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

Deprecate reg:linear' in favor of reg:squarederror'. (#4267)

Browse Source 
* Deprecate `reg:linear' in favor of `reg:squarederror'.
* Replace the use of `reg:linear'.
* Replace the use of `silent`.
This commit is contained in:
Jiaming Yuan 2019-03-17 17:55:04 +08:00 committed by GitHub
parent cf8d5b9b76
commit 29a1356669
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
34 changed files with 210 additions and 193 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
R-package/R/utils.R
View File

@ -209,13 +209,14 @@ generate.cv.folds <- function(nfold, nrows, stratified, label, params) {
if (exists('objective', where = params) && if (exists('objective', where = params) &&
is.character(params$objective)) { is.character(params$objective)) {
# If 'objective' provided in params, assume that y is a classification label # If 'objective' provided in params, assume that y is a classification label
# unless objective is reg:linear # unless objective is reg:squarederror
if (params$objective != 'reg:linear') if (params$objective != 'reg:squarederror')
y <- factor(y) y <- factor(y)
} else { } else {
# If no 'objective' given in params, it means that user either wants to use # If no 'objective' given in params, it means that user either wants to
# the default 'reg:linear' objective or has provided a custom obj function. # use the default 'reg:squarederror' objective or has provided a custom
# Here, assume classification setting when y has 5 or less unique values: # obj function. Here, assume classification setting when y has 5 or less
# unique values:
if (length(unique(y)) <= 5) if (length(unique(y)) <= 5)
y <- factor(y) y <- factor(y)
} }

2
R-package/R/xgb.cv.R
View File

@ -6,7 +6,7 @@
#' \itemize{ #' \itemize{
#' \item \code{objective} objective function, common ones are #' \item \code{objective} objective function, common ones are
#' \itemize{ #' \itemize{
#' \item \code{reg:linear} linear regression #' \item \code{reg:squarederror} Regression with squared loss
#' \item \code{binary:logistic} logistic regression for classification #' \item \code{binary:logistic} logistic regression for classification
#' } #' }
#' \item \code{eta} step size of each boosting step #' \item \code{eta} step size of each boosting step

2
R-package/R/xgb.train.R
View File

@ -42,7 +42,7 @@
#' \itemize{ #' \itemize{
#' \item \code{objective} specify the learning task and the corresponding learning objective, users can pass a self-defined function to it. The default objective options are below: #' \item \code{objective} specify the learning task and the corresponding learning objective, users can pass a self-defined function to it. The default objective options are below:
#' \itemize{ #' \itemize{
#' \item \code{reg:linear} linear regression (Default). #' \item \code{reg:squarederror} Regression with squared loss (Default).
#' \item \code{reg:logistic} logistic regression. #' \item \code{reg:logistic} logistic regression.
#' \item \code{binary:logistic} logistic regression for binary classification. Output probability. #' \item \code{binary:logistic} logistic regression for binary classification. Output probability.
#' \item \code{binary:logitraw} logistic regression for binary classification, output score before logistic transformation. #' \item \code{binary:logitraw} logistic regression for binary classification, output score before logistic transformation.

2
R-package/man/xgb.cv.Rd
View File

@ -16,7 +16,7 @@ xgb.cv(params = list(), data, nrounds, nfold, label = NULL,
\itemize{ \itemize{
\item \code{objective} objective function, common ones are \item \code{objective} objective function, common ones are
\itemize{ \itemize{
\item \code{reg:linear} linear regression \item \code{reg:squarederror} Regression with squared loss.
\item \code{binary:logistic} logistic regression for classification \item \code{binary:logistic} logistic regression for classification
} }
\item \code{eta} step size of each boosting step \item \code{eta} step size of each boosting step

10
R-package/man/xgb.train.Rd
View File

@ -56,7 +56,7 @@ xgboost(data = NULL, label = NULL, missing = NA, weight = NULL,
\itemize{ \itemize{
\item \code{objective} specify the learning task and the corresponding learning objective, users can pass a self-defined function to it. The default objective options are below: \item \code{objective} specify the learning task and the corresponding learning objective, users can pass a self-defined function to it. The default objective options are below:
\itemize{ \itemize{
\item \code{reg:linear} linear regression (Default). \item \code{reg:squarederror} Regression with squared loss (Default).
\item \code{reg:logistic} logistic regression. \item \code{reg:logistic} logistic regression.
\item \code{binary:logistic} logistic regression for binary classification. Output probability. \item \code{binary:logistic} logistic regression for binary classification. Output probability.
\item \code{binary:logitraw} logistic regression for binary classification, output score before logistic transformation. \item \code{binary:logitraw} logistic regression for binary classification, output score before logistic transformation.
@ -210,7 +210,7 @@ dtest <- xgb.DMatrix(agaricus.test$data, label = agaricus.test$label)
watchlist <- list(train = dtrain, eval = dtest) watchlist <- list(train = dtrain, eval = dtest)
## A simple xgb.train example: ## A simple xgb.train example:
param <- list(max_depth = 2, eta = 1, silent = 1, nthread = 2, param <- list(max_depth = 2, eta = 1, verbosity = 0, nthread = 2,
objective = "binary:logistic", eval_metric = "auc") objective = "binary:logistic", eval_metric = "auc")
bst <- xgb.train(param, dtrain, nrounds = 2, watchlist) bst <- xgb.train(param, dtrain, nrounds = 2, watchlist)
@ -231,12 +231,12 @@ evalerror <- function(preds, dtrain) {
# These functions could be used by passing them either: # These functions could be used by passing them either:
# as 'objective' and 'eval_metric' parameters in the params list: # as 'objective' and 'eval_metric' parameters in the params list:
param <- list(max_depth = 2, eta = 1, silent = 1, nthread = 2, param <- list(max_depth = 2, eta = 1, verbosity = 0, nthread = 2,
objective = logregobj, eval_metric = evalerror) objective = logregobj, eval_metric = evalerror)
bst <- xgb.train(param, dtrain, nrounds = 2, watchlist) bst <- xgb.train(param, dtrain, nrounds = 2, watchlist)
# or through the ... arguments: # or through the ... arguments:
param <- list(max_depth = 2, eta = 1, silent = 1, nthread = 2) param <- list(max_depth = 2, eta = 1, verbosity = 0, nthread = 2)
bst <- xgb.train(param, dtrain, nrounds = 2, watchlist, bst <- xgb.train(param, dtrain, nrounds = 2, watchlist,
objective = logregobj, eval_metric = evalerror) objective = logregobj, eval_metric = evalerror)
@ -246,7 +246,7 @@ bst <- xgb.train(param, dtrain, nrounds = 2, watchlist,
## An xgb.train example of using variable learning rates at each iteration: ## An xgb.train example of using variable learning rates at each iteration:
param <- list(max_depth = 2, eta = 1, silent = 1, nthread = 2, param <- list(max_depth = 2, eta = 1, verbosity = 0, nthread = 2,
objective = "binary:logistic", eval_metric = "auc") objective = "binary:logistic", eval_metric = "auc")
my_etas <- list(eta = c(0.5, 0.1)) my_etas <- list(eta = c(0.5, 0.1))
bst <- xgb.train(param, dtrain, nrounds = 2, watchlist, bst <- xgb.train(param, dtrain, nrounds = 2, watchlist,

