diff --git a/src/mlpack/tests/CMakeLists.txt b/src/mlpack/tests/CMakeLists.txt
index d3219b9fbc..b300115ced 100644
--- a/src/mlpack/tests/CMakeLists.txt
+++ b/src/mlpack/tests/CMakeLists.txt
@@ -14,7 +14,6 @@ add_executable(mlpack_test
   async_learning_test.cpp
   augmented_rnns_tasks_test.cpp
   bias_svd_test.cpp
-  binarize_test.cpp
   block_krylov_svd_test.cpp
   callback_test.cpp
   cf_test.cpp
@@ -41,7 +40,6 @@ add_executable(mlpack_test
   hoeffding_tree_test.cpp
   hpt_test.cpp
   hyperplane_test.cpp
-  imputation_test.cpp
   init_rules_test.cpp
   kde_test.cpp
   kernel_pca_test.cpp
@@ -94,7 +92,6 @@ add_executable(mlpack_test
   regularized_svd_test.cpp
   reward_clipping_test.cpp
   rl_components_test.cpp
-  scaling_test.cpp
   serialization.cpp
   serialization.hpp
   serialization_test.cpp
@@ -104,7 +101,6 @@ add_executable(mlpack_test
   sparse_autoencoder_test.cpp
   sparse_coding_test.cpp
   spill_tree_test.cpp
-  split_data_test.cpp
   string_encoding_test.cpp
   sumtree_test.cpp
   svd_batch_test.cpp
@@ -156,10 +152,6 @@ add_executable(mlpack_test
   main_tests/nmf_test.cpp
   main_tests/pca_test.cpp
   main_tests/perceptron_test.cpp
-  main_tests/preprocess_binarize_test.cpp
-  main_tests/preprocess_imputer_test.cpp
-  main_tests/preprocess_scale_test.cpp
-  main_tests/preprocess_split_test.cpp
   main_tests/radical_test.cpp
   main_tests/random_forest_test.cpp
   main_tests/range_search_test.cpp
@@ -173,8 +165,16 @@ add_executable(mlpack_catch_test
   serialization_catch.cpp
   serialization_catch.hpp
   test_catch_tools.hpp
+  binarize_test.cpp
   image_load_test.cpp
+  imputation_test.cpp
+  scaling_test.cpp
+  split_data_test.cpp
   main_tests/image_converter_test.cpp
+  main_tests/preprocess_binarize_test.cpp
+  main_tests/preprocess_imputer_test.cpp
+  main_tests/preprocess_scale_test.cpp
+  main_tests/preprocess_split_test.cpp
   main_tests/test_helper.hpp
 )
 
diff --git a/src/mlpack/tests/binarize_test.cpp b/src/mlpack/tests/binarize_test.cpp
index 5d25028be0..e45781752c 100644
--- a/src/mlpack/tests/binarize_test.cpp
+++ b/src/mlpack/tests/binarize_test.cpp
@@ -13,16 +13,14 @@
 #include <mlpack/core/data/binarize.hpp>
 #include <mlpack/core/math/random.hpp>
 
-#include <boost/test/unit_test.hpp>
-#include "test_tools.hpp"
+#include "test_catch_tools.hpp"
+#include "catch.hpp"
 
 using namespace mlpack;
 using namespace arma;
 using namespace mlpack::data;
 
-BOOST_AUTO_TEST_SUITE(BinarizeTest);
-
-BOOST_AUTO_TEST_CASE(BinerizeOneDimension)
+TEST_CASE("BinerizeOneDimension", "[BinarizeTest]")
 {
   mat input;
   input << 1 << 2 << 3 << endr
@@ -34,18 +32,18 @@ BOOST_AUTO_TEST_CASE(BinerizeOneDimension)
   const size_t dimension = 1;
   Binarize<double>(input, output, threshold, dimension);
 
-  BOOST_REQUIRE_CLOSE(output(0, 0), 1, 1e-5); // 1
-  BOOST_REQUIRE_CLOSE(output(0, 1), 2, 1e-5); // 2
-  BOOST_REQUIRE_CLOSE(output(0, 2), 3, 1e-5); // 3
-  BOOST_REQUIRE_SMALL(output(1, 0), 1e-5); // 4 target
-  BOOST_REQUIRE_SMALL(output(1, 1), 1e-5); // 5 target
-  BOOST_REQUIRE_CLOSE(output(1, 2), 1, 1e-5); // 6 target
-  BOOST_REQUIRE_CLOSE(output(2, 0), 7, 1e-5); // 7
-  BOOST_REQUIRE_CLOSE(output(2, 1), 8, 1e-5); // 8
-  BOOST_REQUIRE_CLOSE(output(2, 2), 9, 1e-5); // 9
+  REQUIRE(output(0, 0)== Approx(1.0).epsilon(1e-5 / 100)); // 1
+  REQUIRE(output(0, 1)== Approx(2.0).epsilon(1e-5 / 100)); // 2
+  REQUIRE(output(0, 2)== Approx(3.0).epsilon(1e-5 / 100)); // 3 
+  REQUIRE(output(1, 0) == Approx(0.0).margin(1e-5)); // 4 target
+  REQUIRE(output(1, 1) == Approx(0.0).margin(1e-5)); // 5 target
+  REQUIRE(output(1, 2)== Approx(1.0).epsilon(1e-5 / 100)); // 6 target
+  REQUIRE(output(2, 0)== Approx(7.0).epsilon(1e-5 / 100)); // 7
+  REQUIRE(output(2, 1)== Approx(8.0).epsilon(1e-5 / 100)); // 8
+  REQUIRE(output(2, 2)== Approx(9.0).epsilon(1e-5 / 100)); // 9
 }
 
-BOOST_AUTO_TEST_CASE(BinerizeAll)
+TEST_CASE("BinerizeAll", "[BinarizeTest]")
 {
   mat input;
   input << 1 << 2 << 3 << endr
@@ -57,15 +55,13 @@ BOOST_AUTO_TEST_CASE(BinerizeAll)
 
