Where communities thrive

Join over 1.5M+ people
Join over 100K+ communities
Free without limits
Create your own community

Explore more communities

dmlc/xgboost

eXtreme Gradient Boosting (GBDT, GBRT or GBM) Library for large-scale and distributed machine learning, on single node, hadoop yarn and more.

People

Repo info

Activity

Chris Chow

@ckchow

At least in Java there is a float[][] constructor, and I think there's a numpy constructor in python as well. might be out of luck if you're using the command line version.

lesshaste

@lesshaste

hi... does anyone understand why xgboost is so slow if you have lots of classes? This code shows the problem https://bpaste.net/show/f7573b5a2fb9 RandomForestClassifier takes about 15 seconds

but xgboost never terminates at all for me

Lyndon White

@oxinabox

I am training a binary classifier.
In the problem I am working on,
I can generate more training data at will.
In that by running a simulation I can (determenistically) determine the correct label for any feature set
Each training case takes a bit to generate (say 0.5 seconds).
The main motivation for training a classifier is that evaluating via simulation takes too long.

Is there a specific way to task advantage of my capacity to generate more data, that I can do in xgboosting,
that I couldn't do with say a SVM?

Its almost an Active Learning problem

Lyndon White

@oxinabox

I'm not sure if there is anything beyond: "Generate more data, both for training and validation , until the validation error hits 0"

KOLANICH

@KOLANICH

Hi everyone! Could anyone explain what are the arguments of a custom loss function?

objective function

Jay Kim (Data Scientist)

@bravekjh

Hi everyone. I joined this room first time today, nice to meet you all

Asbjørn Nilsen Riseth

@anriseth

Is there a built-in way to run XGBoost with a weighted mean square loss function?
Something like

\sum_{i=1}^D w_i(y_i-\hat{y}_i)^2

binyata

@binyata

is there a general reason why xgboost predict returns only nan?

this is for python

xiaolangyuxin

@xiaolangyuxin

multithread

xgboost predict for multithread works bad

on windows xp，i found a lots of issues for xgboost，exspacially，

uaing

using lolibray

Peter M. Landwehr

@pmlandwehr

Anybody have a changelog for 0.7.post4?

AbdealiJK

@AbdealiJK

Check https://github.com/dmlc/xgboost/releases

Rydez

@Rydez

For XGBoost, when considering time series data, is it worth creating features which represent a change in other features? For example, say I have the feature "total_active_users". Would it make sense to have a feature "change_in_total_active_users"? Or, would that just be redundant?

Harshal

@geekyharshal

Hello people

Can someone suggest how to begj with xgboost ?

Harshal

@geekyharshal

begin*

Tommy Yang

@joyang1

I use xgboost4j-0.80.jar predictleaf always return 3 leafindex for one label? is this any error?

@all

@/all

Tommy Yang

@joyang1

have anyone can answer me?

Tommy Yang

@joyang1

:joy: :joy_cat: :joy:

Tommy Yang

@joyang1

:joy:

Tommy Yang

@joyang1

I used xgboost4j-0.80.jar, xgboost train parameter of round is 800 and train data is 2000000. When I use predictleaf to get leafIndex, the jvm crashed.

A fatal error has been detected by the Java Runtime Environment:

SIGSEGV (0xb) at pc=0x00007f42160bf902, pid=880, tid=0x00007f42175f2700

JRE version: Java(TM) SE Runtime Environment (8.0_171-b11) (build 1.8.0_171-b11)

Java VM: Java HotSpot(TM) 64-Bit Server VM (25.171-b11 mixed mode linux-amd64 compressed oops)

Problematic frame:

V [libjvm.so+0x6d6902] jni_SetFloatArrayRegion+0xc2

Core dump written. Default location: /data/suzhe/suzhe-1.0-SNAPSHOT/core or core.880

If you would like to submit a bug report, please visit:

http://bugreport.java.com/bugreport/crash.jsp

Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
j ml.dmlc.xgboost4j.java.XGBoostJNI.XGBoosterPredict(JJII[[F)I+0
j ml.dmlc.xgboost4j.java.Booster.predict(Lml/dmlc/xgboost4j/java/DMatrix;ZIZZ)[[F+45
j ml.dmlc.xgboost4j.java.Booster.predictLeaf(Lml/dmlc/xgboost4j/java/DMatrix;I)[[F+6
j com.jianshu.suzhe.LRTrainer.train()V+23
j com.jianshu.suzhe.LRTrainer.main([Ljava/lang/String;)V+30
v ~StubRoutines::call_stub

Stack: [0x00007f42174f2000,0x00007f42175f3000], sp=0x00007f42175f1590, free space=1021k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, C=native code)
V [libjvm.so+0x6d6902] jni_SetFloatArrayRegion+0xc2
C [libxgboost4j8098523902211486429.so+0x9001c] Java_ml_dmlc_xgboost4j_java_XGBoostJNI_XGBoosterPredict+0x5c

Can someone suggest how to solve this problem?

thx~~~

Adrian Nembach

@AtR1an

Hi, I am having a problem with xgboost4j on linux. More specifically, if I specify eta to be smaller than 1 e.g. 0.3 (which is the default), than the model doesn't seem to learn anything. It almost seems like eta is set to zero somewhere along the line. Does anyone else experience this problem or maybe knows a solution? Thanks in advance.

Lijo Varghese

@lijoev

Hi i am stuck in my work in submitting a spark job to hadoop yarn master in cluster mode
please find my environment setup below

i have a linux machine having 128 GB of RAM, 2TB Hard disk, 2x16 cores.
i have set up cloud era hadoop containers in a docker mount point having 50 GB(this mount point is almost full). i have one datanode, namenode and yarnmaster containers runnings.
i am submitting spark job from my host machine to run Rscript in cluster mode. R server and libraries are set up in datanode.
When i submit the spark job it remains in the accepted state for long time. please find the spark submit command i am using below
spark-submit --master yarn --name RechargeModel --deploy-mode cluster --executor-memory 3G --num-executors 4 rechargemodel.R

camspilly

@camspilly

Hi currently seeing

a segmentation fault when I try to import xgboost

Keith

@DrEhrfurchtgebietend

Are there developers here? Is there interest in adding conformal predictions to the library? The error on the predicted value is an often desired quantity. Here is a link to the paper about how it was done with Random Forrest. https://link.springer.com/article/10.1007/s10994-014-5453-0 This method should work with anything which has our of bag samples.