2
R-package/tests/testthat/test_interactions.R
View File

@ -98,7 +98,7 @@ test_that("SHAP contribution values are not NAN", {
fit <- xgboost( fit <- xgboost(
verbose = 0, verbose = 0,
params = list( params = list(
objective = "reg:linear", objective = "reg:squarederror",
eval_metric = "rmse"), eval_metric = "rmse"),
data = as.matrix(subset(d, fold == 2)[, ivs]), data = as.matrix(subset(d, fold == 2)[, ivs]),
label = subset(d, fold == 2)$y, label = subset(d, fold == 2)$y,

4
demo/regression/README.md
View File

@ -6,9 +6,9 @@ Using XGBoost for regression is very similar to using it for binary classificati
The dataset we used is the [computer hardware dataset from UCI repository](https://archive.ics.uci.edu/ml/datasets/Computer+Hardware). The demo for regression is almost the same as the [binary classification demo](../binary_classification), except a little difference in general parameter: The dataset we used is the [computer hardware dataset from UCI repository](https://archive.ics.uci.edu/ml/datasets/Computer+Hardware). The demo for regression is almost the same as the [binary classification demo](../binary_classification), except a little difference in general parameter:
``` ```
# General parameter # General parameter
# this is the only difference with classification, use reg:linear to do linear regression # this is the only difference with classification, use reg:squarederror to do linear regression
# when labels are in [0,1] we can also use reg:logistic # when labels are in [0,1] we can also use reg:logistic
objective = reg:linear objective = reg:squarederror
... ...
``` ```

4
demo/regression/machine.conf
View File

@ -1,9 +1,9 @@
# General Parameters, see comment for each definition # General Parameters, see comment for each definition
# choose the tree booster, can also change to gblinear # choose the tree booster, can also change to gblinear
booster = gbtree booster = gbtree
# this is the only difference with classification, use reg:linear to do linear classification # this is the only difference with classification, use reg:squarederror to do linear classification
# when labels are in [0,1] we can also use reg:logistic # when labels are in [0,1] we can also use reg:logistic
objective = reg:linear objective = reg:squarederror
# Tree Booster Parameters # Tree Booster Parameters
# step size shrinkage # step size shrinkage

5
demo/yearpredMSD/yearpredMSD.conf
View File

@ -1,9 +1,9 @@
# General Parameters, see comment for each definition # General Parameters, see comment for each definition
# choose the tree booster, can also change to gblinear # choose the tree booster, can also change to gblinear
booster = gbtree booster = gbtree
# this is the only difference with classification, use reg:linear to do linear classification # this is the only difference with classification, use reg:squarederror to do linear classification
# when labels are in [0,1] we can also use reg:logistic # when labels are in [0,1] we can also use reg:logistic
objective = reg:linear objective = reg:squarederror
# Tree Booster Parameters # Tree Booster Parameters
# step size shrinkage # step size shrinkage
@ -27,4 +27,3 @@ data = "yearpredMSD.libsvm.train"
eval[test] = "yearpredMSD.libsvm.test" eval[test] = "yearpredMSD.libsvm.test"
# The path of test data # The path of test data
#test:data = "yearpredMSD.libsvm.test" #test:data = "yearpredMSD.libsvm.test"

2
doc/gpu/index.rst
View File

@ -92,7 +92,7 @@ Most of the objective functions implemented in XGBoost can be run on GPU. Follo
+-----------------+-------------+ +-----------------+-------------+
| Objectives | GPU support | | Objectives | GPU support |
+-----------------+-------------+ +-----------------+-------------+
| reg:linear | |tick| | | reg:squarederror| |tick| |
+-----------------+-------------+ +-----------------+-------------+
| reg:logistic | |tick| | | reg:logistic | |tick| |
+-----------------+-------------+ +-----------------+-------------+

4
doc/parameter.rst
View File

@ -293,9 +293,9 @@ Learning Task Parameters
************************ ************************
Specify the learning task and the corresponding learning objective. The objective options are below: Specify the learning task and the corresponding learning objective. The objective options are below:
* ``objective`` [default=reg:linear] * ``objective`` [default=reg:squarederror]
- ``reg:linear``: linear regression - ``reg:squarederror``: regression with squared loss
- ``reg:logistic``: logistic regression - ``reg:logistic``: logistic regression
- ``binary:logistic``: logistic regression for binary classification, output probability - ``binary:logistic``: logistic regression for binary classification, output probability
- ``binary:logitraw``: logistic regression for binary classification, output score before logistic transformation - ``binary:logitraw``: logistic regression for binary classification, output score before logistic transformation

2
doc/rf.rst
View File

@ -36,7 +36,7 @@ The following parameters must be set to enable random forest training.
Other parameters should be set in a similar way they are set for gradient boosting. For Other parameters should be set in a similar way they are set for gradient boosting. For
instance, ``objective`` will typically be ``reg:linear`` for regression and instance, ``objective`` will typically be ``reg:squarederror`` for regression and
``binary:logistic`` for classification, ``lambda`` should be set according to a desired ``binary:logistic`` for classification, ``lambda`` should be set according to a desired
regularization weight, etc. regularization weight, etc.

8
jvm-packages/xgboost4j-spark/src/main/scala/ml/dmlc/xgboost4j/scala/spark/params/LearningTaskParams.scala
View File

@ -24,8 +24,8 @@ private[spark] trait LearningTaskParams extends Params {
/** /**
* Specify the learning task and the corresponding learning objective. * Specify the learning task and the corresponding learning objective.
* options: reg:linear, reg:logistic, binary:logistic, binary:logitraw, count:poisson, * options: reg:squarederror, reg:logistic, binary:logistic, binary:logitraw, count:poisson,
* multi:softmax, multi:softprob, rank:pairwise, reg:gamma. default: reg:linear * multi:softmax, multi:softprob, rank:pairwise, reg:gamma. default: reg:squarederror
*/ */
final val objective = new Param[String](this, "objective", "objective function used for " + final val objective = new Param[String](this, "objective", "objective function used for " +
s"training, options: {${LearningTaskParams.supportedObjective.mkString(",")}", s"training, options: {${LearningTaskParams.supportedObjective.mkString(",")}",
@ -94,12 +94,12 @@ private[spark] trait LearningTaskParams extends Params {
final def getMaximizeEvaluationMetrics: Boolean = $(maximizeEvaluationMetrics) final def getMaximizeEvaluationMetrics: Boolean = $(maximizeEvaluationMetrics)
setDefault(objective -> "reg:linear", baseScore -> 0.5, setDefault(objective -> "reg:squarederror", baseScore -> 0.5,
trainTestRatio -> 1.0, numEarlyStoppingRounds -> 0) trainTestRatio -> 1.0, numEarlyStoppingRounds -> 0)
} }
private[spark] object LearningTaskParams { private[spark] object LearningTaskParams {
val supportedObjective = HashSet("reg:linear", "reg:logistic", "binary:logistic", val supportedObjective = HashSet("reg:squarederror", "reg:logistic", "binary:logistic",
"binary:logitraw", "count:poisson", "multi:softmax", "multi:softprob", "rank:pairwise", "binary:logitraw", "count:poisson", "multi:softmax", "multi:softprob", "rank:pairwise",
"rank:ndcg", "rank:map", "reg:gamma", "reg:tweedie") "rank:ndcg", "rank:map", "reg:gamma", "reg:tweedie")