   Binarize<double>(input, output, threshold);
 
-  BOOST_REQUIRE_SMALL(output(0, 0), 1e-5); // 1
-  BOOST_REQUIRE_SMALL(output(0, 1), 1e-5); // 2
-  BOOST_REQUIRE_SMALL(output(0, 2), 1e-5); // 3
-  BOOST_REQUIRE_SMALL(output(1, 0), 1e-5); // 4
-  BOOST_REQUIRE_SMALL(output(1, 1), 1e-5); // 5
-  BOOST_REQUIRE_CLOSE(output(1, 2), 1.0, 1e-5); // 6
-  BOOST_REQUIRE_CLOSE(output(2, 0), 1.0, 1e-5); // 7
-  BOOST_REQUIRE_CLOSE(output(2, 1), 1.0, 1e-5); // 8
-  BOOST_REQUIRE_CLOSE(output(2, 2), 1.0, 1e-5); // 9
-}
-
-BOOST_AUTO_TEST_SUITE_END();
+  REQUIRE(output(0, 0) == Approx(0.0).margin(1e-5)); // 1
+  REQUIRE(output(0, 1) == Approx(0.0).margin(1e-5)); // 2
+  REQUIRE(output(0, 2) == Approx(0.0).margin(1e-5)); // 3 
+  REQUIRE(output(1, 0) == Approx(0.0).margin(1e-5)); //4
+  REQUIRE(output(1, 1) == Approx(0.0).margin(1e-5)); // 5
+  REQUIRE(output(1, 2)== Approx(1.0).epsilon(1e-5 / 100)); // 6
+  REQUIRE(output(2, 0)== Approx(1.0).epsilon(1e-5 / 100)); // 7
+  REQUIRE(output(2, 1)== Approx(1.0).epsilon(1e-5 / 100)); // 8
+  REQUIRE(output(2, 2)== Approx(1.0).epsilon(1e-5 / 100)); // 9
+ }
diff --git a/src/mlpack/tests/imputation_test.cpp b/src/mlpack/tests/imputation_test.cpp
index 847330530a..13d148b1ea 100644
--- a/src/mlpack/tests/imputation_test.cpp
+++ b/src/mlpack/tests/imputation_test.cpp
@@ -22,21 +22,20 @@
 #include <mlpack/core/data/imputation_methods/mean_imputation.hpp>
 #include <mlpack/core/data/imputation_methods/median_imputation.hpp>
 
-#include <boost/test/unit_test.hpp>
-#include "test_tools.hpp"
+#include "test_catch_tools.hpp"
+#include "catch.hpp"
 
 using namespace mlpack;
 using namespace mlpack::data;
 using namespace std;
 
-BOOST_AUTO_TEST_SUITE(ImputationTest);
 /**
  * 1. Make sure a CSV is loaded correctly with mappings using MissingPolicy.
  * 2. Try Imputer object with CustomImputation method to impute data "a".
  * (It is ok to test on one method since the other ones will be covered in the
  * next cases).
  */
-BOOST_AUTO_TEST_CASE(DatasetMapperImputerTest)
+TEST_CASE("DatasetMapperImputerTest", "[ImputationTest]")
 {
   fstream f;
   f.open("test_file.csv", fstream::out);
@@ -48,23 +47,23 @@ BOOST_AUTO_TEST_CASE(DatasetMapperImputerTest)
   arma::mat input;
   MissingPolicy policy({"a"});
   DatasetMapper<MissingPolicy> info(policy);
-  BOOST_REQUIRE(data::Load("test_file.csv", input, info) == true);
+  REQUIRE(data::Load("test_file.csv", input, info) == true);
 
   // row and column test.
-  BOOST_REQUIRE_EQUAL(input.n_rows, 3);
-  BOOST_REQUIRE_EQUAL(input.n_cols, 3);
+  REQUIRE(input.n_rows == 3);
+  REQUIRE(input.n_cols == 3);
 
   // Load check
   // MissingPolicy should convert strings to nans.
-  BOOST_REQUIRE(std::isnan(input(0, 0)) == true);
-  BOOST_REQUIRE_CLOSE(input(0, 1), 5.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(input(0, 2), 8.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(input(1, 0), 2.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(input(1, 1), 6.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(input(1, 2), 9.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(input(2, 0), 3.0, 1e-5);
-  BOOST_REQUIRE(std::isnan(input(2, 1)) == true);
-  BOOST_REQUIRE_CLOSE(input(2, 2), 10.0, 1e-5);
+  REQUIRE(std::isnan(input(0, 0)) == true);
+  REQUIRE(input(0, 1) == Approx(5.0).epsilon(1e-5 / 100));
+  REQUIRE(input(0, 2) == Approx(8.0).epsilon(1e-5 / 100));
+  REQUIRE(input(1, 0) == Approx(2.0).epsilon(1e-5 / 100));
+  REQUIRE(input(1, 1) == Approx(6.0).epsilon(1e-5 / 100));
+  REQUIRE(input(1, 2) == Approx(9.0).epsilon(1e-5 / 100));
+  REQUIRE(input(2, 0) == Approx(3.0).epsilon(1e-5 / 100));
+  REQUIRE(std::isnan(input(2, 1)) == true);
+  REQUIRE(input(2, 2) == Approx(10.0).epsilon(1e-5 / 100));
 
   // convert missing vals to 99.
   CustomImputation<double> customStrategy(99);
@@ -75,15 +74,15 @@ BOOST_AUTO_TEST_CASE(DatasetMapperImputerTest)
   imputer.Impute(input, "a", 0);
 
