Hendrik/xgboost
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

[Breaking] Require format to be specified in input URI. (#9077)

Browse Source 
Previously, we use `libsvm` as default when format is not specified. However, the dmlc
data parser is not particularly robust against errors, and the most common type of error
is undefined format.

Along with which, we will recommend users to use other data loader instead. We will
continue the maintenance of the parsers as it's currently used for many internal tests
including federated learning.
This commit is contained in:
Jiaming Yuan
2023-04-28 19:45:15 +08:00
committed by GitHub
parent e922004329
commit 1f9a57d17b
58 changed files with 327 additions and 268 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
jvm-packages/xgboost4j-example/src/main/java/ml/dmlc/xgboost4j/java/example/BasicWalkThrough.java
View File

@@ -1,5 +1,5 @@
/*
Copyright (c) 2014-2021 by Contributors
Copyright (c) 2014-2023 by Contributors
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
@@ -62,8 +62,8 @@ public class BasicWalkThrough {
public static void main(String[] args) throws IOException, XGBoostError {
// load file from text file, also binary buffer generated by xgboost4j
DMatrix trainMat = new DMatrix("../../demo/data/agaricus.txt.train");
DMatrix testMat = new DMatrix("../../demo/data/agaricus.txt.test");
DMatrix trainMat = new DMatrix("../../demo/data/agaricus.txt.train?format=libsvm");
DMatrix testMat = new DMatrix("../../demo/data/agaricus.txt.test?format=libsvm");
HashMap<String, Object> params = new HashMap<String, Object>();
params.put("eta", 1.0);
@@ -112,7 +112,8 @@ public class BasicWalkThrough {
System.out.println("start build dmatrix from csr sparse data ...");
//build dmatrix from CSR Sparse Matrix
DataLoader.CSRSparseData spData = DataLoader.loadSVMFile("../../demo/data/agaricus.txt.train");
DataLoader.CSRSparseData spData =
DataLoader.loadSVMFile("../../demo/data/agaricus.txt.train?format=libsvm");
DMatrix trainMat2 = new DMatrix(spData.rowHeaders, spData.colIndex, spData.data,
DMatrix.SparseType.CSR, 127);

4
jvm-packages/xgboost4j-example/src/main/java/ml/dmlc/xgboost4j/java/example/BoostFromPrediction.java
View File

@@ -32,8 +32,8 @@ public class BoostFromPrediction {
System.out.println("start running example to start from a initial prediction");
// load file from text file, also binary buffer generated by xgboost4j
DMatrix trainMat = new DMatrix("../../demo/data/agaricus.txt.train");
DMatrix testMat = new DMatrix("../../demo/data/agaricus.txt.test");
DMatrix trainMat = new DMatrix("../../demo/data/agaricus.txt.train?format=libsvm");
DMatrix testMat = new DMatrix("../../demo/data/agaricus.txt.test?format=libsvm");
//specify parameters
HashMap<String, Object> params = new HashMap<String, Object>();

2
jvm-packages/xgboost4j-example/src/main/java/ml/dmlc/xgboost4j/java/example/CrossValidation.java
View File

@@ -30,7 +30,7 @@ import ml.dmlc.xgboost4j.java.XGBoostError;
public class CrossValidation {
public static void main(String[] args) throws IOException, XGBoostError {
//load train mat
DMatrix trainMat = new DMatrix("../../demo/data/agaricus.txt.train");
DMatrix trainMat = new DMatrix("../../demo/data/agaricus.txt.train?format=libsvm");
//set params
HashMap<String, Object> params = new HashMap<String, Object>();

4
jvm-packages/xgboost4j-example/src/main/java/ml/dmlc/xgboost4j/java/example/CustomObjective.java
View File

@@ -139,9 +139,9 @@ public class CustomObjective {
public static void main(String[] args) throws XGBoostError {
//load train mat (svmlight format)
DMatrix trainMat = new DMatrix("../../demo/data/agaricus.txt.train");
DMatrix trainMat = new DMatrix("../../demo/data/agaricus.txt.train?format=libsvm");
//load valid mat (svmlight format)
DMatrix testMat = new DMatrix("../../demo/data/agaricus.txt.test");
DMatrix testMat = new DMatrix("../../demo/data/agaricus.txt.test?format=libsvm");
HashMap<String, Object> params = new HashMap<String, Object>();
params.put("eta", 1.0);

4
jvm-packages/xgboost4j-example/src/main/java/ml/dmlc/xgboost4j/java/example/EarlyStopping.java
View File

@@ -29,9 +29,9 @@ import ml.dmlc.xgboost4j.java.example.util.DataLoader;
public class EarlyStopping {
public static void main(String[] args) throws IOException, XGBoostError {
DataLoader.CSRSparseData trainCSR =
DataLoader.loadSVMFile("../../demo/data/agaricus.txt.train");
DataLoader.loadSVMFile("../../demo/data/agaricus.txt.train?format=libsvm");
DataLoader.CSRSparseData testCSR =
DataLoader.loadSVMFile("../../demo/data/agaricus.txt.test");
DataLoader.loadSVMFile("../../demo/data/agaricus.txt.test?format=libsvm");
Map<String, Object> paramMap = new HashMap<String, Object>() {
{

