h2o.infogram throws error in R

[ditopcu]

Hi,

I wrote detailes as follows:

Many thanks

H2O version, Operating System and Environment

R is connected to the H2O cluster: H2O cluster uptime: 3 seconds 363 milliseconds H2O data parsing timezone: UTC H2O cluster version: 3.44.0.3 H2O cluster version age: 18 days H2O cluster total nodes: 1 H2O cluster total memory: 3.95 GB H2O cluster total cores: 12 H2O cluster allowed cores: 12 H2O cluster healthy: TRUE H2O Connection ip: localhost H2O Connection port: 54321 H2O Connection proxy: NA H2O Internal Security: FALSE R Version: R version 4.3.2 (2023-10-31 ucrt)

Actual behavior With the given example in the documentation, I got the following error:

h2o.infogram throws error :

Error in .h2o.doSafeREST(h2oRestApiVersion = h2oRestApiVersion, urlSuffix = page, :

ERROR MESSAGE:

Algorithm 'infogram' is not registered. Available algos: [ınfogram,targetencoder,deeplearning,glm,glrm,kmeans,naivebayes,pca,svd,drf,gbm,ısolationforest,extendedısolationforest,aggregator,word2vec,stackedensemble,coxph,generic,gam,anovaglm,psvm,rulefit,upliftdrf,modelselection,ısotonicregression,dt,adaboost]

Steps to reproduce

 
library(h2o)

h2o.init()


f <- "https://erin-data.s3.amazonaws.com/admissible/data/taiwan_credit_card_uci.csv"

col_types <- list(by.col.name = c("SEX", "MARRIAGE", "default_payment_next_month"),
                  types = c("factor", "factor", "factor"))

df <- h2o.importFile(path = f, col.types = col_types)


splits <- h2o.splitFrame(df, seed = 1)
train <- splits[[1]]
test <- splits[[2]]


y <- "default_payment_next_month"
x <- setdiff(names(train), y)


pcols <- c("SEX", "MARRIAGE", "AGE")


ig <- h2o.infogram(y = y, x = x, training_frame = train, protected_columns = pcols)
plot(ig)

Upload logs

ERROR: Unexpected HTTP Status code: 500 Server Error (url = http://localhost:54321/3/ModelBuilders/infogram)