   // Custom imputation result check.
-  BOOST_REQUIRE_CLOSE(input(0, 0), 99.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(input(0, 1), 5.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(input(0, 2), 8.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(input(1, 0), 2.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(input(1, 1), 6.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(input(1, 2), 9.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(input(2, 0), 3.0, 1e-5);
-  BOOST_REQUIRE(std::isnan(input(2, 1)) == true); // remains as NaN
-  BOOST_REQUIRE_CLOSE(input(2, 2), 10.0, 1e-5);
+  REQUIRE(input(0, 0) == Approx(99.0).epsilon(1e-5 / 100));
+  REQUIRE(input(0, 1) == Approx(5.0).epsilon(1e-5 / 100));
+  REQUIRE(input(0, 2) == Approx(8.0).epsilon(1e-5 / 100));
+  REQUIRE(input(1, 0) == Approx(2.0).epsilon(1e-5 / 100));
+  REQUIRE(input(1, 1) == Approx(6.0).epsilon(1e-5 / 100));
+  REQUIRE(input(1, 2) == Approx(9.0).epsilon(1e-5 / 100));
+  REQUIRE(input(2, 0) == Approx(3.0).epsilon(1e-5 / 100));
+  REQUIRE(std::isnan(input(2, 1)) == true); // remains as NaN
+  REQUIRE(input(2, 2) == Approx(10.0).epsilon(1e-5 / 100));
 
   // Remove the file.
   remove("test_file.csv");
@@ -92,7 +91,7 @@ BOOST_AUTO_TEST_CASE(DatasetMapperImputerTest)
 /**
  * Make sure CustomImputation method replaces data 0 to 99.
  */
-BOOST_AUTO_TEST_CASE(CustomImputationTest)
+TEST_CASE("CustomImputationTest", "[ImputationTest]")
 {
   arma::mat columnWiseInput("3.0 0.0 2.0 0.0;"
                   "5.0 6.0 0.0 6.0;"
@@ -106,41 +105,41 @@ BOOST_AUTO_TEST_CASE(CustomImputationTest)
   // column wise
   imputer.Impute(columnWiseInput, mappedValue, 0/*dimension*/, true);
 
-  BOOST_REQUIRE_CLOSE(columnWiseInput(0, 0), 3.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(0, 1), 99.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(0, 2), 2.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(0, 3), 99.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(1, 0), 5.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(1, 1), 6.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(1, 2), 0.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(1, 3), 6.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(2, 0), 9.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(2, 1), 8.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(2, 2), 4.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(2, 3), 8.0, 1e-5);
+  REQUIRE(columnWiseInput(0, 0) == Approx(3.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(0, 1) == Approx(99.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(0, 2) == Approx(2.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(0, 3) == Approx(99.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(1, 0) == Approx(5.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(1, 1) == Approx(6.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(1, 2) == Approx(0.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(1, 3) == Approx(6.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(2, 0) == Approx(9.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(2, 1) == Approx(8.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(2, 2) == Approx(4.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(2, 3) == Approx(8.0).epsilon(1e-5 / 100));
 
   // row wise
   imputer.Impute(rowWiseInput, mappedValue, 1, false);
 
-  BOOST_REQUIRE_CLOSE(rowWiseInput(0, 0), 3.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(0, 1), 99.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(0, 2), 2.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(0, 3), 0.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(1, 0), 5.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(1, 1), 6.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(1, 2), 0.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(1, 3), 6.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(2, 0), 9.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(2, 1), 8.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(2, 2), 4.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(2, 3), 8.0, 1e-5);
+  REQUIRE(rowWiseInput(0, 0) == Approx(3.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(0, 1) == Approx(99.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(0, 2) == Approx(2.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(0, 3) == Approx(0.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(1, 0) == Approx(5.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(1, 1) == Approx(6.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(1, 2) == Approx(0.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(1, 3) == Approx(6.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(2, 0) == Approx(9.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(2, 1) == Approx(8.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(2, 2) == Approx(4.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(2, 3) == Approx(8.0).epsilon(1e-5 / 100));
 }
 
 /**
  * Make sure MeanImputation method replaces data 0 to mean value of each
  * dimensions.
  */
-BOOST_AUTO_TEST_CASE(MeanImputationTest)
+TEST_CASE("MeanImputationTest", "[ImputationTest]")
 {
   arma::mat columnWiseInput("3.0 0.0 2.0 0.0;"
                   "5.0 6.0 0.0 6.0;"
@@ -153,41 +152,41 @@ BOOST_AUTO_TEST_CASE(MeanImputationTest)
   // column wise
   imputer.Impute(columnWiseInput, mappedValue, 0, true);
 
-  BOOST_REQUIRE_CLOSE(columnWiseInput(0, 0), 3.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(0, 1), 2.5, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(0, 2), 2.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(0, 3), 2.5, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(1, 0), 5.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(1, 1), 6.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(1, 2), 0.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(1, 3), 6.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(2, 0), 9.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(2, 1), 8.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(2, 2), 4.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(2, 3), 8.0, 1e-5);
+  REQUIRE(columnWiseInput(0, 0) == Approx(3.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(0, 1) == Approx(2.5).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(0, 2) == Approx(2.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(0, 3) == Approx(2.5).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(1, 0) == Approx(5.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(1, 1) == Approx(6.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(1, 2) == Approx(0.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(1, 3) == Approx(6.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(2, 0) == Approx(9.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(2, 1) == Approx(8.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(2, 2) == Approx(4.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(2, 3) == Approx(8.0).epsilon(1e-5 / 100));
 
   // row wise
   imputer.Impute(rowWiseInput, mappedValue, 1, false);
 
-  BOOST_REQUIRE_CLOSE(rowWiseInput(0, 0), 3.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(0, 1), 7.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(0, 2), 2.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(0, 3), 0.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(1, 0), 5.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(1, 1), 6.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(1, 2), 0.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(1, 3), 6.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(2, 0), 9.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(2, 1), 8.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(2, 2), 4.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(2, 3), 8.0, 1e-5);
+  REQUIRE(rowWiseInput(0, 0) == Approx(3.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(0, 1) == Approx(7.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(0, 2) == Approx(2.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(0, 3) == Approx(0.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(1, 0) == Approx(5.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(1, 1) == Approx(6.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(1, 2) == Approx(0.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(1, 3) == Approx(6.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(2, 0) == Approx(9.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(2, 1) == Approx(8.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(2, 2) == Approx(4.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(2, 3) == Approx(8.0).epsilon(1e-5 / 100));
 }
 