4
jvm-packages/xgboost4j-example/src/main/java/ml/dmlc/xgboost4j/java/example/ExternalMemory.java
View File

@@ -32,8 +32,8 @@ public class ExternalMemory {
//this is the only difference, add a # followed by a cache prefix name
//several cache file with the prefix will be generated
//currently only support convert from libsvm file
DMatrix trainMat = new DMatrix("../../demo/data/agaricus.txt.train#dtrain.cache");
DMatrix testMat = new DMatrix("../../demo/data/agaricus.txt.test#dtest.cache");
DMatrix trainMat = new DMatrix("../../demo/data/agaricus.txt.train?format=libsvm#dtrain.cache");
DMatrix testMat = new DMatrix("../../demo/data/agaricus.txt.test?format=libsvm#dtest.cache");
//specify parameters
HashMap<String, Object> params = new HashMap<String, Object>();

4
jvm-packages/xgboost4j-example/src/main/java/ml/dmlc/xgboost4j/java/example/GeneralizedLinearModel.java
View File

@@ -32,8 +32,8 @@ import ml.dmlc.xgboost4j.java.example.util.CustomEval;
public class GeneralizedLinearModel {
public static void main(String[] args) throws XGBoostError {
// load file from text file, also binary buffer generated by xgboost4j
DMatrix trainMat = new DMatrix("../../demo/data/agaricus.txt.train");
DMatrix testMat = new DMatrix("../../demo/data/agaricus.txt.test");
DMatrix trainMat = new DMatrix("../../demo/data/agaricus.txt.train?format=libsvm");
DMatrix testMat = new DMatrix("../../demo/data/agaricus.txt.test?format=libsvm");
//specify parameters
//change booster to gblinear, so that we are fitting a linear model

4
jvm-packages/xgboost4j-example/src/main/java/ml/dmlc/xgboost4j/java/example/PredictFirstNtree.java
View File

@@ -31,8 +31,8 @@ import ml.dmlc.xgboost4j.java.example.util.CustomEval;
public class PredictFirstNtree {
public static void main(String[] args) throws XGBoostError {
// load file from text file, also binary buffer generated by xgboost4j
DMatrix trainMat = new DMatrix("../../demo/data/agaricus.txt.train");
DMatrix testMat = new DMatrix("../../demo/data/agaricus.txt.test");
DMatrix trainMat = new DMatrix("../../demo/data/agaricus.txt.train?format=libsvm");
DMatrix testMat = new DMatrix("../../demo/data/agaricus.txt.test?format=libsvm");
//specify parameters
HashMap<String, Object> params = new HashMap<String, Object>();

4
jvm-packages/xgboost4j-example/src/main/java/ml/dmlc/xgboost4j/java/example/PredictLeafIndices.java
View File

@@ -31,8 +31,8 @@ import ml.dmlc.xgboost4j.java.XGBoostError;
public class PredictLeafIndices {
public static void main(String[] args) throws XGBoostError {
// load file from text file, also binary buffer generated by xgboost4j
DMatrix trainMat = new DMatrix("../../demo/data/agaricus.txt.train");
DMatrix testMat = new DMatrix("../../demo/data/agaricus.txt.test");
DMatrix trainMat = new DMatrix("../../demo/data/agaricus.txt.train?format=libsvm");
DMatrix testMat = new DMatrix("../../demo/data/agaricus.txt.test?format=libsvm");
//specify parameters
HashMap<String, Object> params = new HashMap<String, Object>();

8
jvm-packages/xgboost4j-example/src/main/scala/ml/dmlc/xgboost4j/scala/example/BasicWalkThrough.scala
View File

@@ -1,5 +1,5 @@
/*
Copyright (c) 2014 by Contributors
Copyright (c) 2014-2023 by Contributors
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
@@ -36,8 +36,8 @@ object BasicWalkThrough {
}
def main(args: Array[String]): Unit = {
val trainMax = new DMatrix("../../demo/data/agaricus.txt.train")
val testMax = new DMatrix("../../demo/data/agaricus.txt.test")
val trainMax = new DMatrix("../../demo/data/agaricus.txt.train?format=libsvm")
val testMax = new DMatrix("../../demo/data/agaricus.txt.test?format=libsvm")
val params = new mutable.HashMap[String, Any]()
params += "eta" -> 1.0
@@ -76,7 +76,7 @@ object BasicWalkThrough {
// build dmatrix from CSR Sparse Matrix
println("start build dmatrix from csr sparse data ...")
val spData = DataLoader.loadSVMFile("../../demo/data/agaricus.txt.train")
val spData = DataLoader.loadSVMFile("../../demo/data/agaricus.txt.train?format=libsvm")
val trainMax2 = new DMatrix(spData.rowHeaders, spData.colIndex, spData.data,
JDMatrix.SparseType.CSR)
trainMax2.setLabel(spData.labels)

4
jvm-packages/xgboost4j-example/src/main/scala/ml/dmlc/xgboost4j/scala/example/BoostFromPrediction.scala
View File

