Comparing NaN

I need to compare two big arrays of double which may contain items set to not-a-number (NaN).
When both arrays contain NaN items, the comparison should yield "equal" if all NaN elements are placed at the same index position and, of course, if all numeric items (elements that are not NaN) are equal.
Unfortunately this is not the standard behavior: if you compare two NaNs, LabVIEW yields "different" with the standard operator.
I have tried "flattening" to string both arrays before the comparison in order to achieve the expected behaviour. However in this case the performance is very poor (350ms to compare 1MB of data).
Finally I had to implement an external DLL function that merely yields the result of the "memcmp" C function. In this way comparing 1M
b of data requires at the most 100ms.
Isn't there a more elegant way in LabVIEW to obtain the same result?

> I need to compare two big arrays of double which may contain items set
> to not-a-number (NaN).
> When both arrays contain NaN items, the comparison should yield
> "equal" if all NaN elements are placed at the same index position and,
The equals node in LV act the way it does because that is the way IEEE
defines it. The string comparison is a good idea, but most of the time
is being spent converting the data into a string and possibly swapping
bytes.
Another approach, that will probably be faster is to do a simple
comparison first, but when it returns false, look at the elements that
are FALSE and see if both are NaNs. If there aren't that many NaNs, it
should go pretty quick. To get the more detailed equals data, popup on
the equals node and turn
on Compare Aggregates. This will now return an
array of Booleans instead of a single one. You can either loop through
or search for FALSE. The index returned can be used to index the input
elements and see if they are both NaN using the IsNan function.
Of course if you want to use a DLL and C code to do the comparison, that
will work too and will probably be even faster.
Greg McKaskle

Similar Messages

Comparing double NaN

It seems that there is a bug in the way the virtual machine compares floating point NaN's. My understanding of the spec is that dcmpg and fcmpg should return +1 if one of the stack elements are NaN. As far as I can see -1 is returned (which is what dcmpl and fcmpl does).
I have tried 1.4.2 on win2000 and kvm cldc1.1 on win2000.
Any comments?

Interesting, but if you look at:
http://java.sun.com/docs/books/vmspec/2nd-edition/html/
nstructions2.doc3.html
the spec for dcmp says that it should push 1 to the
stack, which it does not. The VM and language spec
seems to say different things. And the VM is following
the language spec and not its own spec.
I don't understand what you are talking about. The code that you posted is java - not byte codes. The java code should return nothing which it does.
But looking a the bytes codes for the above (using 1.4.0_01-b03) I get the following partial byte code listing for the first expression ...
    14 dcmpg
    15 ifge 26
    18 getstatic #15 <Field java.io.PrintStream out>
    21 ldc #117 <String "one">
    23 invokevirtual #23 <Method void println(java.lang.String)>From the VM spec for ifge...
<if>ge succeeds if and only if value >= 0
And your code doesn't print 'one'. That means the branch is being taken.
That means only one of the following is true.
1. There is a '1' (or some positive value on the stack.)
2. The documentation for ifge is wrong.
So are you also stating that the documentation for ifge is wrong?
If not then you need to provide exact information on how you determine that there is a '-1' on the stack instead of a '1'. And perhaps provide the real context that demonstrates this.

Multiple Comparisons on Single Data Source: NaN vs. Large Number

Hello All,
Working in TestStand 2013, and came across a situation with DMM measurements, in which several values would be acceptable for the limit comparison. For example, if I am making a resistance measurement, and I expect it to be a very high value, say greater than 1M, I would insert that as the lower limit. However, there are also certain situations when I may see the result of "NaN" at the meter. For the purposes of the test, any number greater that 1M would be acceptable, but a result of NaN would also be acceptable. In an effort to keep the code modular and reusable for many instances, I am curious if there is a way in TestStand to compare the test result against two values at the same time, (i.e. >1M OR =NaN). Please advise. Thanks.
GSinMN
Solved!
Go to Solution.

