Comments (6)
Thanks for bring this issue up. You cannot use base_margin, since it is used for generating prediction.
Here is a solution which I think fits your need, see updated version of example
https://github.com/tqchen/xgboost/blob/master/R-package/demo/custom_objective.R
from xgboost.
For train and a hold data set is easy setting the attr to each DMatrix.
For CV methods is needed modify the xgb.cv.mknfold function for set the attr to each fold data sets (train and test). This is a bit less elegant but works.
Thanks.
from xgboost.
The advantage of doing things in this way is that you can put arbitrary attributes along with instance with no modification of the DMatrix so it is good for extending various ideas.
For the cv, I suppose we can change mknfold to automatically split all the attributes that are vector, like you mentioned, which will solves the problem for all vector type attributes?
from xgboost.
Sorry, I think the update should be slice in https://github.com/tqchen/xgboost/blob/master/R-package/R/slice.xgb.DMatrix.R
@hetong007 can you look into it?
from xgboost.
Yes, this would be very useful.
from xgboost.
Finally it is fixed in this commit.
Please test it with the following code:
require(xgboost)
data(agaricus.train, package='xgboost')
dtrain <- xgb.DMatrix(agaricus.train$data, label = agaricus.train$label)
attr(dtrain, 'test') <- getinfo(dtrain, 'label')
str(attributes(dtrain))
slice(dtrain,1:20)
from xgboost.
Related Issues (20)
- Horizontal Federated Learning with Secure Features RFC
- [bug] Python - Cuda error (without using Cuda) HOT 5
- Pandas 2.2: Index.format is deprecated
- ArrayInterface handler for cuDF DataFrame cannot yet handle Boolean columns HOT 1
- src/metric/auc.cc:322: Check failed: auc <= local_area HOT 1
- XGBoost4j-spark CrossValidation train FAILED on multi-GPU environment: : Multiple processes running on same CUDA device is not supported! HOT 1
- [jvm-packages] Scaladoc is not working in latest XGBoost
- [CI] Tracker for improving build and CI/CD infrastructure
- [CI] Set up a nightly pipeline to test with dev versions of RAPIDS
- xgboost predict takes a long time
- xgboost4j_2.12:1.7.6 's (ml/dmlc/xgboost4j/java/XGBoostJNI.XGBoosterPredict) much slower than 0.90 in some model HOT 6
- NumPy 2.0 support HOT 2
- Tutorial on c-api distributed training of xgboost HOT 2
- Python 3.12 `xgboost.core.XGBoostError: Invalid Parameter format for nthread expect int but value='-1'` when `DMatrix` used with `import googlecloudprofiler`. HOT 6
- [CI] Retire Mac Mini worker in BuildKite
- [RFC] New logo for XGBoost HOT 1
- c-ares and BoringSSL version in xgboost 2.0.3 HOT 6
- Federated horizontal result does not align with basic training without federation
- monotone_constraints not working with xgb.regressor (python)
- xgboost 2.0.1 is breaking on rootless docker. HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from xgboost.