gmane.comp.apache.uima.general

Testing UIMA

Burt Buchman — 2012-05-16T14:20:02

Hi all:

We are interested in testing UIMA and wondering if there is any site
available for that? Or would we have to install it to then test it out?
Would appreciate any insight. Thanks!

 

Burt Buchman

Solution Architect - Editorial

EBSCO Publishing

Phone: 978-356-6500, ext 2346

bbuchman-TsyAZbWhrljby3iVrkZq2A< at >public.gmane.org

Re: AW: uimaFIT 1.3.1 - AnalysisEngineFactory.setConfigurationParameters deprecated

Richard Eckart de Castilho — 2012-05-16T06:35:57

Hello Armin,


Thanks for pointing that out. Looks like there is a typo in the JavaDoc comment "< at >Deprecated" vs "< at >deprecated". Shall be fixed.

Best,

AW: uimaFIT 1.3.1 - AnalysisEngineFactory.setConfigurationParameters deprecated

2012-05-16T05:45:52

Hello Richard,

Thank you. I missed it in the attached JavaDoc because I've only read http://uimafit.googlecode.com/svn/tags/uimafit-parent-1.3.1/apidocs/index.html. It's not in there.

Regards,

Armin

-----Ursprüngliche Nachricht-----
Von: Richard Eckart de Castilho [mailto:eckart-S7Fx768l4CqSHq+C5vT0LZQlNPQFfrerqZSUQi4AVrg< at >public.gmane.org] 
Gesendet: Dienstag, 15. Mai 2012 21:06
An: <user-HHKbSdsD6WQPKjDvHGQMeg< at >public.gmane.org>
Betreff: Re: uimaFIT 1.3.1 - AnalysisEngineFactory.setConfigurationParameters deprecated

Hello Armin,


The method has been moved to ResourceCreationSpecifierFactory since it is not specific to analysis engines. I can see a corresponding JavaDoc comment in the sources. If you use Maven, it should automatically deliver you the sources/JavaDoc and provide such information. If you do not use Maven, you can download the JavaDoc JAR from Maven Central and attach it manually to the uimaFIT JAR in your Eclipse to get access to the JavaDoc.

Cheers,

Re: uimaFIT 1.3.1 - AnalysisEngineFactory.setConfigurationParameters deprecated

Richard Eckart de Castilho — 2012-05-15T19:06:10

Hello Armin,


The method has been moved to ResourceCreationSpecifierFactory since it is not specific to analysis engines. I can see a corresponding JavaDoc comment in the sources. If you use Maven, it should automatically deliver you the sources/JavaDoc and provide such information. If you do not use Maven, you can download the JavaDoc JAR from Maven Central and attach it manually to the uimaFIT JAR in your Eclipse to get access to the JavaDoc.

Cheers,

UIMA Addons components

Arash Hossein Abady — 2012-05-14T05:07:25

Hi,

How i can add-on components (annotators and consumes) in UIMA?

list of components:
http://uima.apache.org/sandbox.html#UIMA%20Addons%20components

I run the UIMA in eclipse.

Thanks

uimaFIT 1.3.1 - AnalysisEngineFactory.setConfigurationParameters deprecated

2012-05-15T11:04:10

Hi,

Since version 1.3.1 of uimaFIT
org.uimafit.factory.AnalysisEngineFactory.setConfigurationParameters is
marked as deprecated. What should I use instead?

Regards,

Armin

Re: Maven UIMA and import by name

Erik Fäßler — 2012-05-11T18:21:32

Hey you all,

thanks a lot for your answers. I had tried to add the desc/ directory as a maven resource. However, the UIMA component editor still wouldn't show me my descriptors to choose from. I have to remove the "exclude: **" entry in the build path first. Then, the whole directory is copied to the target/classes directory. I didn't like this approach too much because now I have each descriptor doubled. But I guess I can exclude the duplicates from any builds by configuring the appropriate build / package plugin.
I think I will also change it from being a resource to being a source folder. This way I don't get even a third copy of all descriptors in my target/resources directory.

Thanks again and a nice weekend to you!