Probably because you are trying to set the Low limit property. You need to set the Low Limit Expression property and set the Use Expression flag to true.
Consider the screen shot attached. If UseLowExpr is False then the Low numeric property is used. If it is True then the LowExpr property is used.
Hope this helps,
jigg
CTA, CLA
teststandhelp.com
~Will work for kudos and/or BBQ~
Attachments:
LowLimitExpression.png ‏19 KB

File path returning NaN

Hello!
I'm currently involved with a project that requires analysis of images from several hundred stations along the Alaskan seafloor. Each station has three top down (PV) images of the seafloor taken and three images (SPI) that are taken below the seafloor (it's pretty cool).
It is often necessary to compare the SPI and PV images side by side in order to make a proper assessment of the seafloor at each station. I find myself opening the SPI images most often and then needing to dig around in the PV folder to find the appropriate image pairings. This can be time consuming so I've decided to write a script that will query my active document for a name and path and then use that information automatically open up its counterpart.
I'm close to getting it working but I keep returning an error that says "error opening NaN/PV/CH_PV_Analysis/CH-14B1-PV-41-a.psd".
The error returns a perfect file path from "/PV/" onward but for some reasons begins the path with "NaN" when it should pull my file path.   Why is this? I'm using the same line of code for another automation script on this project and can't seem to figure out why it's behaving differently here.
I should mention as well that I'm accessing the files via a server.
My code looks like this:
// call the current document
var srcDoc = app.activeDocument;
// ask user input for PV replicate letter
var repLetter =prompt("Which PV replicate would you like to open? (A.. B.. C.. etc.)");
// get the info out of the source doc
var fileName = srcDoc.name;
var docName = fileName.substring(0,fileName.length -4);
var filePath = srcDoc.path.toString();
var fileExt = fileName.substring(fileName.length -4, fileName.length);
var nameCheck = fileName.substring(0,fileName.indexOf("CH-14B1-SPI-"));
if (nameCheck <1)
   var fileNum = fileName.substring(12,fileName.length -5) + repLetter;
   // no underscore so we need to open it's namesake
   // alert(nameCheck)
   var filePair = filePath -"/SPI/CH_SPI_Analysis/" + "/PV/CH_PV_Analysis/" + "CH-14B1-SPI-PV-" + fileNum + fileExt;
   openThisFile(filePair)
function openThisFile(masterFileNameAndPath)
var fileRef = new File(masterFileNameAndPath)
if (fileRef.exists)
//open that doc
    app.open(fileRef);
else
    alert("error opening " + masterFileNameAndPath)

var filePair = filePath -"/SPI/CH_SPI_Analysis/"
should probably be
var filePair = filePath + "/SPI/CH_SPI_Analysis/"

How to set 0 for "Nan" Value in the Column

Hi All,
I have 3 imp columns with name Build %,Deploy%,Validation % and I would like to set the value as "0" or "1" using formula and below is the formula . But I am getting the values as "Nan" due to that it's reflecting the
Pivot Table & Report which show in below 2nd image.
And I had written formula as "=SWITCH(EDSCIMonthlyReport[Build %] , "Nan", 0)" to copy all the values of the Build % Field to another calculated Column and set the value "0" if found "Nan" ,But I am getting
error as"Function 'SWITCH' attempted to compare values of incompatible types. Ensure that both values are Numeric, String, or Boolean(True/False). Consider using the VALUE or FORMAT function to convert one of the values".Can any one help me how can
I solve this issue
Thanks,
Sid

Hello Sid,
Again TFS? The data type of the column is "text" and now you want to convert it to a numeric to use it as a measure, right?
- First you have to change "NaN" to "0" (<= 0 as type "Text"!, see your error msg); you can do it with an IF expression
- Then you have to remove the % sign; it's a text as well. For this you can use the SUBSTITUE function
- Then you can convert it to a numeric using the VALUE function
As one complete expression it will look like this:
= VALUE(SUBSTITUTE(If(Tabelle3[Build] = "NaN"; "0"; Tabelle3[Build]); "%"; ""))
As you can see in below screenshot PowerPivot automatically detects the data type as "Decimal"; now you can use it as a measure value.
Olaf Helper
[ Blog] [ Xing] [ MVP]

Best fit returns NaN

I've encountered a problem with the "Linear Fit" function.
If I give it the following data then it returns NaN for Intercept and Gradient.
This is when set to "bisquare" method.
I can find similar examples if I use "least absolute residual" method.
"Least squares" functions fine.
x values
1
2
3
4
4
5
6
y values
-0.337
-0.234
-0.140
-0.034
0.34
0.05
0.14
This is clearly the wrong solution. If I adjust any of the y values by 0.001 then it will work.
A much simpler example of what it can't cope with is to do a "least absolute residual" through the data
x=0,0,1
y=0,0,1
Attachments:
cantfit.jpg ‏121 KB
Cantfit.vi ‏16 KB
Cantfit2.vi ‏16 KB

