>>>>>> I adapted code from this page:
>>>>>>>>>
>>>>>>>>> https://www.ncl.ucar.edu/Applications/histo.shtml
>>>>>>>>>
>>>>>>>>> Example # 6.
>>>>>>>>>
>>>>>>>>> I set x = "Obs" and y = "Forecast" below.
>>>>>>>>>
>>>>>>>>> The plot I get spreads the obs and forecast data on the x axis
>>>>>>>>> (where the x axis conforms to the value of each).
>>>>>>>>>
>>>>>>>>> The y axis is labeled "Frequency."
>>>>>>>>>
>>>>>>>>> However, this is not a histogram/bar graph I am familiar with. I
>>>>>>>>> am not sure how frequency is calculated (perhaps out of the total). In any
>>>>>>>>> case, I want to plot the x axis as intervals of 0 to 100 (by 10). I want
>>>>>>>>> the y-axis to show the values in the obs/forecasts that conform to these
>>>>>>>>> intervals. (Kaleidagraph can do this easily, but that's not the point
>>>>>>>>> (obviously)).
>>>>>>>>>
>>>>>>>>> I see no such example of how to do this -- even though this is
>>>>>>>>> often how bar graphs are presented.
>>>>>>>>>
>>>>>>>>> Thank you for your suggestions.
>>>>>>>>>
>>>>>>>>> Barry
>>>>>>>>>
>>>>>>>>> z = new((/2,dimsizes(x)/),integer)
>>>>>>>>> z(0,:) = x
>>>>>>>>> z(1,:) = y
>>>>>>>>>
>>>>>>>>> Percent, Obs, Forecast
>>>>>>>>>
>>>>>>>>> 0 0.0, 0.0,
>>>>>>>>>
>>>>>>>>> 1 227.0, 212.2,
>>>>>>>>>
>>>>>>>>> 4 375.0, 176.7,
>>>>>>>>>
>>>>>>>>> 8 201.0, 141.1,
>>>>>>>>>
>>>>>>>>> 12 62.0, 143.2,
>>>>>>>>>
>>>>>>>>> 18 26.0, 327.5,
>>>>>>>>>
>>>>>>>>> 24 6.0, 199.5,
>>>>>>>>>
>>>>>>>>> 30 1.0, 50.2,
>>>>>>>>>
>>>>>>>>> 36 2.0, 9.4,
>>>>>>>>>
>>>>>>>>> 100 0.0, 4.0,
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> --
>>>>>>>>> Barry H. Lynn, Ph.D
>>>>>>>>> Senior Lecturer,
>>>>>>>>> The Institute of the Earth Science,
>>>>>>>>> The Hebrew University of Jerusalem,
>>>>>>>>> Givat Ram, Jerusalem 91904, Israel
>>>>>>>>> Tel: 972 547 231 170
>>>>>>>>> Fax: (972)-25662581
>>>>>>>>>
>>>>>>>>> C.E.O, Weather It Is, LTD
>>>>>>>>> Weather and Climate Focus
>>>>>>>>> http://weather-it-is.com
>>>>>>>>> Jerusalem, Israel
>>>>>>>>> Local: 02 930 9525
>>>>>>>>> Cell: 054 7 231 170
>>>>>>>>> Int-IS: x972 2 930 9525
>>>>>>>>> US 914 432 3108 <(914)%20432-3108>
>>>>>>>>>
>>>>>>>>> _______________________________________________
>>>>>>>>> ncl-talk mailing list
>>>>>>>>> ncl-talk at ucar.edu
>>>>>>>>> List instructions, subscriber options, unsubscribe:
>>>>>>>>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> --
>>>>>>> Barry H. Lynn, Ph.D
>>>>>>> Senior Lecturer,
>>>>>>> The Institute of the Earth Science,
>>>>>>> The Hebrew University of Jerusalem,
>>>>>>> Givat Ram, Jerusalem 91904, Israel
>>>>>>> Tel: 972 547 231 170
>>>>>>> Fax: (972)-25662581
>>>>>>>
>>>>>>> C.E.O, Weather It Is, LTD
>>>>>>> Weather and Climate Focus
>>>>>>> http://weather-it-is.com
>>>>>>> Jerusalem, Israel
>>>>>>> Local: 02 930 9525
>>>>>>> Cell: 054 7 231 170
>>>>>>> Int-IS: x972 2 930 9525
>>>>>>> US 914 432 3108 <(914)%20432-3108>
>>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> Barry H. Lynn, Ph.D
>>>>> Senior Lecturer,
>>>>> The Institute of the Earth Science,
>>>>> The Hebrew University of Jerusalem,
>>>>> Givat Ram, Jerusalem 91904, Israel
>>>>> Tel: 972 547 231 170
>>>>> Fax: (972)-25662581
>>>>>
>>>>> C.E.O, Weather It Is, LTD
>>>>> Weather and Climate Focus
>>>>> http://weather-it-is.com
>>>>> Jerusalem, Israel
>>>>> Local: 02 930 9525
>>>>> Cell: 054 7 231 170
>>>>> Int-IS: x972 2 930 9525
>>>>> US 914 432 3108 <(914)%20432-3108>
>>>>>
>>>>> _______________________________________________
>>>>> ncl-talk mailing list
>>>>> ncl-talk at ucar.edu
>>>>> List instructions, subscriber options, unsubscribe:
>>>>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>>>>>
>>>>>
>>>>
>>>
>>>
>>> --
>>> Barry H. Lynn, Ph.D
>>> Senior Lecturer,
>>> The Institute of the Earth Science,
>>> The Hebrew University of Jerusalem,
>>> Givat Ram, Jerusalem 91904, Israel
>>> Tel: 972 547 231 170
>>> Fax: (972)-25662581
>>>
>>> C.E.O, Weather It Is, LTD
>>> Weather and Climate Focus
>>> http://weather-it-is.com
>>> Jerusalem, Israel
>>> Local: 02 930 9525
>>> Cell: 054 7 231 170
>>> Int-IS: x972 2 930 9525
>>> US 914 432 3108 <(914)%20432-3108>
>>>
>>
>>
>>
>> --
>> Barry H. Lynn, Ph.D
>> Senior Lecturer,
>> The Institute of the Earth Science,
>> The Hebrew University of Jerusalem,
>> Givat Ram, Jerusalem 91904, Israel
>> Tel: 972 547 231 170
>> Fax: (972)-25662581
>>
>> C.E.O, Weather It Is, LTD
>> Weather and Climate Focus
>> http://weather-it-is.com
>> Jerusalem, Israel
>> Local: 02 930 9525
>> Cell: 054 7 231 170
>> Int-IS: x972 2 930 9525
>> US 914 432 3108 <(914)%20432-3108>
>>
>
>
--
Barry H. Lynn, Ph.D
Senior Lecturer,
The Institute of the Earth Science,
The Hebrew University of Jerusalem,
Givat Ram, Jerusalem 91904, Israel
Tel: 972 547 231 170
Fax: (972)-25662581
C.E.O, Weather It Is, LTD
Weather and Climate Focus
http://weather-it-is.com
Jerusalem, Israel
Local: 02 930 9525
Cell: 054 7 231 170
Int-IS: x972 2 930 9525
US 914 432 3108
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20170503/791b66cf/attachment.html
More information about the ncl-talk
mailing list