Reduce compile warnings (#6198)

Co-authored-by: Hyunsu Cho <chohyu01@cs.washington.edu>

include/xgboost/tree_updater.h

@@ -72,6 +72,9 @@ class TreeUpdater : public Configurable {
    */
   virtual bool UpdatePredictionCache(const DMatrix* data,
                                      HostDeviceVector<bst_float>* out_preds) {
+    // Remove unused parameter compiler warning.
+    (void) data;
+    (void) out_preds;
     return false;
   }
 

src/common/survival_util.h

@@ -240,6 +240,9 @@ namespace aft {
 template <>
 XGBOOST_DEVICE inline double
 GetLimitGradAtInfPred<NormalDistribution>(CensoringType censor_type, bool sign, double sigma) {
+  // Remove unused parameter compiler warning.
+  (void) sigma;
+
   switch (censor_type) {
   case CensoringType::kUncensored:
     return sign ? kMinGradient : kMaxGradient;
@@ -288,6 +291,10 @@ GetLimitGradAtInfPred<LogisticDistribution>(CensoringType censor_type, bool sign
 template <>
 XGBOOST_DEVICE inline double
 GetLimitHessAtInfPred<LogisticDistribution>(CensoringType censor_type, bool sign, double sigma) {
+  // Remove unused parameter compiler warning.
+  (void) sign;
+  (void) sigma;
+
   switch (censor_type) {
   case CensoringType::kUncensored:
   case CensoringType::kRightCensored:
@@ -317,6 +324,9 @@ GetLimitGradAtInfPred<ExtremeDistribution>(CensoringType censor_type, bool sign,
 template <>
 XGBOOST_DEVICE inline double
 GetLimitHessAtInfPred<ExtremeDistribution>(CensoringType censor_type, bool sign, double sigma) {
+  // Remove unused parameter compiler warning.
+  (void) sigma;
+
   switch (censor_type) {
   case CensoringType::kUncensored:
   case CensoringType::kRightCensored:

src/common/transform.h

@@ -157,7 +157,10 @@ class Transform {
     /*! \brief Dummy funtion defined when compiling for CPU.  */
     template <typename std::enable_if<!CompiledWithCuda>::type* = nullptr,
               typename... HDV>
-    void LaunchCUDA(Functor _func, HDV*... _vectors) const {
+    void LaunchCUDA(Functor _func, HDV*...) const {
+      // Remove unused parameter compiler warning.
+      (void) _func;
+
       LOG(FATAL) << "Not part of device code. WITH_CUDA: " << WITH_CUDA();
     }
 #endif  // defined(__CUDACC__)

src/linear/coordinate_common.h

@@ -220,10 +220,10 @@ class FeatureSelector {
    * \param lambda Regularisation lambda.
    * \param param  A parameter with algorithm-dependent use.
    */
-  virtual void Setup(const gbm::GBLinearModel &model,
+  virtual void Setup(const gbm::GBLinearModel &,
-                     const std::vector<GradientPair> &gpair,
+                     const std::vector<GradientPair> &,
-                     DMatrix *p_fmat,
+                     DMatrix *,
-                     float alpha, float lambda, int param) {}
+                     float , float , int ) {}
   /**
    * \brief Select next coordinate to update.
    *
@@ -250,8 +250,8 @@ class FeatureSelector {
 class CyclicFeatureSelector : public FeatureSelector {
  public:
   int NextFeature(int iteration, const gbm::GBLinearModel &model,
-                  int group_idx, const std::vector<GradientPair> &gpair,
+                  int , const std::vector<GradientPair> &,
-                  DMatrix *p_fmat, float alpha, float lambda) override {
+                  DMatrix *, float, float) override {
     return iteration % model.learner_model_param->num_feature;
   }
 };
@@ -263,8 +263,8 @@ class CyclicFeatureSelector : public FeatureSelector {
 class ShuffleFeatureSelector : public FeatureSelector {
  public:
   void Setup(const gbm::GBLinearModel &model,
-             const std::vector<GradientPair> &gpair,
+             const std::vector<GradientPair> &g,
-             DMatrix *p_fmat, float alpha, float lambda, int param) override {
+             DMatrix *, float, float, int) override {
     if (feat_index_.size() == 0) {
       feat_index_.resize(model.learner_model_param->num_feature);
       std::iota(feat_index_.begin(), feat_index_.end(), 0);
@@ -273,8 +273,8 @@ class ShuffleFeatureSelector : public FeatureSelector {
   }
 