@@ -24,8 +24,8 @@ object BoostFromPrediction {
def main(args: Array[String]): Unit = {
println("start running example to start from a initial prediction")
val trainMat = new DMatrix("../../demo/data/agaricus.txt.train")
val testMat = new DMatrix("../../demo/data/agaricus.txt.test")
val trainMat = new DMatrix("../../demo/data/agaricus.txt.train?format=libsvm")
val testMat = new DMatrix("../../demo/data/agaricus.txt.test?format=libsvm")
val params = new mutable.HashMap[String, Any]()
params += "eta" -> 1.0

2
jvm-packages/xgboost4j-example/src/main/scala/ml/dmlc/xgboost4j/scala/example/CrossValidation.scala
View File

@@ -21,7 +21,7 @@ import ml.dmlc.xgboost4j.scala.{XGBoost, DMatrix}
object CrossValidation {
def main(args: Array[String]): Unit = {
val trainMat: DMatrix = new DMatrix("../../demo/data/agaricus.txt.train")
val trainMat: DMatrix = new DMatrix("../../demo/data/agaricus.txt.train?format=libsvm")
// set params
val params = new mutable.HashMap[String, Any]

4
jvm-packages/xgboost4j-example/src/main/scala/ml/dmlc/xgboost4j/scala/example/CustomObjective.scala
View File

@@ -138,8 +138,8 @@ object CustomObjective {
}
def main(args: Array[String]): Unit = {
val trainMat = new DMatrix("../../demo/data/agaricus.txt.train")
val testMat = new DMatrix("../../demo/data/agaricus.txt.test")
val trainMat = new DMatrix("../../demo/data/agaricus.txt.train?format=libsvm")
val testMat = new DMatrix("../../demo/data/agaricus.txt.test?format=libsvm")
val params = new mutable.HashMap[String, Any]()
params += "eta" -> 1.0
params += "max_depth" -> 2

4
jvm-packages/xgboost4j-example/src/main/scala/ml/dmlc/xgboost4j/scala/example/ExternalMemory.scala
View File

@@ -25,8 +25,8 @@ object ExternalMemory {
// this is the only difference, add a # followed by a cache prefix name
// several cache file with the prefix will be generated
// currently only support convert from libsvm file
val trainMat = new DMatrix("../../demo/data/agaricus.txt.train#dtrain.cache")
val testMat = new DMatrix("../../demo/data/agaricus.txt.test#dtest.cache")
val trainMat = new DMatrix("../../demo/data/agaricus.txt.train?format=libsvm#dtrain.cache")
val testMat = new DMatrix("../../demo/data/agaricus.txt.test?format=libsvm#dtest.cache")
val params = new mutable.HashMap[String, Any]()
params += "eta" -> 1.0

4
jvm-packages/xgboost4j-example/src/main/scala/ml/dmlc/xgboost4j/scala/example/GeneralizedLinearModel.scala
View File

@@ -27,8 +27,8 @@ import ml.dmlc.xgboost4j.scala.example.util.CustomEval
*/
object GeneralizedLinearModel {
def main(args: Array[String]): Unit = {
val trainMat = new DMatrix("../../demo/data/agaricus.txt.train")
val testMat = new DMatrix("../../demo/data/agaricus.txt.test")
val trainMat = new DMatrix("../../demo/data/agaricus.txt.train?format=libsvm")
val testMat = new DMatrix("../../demo/data/agaricus.txt.test?format=libsvm")
// specify parameters
// change booster to gblinear, so that we are fitting a linear model

4
jvm-packages/xgboost4j-example/src/main/scala/ml/dmlc/xgboost4j/scala/example/PredictFirstNTree.scala
View File

@@ -23,8 +23,8 @@ import ml.dmlc.xgboost4j.scala.{XGBoost, DMatrix}
object PredictFirstNTree {
def main(args: Array[String]): Unit = {
val trainMat = new DMatrix("../../demo/data/agaricus.txt.train")
val testMat = new DMatrix("../../demo/data/agaricus.txt.test")
val trainMat = new DMatrix("../../demo/data/agaricus.txt.train?format=libsvm")
val testMat = new DMatrix("../../demo/data/agaricus.txt.test?format=libsvm")
val params = new mutable.HashMap[String, Any]()
params += "eta" -> 1.0

4
jvm-packages/xgboost4j-example/src/main/scala/ml/dmlc/xgboost4j/scala/example/PredictLeafIndices.scala
View File

@@ -25,8 +25,8 @@ import ml.dmlc.xgboost4j.scala.{XGBoost, DMatrix}
object PredictLeafIndices {
def main(args: Array[String]): Unit = {
val trainMat = new DMatrix("../../demo/data/agaricus.txt.train")
val testMat = new DMatrix("../../demo/data/agaricus.txt.test")
val trainMat = new DMatrix("../../demo/data/agaricus.txt.train?format=libsvm")
val testMat = new DMatrix("../../demo/data/agaricus.txt.test?format=libsvm")
val params = new mutable.HashMap[String, Any]()
params += "eta" -> 1.0