2
jvm-packages/xgboost4j-spark/src/test/scala/ml/dmlc/xgboost4j/scala/spark/PersistenceSuite.scala
View File

@ -96,7 +96,7 @@ class PersistenceSuite extends FunSuite with PerTest with BeforeAndAfterAll {
val testDM = new DMatrix(Regression.test.iterator) val testDM = new DMatrix(Regression.test.iterator)
val paramMap = Map("eta" -> "0.1", "max_depth" -> "6", "silent" -> "1", val paramMap = Map("eta" -> "0.1", "max_depth" -> "6", "silent" -> "1",
"objective" -> "reg:linear", "num_round" -> "10", "num_workers" -> numWorkers) "objective" -> "reg:squarederror", "num_round" -> "10", "num_workers" -> numWorkers)
val xgbr = new XGBoostRegressor(paramMap) val xgbr = new XGBoostRegressor(paramMap)
val xgbrPath = new File(tempDir, "xgbr").getPath val xgbrPath = new File(tempDir, "xgbr").getPath
xgbr.write.overwrite().save(xgbrPath) xgbr.write.overwrite().save(xgbrPath)

18
jvm-packages/xgboost4j-spark/src/test/scala/ml/dmlc/xgboost4j/scala/spark/XGBoostRegressorSuite.scala
View File

@ -36,7 +36,7 @@ class XGBoostRegressorSuite extends FunSuite with PerTest {
"eta" -> "1", "eta" -> "1",
"max_depth" -> "6", "max_depth" -> "6",
"silent" -> "1", "silent" -> "1",
"objective" -> "reg:linear") "objective" -> "reg:squarederror")
val model1 = ScalaXGBoost.train(trainingDM, paramMap, round) val model1 = ScalaXGBoost.train(trainingDM, paramMap, round)
val prediction1 = model1.predict(testDM) val prediction1 = model1.predict(testDM)
@ -69,7 +69,7 @@ class XGBoostRegressorSuite extends FunSuite with PerTest {
"eta" -> "1", "eta" -> "1",
"max_depth" -> "6", "max_depth" -> "6",
"silent" -> "1", "silent" -> "1",
"objective" -> "reg:linear", "objective" -> "reg:squarederror",
"num_round" -> round, "num_round" -> round,
"num_workers" -> numWorkers) "num_workers" -> numWorkers)
@ -80,7 +80,7 @@ class XGBoostRegressorSuite extends FunSuite with PerTest {
.setEta(1) .setEta(1)
.setMaxDepth(6) .setMaxDepth(6)
.setSilent(1) .setSilent(1)
.setObjective("reg:linear") .setObjective("reg:squarederror")
.setNumRound(round) .setNumRound(round)
.setNumWorkers(numWorkers) .setNumWorkers(numWorkers)
.fit(trainingDF) .fit(trainingDF)
@ -108,7 +108,7 @@ class XGBoostRegressorSuite extends FunSuite with PerTest {
test("use weight") { test("use weight") {
val paramMap = Map("eta" -> "1", "max_depth" -> "6", "silent" -> "1", val paramMap = Map("eta" -> "1", "max_depth" -> "6", "silent" -> "1",
"objective" -> "reg:linear", "num_round" -> 5, "num_workers" -> numWorkers) "objective" -> "reg:squarederror", "num_round" -> 5, "num_workers" -> numWorkers)
val getWeightFromId = udf({id: Int => if (id == 0) 1.0f else 0.001f}, DataTypes.FloatType) val getWeightFromId = udf({id: Int => if (id == 0) 1.0f else 0.001f}, DataTypes.FloatType)
val trainingDF = buildDataFrame(Regression.train) val trainingDF = buildDataFrame(Regression.train)
@ -123,7 +123,7 @@ class XGBoostRegressorSuite extends FunSuite with PerTest {
test("test predictionLeaf") { test("test predictionLeaf") {
val paramMap = Map("eta" -> "1", "max_depth" -> "6", "silent" -> "1", val paramMap = Map("eta" -> "1", "max_depth" -> "6", "silent" -> "1",
"objective" -> "reg:linear", "num_round" -> 5, "num_workers" -> numWorkers) "objective" -> "reg:squarederror", "num_round" -> 5, "num_workers" -> numWorkers)
val training = buildDataFrame(Regression.train) val training = buildDataFrame(Regression.train)
val testDF = buildDataFrame(Regression.test) val testDF = buildDataFrame(Regression.test)
val groundTruth = testDF.count() val groundTruth = testDF.count()
@ -137,7 +137,7 @@ class XGBoostRegressorSuite extends FunSuite with PerTest {
test("test predictionLeaf with empty column name") { test("test predictionLeaf with empty column name") {
val paramMap = Map("eta" -> "1", "max_depth" -> "6", "silent" -> "1", val paramMap = Map("eta" -> "1", "max_depth" -> "6", "silent" -> "1",
"objective" -> "reg:linear", "num_round" -> 5, "num_workers" -> numWorkers) "objective" -> "reg:squarederror", "num_round" -> 5, "num_workers" -> numWorkers)
val training = buildDataFrame(Regression.train) val training = buildDataFrame(Regression.train)
val testDF = buildDataFrame(Regression.test) val testDF = buildDataFrame(Regression.test)
val xgb = new XGBoostRegressor(paramMap) val xgb = new XGBoostRegressor(paramMap)
@ -149,7 +149,7 @@ class XGBoostRegressorSuite extends FunSuite with PerTest {
test("test predictionContrib") { test("test predictionContrib") {
val paramMap = Map("eta" -> "1", "max_depth" -> "6", "silent" -> "1", val paramMap = Map("eta" -> "1", "max_depth" -> "6", "silent" -> "1",
"objective" -> "reg:linear", "num_round" -> 5, "num_workers" -> numWorkers) "objective" -> "reg:squarederror", "num_round" -> 5, "num_workers" -> numWorkers)
val training = buildDataFrame(Regression.train) val training = buildDataFrame(Regression.train)
val testDF = buildDataFrame(Regression.test) val testDF = buildDataFrame(Regression.test)
val groundTruth = testDF.count() val groundTruth = testDF.count()
@ -163,7 +163,7 @@ class XGBoostRegressorSuite extends FunSuite with PerTest {
test("test predictionContrib with empty column name") { test("test predictionContrib with empty column name") {
val paramMap = Map("eta" -> "1", "max_depth" -> "6", "silent" -> "1", val paramMap = Map("eta" -> "1", "max_depth" -> "6", "silent" -> "1",
"objective" -> "reg:linear", "num_round" -> 5, "num_workers" -> numWorkers) "objective" -> "reg:squarederror", "num_round" -> 5, "num_workers" -> numWorkers)
val training = buildDataFrame(Regression.train) val training = buildDataFrame(Regression.train)
val testDF = buildDataFrame(Regression.test) val testDF = buildDataFrame(Regression.test)
val xgb = new XGBoostRegressor(paramMap) val xgb = new XGBoostRegressor(paramMap)
@ -175,7 +175,7 @@ class XGBoostRegressorSuite extends FunSuite with PerTest {
test("test predictionLeaf and predictionContrib") { test("test predictionLeaf and predictionContrib") {
val paramMap = Map("eta" -> "1", "max_depth" -> "6", "silent" -> "1", val paramMap = Map("eta" -> "1", "max_depth" -> "6", "silent" -> "1",
"objective" -> "reg:linear", "num_round" -> 5, "num_workers" -> numWorkers) "objective" -> "reg:squarederror", "num_round" -> 5, "num_workers" -> numWorkers)
val training = buildDataFrame(Regression.train) val training = buildDataFrame(Regression.train)
val testDF = buildDataFrame(Regression.test) val testDF = buildDataFrame(Regression.test)
val groundTruth = testDF.count() val groundTruth = testDF.count()