Best,

Erik

Am 11.05.2012 um 18:05 schrieb Thomas Ginter:

Re: Maven UIMA and import by name

Thomas Ginter — 2012-05-11T16:05:49

We use maven for our UIMA-AS projects.  Here is the build section from our standard POM entries:

<build>
    <resources>
      <resource>
        <directory>src/main/desc/</directory>
      </resource>
      <resource>
        <directory>src/main/resources/</directory>
      </resource>
    </resources>
    <pluginManagement>
        <plugins>
         <plugin>
          <groupId>org.apache.maven.plugins</groupId>
          <artifactId>maven-compiler-plugin</artifactId>
          <configuration>
            <source>1.6</source>
            <target>1.6</target>
          </configuration>
        </plugin>
      </plugins>
    </pluginManagement>
   </build>

This adds the desc and resources directories as source directories that allow you to resolve the import of descriptors by name.

Thanks,

Thomas Ginter
801-448-7676
thomas.ginter-wbocuHtxKic< at >public.gmane.org




On May 11, 2012, at 9:56 AM, Erik Fäßler wrote:

Re: Maven UIMA and import by name

Tommaso Teofili — 2012-05-11T16:04:44

Hi Erik,
I mostly edit XMLs by hand (also because I use IntelliJ IDEA) but I think
you may tell Maven that desc is a resource folder:


<build>
    ...
    <resources>
      <resource>
        <directory>src/main/resources</directory>

        <directory>desc</directory>

</resource> ... </build>


HTH,
Tommaso


2012/5/11 Erik Fäßler <erik.faessler-S+RBGtoBSY4b1SvskN2V4Q< at >public.gmane.org>

Re: Maven UIMA and import by name

Richard Eckart de Castilho — 2012-05-11T16:04:18

Hi Erik.

you can configure Maven with an extra "resources" folder.

http://maven.apache.org/pom.html#Resources

We usually keep our descriptors under "src/main/resources" to have them on the classpath. But we do not care about being PEAR compatible. So adding "desc" as a resources folder in your POM is probably the right thing for you.

Maven UIMA and import by name

Erik Fäßler — 2012-05-11T15:56:09

Hello all,

I have a question on how you deal with a specific use case and would like to know if you have some suggestions for me.

I use Maven for all my Java projects and so I do for my UIMA related projects. Now I have a quite large pipeline with lots of descriptors. They reside in (or subdirectories of) the 'desc' directory of the 'UIMA nature' structure.
Currently I am about to pack these single-AE descriptors into aggregates. For importing all single-AEs into the AAE descriptor, I would like to use "import by name". However, the 'desc' directory is not a library for eclipse and thus, the AAE descriptor editor doesn't list the descriptors residing in this directory - I can't add them (and when I edit the XML, I get error messages about descriptors not found).

I would like to just add the 'desc' directory to the build path as an "class folder" (not a source folder, this won't work), i.e. as a library. When I do this manually, Maven would overwrite it the next time it updates my project configuration.

H

Re: programmatically setting parameters on CpeCasProcessor objects

Andrew Fast — 2012-05-10T19:59:51

I've had the same problem.  Here is what has worked for me:

1) Create a blank ConfigurationParameterSettings object using the 
CPE factory:

CasProcessorConfigurationParameterSettings settings =             
   CpeDescriptorFactory.produceCasProcessorConfigurationParameterSettings();

2) Set your settings, e.g.,:

settings.setParameterValue("PARAM", value);

3) Then, set the settings of your processor

processor.setConfigurationParameterSettings(settings)

4) Finally, add the processor to your CPE

cpe.addCasProcessor(processor);

The CPE xml only ever contains a path to each component descriptor and 
doesn't actually know about the settings until runtime.

hth,

Andy

How to load XMI CAS from a DB inside a Reader AE?

Jörn Kottmann — 2012-05-08T17:03:31

Hi all,

I have a database which contains XMIs and would like to process
these with UIMA-AS.

My pipeline looks like this:
Database Reader AE
Result Writer AE

I made an AAE composed out of the two AEs, the first gets a CAS with an
ID encoded into its sofa. It then fetches the XMI from the DB, and calls
XmiCasDeserializer.deserialize to load it into the CAS.
After that it inserts an ID FeatureStructure into the CAS (via createFS 
and addFsToIndexes).

The second AE should just access the content of the CAS, but the ID FS added
by the first AE is missing. Why is it missing? (input/output 
capabilities are not specified)
The content which is in the XMI can be accessed.