 /**
  * Make sure MedianImputation method replaces data 0 to median value of each
  * dimensions.
  */
-BOOST_AUTO_TEST_CASE(MedianImputationTest)
+TEST_CASE("MedianImputationTest", "[ImputationTest]")
 {
   arma::mat columnWiseInput("3.0 0.0 2.0 0.0;"
                   "5.0 6.0 0.0 6.0;"
@@ -200,41 +199,41 @@ BOOST_AUTO_TEST_CASE(MedianImputationTest)
   // column wise
   imputer.Impute(columnWiseInput, mappedValue, 1, true);
 
-  BOOST_REQUIRE_CLOSE(columnWiseInput(0, 0), 3.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(0, 1), 0.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(0, 2), 2.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(0, 3), 0.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(1, 0), 5.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(1, 1), 6.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(1, 2), 6.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(1, 3), 6.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(2, 0), 9.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(2, 1), 8.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(2, 2), 4.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(2, 3), 8.0, 1e-5);
+  REQUIRE(columnWiseInput(0, 0) == Approx(3.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(0, 1) == Approx(0.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(0, 2) == Approx(2.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(0, 3) == Approx(0.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(1, 0) == Approx(5.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(1, 1) == Approx(6.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(1, 2) == Approx(6.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(1, 3) == Approx(6.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(2, 0) == Approx(9.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(2, 1) == Approx(8.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(2, 2) == Approx(4.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(2, 3) == Approx(8.0).epsilon(1e-5 / 100));
 
   // row wise
   imputer.Impute(rowWiseInput, mappedValue, 1, false);
 
-  BOOST_REQUIRE_CLOSE(rowWiseInput(0, 0), 3.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(0, 1), 7.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(0, 2), 2.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(0, 3), 0.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(1, 0), 5.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(1, 1), 6.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(1, 2), 0.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(1, 3), 6.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(2, 0), 9.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(2, 1), 8.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(2, 2), 4.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(2, 3), 8.0, 1e-5);
+  REQUIRE(rowWiseInput(0, 0) == Approx(3.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(0, 1) == Approx(7.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(0, 2) == Approx(2.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(0, 3) == Approx(0.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(1, 0) == Approx(5.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(1, 1) == Approx(6.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(1, 2) == Approx(0.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(1, 3) == Approx(6.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(2, 0) == Approx(9.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(2, 1) == Approx(8.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(2, 2) == Approx(4.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(2, 3) == Approx(8.0).epsilon(1e-5 / 100));
 }
 
 /**
  * Make sure ListwiseDeletion method deletes the whole column (if column wise)
  * or the row (if row wise) containing value of 0.
  */
-BOOST_AUTO_TEST_CASE(ListwiseDeletionTest)
+TEST_CASE("ListwiseDeletionTest", "[ImputationTest]")
 {
   arma::mat columnWiseInput("3.0 0.0 2.0 0.0;"
                   "5.0 6.0 0.0 6.0;"
@@ -247,30 +246,30 @@ BOOST_AUTO_TEST_CASE(ListwiseDeletionTest)
   // column wise
   imputer.Impute(columnWiseInput, mappedValue, 0, true); // column wise
 
-  BOOST_REQUIRE_CLOSE(columnWiseInput(0, 0), 3.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(0, 1), 2.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(1, 0), 5.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(1, 1), 0.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(2, 0), 9.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(columnWiseInput(2, 1), 4.0, 1e-5);
+  REQUIRE(columnWiseInput(0, 0) == Approx(3.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(0, 1) == Approx(2.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(1, 0) == Approx(5.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(1, 1) == Approx(0.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(2, 0) == Approx(9.0).epsilon(1e-5 / 100));
+  REQUIRE(columnWiseInput(2, 1) == Approx(4.0).epsilon(1e-5 / 100));
 
   // row wise
   imputer.Impute(rowWiseInput, mappedValue, 1, false); // row wise
 
-  BOOST_REQUIRE_CLOSE(rowWiseInput(0, 0), 5.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(0, 1), 6.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(0, 2), 0.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(0, 3), 6.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(1, 0), 9.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(1, 1), 8.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(1, 2), 4.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(rowWiseInput(1, 3), 8.0, 1e-5);
+  REQUIRE(rowWiseInput(0, 0) == Approx(5.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(0, 1) == Approx(6.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(0, 2) == Approx(0.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(0, 3) == Approx(6.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(1, 0) == Approx(9.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(1, 1) == Approx(8.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(1, 2) == Approx(4.0).epsilon(1e-5 / 100));
+  REQUIRE(rowWiseInput(1, 3) == Approx(8.0).epsilon(1e-5 / 100));
 }
 
 /**
  * Make sure we can map non-strings.
  */
-BOOST_AUTO_TEST_CASE(DatasetMapperNonStringMapping)
+TEST_CASE("DatasetMapperNonStringMapping", "[ImputationTest]")
 {
   IncrementPolicy incr(true);
   DatasetMapper<IncrementPolicy, double> dm(incr, 1);
@@ -278,23 +277,23 @@ BOOST_AUTO_TEST_CASE(DatasetMapperNonStringMapping)
   dm.MapString<size_t>(4.3, 0);
   dm.MapString<size_t>(1.1, 0);
 
-  BOOST_REQUIRE_EQUAL(dm.NumMappings(0), 3);
+  REQUIRE(dm.NumMappings(0) == 3);
 