I've just been reading this: http://en.wikipedia.org/wiki/Least_absolute_deviations
It has to use a numerical method of some sort and I guess that some combinations of points combined with rounding of numbers might give nonsensical output.
Since I don't have many data points I might make my own vi to do this using:
"Checking all combinations of lines traversing any two (x,y) data points
is another method of finding the least absolute deviations line. Since
it is known that at least one least absolute deviations line traverses
at least two data points, this method will find a line by comparing the
SAE of each line, and choosing the line with the smallest SAE. In
addition, if multiple lines have the same, smallest SAE, then the lines
outline the region of multiple solutions. Though simple, this final
method is inefficient for large sets of dat"

NaN in Quiz Results page

I created a 5-question quiz for a project in Captivate 7. I want the results to show like this:
In preview, I took the quiz and entered all correct answers. This is what showed in the results page:
It says I only got 1 question correct and says NaN for total questions and accuracy. How do I fix this?
Thanks!

Thanks for the response. I have both a pretest and posttest in this project. I have the pretest score variable set to appear in the results page so the learner can compare pretest and posttest scores. I would like the post test results to appear on the results page as well, but those are displaying as NaN.
These are my Advanced Interaction pages. Not sure what I need to change.
Pretest are slides 10-14.
This is the quiz properties for one of the pretest questions.
Posttest Slides are slides 34, 37, 40, 43, and 46. (I created feedback slides in between.)
This is quiz poperties for one of the posttest questions:
What do I need to do fix the NaN?

Merge Waveforms with NaN Padding

I'm looking for a more efficient way to merge an array of waveforms that are padded with NaN's. Each waveform has the same number of samples and dt, but each one contains a fragment of a signal, padded by NaN's. The actual signal fragments do not overlap. It seems like there has got to be a more efficient way to do this, but I'm stumped, and it's Friday.
The graph in the example (LV8.5) contains each individual waveform, and then a merged waveform.
Any suggestions?
Chris
Message Edited by C. Minnella on 12-12-2008 04:32 PM
Attachments:
Example Merge Waveforms.vi ‏25 KB
Merge Waveforms.png ‏34 KB

Thanks for the link, I must have missed that. It is not entirely the same, because it does not touch the situation where there are NaN and valid values present at the same time.
Clearly, there is some error handling, because the indexes are "-1" in the case of 100% NaNs. But if there is a single valid value in the array, it will be returned as min and max, thus the NaNs are ignored. Many times we are only interested in the max value and not its position so the index of -1 can be easily missed. I strongly feel that is incorrect to return a valid output (zero) for invalid inputs. I would feel much better if it would return NaN as min and max in this case.
Another edge case is an input of an empty array. This also returns zeroes with -1 indexes. Even here, NaN would be more approriate.
It is probably OK to ignore NaNs and operate on the valid values as if nothing is wrong if at least one value is valid. This behavior should be retained.
The output of -1 under problem conditions in not documented in the LabVIEW help. In addition, the labview help contains the following sentence:
"The function compares each data type according to the rules for Array Comparison.".
This seems misplaced and should be deleted, because the rules for array comparisons talks about aggregate vs elements and boolean outputs.
All this is completely irrelevant for array min&max. Right?
In summary:
The function should return NaN for empty array or arrays containing 100% NaN. (unless there is some common standard that needs to be followed)
LabVIEW help page for "array min&max" needs to document the "-1" index output for invalid inputs
The "comparison" sentence needs to be deleted from the help page on "array min&max".
LabVIEW Champion . Do more with less code and in less time .
Attachments:
MinMaxNaN.png ‏14 KB

Why not Integer.NaN?