12
src/objective/regression_obj.cu
View File

@ -128,8 +128,8 @@ class RegLossObj : public ObjFunction {
// register the objective functions // register the objective functions
DMLC_REGISTER_PARAMETER(RegLossParam); DMLC_REGISTER_PARAMETER(RegLossParam);
XGBOOST_REGISTER_OBJECTIVE(LinearRegression, "reg:linear") XGBOOST_REGISTER_OBJECTIVE(SquaredLossRegression, "reg:squarederror")
.describe("Linear regression.") .describe("Regression with squared error.")
.set_body([]() { return new RegLossObj<LinearSquareLoss>(); }); .set_body([]() { return new RegLossObj<LinearSquareLoss>(); });
XGBOOST_REGISTER_OBJECTIVE(LogisticRegression, "reg:logistic") XGBOOST_REGISTER_OBJECTIVE(LogisticRegression, "reg:logistic")
@ -145,7 +145,13 @@ XGBOOST_REGISTER_OBJECTIVE(LogisticRaw, "binary:logitraw")
"before logistic transformation.") "before logistic transformation.")
.set_body([]() { return new RegLossObj<LogisticRaw>(); }); .set_body([]() { return new RegLossObj<LogisticRaw>(); });
// Deprecated GPU functions // Deprecated functions
XGBOOST_REGISTER_OBJECTIVE(LinearRegression, "reg:linear")
.describe("Regression with squared error.")
.set_body([]() {
LOG(WARNING) << "reg:linear is now deprecated in favor of reg:squarederror.";
return new RegLossObj<LinearSquareLoss>(); });
XGBOOST_REGISTER_OBJECTIVE(GPULinearRegression, "gpu:reg:linear") XGBOOST_REGISTER_OBJECTIVE(GPULinearRegression, "gpu:reg:linear")
.describe("Deprecated. Linear regression (computed on GPU).") .describe("Deprecated. Linear regression (computed on GPU).")
.set_body([]() { .set_body([]() {

3
tests/cpp/objective/test_objective.cc
View File

@ -1,4 +1,5 @@
// Copyright by Contributors // Copyright by Contributors
#include <gtest/gtest.h>
#include <xgboost/objective.h> #include <xgboost/objective.h>
#include "../helpers.h" #include "../helpers.h"
@ -6,7 +7,7 @@
TEST(Objective, UnknownFunction) { TEST(Objective, UnknownFunction) {
xgboost::ObjFunction* obj = nullptr; xgboost::ObjFunction* obj = nullptr;
EXPECT_ANY_THROW(obj = xgboost::ObjFunction::Create("unknown_name")); EXPECT_ANY_THROW(obj = xgboost::ObjFunction::Create("unknown_name"));
EXPECT_NO_THROW(obj = xgboost::ObjFunction::Create("reg:linear")); EXPECT_NO_THROW(obj = xgboost::ObjFunction::Create("reg:squarederror"));
if (obj) { if (obj) {
delete obj; delete obj;
} }

5
tests/cpp/objective/test_regression_obj.cc
View File

@ -1,12 +1,13 @@
/*! /*!
* Copyright 2017-2018 XGBoost contributors * Copyright 2017-2019 XGBoost contributors
*/ */
#include <gtest/gtest.h>
#include <xgboost/objective.h> #include <xgboost/objective.h>
#include "../helpers.h" #include "../helpers.h"
TEST(Objective, DeclareUnifiedTest(LinearRegressionGPair)) { TEST(Objective, DeclareUnifiedTest(LinearRegressionGPair)) {
xgboost::ObjFunction * obj = xgboost::ObjFunction::Create("reg:linear"); xgboost::ObjFunction * obj = xgboost::ObjFunction::Create("reg:squarederror");
std::vector<std::pair<std::string, std::string> > args; std::vector<std::pair<std::string, std::string> > args;
obj->Configure(args); obj->Configure(args);
CheckObjFunction(obj, CheckObjFunction(obj,

11
tests/python/regression_test_utilities.py
View File

@ -132,15 +132,16 @@ def run_suite(param, num_rounds=10, select_datasets=None, scale_features=False):
Run the given parameters on a range of datasets. Objective and eval metric will be automatically set Run the given parameters on a range of datasets. Objective and eval metric will be automatically set
""" """
datasets = [ datasets = [
Dataset("Boston", get_boston, "reg:linear", "rmse"), Dataset("Boston", get_boston, "reg:squarederror", "rmse"),
Dataset("Digits", get_digits, "multi:softmax", "merror"), Dataset("Digits", get_digits, "multi:softmax", "merror"),
Dataset("Cancer", get_cancer, "binary:logistic", "error"), Dataset("Cancer", get_cancer, "binary:logistic", "error"),
Dataset("Sparse regression", get_sparse, "reg:linear", "rmse"), Dataset("Sparse regression", get_sparse, "reg:squarederror", "rmse"),
Dataset("Sparse regression with weights", get_sparse_weights, Dataset("Sparse regression with weights", get_sparse_weights,
"reg:linear", "rmse", has_weights=True), "reg:squarederror", "rmse", has_weights=True),
Dataset("Small weights regression", get_small_weights, Dataset("Small weights regression", get_small_weights,
"reg:linear", "rmse", has_weights=True), "reg:squarederror", "rmse", has_weights=True),
Dataset("Boston External Memory", get_boston, "reg:linear", "rmse", Dataset("Boston External Memory", get_boston,
"reg:squarederror", "rmse",
use_external_memory=True) use_external_memory=True)
] ]