Is it safe to use XmiCasDeserializer.deserialize? Or should it be done
in some other way?

Jörn

Re: obtaining xmi:id from tag.

Marshall Schor — 2012-05-07T13:51:05


On 5/3/2012 2:49 PM, James Moliere wrote:

I don't think this is usually done.  If some kind of unique id (for each feature 
structure) is required by the application, it of course could be done, but I 
think that requirement is not the usual case for users of UIMA.

In UIMA, it is much more common to have a feature structure hold a reference 
directly to another feature structure, rather than through a unique ID field.

-Marshall

Submission Deadline Extension

VHPC 12 — 2012-05-06T17:39:44

we apologize if you receive multiple copies of this CFP

===================================================================

CALL FOR PAPERS

7th Workshop on

Virtualization in High-Performance Cloud Computing

VHPC '12

as part of Euro-Par 2012, Rhodes Island, Greece

===================================================================

Date: August 28, 2012

Workshop URL: http://vhpc.org

SUBMISSION DEADLINE:

June 11, 2012 - Full paper submission (extended)


SCOPE:

Virtualization has become a common abstraction layer in modern
data centers, enabling resource owners to manage complex
infrastructure independently of their applications. Conjointly,
virtualization is becoming a driving technology for a manifold of
industry grade IT services. The cloud concept includes the notion
of a separation between resource owners and users, adding  services
such as hosted application frameworks and queueing. Utilizing the
same infrastructure, clouds carry significant potential for use in
high-performance scientific com

Re: obtaining xmi:id from tag.

James Moliere — 2012-05-03T18:49:43


I'm a bit new to UIMA. Do developers typically generate their own unique 
IDs andsimply add the ID attribute to each element in the file?

Re: obtaining xmi:id from tag.

Marshall Schor — 2012-05-03T18:09:30

The xmi:id is not a part of the CAS representation, it only exists when the file 
is persisted in the xmi format.

-Marshall

On 5/3/2012 1:05 PM, James Moliere wrote:

Re: how to get the original filename of the input document?

Marshall Schor — 2012-05-03T18:02:26


On 5/2/2012 6:30 PM, michelangelo wrote:

Because the string heap is nothing more that a pile (heap) of strings that 
arbitrary user-written annotators have added as values of slots in Feature 
Structures.  It's like the runtime memory.  UIMA has no knowledge of what the 
meaning of any string is in the heap.  It's completely up to what ever 
annotators are being run.  It's quite possible for a particular run of UIMA to 
run with annotators that don't use files at all -- maybe they use urls or 
internally generated data or get messages sent to them via a web interface, etc. 
etc.

-Marshall

obtaining xmi:id from tag.

James Moliere — 2012-05-03T17:05:32

Hello,
I'm having difficulty extracting the xmi:id attribute value from each of my 
elements in my xmi file.  Assuming there are xmi:id attribute values, how do I 
extract it's value from the method below

public void processCas(CAS aCAS) throws ResourceProcessException {
    ???
}

James

Advice for running UIMA AS as Windows Apache Procrun Service

Spico Florin — 2012-05-03T13:57:46

Hello!
   I would like to use the Apache Procrun project (
http://commons.apache.org/daemon/procrun.html) to run the UIMA AS pipeline
as Windows Service. :
I.A first solution is to Try is to modify the current UIMA_Service class
and to adapt it to the Apache Procrun requirements (adding start and stop
methods) and use this altered class as Java Class for the Windows Service.
     But, looking up to the current implementation, I've noticed that the
UIMA_Service is launched by org.apache.uima.bootstrap.UimaBootstrap class.
     I'm wondering if this launching part is a requirement of the UIMA AS
framework. Can I skip this step and launch directly the My_UIMA_Service
main class?


  Can you please advice me, how I can implement my use case according to
the UIMA AS specifications?

II. A second solution is trying to use somehow the  current starting script
deployAsyncService.cmd to launch it as windows service.But the following
issues raised:
   1. Is this solution feasible? The script starts the UIMA in command

Re: how to get the original filename of the input document?

michelangelo — 2012-05-02T22:30:04

thank you for elucidation. But in the CASImpl's inner private class
SharedViewData there is a stringheap that contains the filepath.
Why no access method?
thanks 
Michelangelo