Hello
I wonder why Integer does not supply a NaN like Double.NaN. I find the Double.Nan to be very useful.
Regards
JT

:) I think what you mean is that integers are
normally only used with the basic arithmetic
operations: +, -, *, and / (I miss having Python's
built-in exponentiation operator). These operations
never return infinity or NaN, with the single notable
exception of division by zero. And dividing by zero
is almost always an accident with integers, and is
easy to check for.Yes, that's what I meant. My post was so dumb even I have to laugh at it. First, I didn't say integers are closed over multiplication, and then I said modulo and remainder like they're two different things. And integers aren't closed under integer division because of 0, as you said.
Floating point numbers, on the other hand, are often
used with transcendental functions and fractional
exponents and things, which frequently result in
infinity or NaN. Moreover, it's important to be able
determine that the result is one of those things, so
we have constants for them. Also, it's often hard to
check these things ahead of time with these
functions.
You dig?
As an aside, I wonder if an operator-overloading
framework could be developed for Java by explicitly
basing the use of operators on interfaces. For
instance, to use the arithmetic operators, you would
implement an interface called Field. To use
the comparison operators, you would implement an
interface called Ordered (which would
basically be identical to the Comparable interface).
You get the idea. The premise would be that it
t would the operators to conform with their standard
mathematical definitions.As someone adamantly opposed to operator overloading in Java, this proposal is ok to me. If you're familiar with the matlab language, operator overloading is done there by defining functions like "plus()", etc... so its sort of similar I think. I think C++ gets difficult because you can overload operators like the function call operator (), and the member of operator(s) which gets confusing really quick.

Get the values from a table comparing fields of different data types

Hi Experts,
I want to fetch AFVC-AUFPL and AFVC- APLZL by passing WBS element from a ztable.
select aufpl aplzl
from afvc
into TABLE gt_afvc
FOR ALL ENTRIES IN gt_boq
where projn = gt_boq-posid.
AFVC-PROJN - NUMC 8 (WBC ELEMENT)
ZTABLE-POSID - CHAR24 (WBC ELEMENT)
How to get aufpl and aplzl by comparing those 2?
Regards
Mani

WBS element does have an conversion routine at the domain level. So sometimes you work with one format (8 digits) and sometimes with 24 digits.
You use FMs
CONVERSION_EXIT_ABPSP_INPUT
CONVERSION_EXIT_ABPSP_OUTPUT
to convert between values. Do your conversion before your select and do not forget to use a large enough data element for the field, which gets the 24-digits long value of the field.

Since installing LTR 5.4, which I've now upgraded to 5.6, I've encountered repeated slowness and malfunctions in operations, especially when using the Compare View function and the Tab key to open and close the right and left side panels. Such problems n

Since installing LTR 5.4, which I've now upgraded to 5.6, I've encountered repeated slowness and malfunctions in operations, especially when using the Compare View function and the Tab key to open and close the right and left side panels. Such problems never arose during two years of using LTR-4 and nothing else has changed on my computer. I have a pretty simple system with only a few plug-ins, which are usually not in operation. I have 12GB of RAM in my Windows 7 PC. I could illustrate these problems with screen shots if you would tell me how to submit screen shots. Otherwise I will try to describe the problems in words.
The problem is clearly cumulative, growing worse as usage time passes. Compare View feature gradually slows down and eventually seems to choke as my work session proceeds. If I Exit LTR and re-enter and start all over, things will work normally for maybe 30 minutes, but then the Compare View feature begins to become very slow to respond. In a recent example with my screen full of thumbnails in Library mode I highlighted two images to compare. LTR started to open the Compare View screen by first having the top row of thumbnails disappear to be replaced by the "SELECT" and "CANDIDATE" words in their spaces (but no images), but Compare View never succeeded in gaining control of the screen. After some seconds the top row of thumbnails reasserted its position and the Compare View windows disappeared. But LTR kept trying to bring them back. Again the top row of thumbnails would go away, Select and candidate would reappear, try again, and give up. This went on for at least 2-3 minutes before I tried to choose File and Exit, but even that did not initially want to respond. It doesn't like to accept other commands when it's trying to open Compare View. Finally it allowed me to exit.
To experiment I created a new catalog of 1100 images. After 30-40 minutes, the Compare View function began to operate very slowly. With left and right side panels visible and two thumbnails highlighted, hitting Compare View can take half a minute before the two mid-size images open in their respective SELECT and CANDIDATE windows. When the side panels are open and two images are in the Select/Candidate spaces, hitting the Tab button to close the side panels produces a very delayed response--25-30 seconds to close them, a few more seconds to enlarge the two images to full size. To reverse the process (i.e., to recall the two side panels), hitting Tab would make the two sides of the screen go black for up to a minute, with no words visible. Eventually the info fields in the panels would open up.
I also created a new user account and imported a folder of 160 images. After half an hour Compare View began mis-placing data. (I have a screen shot to show this.) CANDIDATE appears on the left side of SELECT, whereas it should be on the right. The accompanying camera exposure data appears almost entirely to the left of the mid-screen dividing line. Although the Candidate and Select headings were transposed, the image exposure data was not, but the data for the image on the right was almost entirely to the left of the line dividing the screen in two.
Gurus in The Lightroom Forum have examined Task Manager data showing Processes running and Performance indicators and they see nothing wrong. I could also send screen shots of this data.
At this point, the only way I can process my images is to work 30-40 minutes and then shut down everything, exit, and re-start LTR. This is not normal. I hope you can find the cause, and then the solution. If you would like to see my screen shots, tell me how to submit them.
Ollie
[email protected]

