Disable parameter validation for Scikit-Learn interface. (#5167)

* Disable parameter validation for now.

Scikit-Learn passes all parameters down to XGBoost, whether they are used or
not.

* Add option `validate_parameters`.

tests/cpp/common/test_monitor.cc

@@ -30,8 +30,6 @@ TEST(Monitor, Logging) {
   run_monitor();
   output = testing::internal::GetCapturedStderr();
   ASSERT_EQ(output.size(), 0);
-
-  ConsoleLogger::Configure(Args{{"verbosity", "1"}});
 }
 }  // namespace common
 }  // namespace xgboost

tests/cpp/test_learner.cc

@@ -31,12 +31,14 @@ TEST(Learner, Basic) {
 }
 
 TEST(Learner, ParameterValidation) {
+  ConsoleLogger::Configure({{"verbosity", "2"}});
   size_t constexpr kRows = 1;
   size_t constexpr kCols = 1;
   auto pp_mat = CreateDMatrix(kRows, kCols, 0);
   auto& p_mat = *pp_mat;
 
   auto learner = std::unique_ptr<Learner>(Learner::Create({p_mat}));
+  learner->SetParam("validate_parameters", "1");
   learner->SetParam("Knock Knock", "Who's there?");
   learner->SetParam("Silence", "....");
   learner->SetParam("tree_method", "exact");
@@ -45,7 +47,7 @@ TEST(Learner, ParameterValidation) {
   learner->Configure();
   std::string output = testing::internal::GetCapturedStderr();
 
-  ASSERT_TRUE(output.find("Parameters: { Knock Knock, Silence } are not used.") != std::string::npos);
+  ASSERT_TRUE(output.find("Parameters: { Knock Knock, Silence }") != std::string::npos);
   delete pp_mat;
 }
 

tests/cpp/test_serialization.cc

@@ -334,7 +334,7 @@ TEST_F(SerializationTest, ConfigurationCount) {
   }
   ASSERT_EQ(occureences, 2);
 
-  xgboost::ConsoleLogger::Configure({{"verbosity", "1"}});
+  xgboost::ConsoleLogger::Configure({{"verbosity", "2"}});
 }
 
 TEST_F(SerializationTest, GPU_CoordDescent) {

tests/python-gpu/test_gpu_pickling.py

@@ -92,6 +92,8 @@ class TestPickling(unittest.TestCase):
         status = subprocess.call(args, env=env)
         assert status == 0
 
+        os.remove(model_path)
+
     def test_pickled_predictor(self):
         x, y = build_dataset()
         train_x = xgb.DMatrix(x, label=y)
@@ -129,6 +131,8 @@ class TestPickling(unittest.TestCase):
         status = subprocess.call(args, env=env)
         assert status == 0
 
+        os.remove(model_path)
+
     def test_predict_sklearn_pickle(self):
         x, y = build_dataset()