Makoto Yui created HIVEMALL-287:
-----------------------------------
Summary: disable xgboost early stopping when training/test contain
only 1 class
Key: HIVEMALL-287
URL: https://issues.apache.org/jira/browse/HIVEMALL-287
Project: Hivemall
Issue Type: Bug
Affects Versions: 0.6.0
Reporter: Makoto Yui
Assignee: Makoto Yui
Fix For: 0.6.2
{code:java}
WITH data as (
select array('1:4.9','2:3.0','3:1.4','4:0.2') as features, 0 as label
union all
select array('1:5.4','2:3.7','3:1.5','4:0.2') as features, 1 as label
union all
select array('1:5.1','2:3.5','3:1.4','4:0.3') as features, 0 as label
)
select
train_xgboost(features, label, '-objective binary:logistic -num_round 10
-num_early_stopping_rounds 3')
from
data;{code}
Diagnostic Messages for this Task:
java.lang.RuntimeException: Hive Runtime Error while closing operators
at
org.apache.hadoop.hive.ql.exec.mr.ExecMapper.close(ExecMapper.java:207)
at org.apache.hadoop.mapred.MapRunner.run(MapRunner.java:61)
at org.apache.hadoop.mapred.MapTask.runOldMapper(MapTask.java:453)
at org.apache.hadoop.mapred.MapTask.run(MapTask.java:343)
at
org.apache.hadoop.mapred.LocalContainerLauncher$EventHandler.runSubtask(LocalContainerLauncher.java:380)
at
org.apache.hadoop.mapred.LocalContainerLauncher$EventHandler.runTask(LocalContainerLauncher.java:301)
at
org.apache.hadoop.mapred.LocalContainerLauncher$EventHandler.access$200(LocalContainerLauncher.java:187)
at
org.apache.hadoop.mapred.LocalContainerLauncher$EventHandler$1.run(LocalContainerLauncher.java:230)
at
java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
at java.util.concurrent.FutureTask.run(FutureTask.java:266)
at
java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:748)
Caused by: org.apache.hadoop.hive.ql.metadata.HiveException:
ml.dmlc.xgboost4j.java.XGBoostError: [10:17:26]
/home/travis/build/myui/build-xgboost-jvm/xgboost/src/metric/elementwise_metric.cu:325:
Check failed: info.labels_.Size() != 0U (0 vs. 0) : label set cannot be empty
Stack trace:
[bt] (0)
/mnt2/hadoop/yarn/cache/yarn/nm-local-dir/usercache/1/appcache/application_1575881086268_0296/container_1575881086268_0296_01_000001/tmp/libxgboost4j4823645727987632626.so(+0x15ac57)
[0x7f43d7e49c57]
[bt] (1)
/mnt2/hadoop/yarn/cache/yarn/nm-local-dir/usercache/1/appcache/application_1575881086268_0296/container_1575881086268_0296_01_000001/tmp/libxgboost4j4823645727987632626.so(+0x1eaf46)
[0x7f43d7ed9f46]
[bt] (2)
/mnt2/hadoop/yarn/cache/yarn/nm-local-dir/usercache/1/appcache/application_1575881086268_0296/container_1575881086268_0296_01_000001/tmp/libxgboost4j4823645727987632626.so(+0x207ccd)
[0x7f43d7ef6ccd]
[bt] (3)
/mnt2/hadoop/yarn/cache/yarn/nm-local-dir/usercache/1/appcache/application_1575881086268_0296/container_1575881086268_0296_01_000001/tmp/libxgboost4j4823645727987632626.so(+0x1f431a)
[0x7f43d7ee331a]
[bt] (4)
/mnt2/hadoop/yarn/cache/yarn/nm-local-dir/usercache/1/appcache/application_1575881086268_0296/container_1575881086268_0296_01_000001/tmp/libxgboost4j4823645727987632626.so(Java_ml_dmlc_xgboost4j_java_XGBoostJNI_XGBoosterEvalOneIter+0x2be)
[0x7f43d7e47fa1]
[bt] (5) [0x7f43ed017774] at
hivemall.xgboost.XGBoostTrainUDTF.close(XGBoostTrainUDTF.java:551)
at
org.apache.hadoop.hive.ql.exec.UDTFOperator.closeOp(UDTFOperator.java:152)
at org.apache.hadoop.hive.ql.exec.Operator.close(Operator.java:697)
at org.apache.hadoop.hive.ql.exec.Operator.close(Operator.java:711)
at org.apache.hadoop.hive.ql.exec.Operator.close(Operator.java:711)
at org.apache.hadoop.hive.ql.exec.Operator.close(Operator.java:711)
at org.apache.hadoop.hive.ql.exec.Operator.close(Operator.java:711)
at org.apache.hadoop.hive.ql.exec.Operator.close(Operator.java:711)
at
org.apache.hadoop.hive.ql.exec.mr.ExecMapper.close(ExecMapper.java:189)
... 12 more
Caused by: ml.dmlc.xgboost4j.java.XGBoostError: [10:17:26]
/home/travis/build/myui/build-xgboost-jvm/xgboost/src/metric/elementwise_metric.cu:325:
Check failed: info.labels_.Size() != 0U (0 vs. 0) : label set cannot be empty
Stack trace:
[bt] (0)
/mnt2/hadoop/yarn/cache/yarn/nm-local-dir/usercache/1/appcache/application_1575881086268_0296/container_1575881086268_0296_01_000001/tmp/libxgboost4j4823645727987632626.so(+0x15ac57)
[0x7f43d7e49c57]
[bt] (1)
/mnt2/hadoop/yarn/cache/yarn/nm-local-dir/usercache/1/appcache/application_1575881086268_0296/container_1575881086268_0296_01_000001/tmp/libxgboost4j4823645727987632626.so(+0x1eaf46)
[0x7f43d7ed9f46]
[bt] (2)
/mnt2/hadoop/yarn/cache/yarn/nm-local-dir/usercache/1/appcache/application_1575881086268_0296/container_1575881086268_0296_01_000001/tmp/libxgboost4j4823645727987632626.so(+0x207ccd)
[0x7f43d7ef6ccd]
[bt] (3)
/mnt2/hadoop/yarn/cache/yarn/nm-local-dir/usercache/1/appcache/application_1575881086268_0296/container_1575881086268_0296_01_000001/tmp/libxgboost4j4823645727987632626.so(+0x1f431a)
[0x7f43d7ee331a]
[bt] (4)
/mnt2/hadoop/yarn/cache/yarn/nm-local-dir/usercache/1/appcache/application_1575881086268_0296/container_1575881086268_0296_01_000001/tmp/libxgboost4j4823645727987632626.so(Java_ml_dmlc_xgboost4j_java_XGBoostJNI_XGBoosterEvalOneIter+0x2be)
[0x7f43d7e47fa1]
[bt] (5) [0x7f43ed017774] at
ml.dmlc.xgboost4j.java.XGBoostJNI.checkCall(XGBoostJNI.java:48)
at ml.dmlc.xgboost4j.java.Booster.evalSet(Booster.java:226)
at ml.dmlc.xgboost4j.java.Booster.evalSet(Booster.java:243)
at hivemall.xgboost.XGBoostTrainUDTF.train(XGBoostTrainUDTF.java:591)
at hivemall.xgboost.XGBoostTrainUDTF.close(XGBoostTrainUDTF.java:534)
... 20 more
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