Since installing LTR 5.4, which I've now upgraded to 5.6, I've encountered repeated slowness and malfunctions in operations, especially when using the Compare View function and the Tab key to open and close the right and left side panels. Such problems never arose during two years of using LTR-4 and nothing else has changed on my computer. I have a pretty simple system with only a few plug-ins, which are usually not in operation. I have 12GB of RAM in my Windows 7 PC. I could illustrate these problems with screen shots if you would tell me how to submit screen shots. Otherwise I will try to describe the problems in words.
The problem is clearly cumulative, growing worse as usage time passes. Compare View feature gradually slows down and eventually seems to choke as my work session proceeds. If I Exit LTR and re-enter and start all over, things will work normally for maybe 30 minutes, but then the Compare View feature begins to become very slow to respond. In a recent example with my screen full of thumbnails in Library mode I highlighted two images to compare. LTR started to open the Compare View screen by first having the top row of thumbnails disappear to be replaced by the "SELECT" and "CANDIDATE" words in their spaces (but no images), but Compare View never succeeded in gaining control of the screen. After some seconds the top row of thumbnails reasserted its position and the Compare View windows disappeared. But LTR kept trying to bring them back. Again the top row of thumbnails would go away, Select and candidate would reappear, try again, and give up. This went on for at least 2-3 minutes before I tried to choose File and Exit, but even that did not initially want to respond. It doesn't like to accept other commands when it's trying to open Compare View. Finally it allowed me to exit.
To experiment I created a new catalog of 1100 images. After 30-40 minutes, the Compare View function began to operate very slowly. With left and right side panels visible and two thumbnails highlighted, hitting Compare View can take half a minute before the two mid-size images open in their respective SELECT and CANDIDATE windows. When the side panels are open and two images are in the Select/Candidate spaces, hitting the Tab button to close the side panels produces a very delayed response--25-30 seconds to close them, a few more seconds to enlarge the two images to full size. To reverse the process (i.e., to recall the two side panels), hitting Tab would make the two sides of the screen go black for up to a minute, with no words visible. Eventually the info fields in the panels would open up.
I also created a new user account and imported a folder of 160 images. After half an hour Compare View began mis-placing data. (I have a screen shot to show this.) CANDIDATE appears on the left side of SELECT, whereas it should be on the right. The accompanying camera exposure data appears almost entirely to the left of the mid-screen dividing line. Although the Candidate and Select headings were transposed, the image exposure data was not, but the data for the image on the right was almost entirely to the left of the line dividing the screen in two.
Gurus in The Lightroom Forum have examined Task Manager data showing Processes running and Performance indicators and they see nothing wrong. I could also send screen shots of this data.
At this point, the only way I can process my images is to work 30-40 minutes and then shut down everything, exit, and re-start LTR. This is not normal. I hope you can find the cause, and then the solution. If you would like to see my screen shots, tell me how to submit them.
Ollie
[email protected]

How to compare delivery date in schedule lines and PGI in outbound delivery

Hi,
I want to create a report to monitor and compare different dates in sales and delivery process.
One of my key figure is the comparison between date confirmed to the customer, that I can find in the schedule line of the sales order, and the actual goods issue date from the outbound delivery.
But I can't succeed in making the link between delivery and schedule lines, means between LIKP and VBEP.
Would you have any idea?
Thank you.