14
tests/python/test_basic.py
View File

@ -38,7 +38,7 @@ class TestBasic(unittest.TestCase):
def test_basic(self): def test_basic(self):
dtrain = xgb.DMatrix(dpath + 'agaricus.txt.train') dtrain = xgb.DMatrix(dpath + 'agaricus.txt.train')
dtest = xgb.DMatrix(dpath + 'agaricus.txt.test') dtest = xgb.DMatrix(dpath + 'agaricus.txt.test')
param = {'max_depth': 2, 'eta': 1, 'silent': 1, param = {'max_depth': 2, 'eta': 1, 'verbosity': 0,
'objective': 'binary:logistic'} 'objective': 'binary:logistic'}
# specify validations set to watch performance # specify validations set to watch performance
watchlist = [(dtest, 'eval'), (dtrain, 'train')] watchlist = [(dtest, 'eval'), (dtrain, 'train')]
@ -85,7 +85,7 @@ class TestBasic(unittest.TestCase):
def test_record_results(self): def test_record_results(self):
dtrain = xgb.DMatrix(dpath + 'agaricus.txt.train') dtrain = xgb.DMatrix(dpath + 'agaricus.txt.train')
dtest = xgb.DMatrix(dpath + 'agaricus.txt.test') dtest = xgb.DMatrix(dpath + 'agaricus.txt.test')
param = {'max_depth': 2, 'eta': 1, 'silent': 1, param = {'max_depth': 2, 'eta': 1, 'verbosity': 0,
'objective': 'binary:logistic'} 'objective': 'binary:logistic'}
# specify validations set to watch performance # specify validations set to watch performance
watchlist = [(dtest, 'eval'), (dtrain, 'train')] watchlist = [(dtest, 'eval'), (dtrain, 'train')]
@ -102,7 +102,7 @@ class TestBasic(unittest.TestCase):
def test_multiclass(self): def test_multiclass(self):
dtrain = xgb.DMatrix(dpath + 'agaricus.txt.train') dtrain = xgb.DMatrix(dpath + 'agaricus.txt.train')
dtest = xgb.DMatrix(dpath + 'agaricus.txt.test') dtest = xgb.DMatrix(dpath + 'agaricus.txt.test')
param = {'max_depth': 2, 'eta': 1, 'silent': 1, 'num_class': 2} param = {'max_depth': 2, 'eta': 1, 'verbosity': 0, 'num_class': 2}
# specify validations set to watch performance # specify validations set to watch performance
watchlist = [(dtest, 'eval'), (dtrain, 'train')] watchlist = [(dtest, 'eval'), (dtrain, 'train')]
num_round = 2 num_round = 2
@ -273,7 +273,7 @@ class TestBasic(unittest.TestCase):
def test_cv(self): def test_cv(self):
dm = xgb.DMatrix(dpath + 'agaricus.txt.train') dm = xgb.DMatrix(dpath + 'agaricus.txt.train')
params = {'max_depth': 2, 'eta': 1, 'silent': 1, params = {'max_depth': 2, 'eta': 1, 'verbosity': 0,
'objective': 'binary:logistic'} 'objective': 'binary:logistic'}
# return np.ndarray # return np.ndarray
@ -283,7 +283,7 @@ class TestBasic(unittest.TestCase):
def test_cv_no_shuffle(self): def test_cv_no_shuffle(self):
dm = xgb.DMatrix(dpath + 'agaricus.txt.train') dm = xgb.DMatrix(dpath + 'agaricus.txt.train')
params = {'max_depth': 2, 'eta': 1, 'silent': 1, params = {'max_depth': 2, 'eta': 1, 'verbosity': 0,
'objective': 'binary:logistic'} 'objective': 'binary:logistic'}
# return np.ndarray # return np.ndarray
@ -294,7 +294,7 @@ class TestBasic(unittest.TestCase):
def test_cv_explicit_fold_indices(self): def test_cv_explicit_fold_indices(self):
dm = xgb.DMatrix(dpath + 'agaricus.txt.train') dm = xgb.DMatrix(dpath + 'agaricus.txt.train')
params = {'max_depth': 2, 'eta': 1, 'silent': 1, 'objective': params = {'max_depth': 2, 'eta': 1, 'verbosity': 0, 'objective':
'binary:logistic'} 'binary:logistic'}
folds = [ folds = [
# Train Test # Train Test
@ -310,7 +310,7 @@ class TestBasic(unittest.TestCase):
def test_cv_explicit_fold_indices_labels(self): def test_cv_explicit_fold_indices_labels(self):
params = {'max_depth': 2, 'eta': 1, 'verbosity': 0, 'objective': params = {'max_depth': 2, 'eta': 1, 'verbosity': 0, 'objective':
'reg:linear'} 'reg:squarederror'}
N = 100 N = 100
F = 3 F = 3
dm = xgb.DMatrix(data=np.random.randn(N, F), label=np.arange(N)) dm = xgb.DMatrix(data=np.random.randn(N, F), label=np.arange(N))

