News from the SPSS Community
Locked
12
12
 
|
A new extension command, STATS CARTPROD,
has been uploaded to the SPSS Community website (www.ibm.com/developerworks/spssdevcentral)
in the Extension Commands collection. It computes the Cartesian product
of two sets of variables in the same or separate datasets and writes the
result to a sav file and opens it. While not as fast as this operation
would be in a database, it is quite general and preserves all variable
properties. (All the data operations are carried out with traditional
Statistics syntax, but Python code is used to generate that syntax.)
This command requires the Python Essentials. The Essentials can be obtained via the Community site or the Statistics product materials, depending on the Statistics version. The extension command can be downloaded from the Extension Commands collection or installed directly using Utilities > Download and Install Extension Bundles with Statistics version 22. Regards, Jon Peck (no "h") aka Kim Senior Software Engineer, IBM [hidden email] phone: 720-342-5621 |
Re: News from the SPSS Community
|
Jon,
1) Cartesian product is simple to do in MATRIX. MATRIX. COMPUTE X1= {'A';'B';'C'}. COMPUTE X2= {'1';'2';'3';'4'}. COMPUTE X1X2= {KRONEKER(X1,MAKE(NROW(X2),1,1)),KRONEKER(MAKE(NROW(X1),1,1),X2)}. PRINT X1X2 /format= a1. END MATRIX. Please, name the advantages of your extension command over the above syntax. 2) Is it possible to get into the actual SPSS syntax which hides behind the extension command? Because I'm interested. |
Re: News from the SPSS Community
|
|
In reply to this post by Jon K Peck
P.S.
Cartesian product of 3+ sets is easily done by the same approach of consequtive products of 2 sets. For example, 3 sets:. MATRIX. COMPUTE X1= {'A';'B';'C'}. COMPUTE X2= {'1';'2';'3';'4'}. COMPUTE X3= {'+';'-'}. COMPUTE X2X3= {KRONEKER(X2,MAKE(NROW(X3),1,1)),KRONEKER(MAKE(NROW(X2),1,1),X3)}. COMPUTE X1X2X3= {KRONEKER(X1,MAKE(NROW(X2X3),1,1)),KRONEKER(MAKE(NROW(X1),1,1),X2X3)}. PRINT X1X2X3 /format= a1. END MATRIX. Clearly, this task for any number of sets is looping task (macro or usual), so one can write a function. |
Re: News from the SPSS Community
|
|
In reply to this post by Jon K Peck
Nice. I rarely need a Cartesian product though; in R it sometimes happens when I make a mistake. Nevertheless I thought about creating such an extension command before, by using SQLite. You have the option between in-memory and file-based with Sqlite. But I suppose that pushing and fetching the data is not very efficient, even with an in-memory database.
Regards, Albert-Jan ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ All right, but apart from the sanitation, the medicine, education, wine, public order, irrigation, roads, a fresh water system, and public health, what have the Romans ever done for us? ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
|
Re: News from the SPSS Community
|
|
In reply to this post by Jon K Peck
P.P.S.
And here comes the promised MATRIX macrofunction. I'll add it to my site. !cartprod(PREFIX%NSETS%RESULT). *PREFIX - prefix in the names of the sets (may be quoted), NSETS - number of the sets. *For example, X and 3 means sets X1 X2 X3. *A set may consist of any number of columns (it's not important). Elements of a set are its ROWS. define !cartprod(!pos= !token(1) /!pos= !charend('%') /!pos= !charend('%') /!pos= !charend(')')) comp !4= !conc(!unquote(!2),!3). !do !i= !3 !to 1 !by -1 !if (!i<>!3) !then !let !set= !conc(!unquote(!2),!i) - comp !4= {kroneker(!set,make(nrow(!4),1,1)),kroneker(make(nrow(!set),1,1),!4)}. !ifend !doend !enddefine. *Example. MATRIX. COMPUTE X1= {'A';'B';'C'}. COMPUTE X2= {'1';'2';'3';'4'}. COMPUTE X3= {'+';'-'}. !cartprod('x'%3%prod). PRINT prod /format= a4. end matrix. |
Administrator
|
In reply to this post by Kirill Orlov
P.S.S.
