combining output to excel

Tue Mar 1 14:33:54 UTC 2016

Hi Leonid,

My responses are in line below.

Thanks!
Brian

On Monday, February 29, 2016 at 4:40:50 PM UTC-5, Spektor, Leonid: CMU 
wrote:
>
> Brian.
>
>     I can't answer your question 1). I have created my own batch file and 
> when I ran it I did not get any error messages.  I ran it on both Mac and 
> Windows 8.1 PC. If you could email me directly some *.cha files you use as 
> input and the CutFileAllCodes.cut file, then I will have better chance of 
> replicating the problem.
>
> I will send some immediately after posting.  As I said, the commands work 
when I plug them individually into the command window.  Seems weird that it 
could be either my chat files or the cut file, but anything is possible I 
suppose... I assumed maybe there was something wrong with the formatting of 
the batch file or the way I called it (I literally type in "batch 
BatchCommands.cex" into the command window).  Nonetheless, this is a 
problem I would definitely like to figure out.

> 2).  When FREQ creates a SPREADSHEET it does not append data to existing 
> output files and certainly it can't add it to the end of each row of each 
> input file's output. Your output consists of only four Excel files, so it 
> should be easy to append each consecutive FREQ file to the right of the 
> previous FREQ file's output by hand using Excel application. 
>

My output from the example commands I posted is only 4 spreadsheets, but I 
will be running these commands with at least 7 groups of words (so 28+ 
spreadsheets) and inevitably there will be other groups of words/codes we 
will want to export.  This still isn't a *huge* problem, but if clan had a 
way to automatically combine them I wanted to make sure I was doing the 
easiest thing.

> All rows for each FREQ command's output Excel file should align 
> automatically since, every input file will have exactly one corresponding 
> row in Excel file and every FREQ command will have exactly the same number 
> of rows. You said that some of your input CHAT files do not have Toy/IPA 
> speaker, but this can be fixed by creating @ID header for Toy speaker in 
> all files and in files that currently do not have any Toy transcription you 
> would just create a one dummy tier for that speaker. For example, if "Toy" 
> speaker code is "IPA", that just add one dummy tier "*IPA:    0." to files 
> that do not have IPA speakers now.  Otherwise, FREQ does not create output 
> for files that do not have speakers specified on command line. 
>

Adding an IPA line should drastically help with combining them and, in 
fact, should make it really easy to write a copy/paste macro for the excel 
files that will do it in a few seconds.  Since I created the chat files 
from excel files to begin with, they actually all already have the @ID 
header for toy.  Just need to add the one dummy line.  Thanks for the idea. 
 I think this will take care of my biggest problem.

> The other problem I see you have is command:
>
> Freq +d5 +d2 +t at ID=”*|Target_Adult|*” +t at ID=”*|Toy|*” +o3 
> +s at CutFileAllCodes.cut +fPARIPA_AllCodes *.cha
>
> This command will produce two rows in Excel for every input file that has 
> both "Target_Adult” and "Toy". If you want FREQ to combine results of those 
> two speakers into one row, then add +o3 option to the FREQ command line.
>

Maybe I'm misunderstanding, but the +o3 is after "Toy" in the command and 
seems to be combining just fine when I copy/paste the command by itself 
into the command window.

>
> If I misunderstood what you are trying to accomplish, then please give me 
> directly a more specific description and some examples would help a lot too.
>

I think you mostly interpreted correctly. If I can get the batch running 
issue fixed I think I'll be back in business.