   int NextFeature(int iteration, const gbm::GBLinearModel &model,
-                  int group_idx, const std::vector<GradientPair> &gpair,
+                  int, const std::vector<GradientPair> &,
-                  DMatrix *p_fmat, float alpha, float lambda) override {
+                  DMatrix *, float, float) override {
     return feat_index_[iteration % model.learner_model_param->num_feature];
   }
 
@@ -288,9 +288,9 @@ class ShuffleFeatureSelector : public FeatureSelector {
  */
 class RandomFeatureSelector : public FeatureSelector {
  public:
-  int NextFeature(int iteration, const gbm::GBLinearModel &model,
+  int NextFeature(int, const gbm::GBLinearModel &model,
-                  int group_idx, const std::vector<GradientPair> &gpair,
+                  int, const std::vector<GradientPair> &,
-                  DMatrix *p_fmat, float alpha, float lambda) override {
+                  DMatrix *, float, float) override {
     return common::GlobalRandom()() % model.learner_model_param->num_feature;
   }
 };
@@ -307,8 +307,8 @@ class RandomFeatureSelector : public FeatureSelector {
 class GreedyFeatureSelector : public FeatureSelector {
  public:
   void Setup(const gbm::GBLinearModel &model,
-             const std::vector<GradientPair> &gpair,
+             const std::vector<GradientPair> &,
-             DMatrix *p_fmat, float alpha, float lambda, int param) override {
+             DMatrix *, float, float, int param) override {
     top_k_ = static_cast<bst_uint>(param);
     const bst_uint ngroup = model.learner_model_param->num_output_group;
     if (param <= 0) top_k_ = std::numeric_limits<bst_uint>::max();
@@ -321,7 +321,7 @@ class GreedyFeatureSelector : public FeatureSelector {
     }
   }
 
-  int NextFeature(int iteration, const gbm::GBLinearModel &model,
+  int NextFeature(int, const gbm::GBLinearModel &model,
                   int group_idx, const std::vector<GradientPair> &gpair,
                   DMatrix *p_fmat, float alpha, float lambda) override {
     // k-th selected feature for a group
@@ -438,9 +438,9 @@ class ThriftyFeatureSelector : public FeatureSelector {
     }
   }
 
-  int NextFeature(int iteration, const gbm::GBLinearModel &model,
+  int NextFeature(int, const gbm::GBLinearModel &model,
-                  int group_idx, const std::vector<GradientPair> &gpair,
+                  int group_idx, const std::vector<GradientPair> &,
-                  DMatrix *p_fmat, float alpha, float lambda) override {
+                  DMatrix *, float, float) override {
     // k-th selected feature for a group
     auto k = counter_[group_idx]++;
     // stop after either reaching top-N or going through all the features in a group

src/objective/multiclass_obj.cu

@@ -49,6 +49,9 @@ class SoftmaxMultiClassObj : public ObjFunction {
                    const MetaInfo& info,
                    int iter,
                    HostDeviceVector<GradientPair>* out_gpair) override {
+    // Remove unused parameter compiler warning.
+    (void) iter;
+
     if (info.labels_.Size() == 0) {
       return;
     }

src/tree/tree_model.cc

@@ -787,7 +787,7 @@ void RegTree::LoadCategoricalSplit(Json const& in) {
   if (!categories_nodes.empty()) {
     last_cat_node = get<Integer const>(categories_nodes[cnt]);
   }
-  for (size_t nidx = 0; nidx < param.num_nodes; ++nidx) {
+  for (bst_node_t nidx = 0; nidx < param.num_nodes; ++nidx) {
     if (nidx == last_cat_node) {
       auto j_begin = get<Integer const>(categories_segments[cnt]);
       auto j_end = get<Integer const>(categories_sizes[cnt]) + j_begin;