-  BOOST_REQUIRE(dm.Type(0) == data::Datatype::categorical);
+  REQUIRE(dm.Type(0) == data::Datatype::categorical);
 
-  BOOST_REQUIRE_EQUAL(dm.UnmapValue(5.0, 0), 0);
-  BOOST_REQUIRE_EQUAL(dm.UnmapValue(4.3, 0), 1);
-  BOOST_REQUIRE_EQUAL(dm.UnmapValue(1.1, 0), 2);
+  REQUIRE(dm.UnmapValue(5.0, 0) == 0);
+  REQUIRE(dm.UnmapValue(4.3, 0) == 1);
+  REQUIRE(dm.UnmapValue(1.1, 0) == 2);
 
-  BOOST_REQUIRE_EQUAL(dm.UnmapString(0, 0), 5.0);
-  BOOST_REQUIRE_EQUAL(dm.UnmapString(1, 0), 4.3);
-  BOOST_REQUIRE_EQUAL(dm.UnmapString(2, 0), 1.1);
+  REQUIRE(dm.UnmapString(0, 0) == 5.0);
+  REQUIRE(dm.UnmapString(1, 0) == 4.3);
+  REQUIRE(dm.UnmapString(2, 0) == 1.1);
 }
 
 /**
  * Make sure we can map strange types.
  */
-BOOST_AUTO_TEST_CASE(DatasetMapperPointerMapping)
+TEST_CASE("DatasetMapperPointerMapping", "[ImputationTest]")
 {
   int a = 1, b = 2, c = 3;
   IncrementPolicy incr(true);
@@ -304,15 +303,13 @@ BOOST_AUTO_TEST_CASE(DatasetMapperPointerMapping)
   dm.MapString<size_t>(&b, 0);
   dm.MapString<size_t>(&c, 0);
 
-  BOOST_REQUIRE_EQUAL(dm.NumMappings(0), 3);
+  REQUIRE(dm.NumMappings(0) == 3);
 
-  BOOST_REQUIRE_EQUAL(dm.UnmapValue(&a, 0), 0);
-  BOOST_REQUIRE_EQUAL(dm.UnmapValue(&b, 0), 1);
-  BOOST_REQUIRE_EQUAL(dm.UnmapValue(&c, 0), 2);
+  REQUIRE(dm.UnmapValue(&a, 0) == 0);
+  REQUIRE(dm.UnmapValue(&b, 0) == 1);
+  REQUIRE(dm.UnmapValue(&c, 0) == 2);
 
-  BOOST_REQUIRE_EQUAL(dm.UnmapString(0, 0), &a);
-  BOOST_REQUIRE_EQUAL(dm.UnmapString(1, 0), &b);
-  BOOST_REQUIRE_EQUAL(dm.UnmapString(2, 0), &c);
+  REQUIRE(dm.UnmapString(0, 0) == &a);
+  REQUIRE(dm.UnmapString(1, 0) == &b);
+  REQUIRE(dm.UnmapString(2, 0) == &c);
 }
-
-BOOST_AUTO_TEST_SUITE_END();
diff --git a/src/mlpack/tests/main_tests/preprocess_binarize_test.cpp b/src/mlpack/tests/main_tests/preprocess_binarize_test.cpp
index 3bb1106c7b..e1e9d38860 100644
--- a/src/mlpack/tests/main_tests/preprocess_binarize_test.cpp
+++ b/src/mlpack/tests/main_tests/preprocess_binarize_test.cpp
@@ -18,8 +18,8 @@ static const std::string testName = "PreprocessBinarize";
 #include <mlpack/methods/preprocess/preprocess_binarize_main.cpp>
 
 #include "test_helper.hpp"
-#include <boost/test/unit_test.hpp>
-#include "../test_tools.hpp"
+#include "../test_catch_tools.hpp"
+#include "../catch.hpp"
 
 using namespace mlpack;
 
@@ -40,13 +40,12 @@ struct PreprocessBinarizeTestFixture
   }
 };
 
-BOOST_FIXTURE_TEST_SUITE(PreprocessBinarizeMainTest,
-                         PreprocessBinarizeTestFixture);
-
 /**
  * Check that input and output have same dimensions.
  */
-BOOST_AUTO_TEST_CASE(PreprocessBinarizeDimensionTest)
+TEST_CASE_METHOD(
+    PreprocessBinarizeTestFixture, "PreprocessBinarizeDimensionTest",
+    "PreprocessBinarizeMainTest")
 {
   // Create a synthetic dataset.
   arma::mat inputData = arma::randu<arma::mat>(2, 5);
@@ -62,14 +61,16 @@ BOOST_AUTO_TEST_CASE(PreprocessBinarizeDimensionTest)
   mlpackMain();
 
   // Now check that the output has desired dimensions.
-  BOOST_REQUIRE_EQUAL(IO::GetParam<arma::mat>("output").n_rows, 2);
-  BOOST_REQUIRE_EQUAL(IO::GetParam<arma::mat>("output").n_cols, inputSize);
+  REQUIRE(IO::GetParam<arma::mat>("output").n_rows == 2);
+  REQUIRE(IO::GetParam<arma::mat>("output").n_cols == inputSize);
 }
 
 /**
  * Check that specified dimension is non-negative.
  */
-BOOST_AUTO_TEST_CASE(PreprocessBinarizeNegativeDimensionTest)
+TEST_CASE_METHOD(
+    PreprocessBinarizeTestFixture, "PreprocessBinarizeNegativeDimensionTest",
+    "PreprocessBinarizeMainTest")
 {
   arma::mat inputData = arma::randu<arma::mat>(2, 2);
 
@@ -78,14 +79,16 @@ BOOST_AUTO_TEST_CASE(PreprocessBinarizeNegativeDimensionTest)
   SetInputParam("dimension", (int) -2); // Invalid.
 
