Release 1.4.1 (#6874)

Fix GPU AUC.

.github/workflows/main.yml

@@ -279,7 +279,7 @@ jobs:
       run: |
         cd build/
         tar cvjf ${{ steps.extract_branch.outputs.branch }}.tar.bz2 doc_doxygen/
-        python -m awscli s3 cp ./${{ steps.extract_branch.outputs.branch }}.tar.bz2 s3://xgboost-docs/ --acl public-read
+        python -m awscli s3 cp ./${{ steps.extract_branch.outputs.branch }}.tar.bz2 s3://xgboost-docs/doxygen/ --acl public-read
       if: github.ref == 'refs/heads/master' || contains(github.ref, 'refs/heads/release_')
       env:
         AWS_ACCESS_KEY_ID: ${{ secrets.AWS_ACCESS_KEY_ID_IAM_S3_UPLOADER }}

.github/workflows/r_tests.yml

@@ -100,7 +100,7 @@ jobs:
 
     - name: Install system packages
       run: |
-        sudo apt-get update && sudo apt-get install libcurl4-openssl-dev libssl-dev libssh2-1-dev libgit2-dev
+        sudo apt-get update && sudo apt-get install libcurl4-openssl-dev libssl-dev libssh2-1-dev libgit2-dev pandoc pandoc-citeproc
 
     - name: Install dependencies
       shell: Rscript {0}

CMakeLists.txt

@@ -1,5 +1,5 @@
 cmake_minimum_required(VERSION 3.13)
-project(xgboost LANGUAGES CXX C VERSION 1.4.0)
+project(xgboost LANGUAGES CXX C VERSION 1.4.1)
 include(cmake/Utils.cmake)
 list(APPEND CMAKE_MODULE_PATH "${xgboost_SOURCE_DIR}/cmake/modules")
 cmake_policy(SET CMP0022 NEW)

R-package/DESCRIPTION

@@ -1,7 +1,7 @@
 Package: xgboost
 Type: Package
 Title: Extreme Gradient Boosting
-Version: 1.4.0.1
+Version: 1.4.1.1
 Date: 2020-08-28
 Authors@R: c(
   person("Tianqi", "Chen", role = c("aut"),

include/xgboost/version_config.h

@@ -6,6 +6,6 @@
 
 #define XGBOOST_VER_MAJOR 1
 #define XGBOOST_VER_MINOR 4
-#define XGBOOST_VER_PATCH 0
+#define XGBOOST_VER_PATCH 1
 
 #endif  // XGBOOST_VERSION_CONFIG_H_

jvm-packages/pom.xml

@@ -6,7 +6,7 @@
 
     <groupId>ml.dmlc</groupId>
     <artifactId>xgboost-jvm_2.12</artifactId>
-    <version>1.4.0</version>
+    <version>1.4.1</version>
     <packaging>pom</packaging>
     <name>XGBoost JVM Package</name>
     <description>JVM Package for XGBoost</description>

jvm-packages/xgboost4j-example/pom.xml

@@ -6,10 +6,10 @@
     <parent>
         <groupId>ml.dmlc</groupId>
         <artifactId>xgboost-jvm_2.12</artifactId>
-        <version>1.4.0</version>
+        <version>1.4.1</version>
     </parent>
     <artifactId>xgboost4j-example_2.12</artifactId>
-    <version>1.4.0</version>
+    <version>1.4.1</version>
     <packaging>jar</packaging>
     <build>
         <plugins>
@@ -26,7 +26,7 @@
         <dependency>
             <groupId>ml.dmlc</groupId>
             <artifactId>xgboost4j-spark_${scala.binary.version}</artifactId>
-            <version>1.4.0</version>
+            <version>1.4.1</version>
         </dependency>
         <dependency>
             <groupId>org.apache.spark</groupId>
@@ -37,7 +37,7 @@
         <dependency>
             <groupId>ml.dmlc</groupId>
             <artifactId>xgboost4j-flink_${scala.binary.version}</artifactId>
-            <version>1.4.0</version>
+            <version>1.4.1</version>
         </dependency>
         <dependency>
             <groupId>org.apache.commons</groupId>