DEFINE KronSets (!POS !TOKENS(1) / !POS !CMDEND). !LET !M1=!HEAD(!2) !LET !M2=!HEAD(!TAIL(!2)) COMPUTE !1={KRONEKER(!M1,MAKE(NROW(!M2),1,1)),KRONEKER(MAKE(NROW(!M1),1,1),!M2)}. !IF (!TAIL(!TAIL(!2)) !NE !NULL ) !THEN KronSets !1 !1 !TAIL(!TAIL(!2)). !IFEND !ENDDEFINE . MATRIX. COMPUTE X1= {'A';'B';'C'}. COMPUTE X2= {'1';'2';'3';'4'}. COMPUTE X3= {'+';'-'}. COMPUTE X4= {'x';'y';'z'}. COMPUTE X5= {'@';'!';'%'}. KronSets OUTMAT X1 X2 X3 X4 X5. PRINT OUTMAT / FORMAT "A1". END MATRIX. OUTMAT A 1 + x @ A 1 + x ! A 1 + x % A 1 + y @ A 1 + y ! A 1 + y % A 1 + z @ A 1 + z ! A 1 + z % A 1 - x @ A 1 - x ! A 1 - x % A 1 - y @ A 1 - y ! A 1 - y % A 1 - z @ A 1 - z ! A 1 - z % A 2 + x @ A 2 + x ! A 2 + x % A 2 + y @ A 2 + y ! A 2 + y % A 2 + z @ A 2 + z ! A 2 + z % A 2 - x @ A 2 - x ! A 2 - x % A 2 - y @ A 2 - y ! A 2 - y % A 2 - z @ A 2 - z ! A 2 - z % A 3 + x @ A 3 + x ! A 3 + x % A 3 + y @ A 3 + y ! A 3 + y % A 3 + z @ A 3 + z ! A 3 + z % A 3 - x @ A 3 - x ! A 3 - x % A 3 - y @ A 3 - y ! A 3 - y % A 3 - z @ A 3 - z ! A 3 - z % A 4 + x @ A 4 + x ! A 4 + x % A 4 + y @ A 4 + y ! A 4 + y % A 4 + z @ A 4 + z ! A 4 + z % A 4 - x @ A 4 - x ! A 4 - x % A 4 - y @ A 4 - y ! A 4 - y % A 4 - z @ A 4 - z ! A 4 - z % B 1 + x @ B 1 + x ! B 1 + x % B 1 + y @ B 1 + y ! B 1 + y % B 1 + z @ B 1 + z ! B 1 + z % B 1 - x @ B 1 - x ! B 1 - x % B 1 - y @ B 1 - y ! B 1 - y % B 1 - z @ B 1 - z ! B 1 - z % B 2 + x @ B 2 + x ! B 2 + x % B 2 + y @ B 2 + y ! B 2 + y % B 2 + z @ B 2 + z ! B 2 + z % B 2 - x @ B 2 - x ! B 2 - x % B 2 - y @ B 2 - y ! B 2 - y % B 2 - z @ B 2 - z ! B 2 - z % B 3 + x @ B 3 + x ! B 3 + x % B 3 + y @ B 3 + y ! B 3 + y % B 3 + z @ B 3 + z ! B 3 + z % B 3 - x @ B 3 - x ! B 3 - x % B 3 - y @ B 3 - y ! B 3 - y % B 3 - z @ B 3 - z ! B 3 - z % B 4 + x @ B 4 + x ! B 4 + x % B 4 + y @ B 4 + y ! B 4 + y % B 4 + z @ B 4 + z ! B 4 + z % B 4 - x @ B 4 - x ! B 4 - x % B 4 - y @ B 4 - y ! B 4 - y % B 4 - z @ B 4 - z ! B 4 - z % C 1 + x @ C 1 + x ! C 1 + x % C 1 + y @ C 1 + y ! C 1 + y % C 1 + z @ C 1 + z ! C 1 + z % C 1 - x @ C 1 - x ! C 1 - x % C 1 - y @ C 1 - y ! C 1 - y % C 1 - z @ C 1 - z ! C 1 - z % C 2 + x @ C 2 + x ! C 2 + x % C 2 + y @ C 2 + y ! C 2 + y % C 2 + z @ C 2 + z ! C 2 + z % C 2 - x @ C 2 - x ! C 2 - x % C 2 - y @ C 2 - y ! C 2 - y % C 2 - z @ C 2 - z ! C 2 - z % C 3 + x @ C 3 + x ! C 3 + x % C 3 + y @ C 3 + y ! C 3 + y % C 3 + z @ C 3 + z ! C 3 + z % C 3 - x @ C 3 - x ! C 3 - x % C 3 - y @ C 3 - y ! C 3 - y % C 3 - z @ C 3 - z ! C 3 - z % C 4 + x @ C 4 + x ! C 4 + x % C 4 + y @ C 4 + y ! C 4 + y % C 4 + z @ C 4 + z ! C 4 + z % C 4 - x @ C 4 - x ! C 4 - x % C 4 - y @ C 4 - y ! C 4 - y % C 4 - z @ C 4 - z ! C 4 - z %
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me. --- "Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis." Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?" |
Re: News from the SPSS Community
|
|
I see, David. You could not resist applying the recursive macro idea
of which you are an authority. +1 for that. However, it is
questionable if this refined approach is more efficient than simple
cycling, given that the task is elementary.
05.03.2014 13:41, David Marso пишет:
|
|
Administrator
|
This has nothing to do with efficiency per se. I would argue for elegance and aesthetics ;-)