   Log::Fatal.ignoreInput = true;
-  BOOST_REQUIRE_THROW(mlpackMain(), std::runtime_error);
+  REQUIRE_THROWS_AS(mlpackMain(), std::runtime_error);
   Log::Fatal.ignoreInput = false;
 }
 
 /**
  * Check that we can't specify a dimension larger than input.
  */
-BOOST_AUTO_TEST_CASE(PreprocessBinarizelargerDimensionTest)
+TEST_CASE_METHOD(
+    PreprocessBinarizeTestFixture, "PreprocessBinarizelargerDimensionTest",
+    "PreprocessBinarizeMainTest")
 {
   arma::mat inputData = arma::randu<arma::mat>(2, 2);
 
@@ -94,14 +97,16 @@ BOOST_AUTO_TEST_CASE(PreprocessBinarizelargerDimensionTest)
   SetInputParam("dimension", (int) 6); // Invalid.
 
   Log::Fatal.ignoreInput = true;
-  BOOST_REQUIRE_THROW(mlpackMain(), std::runtime_error);
+  REQUIRE_THROWS_AS(mlpackMain(), std::runtime_error);
   Log::Fatal.ignoreInput = false;
 }
 
 /**
  * Check that binarization took place for the specified dimension.
  */
-BOOST_AUTO_TEST_CASE(PreprocessBinarizeVerificationTest)
+TEST_CASE_METHOD(
+    PreprocessBinarizeTestFixture, "PreprocessBinarizeVerificationTest",
+    "PreprocessBinarizeMainTest")
 {
   arma::mat inputData({{7.0, 4.0, 5.0}, {2.0, 5.0, 9.0}, {7.0, 3.0, 8.0}});
 
@@ -115,25 +120,27 @@ BOOST_AUTO_TEST_CASE(PreprocessBinarizeVerificationTest)
   output = std::move(IO::GetParam<arma::mat>("output"));
 
   // All values dimension should remain unchanged.
-  BOOST_REQUIRE_CLOSE(output(0, 0), 7.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(output(0, 1), 4.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(output(0, 2), 5.0, 1e-5);
+  REQUIRE(output(0, 0)== Approx(7.0).epsilon(1e-5 / 100));
+  REQUIRE(output(0, 1)== Approx(4.0).epsilon(1e-5 / 100));
+  REQUIRE(output(0, 2)== Approx(5.0).epsilon(1e-5 / 100));
 
   // All values should be binarized according to the threshold.
-  BOOST_REQUIRE_SMALL(output(1, 0), 1e-5);
-  BOOST_REQUIRE_SMALL(output(1, 1), 1e-5);
-  BOOST_REQUIRE_CLOSE(output(1, 2), 1.0, 1e-5);
+  REQUIRE(output(1, 0) == Approx(0.0).margin(1e-5));
+  REQUIRE(output(1, 1) == Approx(0.0).margin(1e-5));
+  REQUIRE(output(1, 2)== Approx(1.0).epsilon(1e-5 / 100));
 
   // All values dimension should remain unchanged.
-  BOOST_REQUIRE_CLOSE(output(2, 0), 7.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(output(2, 1), 3.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(output(2, 2), 8.0, 1e-5);
+  REQUIRE(output(2, 0)== Approx(7.0).epsilon(1e-5 / 100));
+  REQUIRE(output(2, 1)== Approx(3.0).epsilon(1e-5 / 100));
+  REQUIRE(output(2, 2)== Approx(8.0).epsilon(1e-5 / 100));
 }
 
 /**
  * Check that all dimensions are binarized when dimension is not specified.
  */
-BOOST_AUTO_TEST_CASE(PreprocessBinarizeDimensionLessVerTest)
+TEST_CASE_METHOD(
+    PreprocessBinarizeTestFixture, "PreprocessBinarizeDimensionLessVerTest",
+    "PreprocessBinarizeMainTest")
 {
   arma::mat inputData({{7.0, 4.0, 5.0}, {2.0, 5.0, 9.0}, {7.0, 3.0, 8.0}});
 
@@ -146,15 +153,13 @@ BOOST_AUTO_TEST_CASE(PreprocessBinarizeDimensionLessVerTest)
   output = std::move(IO::GetParam<arma::mat>("output"));
 
   // All values should be binarized according to the threshold.
-  BOOST_REQUIRE_CLOSE(output(0, 0), 1.0, 1e-5);
-  BOOST_REQUIRE_SMALL(output(0, 1), 1e-5);
-  BOOST_REQUIRE_SMALL(output(0, 2), 1e-5);
-  BOOST_REQUIRE_SMALL(output(1, 0), 1e-5);
-  BOOST_REQUIRE_SMALL(output(1, 1), 1e-5);
-  BOOST_REQUIRE_CLOSE(output(1, 2), 1.0, 1e-5);
-  BOOST_REQUIRE_CLOSE(output(2, 0), 1.0, 1e-5);
-  BOOST_REQUIRE_SMALL(output(2, 1), 1e-5);
-  BOOST_REQUIRE_CLOSE(output(2, 2), 1.0, 1e-5);
+  REQUIRE(output(0, 0)== Approx(1.0).epsilon(1e-5 / 100));
+  REQUIRE(output(0, 1) == Approx(0.0).margin(1e-5));
+  REQUIRE(output(0, 2) == Approx(0.0).margin(1e-5));
+  REQUIRE(output(1, 0) == Approx(0.0).margin(1e-5));
+  REQUIRE(output(1, 1) == Approx(0.0).margin(1e-5));
+  REQUIRE(output(1, 2)== Approx(1.0).epsilon(1e-5 / 100));
+  REQUIRE(output(2, 0)== Approx(1.0).epsilon(1e-5 / 100));
+  REQUIRE(output(2, 1) == Approx(0.0).margin(1e-5));
+  REQUIRE(output(2, 2)== Approx(1.0).epsilon(1e-5 / 100));
 }
-
-BOOST_AUTO_TEST_SUITE_END();
diff --git a/src/mlpack/tests/main_tests/preprocess_imputer_test.cpp b/src/mlpack/tests/main_tests/preprocess_imputer_test.cpp
index 8645bc5466..fe03542e24 100644
--- a/src/mlpack/tests/main_tests/preprocess_imputer_test.cpp
+++ b/src/mlpack/tests/main_tests/preprocess_imputer_test.cpp
@@ -18,8 +18,8 @@ static const std::string testName = "PreprocessImputer";
 #include <mlpack/methods/preprocess/preprocess_imputer_main.cpp>
 