jvm-packages/xgboost4j-flink/pom.xml

@@ -6,10 +6,10 @@
     <parent>
         <groupId>ml.dmlc</groupId>
         <artifactId>xgboost-jvm_2.12</artifactId>
-        <version>1.4.0</version>
+        <version>1.4.1</version>
     </parent>
     <artifactId>xgboost4j-flink_2.12</artifactId>
-    <version>1.4.0</version>
+    <version>1.4.1</version>
     <build>
         <plugins>
             <plugin>
@@ -26,7 +26,7 @@
         <dependency>
             <groupId>ml.dmlc</groupId>
             <artifactId>xgboost4j_${scala.binary.version}</artifactId>
-            <version>1.4.0</version>
+            <version>1.4.1</version>
         </dependency>
         <dependency>
             <groupId>org.apache.commons</groupId>

jvm-packages/xgboost4j-gpu/pom.xml

@@ -6,10 +6,10 @@
     <parent>
         <groupId>ml.dmlc</groupId>
         <artifactId>xgboost-jvm_2.12</artifactId>
-        <version>1.4.0</version>
+        <version>1.4.1</version>
     </parent>
     <artifactId>xgboost4j-gpu_2.12</artifactId>
-    <version>1.4.0</version>
+    <version>1.4.1</version>
     <packaging>jar</packaging>
 
     <dependencies>

jvm-packages/xgboost4j-spark-gpu/pom.xml

@@ -6,7 +6,7 @@
     <parent>
         <groupId>ml.dmlc</groupId>
         <artifactId>xgboost-jvm_2.12</artifactId>
-        <version>1.4.0</version>
+        <version>1.4.1</version>
     </parent>
     <artifactId>xgboost4j-spark-gpu_2.12</artifactId>
     <build>
@@ -24,7 +24,7 @@
         <dependency>
             <groupId>ml.dmlc</groupId>
             <artifactId>xgboost4j-gpu_${scala.binary.version}</artifactId>
-            <version>1.4.0</version>
+            <version>1.4.1</version>
         </dependency>
         <dependency>
             <groupId>org.apache.spark</groupId>

jvm-packages/xgboost4j-spark/pom.xml

@@ -6,7 +6,7 @@
     <parent>
         <groupId>ml.dmlc</groupId>
         <artifactId>xgboost-jvm_2.12</artifactId>
-        <version>1.4.0</version>
+        <version>1.4.1</version>
     </parent>
     <artifactId>xgboost4j-spark_2.12</artifactId>
     <build>
@@ -24,7 +24,7 @@
         <dependency>
             <groupId>ml.dmlc</groupId>
             <artifactId>xgboost4j_${scala.binary.version}</artifactId>
-            <version>1.4.0</version>
+            <version>1.4.1</version>
         </dependency>
         <dependency>
             <groupId>org.apache.spark</groupId>

jvm-packages/xgboost4j/pom.xml

@@ -6,10 +6,10 @@
     <parent>
         <groupId>ml.dmlc</groupId>
         <artifactId>xgboost-jvm_2.12</artifactId>
-        <version>1.4.0</version>
+        <version>1.4.1</version>
     </parent>
     <artifactId>xgboost4j_2.12</artifactId>
-    <version>1.4.0</version>
+    <version>1.4.1</version>
     <packaging>jar</packaging>
 
     <dependencies>

python-package/xgboost/VERSION

@@ -1 +1 @@
-1.4.0
+1.4.1

src/common/device_helpers.cuh

@@ -1321,15 +1321,16 @@ void ArgSort(xgboost::common::Span<U> keys, xgboost::common::Span<IdxT> sorted_i
   TemporaryArray<KeyT> out(keys.size());
   cub::DoubleBuffer<KeyT> d_keys(const_cast<KeyT *>(keys.data()),
                                  out.data().get());
+  TemporaryArray<IdxT> sorted_idx_out(sorted_idx.size());
   cub::DoubleBuffer<ValueT> d_values(const_cast<ValueT *>(sorted_idx.data()),
-                                     sorted_idx.data());
+                                     sorted_idx_out.data().get());
 