23
tests/python/test_basic_models.py
View File

@ -11,7 +11,7 @@ rng = np.random.RandomState(1994)
class TestModels(unittest.TestCase): class TestModels(unittest.TestCase):
def test_glm(self): def test_glm(self):
param = {'silent': 1, 'objective': 'binary:logistic', param = {'verbosity': 0, 'objective': 'binary:logistic',
'booster': 'gblinear', 'alpha': 0.0001, 'lambda': 1, 'nthread': 1} 'booster': 'gblinear', 'alpha': 0.0001, 'lambda': 1, 'nthread': 1}
watchlist = [(dtest, 'eval'), (dtrain, 'train')] watchlist = [(dtest, 'eval'), (dtrain, 'train')]
num_round = 4 num_round = 4
@ -26,7 +26,7 @@ class TestModels(unittest.TestCase):
def test_dart(self): def test_dart(self):
dtrain = xgb.DMatrix(dpath + 'agaricus.txt.train') dtrain = xgb.DMatrix(dpath + 'agaricus.txt.train')
dtest = xgb.DMatrix(dpath + 'agaricus.txt.test') dtest = xgb.DMatrix(dpath + 'agaricus.txt.test')
param = {'max_depth': 5, 'objective': 'binary:logistic', 'booster': 'dart', 'silent': False} param = {'max_depth': 5, 'objective': 'binary:logistic', 'booster': 'dart', 'verbosity': 1}
# specify validations set to watch performance # specify validations set to watch performance
watchlist = [(dtest, 'eval'), (dtrain, 'train')] watchlist = [(dtest, 'eval'), (dtrain, 'train')]
num_round = 2 num_round = 2
@ -51,7 +51,7 @@ class TestModels(unittest.TestCase):
# check whether sample_type and normalize_type work # check whether sample_type and normalize_type work
num_round = 50 num_round = 50
param['silent'] = True param['verbosity'] = 0
param['learning_rate'] = 0.1 param['learning_rate'] = 0.1
param['rate_drop'] = 0.1 param['rate_drop'] = 0.1
preds_list = [] preds_list = []
@ -74,7 +74,8 @@ class TestModels(unittest.TestCase):
# learning_rates as a list # learning_rates as a list
# init eta with 0 to check whether learning_rates work # init eta with 0 to check whether learning_rates work
param = {'max_depth': 2, 'eta': 0, 'silent': 1, 'objective': 'binary:logistic'} param = {'max_depth': 2, 'eta': 0, 'verbosity': 0,
'objective': 'binary:logistic'}
evals_result = {} evals_result = {}
bst = xgb.train(param, dtrain, num_round, watchlist, learning_rates=[0.8, 0.7, 0.6, 0.5], bst = xgb.train(param, dtrain, num_round, watchlist, learning_rates=[0.8, 0.7, 0.6, 0.5],
evals_result=evals_result) evals_result=evals_result)
@ -84,7 +85,8 @@ class TestModels(unittest.TestCase):
assert eval_errors[0] > eval_errors[-1] assert eval_errors[0] > eval_errors[-1]
# init learning_rate with 0 to check whether learning_rates work # init learning_rate with 0 to check whether learning_rates work
param = {'max_depth': 2, 'learning_rate': 0, 'silent': 1, 'objective': 'binary:logistic'} param = {'max_depth': 2, 'learning_rate': 0, 'verbosity': 0,
'objective': 'binary:logistic'}
evals_result = {} evals_result = {}
bst = xgb.train(param, dtrain, num_round, watchlist, learning_rates=[0.8, 0.7, 0.6, 0.5], bst = xgb.train(param, dtrain, num_round, watchlist, learning_rates=[0.8, 0.7, 0.6, 0.5],
evals_result=evals_result) evals_result=evals_result)
@ -94,7 +96,7 @@ class TestModels(unittest.TestCase):
assert eval_errors[0] > eval_errors[-1] assert eval_errors[0] > eval_errors[-1]
# check if learning_rates override default value of eta/learning_rate # check if learning_rates override default value of eta/learning_rate
param = {'max_depth': 2, 'silent': 1, 'objective': 'binary:logistic'} param = {'max_depth': 2, 'verbosity': 0, 'objective': 'binary:logistic'}
evals_result = {} evals_result = {}
bst = xgb.train(param, dtrain, num_round, watchlist, learning_rates=[0, 0, 0, 0], bst = xgb.train(param, dtrain, num_round, watchlist, learning_rates=[0, 0, 0, 0],
evals_result=evals_result) evals_result=evals_result)
@ -111,7 +113,7 @@ class TestModels(unittest.TestCase):
assert isinstance(bst, xgb.core.Booster) assert isinstance(bst, xgb.core.Booster)
def test_custom_objective(self): def test_custom_objective(self):
param = {'max_depth': 2, 'eta': 1, 'silent': 1} param = {'max_depth': 2, 'eta': 1, 'verbosity': 0}
watchlist = [(dtest, 'eval'), (dtrain, 'train')] watchlist = [(dtest, 'eval'), (dtrain, 'train')]
num_round = 2 num_round = 2
@ -152,7 +154,8 @@ class TestModels(unittest.TestCase):
def test_multi_eval_metric(self): def test_multi_eval_metric(self):
watchlist = [(dtest, 'eval'), (dtrain, 'train')] watchlist = [(dtest, 'eval'), (dtrain, 'train')]
param = {'max_depth': 2, 'eta': 0.2, 'silent': 1, 'objective': 'binary:logistic'} param = {'max_depth': 2, 'eta': 0.2, 'verbosity': 0,
'objective': 'binary:logistic'}
param['eval_metric'] = ["auc", "logloss", 'error'] param['eval_metric'] = ["auc", "logloss", 'error']
evals_result = {} evals_result = {}
bst = xgb.train(param, dtrain, 4, watchlist, evals_result=evals_result) bst = xgb.train(param, dtrain, 4, watchlist, evals_result=evals_result)
@ -161,7 +164,7 @@ class TestModels(unittest.TestCase):
assert set(evals_result['eval'].keys()) == {'auc', 'error', 'logloss'} assert set(evals_result['eval'].keys()) == {'auc', 'error', 'logloss'}
def test_fpreproc(self): def test_fpreproc(self):
param = {'max_depth': 2, 'eta': 1, 'silent': 1, param = {'max_depth': 2, 'eta': 1, 'verbosity': 0,
'objective': 'binary:logistic'} 'objective': 'binary:logistic'}
num_round = 2 num_round = 2
@ -175,7 +178,7 @@ class TestModels(unittest.TestCase):
metrics={'auc'}, seed=0, fpreproc=fpreproc) metrics={'auc'}, seed=0, fpreproc=fpreproc)
def test_show_stdv(self): def test_show_stdv(self):
param = {'max_depth': 2, 'eta': 1, 'silent': 1, param = {'max_depth': 2, 'eta': 1, 'verbosity': 0,
'objective': 'binary:logistic'} 'objective': 'binary:logistic'}
num_round = 2 num_round = 2
xgb.cv(param, dtrain, num_round, nfold=5, xgb.cv(param, dtrain, num_round, nfold=5,

2
tests/python/test_early_stopping.py
View File

@ -52,7 +52,7 @@ class TestEarlyStopping(unittest.TestCase):
X = digits['data'] X = digits['data']
y = digits['target'] y = digits['target']
dm = xgb.DMatrix(X, label=y) dm = xgb.DMatrix(X, label=y)
params = {'max_depth': 2, 'eta': 1, 'silent': 1, params = {'max_depth': 2, 'eta': 1, 'verbosity': 0,
'objective': 'binary:logistic'} 'objective': 'binary:logistic'}
cv = xgb.cv(params, dm, num_boost_round=10, nfold=10, cv = xgb.cv(params, dm, num_boost_round=10, nfold=10,

