Problem with chip command

[Zhuo Chen]

Dear ChiBolts,

I was using chip command in CLAN. The check command said the files are good, but when I ran the chip command, it complained that the %mor tier does not have a one-to-one correspondence with its speaker tier. Could you help me with this? 

In the attachment you can find an example of this error. 

I found that if I get rid of the + sign in the %mor tier, chip command will work on this particular utterance. Otherwise, I will get the error message. 

Thank you very much!

[Leonid Spektor]

Zhuo,

Could you please email to me directly the data file that causes this error? From the error message that you've attached to your email I don’t see any reason this error message. CLAN does verify the one-to-one correspondence of %mor tier and speaker tier. CHECK should have complained about it too. Having your data file will help a lot to figure out what the problem may be and if it is a bug in CLAN it will help me fix it.

Leonid.

--
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...@googlegroups.com.
To post to this group, send email to chib...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/chibolts/cd08eb18-4ec3-4372-b1d3-caa623665a3b%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
<chip.tiff>

[Zhuo Chen]

Dear Leonid,

Thank you for your prompt reply!

I was running chip command on the Valian Corpus and actually chip command complained in the very first file and it would not continue running after the first error. 

Attached is the first file from Valian Corpus. 

The first error happened in line 88. 

Thank you very much!

[Brian MacWhinney]

Dear Zhou, 

     You are using an old version of the Valian file in question.  Please get a new one from the web and see if it still has this problem.  I am not able to test out your problem in detail, because you didn’t include your actual CHIP command.

—Brian MacWhinney

--
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...@googlegroups.com.
To post to this group, send email to chib...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/chibolts/CALrxph1nJT49yHuMAatAqD9JjR6R9wUBRwJaQAZJ7qu5yZkMfw%40mail.gmail.com.

For more options, visit https://groups.google.com/groups/opt_out.

<01a.cha>

[Leonid Spektor]

Zhou,

There was a bug in CHIP that couldn't handle some utterance delimiters like "+...". I have fixed this bug. But, you also have an old version of Valian Corpus data files. It is important to realize the todays CLAN is linked very closely to the MOR grammar. This means that whenever you get a new CLAN you need to either get the latest version of the data from our servers or you need to get the latest version of MOR grammar and run MOR and POST yourself on your data files. New CLAN is on childes web site.

Leonid.

To view this discussion on the web visit https://groups.google.com/d/msgid/chibolts/95412AAA-372D-4594-B2D2-0C5A17E7A68F%40cmu.edu.

[Zhuo Chen]

Dear Leonid, 

The chip can now handle delimiter "+...". However, I found that if there is a note in square bracket in the speaker tier, CLAN will give me the same error message "There is not a one-to-one correspondence %mor and its speaker tier". I attached the file I was working on and the output file. Could you have a look at it? Thank you very much！

Zhuo

[Leonid Spektor]

Zhuo,

This was another bug that got introduced when I have fixed the +… bug. Hopefully this is the last problem with CHIP. It is a very complicated program. New CLAN is on childes web site.

Leonid.

To view this discussion on the web visit https://groups.google.com/d/msgid/chibolts/76b5eb25-aee1-4128-95aa-42e297bd0271%40googlegroups.com.

For more options, visit https://groups.google.com/groups/opt_out.

<01b.cha><01b.notmit.cex>

[Zhuo Chen]

Dear Leonid,

I have a last problem with CHIP command. When there is "0" in the speaker tier, the same error message popped up. As long as I get rid of the "0", CHIP works perfectly. Actually, this problem would not have any effect on the calculation of the repetition ratio of the children, but I just want to let you know so that CHIP command will not have any more problem. I attached the file I was working on and the CLAN output file. Thank you!

Zhuo

[Leonid Spektor]

Zhuo,

I can only apologize for not catching all those error myself. I really appreciate your help in letting us fix CHIP. This is the last command that we updated to the rest of CLAN's standard and it turned out to present the most challenges. I have fixed the “0” word error. Please get new CLAN from childes web site.

Leonid.

To view this discussion on the web visit https://groups.google.com/d/msgid/chibolts/72b63b4a-8082-4924-ae34-e00acb6b31d4%40googlegroups.com.

For more options, visit https://groups.google.com/groups/opt_out.

<04c.cha><04c.notimit.cex>

[Zhuo Chen]

Dear Leonid,

I really appreciate your help with the CHIP command! It runs perfectly now! I have another question to ask. Is there a way to send the output of CHIP command to an output file in CHAT format? I read the manual and d1 option does not do so because it only outputs the summary statistics. 

Thank you very much!

Zhuo

[Leonid Spektor]

Zhuo,

The +d option does what I believe you want. If you just type command name in Commands window and press return key you will get listing of all option available and their short description.

Leonid.

To view this discussion on the web visit https://groups.google.com/d/msgid/chibolts/aa71e117-490b-4556-9f08-0d583bd3b0dd%40googlegroups.com.

[Zhuo Chen]

Hi Leonid,

I actually tried d. It only outputs the coding tiers. CHECK command will tell me some errors in the format. Do I have to manually correct those errors? I just want to make sure whether there is an option that can give me output files in perfect CHAT format. 

Zhuo

[Leonid Spektor]

Zhuo,

CHIP adds "%chi:" dependent tier, among other tiers depending on use of +n option, to the output. This dependent tier is not a legal CHAT tier, because it is not specified in depfile.cut. You can add "+t@" option to command line to instruct CHIP to add all header tiers to the output, which will pass most of CHECK tests. If you use a text editor, like CLAN for example, and replace all "%chi:" tier codes with "%xchi:" then CHECK will not find anymore errors.. 

Leonid.

To view this discussion on the web visit https://groups.google.com/d/msgid/chibolts/82ac5ee7-8320-4632-8b61-2195d2fa86bd%40googlegroups.com.

[Zhuo Chen]

It works! Check command does not complain now. Thank you very much!

Zhuo