 #include "test_helper.hpp"
-#include <boost/test/unit_test.hpp>
-#include "../test_tools.hpp"
+#include "../test_catch_tools.hpp"
+#include "../catch.hpp"
 
 #include <cmath>
 
@@ -42,14 +42,13 @@ struct PreprocessImputerTestFixture
   }
 };
 
-BOOST_FIXTURE_TEST_SUITE(PreprocessImputerMainTest,
-                         PreprocessImputerTestFixture);
-
 /**
  * Check that input and output have same dimensions
  * except for listwise_deletion strategy.
  */
-BOOST_AUTO_TEST_CASE(PreprocessImputerDimensionTest)
+TEST_CASE_METHOD(
+    PreprocessImputerTestFixture, "PreprocessImputerDimensionTest",
+    "PreprocessImputerMainTest")
 {
   // Load synthetic dataset.
   arma::mat inputData;
@@ -73,8 +72,8 @@ BOOST_AUTO_TEST_CASE(PreprocessImputerDimensionTest)
 
   // Now check that the output has desired dimensions.
   data::Load(IO::GetParam<std::string>("output_file"), outputData);
-  BOOST_REQUIRE_EQUAL(outputData.n_cols, inputSize);
-  BOOST_REQUIRE_EQUAL(outputData.n_rows, 3); // Input Dimension.
+  REQUIRE(outputData.n_cols == inputSize);
+  REQUIRE(outputData.n_rows == 3); // Input Dimension.
 
   // Reset passed strategy.
   IO::GetSingleton().Parameters()["strategy"].wasPassed = false;
@@ -86,8 +85,8 @@ BOOST_AUTO_TEST_CASE(PreprocessImputerDimensionTest)
 
   // Now check that the output has desired dimensions.
   data::Load(IO::GetParam<std::string>("output_file"), outputData);
-  BOOST_REQUIRE_EQUAL(outputData.n_cols, inputSize);
-  BOOST_REQUIRE_EQUAL(outputData.n_rows, 3); // Input Dimension.
+  REQUIRE(outputData.n_cols == inputSize);
+  REQUIRE(outputData.n_rows == 3); // Input Dimension.
 
   // Reset passed strategy.
   IO::GetSingleton().Parameters()["strategy"].wasPassed = false;
@@ -100,14 +99,16 @@ BOOST_AUTO_TEST_CASE(PreprocessImputerDimensionTest)
 
   // Now check that the output has desired dimensions.
   data::Load(IO::GetParam<std::string>("output_file"), outputData);
-  BOOST_REQUIRE_EQUAL(outputData.n_cols, inputSize);
-  BOOST_REQUIRE_EQUAL(outputData.n_rows, 3); // Input Dimension.
+  REQUIRE(outputData.n_cols == inputSize);
+  REQUIRE(outputData.n_rows == 3); // Input Dimension.
 }
 
 /**
  * Check that output has fewer points in case of listwise_deletion strategy.
  */
-BOOST_AUTO_TEST_CASE(PreprocessImputerListwiseDimensionTest)
+TEST_CASE_METHOD(
+    PreprocessImputerTestFixture, "PreprocessImputerListwiseDimensionTest",
+    "PreprocessImputerMainTest")
 {
   // Load synthetic dataset.
   arma::mat inputData;
@@ -140,14 +141,16 @@ BOOST_AUTO_TEST_CASE(PreprocessImputerListwiseDimensionTest)
   // Now check that the output has desired dimensions.
   arma::mat outputData;
   data::Load(IO::GetParam<std::string>("output_file"), outputData);
-  BOOST_REQUIRE_EQUAL(outputData.n_cols + countNaN, inputSize);
-  BOOST_REQUIRE_EQUAL(outputData.n_rows, 3); // Input Dimension.
+  REQUIRE(outputData.n_cols + countNaN == inputSize);
+  REQUIRE(outputData.n_rows == 3); // Input Dimension.
 }
 
 /**
  * Check that invalid strategy can't be specified.
  */
-BOOST_AUTO_TEST_CASE(PreprocessImputerStrategyTest)
+TEST_CASE_METHOD(
+    PreprocessImputerTestFixture, "PreprocessImputerStrategyTest",
+    "PreprocessImputerMainTest")
 {
   // Load synthetic dataset.
   arma::mat inputData;
@@ -159,8 +162,6 @@ BOOST_AUTO_TEST_CASE(PreprocessImputerStrategyTest)
   SetInputParam("strategy", (std::string) "notmean"); // Invalid.
 
   Log::Fatal.ignoreInput = true;
-  BOOST_REQUIRE_THROW(mlpackMain(), std::runtime_error);
+  REQUIRE_THROWS_AS(mlpackMain(), std::runtime_error);
   Log::Fatal.ignoreInput = false;
 }
-
-BOOST_AUTO_TEST_SUITE_END();
diff --git a/src/mlpack/tests/main_tests/preprocess_scale_test.cpp b/src/mlpack/tests/main_tests/preprocess_scale_test.cpp
index 8d60e6aae7..71fe49fb7b 100644
--- a/src/mlpack/tests/main_tests/preprocess_scale_test.cpp
+++ b/src/mlpack/tests/main_tests/preprocess_scale_test.cpp
@@ -18,14 +18,15 @@ static const std::string testName = "PreprocessScale";
 #include <mlpack/methods/preprocess/preprocess_scale_main.cpp>
 