>
> Leonid.
>
>
> On 29-02-16 15:17, Brian Verdine wrote:
>
> Hi, 
>
> I am trying to figure out if there is a simple way to run and compile data 
> from multiple freq commands.  
>
> Most of the transcripts I have contain only 2 speakers (a child and a 
> parent).  However, about 1/3 also have a toy (an iPad) that talks and that 
> we transcribed.  I am trying to count the types, tokens, and get TTR for 
> different groups of words for these 3 speaker tiers individually.  In 
> addition, for those that heard the iPad, I need to produce type/token stats 
> for a combination of the parent and iPad.  By getting that data we will be 
> able to look at all of the speech a child hears (whether from parent or 
> iPad).  These pieces of data are generated into excel files by the 4 
> commands below.  They seem to be generating the data I need into excel 
> files in (generally) the format I need.
>
> Child:
> Freq +d5 +d2 +t at ID=”*|Target_Child|*” +s at CutFileAllCodes.cut 
> +fCHI_AllCodes *.cha
> Parent:
> Freq +d5 +d2 +t at ID=”*|Target_Adult|*” +s at CutFileAllCodes.cut 
> +fPAR_AllCodes *.cha
> iPad:
> Freq +d5 +d2 +t at ID=”*|Toy|*” +s at CutFileAllCodes.cut +fIPA_AllCodes *.cha
> Parent and iPad:
> Freq +d5 +d2 +t at ID=”*|Target_Adult|*” +t at ID=”*|Toy|*” +o3 
> +s at CutFileAllCodes.cut +fPARIPA_AllCodes *.cha
>
> However, I have three problems I am hoping to solve.  I am repeating this 
> set of 4 freq analyses for 7 groups of codes/words (which will replace the 
> cut file referenced in +s).  So I will have at least 28 commands and 28 
> excel files I will need to combine into a single database for analysis.  I 
> would love to automate this data generation and combination as much as 
> possible.
>
>
> 1)  Every time I run these commands separately they work fine.  If I try 
> to run them as a batch by using the attached batch file (command "batch 
> BatchCommands.cex") I get an output that says "Using search file: 
> C:\talkbank\clan\work\CutFileAllCodes.cut" was found.  CAN'T FIND ANY DATA 
> TIERS IN ANY OF INPUT FILES PLEASE PROVIDE A SPECIFIC SPEAKER WITH +t 
> OPTION"  The batch commands results in a repeat of this same warning 4 
> times and produces nothing.  The batch file is in the "work" directory as 
> well as the "CutFileAllCodes.cut" file and all of the chat files.  Not sure 
> why it does not seem to be able to find these.  It would be nice to set 
> these up in a file and just run them all at once.  Plus it will make the 
> stats more easily reproducible and better documented.
>
>
> 2)  Ultimately I would like one database with a single line for each 
> transcription and the counts/stats generated from each of my commands in 
> rows going across.  Is there a way to have CLAN automatically append data 
> and match the data into a single row for each file as each additional 
> command is run?  
>
>
> 3)  If there is nothing I can do for number 2, then it would be really 
> nice if I could solve another problem.  Since the iPad codes are only in 
> about 1/3 of the files, when the excel sheet is produced for the iPad 
> commands, it doesn't produce a row for every analyzed file.  Therefore, the 
> excel file is "shorter" than the rest and when I go to paste this data in 
> with the other data, I have to match the iPad data for each participant 
> with the data output from the rest of the files.  If I could have CLAN 
> output 0's/periods/blanks for the files that are analyzed I can write a *very 
> simple* copy/paste macro to combine the excel files.  Is there a way to 
> tell CLAN to produce output for every analyzed file even if the tier is 
> missing?
>
>
> Thanks for any help!
>
> Brian
>
>
>
> -- 
> You received this message because you are subscribed to the Google Groups 
> "chibolts" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to chibolts+u... at googlegroups.com <javascript:>.
> To post to this group, send email to chib... at googlegroups.com 
> <javascript:>.
> To view this discussion on the web visit 
> <https://groups.google.com/d/msgid/chibolts/f739fe4d-d3f2-4a52-9f7a-95225a4e3a68%40googlegroups.com?utm_medium=email&utm_source=footer>
> https://groups.google.com/d/msgid/chibolts/f739fe4d-d3f2-4a52-9f7a-95225a4e3a68%40googlegroups.com
> .
> For more options, visit https://groups.google.com/d/optout.
>
>
>

-- 
You received this message because you are subscribed to the Google Groups "chibolts" group.
To unsubscribe from this group and stop receiving emails from it, send an email to chibolts+unsubscribe at googlegroups.com.
To post to this group, send email to chibolts at googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/chibolts/86178599-a5c9-4b36-8ade-8186f8b87424%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/chibolts/attachments/20160301/6de09bf3/attachment.htm>