BTW: I posted my recursive solution before I had read your cycling version. --- Here is a slight refinement of my original version. DEFINE RecurKronProd (!POS !TOKENS(1) / !POS !CMDEND). COMPUTE !1={KRONEKER(!HEAD(!2),MAKE(NROW(!HEAD(!TAIL(!2))),1,1)), KRONEKER(MAKE(NROW(!HEAD(!2)),1,1),!HEAD(!TAIL(!2)))}. !IF (!TAIL(!TAIL(!2)) !NE !NULL ) !THEN RecurKronProd !1 !1 !TAIL(!TAIL(!2)). !IFEND !ENDDEFINE . MATRIX. COMPUTE X1= {'A';'B'}. COMPUTE X2= {'1';'2';'3'}. COMPUTE X3= {'+';'-'}. COMPUTE X4= {'x';'y';'z'}. COMPUTE X5= {'@';'!';'%'}. RecurKronProd OUTMAT X1 X2 X3 X4 X5. PRINT OUTMAT / FORMAT "A1". END MATRIX.
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me. --- "Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis." Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?" |
|
|
In reply to this post by Kirill Orlov
No doubt MATRIX works well in some cases,
and would be faster than STATS CARTPROD when applicable.
Advantages of the extension command: - no memory limits - no variable type or name restrictions - preserves all metadata - requires only standard syntax - no MATRIX or macro knowledge required - includes syntax help via STATS CARTPROD /HELP. - clear error messages - log shows only the high level syntax making it easier to read the job output - packaging means it is automatically available in any session including dialog box interface once installed Since the extension actually operates through regular Statistics transformation commands, it is only a convenience, although Python code is able to get the case counts easily which is difficult to do with standard syntax. The Statistics syntax it generates is not directly visible, but if you examine the STATS_CARTPROD.py file that is extracted from the spe when the extension is installed, you can see it pretty clearly in the docart function. p.s. If you use the recently published STATS PROJECT extension command, you could have your macro libraries and settings automatically loaded when Statistics starts. Jon Peck (no "h") aka Kim Senior Software Engineer, IBM [hidden email] phone: 720-342-5621 From: Kirill Orlov <[hidden email]> To: [hidden email], Date: 03/04/2014 11:53 PM Subject: Re: [SPSSX-L] News from the SPSS Community Sent by: "SPSSX(r) Discussion" <[hidden email]> Jon, 1) Cartesian product is simple to do in MATRIX. MATRIX. COMPUTE X1= {'A';'B';'C'}. COMPUTE X2= {'1';'2';'3';'4'}. COMPUTE X1X2= {KRONEKER(X1,MAKE(NROW(X2),1,1)),KRONEKER(MAKE(NROW(X1),1,1),X2)}. PRINT X1X2 /format= a1. END MATRIX. Please, name the advantages of your extension command over the above syntax. 2) Is it possible to get into the actual SPSS syntax which hides behind the extension command? Because I'm interested. |
Re: News from the SPSS Community
|
By the by --
At 09:04 AM 3/5/2014, Jon K Peck wrote: >Since the extension actually operates through regular Statistics >transformation commands, it is only a convenience, although Python >code is able to get the case counts easily which is difficult to do >with standard syntax. Why do you say that? It's quite easy to get case counts with AGGREGATE and MATCH FILES; I do it that way all the time. ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD |
|
|
Because that requires an entire data pass
and creation of a new dataset or replication of the count in a casewise
variable and subsequent retrieval code. Python has an api that returns
the count without those consequences as long as Statistics has that information.