   if (accending) {
     void *d_temp_storage = nullptr;
     safe_cuda((cub::DispatchRadixSort<false, KeyT, ValueT, size_t>::Dispatch(
         d_temp_storage, bytes, d_keys, d_values, sorted_idx.size(), 0,
         sizeof(KeyT) * 8, false, nullptr, false)));
-    dh::TemporaryArray<char> storage(bytes);
+    TemporaryArray<char> storage(bytes);
     d_temp_storage = storage.data().get();
     safe_cuda((cub::DispatchRadixSort<false, KeyT, ValueT, size_t>::Dispatch(
         d_temp_storage, bytes, d_keys, d_values, sorted_idx.size(), 0,
@@ -1339,12 +1340,15 @@ void ArgSort(xgboost::common::Span<U> keys, xgboost::common::Span<IdxT> sorted_i
     safe_cuda((cub::DispatchRadixSort<true, KeyT, ValueT, size_t>::Dispatch(
         d_temp_storage, bytes, d_keys, d_values, sorted_idx.size(), 0,
         sizeof(KeyT) * 8, false, nullptr, false)));
-    dh::TemporaryArray<char> storage(bytes);
+    TemporaryArray<char> storage(bytes);
     d_temp_storage = storage.data().get();
     safe_cuda((cub::DispatchRadixSort<true, KeyT, ValueT, size_t>::Dispatch(
         d_temp_storage, bytes, d_keys, d_values, sorted_idx.size(), 0,
         sizeof(KeyT) * 8, false, nullptr, false)));
   }
+
+  safe_cuda(cudaMemcpyAsync(sorted_idx.data(), sorted_idx_out.data().get(),
+                            sorted_idx.size_bytes(), cudaMemcpyDeviceToDevice));
 }
 
 namespace detail {
@@ -1379,14 +1383,19 @@ void SegmentedArgSort(xgboost::common::Span<U> values,
   size_t bytes = 0;
   Iota(sorted_idx);
   TemporaryArray<std::remove_const_t<U>> values_out(values.size());
+  TemporaryArray<std::remove_const_t<IdxT>> sorted_idx_out(sorted_idx.size());
+
   detail::DeviceSegmentedRadixSortPair<!accending>(
       nullptr, bytes, values.data(), values_out.data().get(), sorted_idx.data(),
-      sorted_idx.data(), sorted_idx.size(), n_groups, group_ptr.data(),
+      sorted_idx_out.data().get(), sorted_idx.size(), n_groups, group_ptr.data(),
       group_ptr.data() + 1);
-  dh::TemporaryArray<xgboost::common::byte> temp_storage(bytes);
+  TemporaryArray<xgboost::common::byte> temp_storage(bytes);
   detail::DeviceSegmentedRadixSortPair<!accending>(
       temp_storage.data().get(), bytes, values.data(), values_out.data().get(),
-      sorted_idx.data(), sorted_idx.data(), sorted_idx.size(), n_groups,
+      sorted_idx.data(), sorted_idx_out.data().get(), sorted_idx.size(),
-      group_ptr.data(), group_ptr.data() + 1);
+      n_groups, group_ptr.data(), group_ptr.data() + 1);
+
+  safe_cuda(cudaMemcpyAsync(sorted_idx.data(), sorted_idx_out.data().get(),
+                            sorted_idx.size_bytes(), cudaMemcpyDeviceToDevice));
 }
 }  // namespace dh

src/metric/auc.cu

@@ -269,7 +269,7 @@ float GPUMultiClassAUCOVR(common::Span<float const> predts, MetaInfo const &info
       });
 
   // unique values are sparse, so we need a CSR style indptr
-  dh::TemporaryArray<uint32_t> unique_class_ptr(class_ptr.size() + 1);
+  dh::TemporaryArray<uint32_t> unique_class_ptr(class_ptr.size());
   auto d_unique_class_ptr = dh::ToSpan(unique_class_ptr);
   auto n_uniques = dh::SegmentedUniqueByKey(
       thrust::cuda::par(alloc),