It seems there's no exact (table) connection / link between delivery item and schedule line item.
Those who were searching for this thing (link / connection between delivery item and schedule line) were recommended to use FM "RV_SCHEDULE_CHECK_DELIVERIES".
This FM calculates the delivered quantity for each schedule line, so it makes link between the delivery item and schedule lines - however if you check the code it's quite sophisticated...
(if you put a breakpoint in the FM and display a SO in VA03, you can check how the parameters have to be populated with data).
Schedule Line and Delivery Link
Edited by: Csaba Szommer on Aug 5, 2011 12:03 AM

What is new in ECC 5.0 compared to R/3 4.7 version

Hi
what is new in ECC 5.0 compared to R/3 4.7 version
Thanks in Advance
Deepa

Hi
see these links
Refer this link for ECC 5.0 & 6.0:
http://help.sap.com/saphelp_erp2005/helpdata/en/43/68805bb88f297ee10000000a422035/frameset.htm
A similar post
/message/1783778#1783778 [original link is broken]
You can go through the Release Notes for each of the versions after 4.6B (4.6C, 4.7 , ECC 5.0 & ECC 6.0)
http://help.sap.com/saphelp_erp2005/helpdata/en/43/688055b88f297ee10000000a422035/content.htm
For 4.7 SAP R/3 http://help.sap.com/saphelp_47x200/helpdata/en/12/9d78d6d8f74043a32e82be87e433b7/content.htm
Release Notes on SAP Web Application Server 6.30
http://help.sap.com/saphelp_47x200/helpdata/en/2b/64fc3e9d3b6927e10000000a114084/content.htm
http://solutionbrowser.erp.sap.fmpmedia.com/
Reward points if useful
Regards
Anji

SSRS Matrix report. Variance expression by Month. Need to compare month from prior year to current month of current year VS2010

Please help. I have a matrix report. In the report I have row group PO Type. One the Column groups I have a parent group by Fiscal Year, and then a child group by Month. When I run the report, I get two years of data back broken
out by month. Please see below.
Now here is where I am getting stuck. I need to take the variance between the current month of the current year, from the same month of the prior year. So I need to show the difference between Oct , 2014 from Oct, 2013. November, 2014 from November
2013... etc. etc.
In the example below, how do I create a column or row showing the variance for Contracts for October 2014. I need to take the contracts for October 2014 which is 3 and subtract that from October 2013 which is 8. Any suggestions? How do I do that
for each month? Then I need to do it for the quarter... then the year? But I'll be happy if I can just get the month working first.
Any help will be appreciated.
here is what my rdl file looks like.
Here is what my report looks like when I render it.

Hi Adrian_s2012,
According to your description, you want to compare values for the month of current year with the month of prior year and get the variance. Right?
In Reporting Services, we don't have any function to get this "Year to Year" Growth. In this scenario, if you data source is a cube, we suggest you use Analysis Services to achieve your requirement. If this data source is just from database, it will be hardly
to calculate the variance because we need to compare the values within every two different column group and matrix generate adjacent columns one by one. Even we make it by using custom, every time executing the long code when generating result
in a cell will reduce a lot of performance, we really don't suggest to do that in SSRS. Here is a thread with much easier requirement, please take a reference of that:
http://social.msdn.microsoft.com/Forums/office/en-US/842e2dcb-d949-4297-9d91-eac989692cb5/difference-between-the-grouped-column?forum=sqlreportingservices
If you have any question, please feel free to ask.
Best Regards,
Simon Hou

Getting error in XMLP while compare a Tag value to partial value

Hi All,
We are working on 'AP Check Print Report' which is XML based Report and the output is in the PDF format.
We are facing a problem when we are suppose to compare a Tag value (DocumentNumber/ReferenceNumber) to a partial value('%TDS-CM%') as below,
<?if:DocumentNumber/ReferenceNumber NOT LIKE '%TDS-CM%'?>
Can someone help me, how to achieve it as well as whether it can be done or not.
Thanks in advance
Abhishek

Have a look at these threads:
Re: Using Wild Card % in the IF condition of RTF templates
Re: Syntax for IN / NOT IN Commands
Thanks!

Comparing NaN

Similar Messages

Maybe you are looking for