Jon Peck (no "h") aka Kim Senior Software Engineer, IBM [hidden email] phone: 720-342-5621 From: Richard Ristow <[hidden email]> To: Jon K Peck/Chicago/IBM@IBMUS, [hidden email], Date: 03/10/2014 07:24 AM Subject: Re: News from the SPSS Community By the by -- At 09:04 AM 3/5/2014, Jon K Peck wrote: >Since the extension actually operates through regular Statistics >transformation commands, it is only a convenience, although Python >code is able to get the case counts easily which is difficult to do >with standard syntax. Why do you say that? It's quite easy to get case counts with AGGREGATE and MATCH FILES; I do it that way all the time. |
|
Administrator
|
In reply to this post by Richard Ristow
This is a VERY FAST alternative.
DATASET DECLARE nOfCases. MATRIX. GET x / FILE * . SAVE NROW(x) / OUTFILE nOfCases/ VARIABLES N. END MATRIX. DATASET ACTIVATE nOfCases. PRINT / N. EXECUTE. The BUGs in matrix which prevent all of this CARTPROD/Many2Many business from being done quite generally with great ease in MATRIX are the lack of support for reading DATASETS rather than disk or * and the requirement to run SPLIT FILE processing in 'BATCH mode'. Building Richard's 'spine' can be done easily with KRONEKER.
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me. --- "Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis." Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?" |
Re: News from the SPSS Community
|
|
In MATRIX, I believe, functions NROW and NCOLS are automatically
recomputed every time a variable is modified. So these are "fast
functions": value is already there, waiting.
I always said that SPSS should have automatic system variable (constant) $ncases. It would be more convenient than to worry AGGREGATE command when one wants to know the number of cases. 10.03.2014 17:57, David Marso пишет:
This is a VERY FAST alternative. DATASET DECLARE nOfCases. MATRIX. GET x / FILE * . SAVE NROW(x) / OUTFILE nOfCases/ VARIABLES N. END MATRIX. DATASET ACTIVATE nOfCases. PRINT / N. EXECUTE. The BUGs in matrix which prevent all of this CARTPROD/Many2Many business from being done quite generally with great ease in MATRIX are the lack of support for reading DATASETS rather than disk or * and the requirement to run SPLIT FILE processing in 'BATCH mode'. Building Richard's 'spine' can be done easily with KRONEKER. |
|
|
The MATRIX approach requires a data pass,
and the value would be awkward to use in syntax. MATRIX, of course,
would also be memory limited.
