Pages: [1]
  Print  
Author Topic: XVPrediction and LinearRegression  (Read 2256 times)
bfranke
Newbie
*
Posts: 3


« on: December 12, 2008, 12:41:44 PM »

Hello,

I'm having a problem with XVPrediction and the LinearRegression operator. What I want to achieve is to learn a linear regression model and then apply the model to some test data. Initially, I use cross-validation to evaluate my approach and as I am interested in the actual predictions I am using XVPrediction rather than XValidation. While I can see a new attribute "Prediction(...)" in the resulting data set, all values in this column are "unknown".

My process looks like this:

<?xml version="1.0" encoding="MacRoman"?>
<process version="4.3">

  <operator name="Root" class="Process" expanded="yes">
      <operator name="CSVExampleSource" class="CSVExampleSource">
          <parameter key="filename"   value="/Users/bfranke/HPCA/rapidminer/ARM.csv"/>
          <parameter key="id_name"   value="Filename"/>
          <parameter key="label_name"   value="CNT_cycles"/>
          <parameter key="trim_lines"   value="true"/>
      </operator>
      <operator name="FeatureNameFilter" class="FeatureNameFilter">
          <parameter key="skip_features_with_name"   value="CNT_alloc_OSMs|CNT_retired_OSMs"/>
      </operator>
      <operator name="Numerical2Real" class="Numerical2Real">
      </operator>
      <operator name="XVPrediction" class="XVPrediction" expanded="yes">
          <parameter key="leave_one_out"   value="true"/>
          <parameter key="sampling_type"   value="linear sampling"/>
          <operator name="LinearRegression" class="LinearRegression">
              <parameter key="feature_selection"   value="greedy"/>
              <parameter key="keep_example_set"   value="true"/>
          </operator>
          <operator name="ModelApplier" class="ModelApplier">
              <list key="application_parameters">
              </list>
              <parameter key="create_view"   value="true"/>
              <parameter key="keep_model"   value="true"/>
          </operator>
      </operator>
      <operator name="ResultWriter" class="ResultWriter">
          <parameter key="result_file"   value="/Users/bfranke/HPCA/rapidminer/results.res"/>
      </operator>
  </operator>

</process>


I have a data set with 293 examples and 45 attributes, of which one is the textual/nominal ID and another one the numerical label. All other attributes are also numerical, two of them I filter out using the FeatureNameFilter. I also make sure all regular attributes are converted to real values as some of the attributes are identified as int and others as real. This conversion using the Numerical2Real operator makes sure all attributes are represented using the same real type. Up to this point everything seems to be ok (I set a breakpoint and inspected the data). Hence, the problem seems to be related to XVPrediction and LinearRegression.

Checking the older posts to this forum I haven't found any known issues with either XVPrediction or LinearRegression (the same problem shows up with PolynomialRegression and also GPLearner), so I guess there's something wrong with my process. I've already "experimented" with explicit feature selection to select fewer attributes, but this didn't solve the problem. Any ideas?

Thanks!

Cheers,

   Bjoern
Logged
Sebastian Land
Administrator
Hero Member
*****
Posts: 2426


« Reply #1 on: December 15, 2008, 01:01:31 PM »

Hi Bjoern,
seems to me there is a bug in the XVPrediction. I will check that and keep you informed.

Greetings,
  Sebastian
Logged
Sebastian Land
Administrator
Hero Member
*****
Posts: 2426


« Reply #2 on: December 15, 2008, 03:13:32 PM »

Hi,
I have found bug, but you need to check out the latest developer version from cvs to get a bug free version.

Greetings,
  Sebastian
Logged
steffen
Sr. Member
****
Posts: 376



« Reply #3 on: December 15, 2008, 05:12:47 PM »

Hello

Do you still use the bug tracker? I just ask to avoid spamming Wink

regards,

Steffen
Logged

"I want to make computers do what I mean instead of what I say"
Read The Fantastic Manual
Sebastian Land
Administrator
Hero Member
*****
Posts: 2426


« Reply #4 on: December 15, 2008, 05:36:51 PM »

Hi Steffen,
I give my very best to pay an appropriate amount of attention to any way messages about bugs could occur. But sometimes its simply one way too much to keep all in mind and the emails had been disappeared between some SF spam...
Thanks for reminding  Smiley

Greetings,
  Sebastian
Logged
bfranke
Newbie
*
Posts: 3


« Reply #5 on: December 15, 2008, 10:46:54 PM »

Hello Sebastian,

I have found bug, but you need to check out the latest developer version from cvs to get a bug free version.

Many thanks for your prompt help! Now I've got another question relating to how to get access to this latest version. When I check out the code using anonymous CVS I get an older version 4.2 of RapidMiner. The developer CVS access via SSH does not seem to work for me (password not recognised, permission denied). Do I need to become an "official" developer to check out the latest version and how do I do this? Thanks.

Cheers,

  Björn
Logged
Sebastian Land
Administrator
Hero Member
*****
Posts: 2426


« Reply #6 on: December 16, 2008, 10:43:48 AM »

Hi,
simply switch to the developer branch called Zaniah.

Greetings,
  Sebastian
Logged
bfranke
Newbie
*
Posts: 3


« Reply #7 on: December 16, 2008, 05:08:46 PM »

Ok, thanks!

Cheers,

   Bjoern
Logged
Pages: [1]
  Print  
 
Jump to: