The latest mahout-696.path can work with mahout-trunk rev1104229, which is my
development revision, but when patched with 0.5 release, the following error
occur when run validateAdaptiveLogistic
Exception in thread "main" java.io.IOException: Can't create object
at
org.apache.mahout.classifier.sgd.PolymorphicWritable.read(PolymorphicWritable.java:45)
at
org.apache.mahout.classifier.sgd.OnlineLogisticRegression.readFields(OnlineLogisticRegression.java:162)
at
org.apache.mahout.classifier.sgd.CrossFoldLearner.readFields(CrossFoldLearner.java:317)
at
org.apache.mahout.classifier.sgd.AdaptiveLogisticRegression$Wrapper.readFields(AdaptiveLogisticRegression.java:435)
at
org.apache.mahout.classifier.sgd.PolymorphicWritable.read(PolymorphicWritable.java:51)
at org.apache.mahout.ep.State.readFields(State.java:297)
at
org.apache.mahout.classifier.sgd.PolymorphicWritable.read(PolymorphicWritable.java:51)
at
org.apache.mahout.ep.EvolutionaryProcess.readFields(EvolutionaryProcess.java:222)
at
org.apache.mahout.classifier.sgd.AdaptiveLogisticRegression.readFields(AdaptiveLogisticRegression.java:539)
at
org.apache.mahout.classifier.sgd.AdaptiveLogisticModelParameters.readFields(AdaptiveLogisticModelParameters.java:173)
at
org.apache.mahout.classifier.sgd.AdaptiveLogisticModelParameters.loadFromStream(AdaptiveLogisticModelParameters.java:179)
at
org.apache.mahout.classifier.sgd.AdaptiveLogisticModelParameters.loadFromFile(AdaptiveLogisticModelParameters.java:186)
at
org.apache.mahout.classifier.sgd.ValidateAdaptiveLogistic.main(ValidateAdaptiveLogistic.java:70)
Caused by: java.lang.InstantiationException:
org.apache.mahout.classifier.sgd.ElasticBandPrior
at java.lang.Class.newInstance0(Class.java:340)
at java.lang.Class.newInstance(Class.java:308)
at
org.apache.mahout.classifier.sgd.PolymorphicWritable.read(PolymorphicWritable.java:43)
... 12 more
> -----Original Message-----
> From: Lance Norskog [mailto:[email protected]]
> Sent: Thursday, June 30, 2011 11:26 AM
> To: [email protected]
> Subject: Re: [jira] [Updated] (MAHOUT-696) Command line program for
> AdaptiveLogiscticRegression
>
> Thanks for this project, Xiaobo! I look forward to playing with it.
>
> Lance
>
> On Wed, Jun 29, 2011 at 8:21 PM, XiaoboGu (JIRA) <[email protected]> wrote:
> >
> > [
> > https://issues.apache.org/jira/browse/MAHOUT-696?page=com.atlassian.jira.plugin
> .system.issuetabpanels:all-tabpanel ]
> >
> > XiaoboGu updated MAHOUT-696:
> > ----------------------------
> >
> > Attachment: MAHOUT-696.patch
> >
> > This version has passed the following testes:
> > mahout trainAdaptiveLogistic --input donut.csv --output d:\\model1 --target
> > color
> --categories 2 --predictors x y --types numeric --threads 8
> > mahout trainAdaptiveLogistic --input donut.csv --output d:\\model1 --target
> > color
> --categories 2 --predictors x y --types numeric --threads 8 --showperf
> > mahout trainAdaptiveLogistic --input donut.csv --output d:\\model1 --target
> > color
> --categories 2 --predictors x y --types numeric --threads 8 --passes 1000
> > mahout trainAdaptiveLogistic --input donut.csv --output d:\\model1 --target
> > color
> --categories 2 --predictors x y --types numeric --threads 8 --passes 1000
> --showperf
> > mahout trainAdaptiveLogistic --input donut.csv --output d:\\model1 --target
> > color
> --categories 2 --predictors x y --types numeric --threads 8 --passes 1000
> --showperf
> --features 100
> > mahout trainAdaptiveLogistic --input donut.csv --output d:\\model1 --target
> > color
> --categories 2 --predictors x y --types numeric --threads 8 --passes 1000
> --showperf
> --features 100 --skipperfnum 399
> >
> > mahout trainAdaptiveLogistic --input donut.csv --output d:\\model1 --target
> > color
> --categories 2 --predictors x y --types numeric --threads 8 --passes 1000
> --showperf
> --features 100 --skipperfnum 399 --prior L1
> > mahout trainAdaptiveLogistic --input donut.csv --output d:\\model1 --target
> > color
> --categories 2 --predictors x y --types numeric --threads 8 --passes 1000
> --showperf
> --features 100 --skipperfnum 399 --prior L2
> > mahout trainAdaptiveLogistic --input donut.csv --output d:\\model1 --target
> > color
> --categories 2 --predictors x y --types numeric --threads 8 --passes 1000
> --showperf
> --features 100 --skipperfnum 399 --prior up
> > mahout trainAdaptiveLogistic --input donut.csv --output d:\\model1 --target
> > color
> --categories 2 --predictors x y --types numeric --threads 8 --passes 1000
> --showperf
> --features 100 --skipperfnum 399 --prior tp
> > mahout trainAdaptiveLogistic --input donut.csv --output d:\\model1 --target
> > color
> --categories 2 --predictors x y --types numeric --threads 8 --passes 1000
> --showperf
> --features 100 --skipperfnum 399 --prior ebp
> > mahout trainAdaptiveLogistic --input donut.csv --output d:\\model1 --target
> > color
> --categories 2 --predictors x y --types numeric --threads 8 --passes 1000
> --showperf
> --features 100 --skipperfnum 399 --prior tp --prioroption 2
> > mahout trainAdaptiveLogistic --input donut.csv --output d:\\model1 --target
> > color
> --categories 2 --predictors x y --types numeric --threads 8 --passes 1000
> --showperf
> --features 100 --skipperfnum 399 --prior ebp --prioroption 2
> >
> > mahout trainAdaptiveLogistic --input donut.csv --output d:\\model1 --target
> > color
> --categories 2 --predictors x y --types numeric --threads 8 --passes 1000
> --showperf
> --features 100 --skipperfnum 399 --prior L1 --auc global
> > mahout trainAdaptiveLogistic --input donut.csv --output d:\\model1 --target
> > color
> --categories 2 --predictors x y --types numeric --threads 8 --passes 1000
> --showperf
> --features 100 --skipperfnum 399 --prior L1 --auc grouped
> >
> > mahout validateAdaptiveLogistic --input donut-test.csv --model d:\\model1
> > --auc
> --confusion --scores
> >
> > mahout runAdaptiveLogistic --input donut-test.csv --model d:\\model1
> > --output
> d:\\scores.txt --idcolumn c
> > mahout runAdaptiveLogistic --input donut-test.csv --model d:\\model1
> > --output
> d:\\scores1.txt --idcolumn c --maxscoreonly
> >
> >
> >
> >> Command line program for AdaptiveLogiscticRegression
> >> ----------------------------------------------------
> >>
> >> Key: MAHOUT-696
> >> URL: https://issues.apache.org/jira/browse/MAHOUT-696
> >> Project: Mahout
> >> Issue Type: Improvement
> >> Components: Classification
> >> Affects Versions: 0.5
> >> Reporter: XiaoboGu
> >> Assignee: Ted Dunning
> >> Fix For: 0.6
> >>
> >> Attachments: MAHOUT-696.patch, MAHOUT-696.patch, MAHOUT-696.patch,
> MAHOUT-696.patch, MAHOUT-696.patch, mahout-696-r1.patch, mahout-696-r2.patch,
> mahout-696-r3.patch, mahout-696-r4.patch, mahout-696-r5.patch
> >>
> >>
> >> Suggested by Ted, I'll try to write a command line program for
> AdaptiveLogicticRegression, but as I am not familir with the algorithm, I'll
> try to write a
> prototype for the program from a Java developer's perspactive, hope anyone
> else will help
> with the details of the algorithm.
> >
> > --
> > This message is automatically generated by JIRA.
> > For more information on JIRA, see: http://www.atlassian.com/software/jira
> >
> >
> >
>
>
>
> --
> Lance Norskog
> [email protected]