Statistics does keep the case count as a variable internally, but it is not exposed except through programmability apis. It is, however, not always known. If the data is a non-SPSS source, and the data have not been passed, the value is unknown. Jon Peck (no "h") aka Kim Senior Software Engineer, IBM [hidden email] phone: 720-342-5621 From: Kirill Orlov <[hidden email]> To: [hidden email], Date: 03/10/2014 08:37 AM Subject: Re: [SPSSX-L] News from the SPSS Community Sent by: "SPSSX(r) Discussion" <[hidden email]> In MATRIX, I believe, functions NROW and NCOLS are automatically recomputed every time a variable is modified. So these are "fast functions": value is already there, waiting. I always said that SPSS should have automatic system variable (constant) $ncases. It would be more convenient than to worry AGGREGATE command when one wants to know the number of cases. 10.03.2014 17:57, David Marso пишет: This is a VERY FAST alternative. DATASET DECLARE nOfCases. MATRIX. GET x / FILE * . SAVE NROW(x) / OUTFILE nOfCases/ VARIABLES N. END MATRIX. DATASET ACTIVATE nOfCases. PRINT / N. EXECUTE. The BUGs in matrix which prevent all of this CARTPROD/Many2Many business from being done quite generally with great ease in MATRIX are the lack of support for reading DATASETS rather than disk or * and the requirement to run SPLIT FILE processing in 'BATCH mode'. Building Richard's 'spine' can be done easily with KRONEKER. |
|
Administrator
|
In reply to this post by Jon K Peck
I have the extension installed, but STATS_CARTPROD.py does not seem to appear anywhere on my system.
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me. --- "Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis." Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?" |
|
|
You should see it on the Data menu below
Merge Files. And if the syntax definition is in place and you have
restarted Statistics, you should see STATS CARTPROD in the syntax window
popup when you type STATS.
Jon Peck (no "h") aka Kim Senior Software Engineer, IBM [hidden email] phone: 720-342-5621 From: David Marso <[hidden email]> To: [hidden email], Date: 03/10/2014 09:38 AM Subject: Re: [SPSSX-L] News from the SPSS Community Sent by: "SPSSX(r) Discussion" <[hidden email]> I have the extension installed, but STATS_CARTPROD.py does not seem to appear anywhere on my system. ----- Please reply to the list and not to my personal email. Those desiring my consulting or training services please feel free to email me. --- "Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis." Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?" -- View this message in context: http://spssx-discussion.1045642.n5.nabble.com/News-from-the-SPSS-Community-tp5724729p5724795.html Sent from the SPSSX Discussion mailing list archive at Nabble.com. ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD |
|
|
In reply to this post by David Marso
Bear in mind that (1) Kroneker product
multiplication does not make sense for strings, although MATRIX will do
it.
Jon Peck (no "h") aka Kim Senior Software Engineer, IBM [hidden email] phone: 720-342-5621 From: David Marso <[hidden email]> To: [hidden email], Date: 03/10/2014 07:58 AM Subject: Re: [SPSSX-L] News from the SPSS Community Sent by: "SPSSX(r) Discussion" <[hidden email]> This is a VERY FAST alternative. DATASET DECLARE nOfCases. MATRIX. GET x / FILE * . SAVE NROW(x) / OUTFILE nOfCases/ VARIABLES N. END MATRIX. DATASET ACTIVATE nOfCases. PRINT / N. EXECUTE. The BUGs in matrix which prevent all of this CARTPROD/Many2Many business from being done quite generally with great ease in MATRIX are the lack of support for reading DATASETS rather than disk or * and the requirement to run SPLIT FILE processing in 'BATCH mode'. Building Richard's 'spine' can be done easily with KRONEKER. Richard Ristow wrote > By the by -- > > At 09:04 AM 3/5/2014, Jon K Peck wrote: > >>Since the extension actually operates through regular Statistics >>transformation commands, it is only a convenience, although Python >>code is able to get the case counts easily which is difficult to do >>with standard syntax. > > Why do you say that? It's quite easy to get case counts with > AGGREGATE and MATCH FILES; I do it that way all the time. > > ===================== > To manage your subscription to SPSSX-L, send a message to > LISTSERV@.UGA > (not to SPSSX-L), with no body text except the > command. To leave the list, send the command > SIGNOFF SPSSX-L > For a list of commands to manage subscriptions, send the command > INFO REFCARD ----- Please reply to the list and not to my personal email. Those desiring my consulting or training services please feel free to email me. --- "Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis." Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?" -- View this message in context: http://spssx-discussion.1045642.n5.nabble.com/News-from-the-SPSS-Community-tp5724729p5724790.html Sent from the SPSSX Discussion mailing list archive at Nabble.com. ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD |
|
Administrator
|
In reply to this post by Jon K Peck
Jon,
You had previously replied to Kirill's note: "The Statistics syntax it generates is not directly visible, but if you examine the STATS_CARTPROD.py file that is extracted from the spe when the extension is installed, you can see it pretty clearly in the docart function." I can't locate the .py file. Does it get extracted to any specific location? The menu is misspelled Catesian product. The dialog opens, but I wouldn't mind studying the code to see what's under the hood.