8
tests/python/test_eval_metrics.py
View File

@ -9,25 +9,25 @@ rng = np.random.RandomState(1337)
class TestEvalMetrics(unittest.TestCase): class TestEvalMetrics(unittest.TestCase):
xgb_params_01 = { xgb_params_01 = {
'silent': 1, 'verbosity': 0,
'nthread': 1, 'nthread': 1,
'eval_metric': 'error' 'eval_metric': 'error'
} }
xgb_params_02 = { xgb_params_02 = {
'silent': 1, 'verbosity': 0,
'nthread': 1, 'nthread': 1,
'eval_metric': ['error'] 'eval_metric': ['error']
} }
xgb_params_03 = { xgb_params_03 = {
'silent': 1, 'verbosity': 0,
'nthread': 1, 'nthread': 1,
'eval_metric': ['rmse', 'error'] 'eval_metric': ['rmse', 'error']
} }
xgb_params_04 = { xgb_params_04 = {
'silent': 1, 'verbosity': 0,
'nthread': 1, 'nthread': 1,
'eval_metric': ['error', 'rmse'] 'eval_metric': ['error', 'rmse']
} }

2
tests/python/test_interaction_constraints.py
View File

@ -18,7 +18,7 @@ class TestInteractionConstraints(unittest.TestCase):
X = np.column_stack((x1, x2, x3)) X = np.column_stack((x1, x2, x3))
dtrain = xgboost.DMatrix(X, label=y) dtrain = xgboost.DMatrix(X, label=y)
params = {'max_depth': 3, 'eta': 0.1, 'nthread': 2, 'silent': 1, params = {'max_depth': 3, 'eta': 0.1, 'nthread': 2, 'verbosity': 0,
'interaction_constraints': '[[0, 1]]'} 'interaction_constraints': '[[0, 1]]'}
num_boost_round = 100 num_boost_round = 100
# Fit a model that only allows interaction between x1 and x2 # Fit a model that only allows interaction between x1 and x2

4
tests/python/test_linear.py
View File

@ -30,7 +30,7 @@ def xgb_get_weights(bst):
def assert_regression_result(results, tol): def assert_regression_result(results, tol):
regression_results = [r for r in results if regression_results = [r for r in results if
r["param"]["objective"] == "reg:linear"] r["param"]["objective"] == "reg:squarederror"]
for res in regression_results: for res in regression_results:
X = scale(res["dataset"].X, X = scale(res["dataset"].X,
with_mean=isinstance(res["dataset"].X, np.ndarray)) with_mean=isinstance(res["dataset"].X, np.ndarray))
@ -52,7 +52,7 @@ def assert_regression_result(results, tol):
# TODO: More robust classification tests # TODO: More robust classification tests
def assert_classification_result(results): def assert_classification_result(results):
classification_results = [r for r in results if classification_results = [r for r in results if
r["param"]["objective"] != "reg:linear"] r["param"]["objective"] != "reg:squarederror"]
for res in classification_results: for res in classification_results:
# Check accuracy is reasonable # Check accuracy is reasonable
assert res["eval"][-1] < 0.5, (res["dataset"].name, res["eval"][-1]) assert res["eval"][-1] < 0.5, (res["dataset"].name, res["eval"][-1])

3
tests/python/test_parse_tree.py
View File

@ -16,7 +16,8 @@ class TestTreesToDataFrame(unittest.TestCase):
def build_model(self, max_depth, num_round): def build_model(self, max_depth, num_round):
dtrain = xgb.DMatrix(dpath + 'agaricus.txt.train') dtrain = xgb.DMatrix(dpath + 'agaricus.txt.train')
param = {'max_depth': max_depth, 'objective': 'binary:logistic', 'silent': False} param = {'max_depth': max_depth, 'objective': 'binary:logistic',
'verbosity': 1}
num_round = num_round num_round = num_round
bst = xgb.train(param, dtrain, num_round) bst = xgb.train(param, dtrain, num_round)
return bst return bst

2
tests/python/test_shap.py
View File

@ -51,7 +51,7 @@ class TestSHAP(unittest.TestCase):
def fn(max_depth, num_rounds): def fn(max_depth, num_rounds):
# train # train
params = {'max_depth': max_depth, 'eta': 1, 'silent': 1} params = {'max_depth': max_depth, 'eta': 1, 'verbosity': 0}
bst = xgb.train(params, dtrain, num_boost_round=num_rounds) bst = xgb.train(params, dtrain, num_boost_round=num_rounds)
# predict # predict

2
tests/python/test_sparse_dmatrix.py
View File

@ -4,7 +4,7 @@ from scipy.sparse import rand
rng = np.random.RandomState(1) rng = np.random.RandomState(1)
param = {'max_depth': 3, 'objective': 'binary:logistic', 'silent': 1} param = {'max_depth': 3, 'objective': 'binary:logistic', 'verbosity': 0}
def test_sparse_dmatrix_csr(): def test_sparse_dmatrix_csr():

6
tests/python/test_training_continuation.py
View File

@ -11,18 +11,18 @@ class TestTrainingContinuation(unittest.TestCase):
num_parallel_tree = 3 num_parallel_tree = 3
xgb_params_01 = { xgb_params_01 = {
'silent': 1, 'verbosity': 0,
'nthread': 1, 'nthread': 1,
} }
xgb_params_02 = { xgb_params_02 = {
'silent': 1, 'verbosity': 0,
'nthread': 1, 'nthread': 1,
'num_parallel_tree': num_parallel_tree 'num_parallel_tree': num_parallel_tree
} }
xgb_params_03 = { xgb_params_03 = {
'silent': 1, 'verbosity': 0,
'nthread': 1, 'nthread': 1,
'num_class': 5, 'num_class': 5,
'num_parallel_tree': num_parallel_tree 'num_parallel_tree': num_parallel_tree

9
tests/python/test_tree_regularization.py
View File

@ -10,7 +10,8 @@ train_data = xgb.DMatrix(np.array([[1]]), label=np.array([1]))
class TestTreeRegularization(unittest.TestCase): class TestTreeRegularization(unittest.TestCase):
def test_alpha(self): def test_alpha(self):
params = { params = {
'tree_method': 'exact', 'silent': 1, 'objective': 'reg:linear', 'tree_method': 'exact', 'verbosity': 0,
'objective': 'reg:squarederror',
'eta': 1, 'eta': 1,
'lambda': 0, 'lambda': 0,
'alpha': 0.1 'alpha': 0.1
@ -27,7 +28,8 @@ class TestTreeRegularization(unittest.TestCase):
def test_lambda(self): def test_lambda(self):
params = { params = {
'tree_method': 'exact', 'silent': 1, 'objective': 'reg:linear', 'tree_method': 'exact', 'verbosity': 0,
'objective': 'reg:squarederror',
'eta': 1, 'eta': 1,
'lambda': 1, 'lambda': 1,
'alpha': 0 'alpha': 0
@ -44,7 +46,8 @@ class TestTreeRegularization(unittest.TestCase):
def test_alpha_and_lambda(self): def test_alpha_and_lambda(self):
params = { params = {
'tree_method': 'exact', 'silent': 1, 'objective': 'reg:linear', 'tree_method': 'exact', 'verbosity': 1,
'objective': 'reg:squarederror',
'eta': 1, 'eta': 1,
'lambda': 1, 'lambda': 1,
'alpha': 0.1 'alpha': 0.1