java.lang.IllegalStateException [1] "java.lang.IllegalStateException: Algorithm 'infogram' is not registered. Available algos: [ınfogram,targetencoder,deeplearning,glm,glrm,kmeans,naivebayes,pca,svd,drf,gbm,ısolationforest,extendedısolationforest,aggregator,word2vec,stackedensemble,coxph,generic,gam,anovaglm,psvm,rulefit,upliftdrf,modelselection,ısotonicregression,dt,adaboost]" [2] " hex.ModelBuilder.lambda$make$1(ModelBuilder.java:168)"
[3] " java.base/java.util.Optional.orElseThrow(Optional.java:403)"
[4] " hex.ModelBuilder.make(ModelBuilder.java:163)"
[5] " water.api.ModelBuildersHandler.fetch(ModelBuildersHandler.java:35)"
[6] " java.base/jdk.internal.reflect.DirectMethodHandleAccessor.invoke(DirectMethodHandleAccessor.java:104)"
[7] " java.base/java.lang.reflect.Method.invoke(Method.java:578)"
[8] " water.api.Handler.handle(Handler.java:60)"
[9] " water.api.RequestServer.serve(RequestServer.java:472)"
[10] " water.api.RequestServer.doGeneric(RequestServer.java:303)"
[11] " water.api.RequestServer.doGet(RequestServer.java:225)"
[12] " javax.servlet.http.HttpServlet.service(HttpServlet.java:687)"
[13] " javax.servlet.http.HttpServlet.service(HttpServlet.java:790)"
[14] " org.eclipse.jetty.servlet.ServletHolder.handle(ServletHolder.java:799)"
[15] " org.eclipse.jetty.servlet.ServletHandler.doHandle(ServletHandler.java:554)"
[16] " org.eclipse.jetty.server.handler.ScopedHandler.nextHandle(ScopedHandler.java:233)"
[17] " org.eclipse.jetty.server.handler.ContextHandler.doHandle(ContextHandler.java:1440)"
[18] " org.eclipse.jetty.server.handler.ScopedHandler.nextScope(ScopedHandler.java:188)"
[19] " org.eclipse.jetty.servlet.ServletHandler.doScope(ServletHandler.java:505)"
[20] " org.eclipse.jetty.server.handler.ScopedHandler.nextScope(ScopedHandler.java:186)"
[21] " org.eclipse.jetty.server.handler.ContextHandler.doScope(ContextHandler.java:1355)"
[22] " org.eclipse.jetty.server.handler.ScopedHandler.handle(ScopedHandler.java:141)"
[23] " org.eclipse.jetty.server.handler.HandlerCollection.handle(HandlerCollection.java:146)"
[24] " org.eclipse.jetty.server.handler.HandlerWrapper.handle(HandlerWrapper.java:127)"
[25] " water.webserver.jetty9.Jetty9ServerAdapter$LoginHandler.handle(Jetty9ServerAdapter.java:130)"
[26] " org.eclipse.jetty.server.handler.HandlerCollection.handle(HandlerCollection.java:146)"
[27] " org.eclipse.jetty.server.handler.HandlerWrapper.handle(HandlerWrapper.java:127)"
[28] " org.eclipse.jetty.server.Server.handle(Server.java:516)"
[29] " org.eclipse.jetty.server.HttpChannel.lambda$handle$1(HttpChannel.java:487)"
[30] " org.eclipse.jetty.server.HttpChannel.dispatch(HttpChannel.java:732)"
[31] " org.eclipse.jetty.server.HttpChannel.handle(HttpChannel.java:479)"
[32] " org.eclipse.jetty.server.HttpConnection.onFillable(HttpConnection.java:277)"
[33] " org.eclipse.jetty.io.AbstractConnection$ReadCallback.succeeded(AbstractConnection.java:311)"
[34] " org.eclipse.jetty.io.FillInterest.fillable(FillInterest.java:105)"
[35] " org.eclipse.jetty.io.ChannelEndPoint$1.run(ChannelEndPoint.java:104)"
[36] " org.eclipse.jetty.util.thread.strategy.EatWhatYouKill.runTask(EatWhatYouKill.java:338)"
[37] " org.eclipse.jetty.util.thread.strategy.EatWhatYouKill.doProduce(EatWhatYouKill.java:315)"
[38] " org.eclipse.jetty.util.thread.strategy.EatWhatYouKill.tryProduce(EatWhatYouKill.java:173)"
[39] " org.eclipse.jetty.util.thread.strategy.EatWhatYouKill.run(EatWhatYouKill.java:131)"
[40] " org.eclipse.jetty.util.thread.ReservedThreadExecutor$ReservedThread.run(ReservedThreadExecutor.java:409)"
[41] " org.eclipse.jetty.util.thread.QueuedThreadPool.runJob(QueuedThreadPool.java:883)"
[42] " org.eclipse.jetty.util.thread.QueuedThreadPool$Runner.run(QueuedThreadPool.java:1034)"
[43] " java.base/java.lang.Thread.run(Thread.java:1623)"

Error in .h2o.doSafeREST(h2oRestApiVersion = h2oRestApiVersion, urlSuffix = page, :

ERROR MESSAGE:

Algorithm 'infogram' is not registered. Available algos: [ınfogram,targetencoder,deeplearning,glm,glrm,kmeans,naivebayes,pca,svd,drf,gbm,ısolationforest,extendedısolationforest,aggregator,word2vec,stackedensemble,coxph,generic,gam,anovaglm,psvm,rulefit,upliftdrf,modelselection,ısotonicregression,dt,adaboost]

[tomasfryda]

Hello @ditopcu, thank you for reporting this issue. I think it's the "Turkish i issue". The workaround is to set different locale. You can do this when starting the h2o like:

h2o.init(jvm_custom_args = "-Duser.language=en -Duser.country=US")

---

For my h2o collegues:

I can reproduce the issue by using Turkish locale.

h2o.init(jvm_custom_args = "-Duser.language=tr -Duser.country=TR")

The issue is likely caused by using toLowerCase and similar functions without fixed locale (IIRC in Turkish I -> ı and İ -> i and since we have Infogram we convert it to ınfogram etc).

[ditopcu]

Hi @tomasfryda,

Many thanks for the fast reply and recommendation. It worked like a charm. I used For clarification, I have exactly used:

h2o.init(jvm_custom_args = "-Duser.language=tr -Duser.country=US")

Cheers Deniz