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me. --- "Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis." Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?" |
|
|
The location of the Python file varies
by your machine configuration and, to some extent, your Statistics version.
Run
SHOW EXT to see the possible locations of the .py file. In the case of multiples, the first matching location on the list is used. The XML file defining the legal syntax will be in that same location. Jon Peck (no "h") aka Kim Senior Software Engineer, IBM [hidden email] phone: 720-342-5621 From: David Marso <[hidden email]> To: [hidden email], Date: 03/10/2014 11:54 AM Subject: Re: [SPSSX-L] News from the SPSS Community Sent by: "SPSSX(r) Discussion" <[hidden email]> Jon, You had previously replied to Kirill's note: "The Statistics syntax it generates is not directly visible, but if you examine the STATS_CARTPROD.py file that is extracted from the spe when the extension is installed, you can see it pretty clearly in the docart function." I can't locate the .py file. Does it get extracted to any specific location? The menu is misspelled Catesian product. The dialog opens, but I wouldn't mind studying the code to see what's under the hood. Jon K Peck wrote > You should see it on the Data menu below Merge Files. And if the syntax > definition is in place and you have restarted Statistics, you should see > STATS CARTPROD in the syntax window popup when you type STATS. > > > Jon Peck (no "h") aka Kim > Senior Software Engineer, IBM > peck@.ibm > phone: 720-342-5621 > > > > > From: David Marso < > david.marso@ > > > To: > SPSSX-L@.uga > , > Date: 03/10/2014 09:38 AM > Subject: Re: [SPSSX-L] News from the SPSS Community > Sent by: "SPSSX(r) Discussion" < > SPSSX-L@.uga > > > > > > I have the extension installed, but STATS_CARTPROD.py does not seem to > appear > anywhere on my system. > > > > ----- > Please reply to the list and not to my personal email. > Those desiring my consulting or training services please feel free to > email me. > --- > "Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos > ne forte conculcent eas pedibus suis." > Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in > abyssum?" > -- > View this message in context: > http://spssx-discussion.1045642.n5.nabble.com/News-from-the-SPSS-Community-tp5724729p5724795.html > > Sent from the SPSSX Discussion mailing list archive at Nabble.com. > > ===================== > To manage your subscription to SPSSX-L, send a message to > LISTSERV@.UGA > (not to SPSSX-L), with no body text except the > command. To leave the list, send the command > SIGNOFF SPSSX-L > For a list of commands to manage subscriptions, send the command > INFO REFCARD ----- Please reply to the list and not to my personal email. Those desiring my consulting or training services please feel free to email me. --- "Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis." Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?" -- View this message in context: http://spssx-discussion.1045642.n5.nabble.com/News-from-the-SPSS-Community-tp5724729p5724803.html Sent from the SPSSX Discussion mailing list archive at Nabble.com. ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD |
Re: News from the SPSS Community
|
|
In reply to this post by Jon K Peck
One more note on extensions:
At 09:04 AM 3/5/2014, Jon K Peck wrote: >Since the extension actually operates through regular Statistics >transformation commands, it is only a convenience, although Python >code is able to get the case counts easily which is difficult to do >with standard syntax. I'd rank Python as more than a convenience, even for extensions that do all data manipulation and statistics through native SPSS. Python's 'scripting' ability -- to take action based on the contents of data dictionaries, data files, or output -- can be invaluable in determining what code to generate, or to run. For example, you can (and you did at least some of these in STATS CARTPROD), . Give intelligible error messages and abort in cases where, for example, an input file that doesn't exist was specified, or where keys or other necessary variables aren't present. . Be cognizant of what variables and datasets currently exist; and automatically select non-conflicting names for scratch storage Besides, you could use the scripting capability to run and then verify a transitive closure :-D ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD |
«
Return to SPSSX Discussion
|
1 view|%1 views
| Free forum by Nabble | Edit this page |