4
tests/python/test_updaters.py
View File

@ -33,7 +33,7 @@ class TestUpdaters(unittest.TestCase):
'max_bin': [2, 256], 'max_bin': [2, 256],
'grow_policy': ['depthwise', 'lossguide'], 'grow_policy': ['depthwise', 'lossguide'],
'max_leaves': [64, 0], 'max_leaves': [64, 0],
'silent': [1]} 'verbosity': [0]}
for param in parameter_combinations(variable_param): for param in parameter_combinations(variable_param):
result = run_suite(param) result = run_suite(param)
assert_results_non_increasing(result, 1e-2) assert_results_non_increasing(result, 1e-2)
@ -45,7 +45,7 @@ class TestUpdaters(unittest.TestCase):
ag_param = {'max_depth': 2, ag_param = {'max_depth': 2,
'tree_method': 'hist', 'tree_method': 'hist',
'eta': 1, 'eta': 1,
'silent': 1, 'verbosity': 0,
'objective': 'binary:logistic', 'objective': 'binary:logistic',
'eval_metric': 'auc'} 'eval_metric': 'auc'}
hist_res = {} hist_res = {}

14
tests/python/test_with_pandas.py
View File

@ -120,7 +120,7 @@ class TestPandas(unittest.TestCase):
def test_cv_as_pandas(self): def test_cv_as_pandas(self):
dm = xgb.DMatrix(dpath + 'agaricus.txt.train') dm = xgb.DMatrix(dpath + 'agaricus.txt.train')
params = {'max_depth': 2, 'eta': 1, 'silent': 1, params = {'max_depth': 2, 'eta': 1, 'verbosity': 0,
'objective': 'binary:logistic'} 'objective': 'binary:logistic'}
cv = xgb.cv(params, dm, num_boost_round=10, nfold=10) cv = xgb.cv(params, dm, num_boost_round=10, nfold=10)
@ -143,19 +143,19 @@ class TestPandas(unittest.TestCase):
u'train-error-mean', u'train-error-std']) u'train-error-mean', u'train-error-std'])
assert cv.columns.equals(exp) assert cv.columns.equals(exp)
params = {'max_depth': 2, 'eta': 1, 'silent': 1, params = {'max_depth': 2, 'eta': 1, 'verbosity': 0,
'objective': 'binary:logistic', 'eval_metric': 'auc'} 'objective': 'binary:logistic', 'eval_metric': 'auc'}
cv = xgb.cv(params, dm, num_boost_round=10, nfold=10, as_pandas=True) cv = xgb.cv(params, dm, num_boost_round=10, nfold=10, as_pandas=True)
assert 'eval_metric' in params assert 'eval_metric' in params
assert 'auc' in cv.columns[0] assert 'auc' in cv.columns[0]
params = {'max_depth': 2, 'eta': 1, 'silent': 1, params = {'max_depth': 2, 'eta': 1, 'verbosity': 0,
'objective': 'binary:logistic', 'eval_metric': ['auc']} 'objective': 'binary:logistic', 'eval_metric': ['auc']}
cv = xgb.cv(params, dm, num_boost_round=10, nfold=10, as_pandas=True) cv = xgb.cv(params, dm, num_boost_round=10, nfold=10, as_pandas=True)
assert 'eval_metric' in params assert 'eval_metric' in params
assert 'auc' in cv.columns[0] assert 'auc' in cv.columns[0]
params = {'max_depth': 2, 'eta': 1, 'silent': 1, params = {'max_depth': 2, 'eta': 1, 'verbosity': 0,
'objective': 'binary:logistic', 'eval_metric': ['auc']} 'objective': 'binary:logistic', 'eval_metric': ['auc']}
cv = xgb.cv(params, dm, num_boost_round=10, nfold=10, cv = xgb.cv(params, dm, num_boost_round=10, nfold=10,
as_pandas=True, early_stopping_rounds=1) as_pandas=True, early_stopping_rounds=1)
@ -163,19 +163,19 @@ class TestPandas(unittest.TestCase):
assert 'auc' in cv.columns[0] assert 'auc' in cv.columns[0]
assert cv.shape[0] < 10 assert cv.shape[0] < 10
params = {'max_depth': 2, 'eta': 1, 'silent': 1, params = {'max_depth': 2, 'eta': 1, 'verbosity': 0,
'objective': 'binary:logistic'} 'objective': 'binary:logistic'}
cv = xgb.cv(params, dm, num_boost_round=10, nfold=10, cv = xgb.cv(params, dm, num_boost_round=10, nfold=10,
as_pandas=True, metrics='auc') as_pandas=True, metrics='auc')
assert 'auc' in cv.columns[0] assert 'auc' in cv.columns[0]
params = {'max_depth': 2, 'eta': 1, 'silent': 1, params = {'max_depth': 2, 'eta': 1, 'verbosity': 0,
'objective': 'binary:logistic'} 'objective': 'binary:logistic'}
cv = xgb.cv(params, dm, num_boost_round=10, nfold=10, cv = xgb.cv(params, dm, num_boost_round=10, nfold=10,
as_pandas=True, metrics=['auc']) as_pandas=True, metrics=['auc'])
assert 'auc' in cv.columns[0] assert 'auc' in cv.columns[0]
params = {'max_depth': 2, 'eta': 1, 'silent': 1, params = {'max_depth': 2, 'eta': 1, 'verbosity': 0,
'objective': 'binary:logistic', 'eval_metric': ['auc']} 'objective': 'binary:logistic', 'eval_metric': ['auc']}
cv = xgb.cv(params, dm, num_boost_round=10, nfold=10, cv = xgb.cv(params, dm, num_boost_round=10, nfold=10,
as_pandas=True, metrics='error') as_pandas=True, metrics='error')

3
tests/python/test_with_sklearn.py
View File

@ -603,7 +603,8 @@ def test_RFECV():
# Regression # Regression
X, y = load_boston(return_X_y=True) X, y = load_boston(return_X_y=True)
bst = xgb.XGBClassifier(booster='gblinear', learning_rate=0.1, bst = xgb.XGBClassifier(booster='gblinear', learning_rate=0.1,
n_estimators=10, n_jobs=1, objective='reg:linear', n_estimators=10, n_jobs=1,
objective='reg:squarederror',
random_state=0, verbosity=0) random_state=0, verbosity=0)
rfecv = RFECV( rfecv = RFECV(
estimator=bst, step=1, cv=3, scoring='neg_mean_squared_error') estimator=bst, step=1, cv=3, scoring='neg_mean_squared_error')