 #include "test_helper.hpp"
-#include <boost/test/unit_test.hpp>
-#include "../test_tools.hpp"
+#include "../test_catch_tools.hpp"
+#include "../catch.hpp"
 
 using namespace mlpack;
 
 struct PreprocessScaleTestFixture
 {
  public:
+  static arma::mat dataset;
   PreprocessScaleTestFixture()
   {
     // Cache in the options for this program.
@@ -40,16 +41,14 @@ struct PreprocessScaleTestFixture
   }
 };
 
-BOOST_FIXTURE_TEST_SUITE(PreprocessScaleMainTest,
-                         PreprocessScaleTestFixture);
-
-arma::mat dataset = "-1 -0.5 0 1;"
-                    "2 6 10 18;";
+arma::mat PreprocessScaleTestFixture::dataset = "-1 -0.5 0 1;"
+                                                "2 6 10 18;";
 
 /**
  * Check that two different scalers give two different output.
  */
-BOOST_AUTO_TEST_CASE(TwoScalerTest)
+TEST_CASE_METHOD(PreprocessScaleTestFixture, "TwoScalerTest",
+                 "PreprocessScaleMainTest")
 {
   // Input custom data points.
   std::string method = "max_abs_scaler";
@@ -75,7 +74,8 @@ BOOST_AUTO_TEST_CASE(TwoScalerTest)
  * Check that two different option for a particular scaler give two
  * different output.
  */
-BOOST_AUTO_TEST_CASE(TwoOptionTest)
+TEST_CASE_METHOD(PreprocessScaleTestFixture, "TwoOptionTest",
+                 "PreprocessScaleMainTest")
 {
   std::string method = "min_max_scaler";
   // Input custom data points.
@@ -101,7 +101,8 @@ BOOST_AUTO_TEST_CASE(TwoOptionTest)
 /**
  * Check that passing unrelated option don't change anything.
  */
-BOOST_AUTO_TEST_CASE(UnrelatedOptionTest)
+TEST_CASE_METHOD(PreprocessScaleTestFixture, "UnrelatedOptionTest",
+                 "PreprocessScaleMainTest")
 {
   std::string method = "standard_scaler";
   // Input custom data points.
@@ -128,7 +129,8 @@ BOOST_AUTO_TEST_CASE(UnrelatedOptionTest)
 /**
  * Check Inverse Scaling is working.
  */
-BOOST_AUTO_TEST_CASE(InverseScalingTest)
+TEST_CASE_METHOD(PreprocessScaleTestFixture, "InverseScalingTest",
+                 "PreprocessScaleMainTest")
 {
   std::string method = "zca_whitening";
   // Input custom data points.
@@ -151,7 +153,8 @@ BOOST_AUTO_TEST_CASE(InverseScalingTest)
 /**
  * Check Saved model is working.
  */
-BOOST_AUTO_TEST_CASE(SavedModelTest)
+TEST_CASE_METHOD(PreprocessScaleTestFixture, "SavedModelTest",
+                 "PreprocessScaleMainTest")
 {
   std::string method = "pca_whitening";
   // Input custom data points.
@@ -173,7 +176,8 @@ BOOST_AUTO_TEST_CASE(SavedModelTest)
 /**
  * Check different epsilon for PCA give two different output.
  */
-BOOST_AUTO_TEST_CASE(EpsilonTest)
+TEST_CASE_METHOD(PreprocessScaleTestFixture, "EpsilonTest",
+                 "PreprocessScaleMainTest")
 {
   std::string method = "pca_whitening";
   // Input custom data points.
@@ -198,7 +202,8 @@ BOOST_AUTO_TEST_CASE(EpsilonTest)
 /**
  * Check for invalid epsilon.
  */
-BOOST_AUTO_TEST_CASE(InvalidEpsilonTest)
+TEST_CASE_METHOD(PreprocessScaleTestFixture, "InvalidEpsilonTest",
+                 "PreprocessScaleMainTest")
 {
   std::string method = "pca_whitening";
   // Input custom data points.
@@ -206,13 +211,14 @@ BOOST_AUTO_TEST_CASE(InvalidEpsilonTest)
   SetInputParam("scaler_method", std::move(method));
   SetInputParam("epsilon", -1.0);
 
-  BOOST_REQUIRE_THROW(mlpackMain(), std::runtime_error);
+  REQUIRE_THROWS_AS(mlpackMain(), std::runtime_error);
 }
 
 /**
  * Check for invalid range in min_max_scaler.
  */
-BOOST_AUTO_TEST_CASE(InvalidRangeTest)
+TEST_CASE_METHOD(PreprocessScaleTestFixture, "InvalidRangeTest",
+                 "PreprocessScaleMainTest")
 {
   std::string method = "min_max_scaler";
   // Input custom data points.
@@ -221,13 +227,14 @@ BOOST_AUTO_TEST_CASE(InvalidRangeTest)
   SetInputParam("min_value", 4);
   SetInputParam("max_value", 2);
 
-  BOOST_REQUIRE_THROW(mlpackMain(), std::runtime_error);
+  REQUIRE_THROWS_AS(mlpackMain(), std::runtime_error);
 }
 
 /**
  * Check for invalid scaler type.
  */
-BOOST_AUTO_TEST_CASE(InvalidScalerTest)
+TEST_CASE_METHOD(PreprocessScaleTestFixture, "InvalidScalerTest",
+                 "PreprocessScaleMainTest")
 {
   std::string method = "invalid_scaler";
   // Input custom data points.
@@ -237,50 +244,53 @@ BOOST_AUTO_TEST_CASE(InvalidScalerTest)
   SetInputParam("max_value", 2);
 
   Log::Fatal.ignoreInput = true;
-  BOOST_REQUIRE_THROW(mlpackMain(), std::runtime_error);
+  REQUIRE_THROWS_AS(mlpackMain(), std::runtime_error);
   Log::Fatal.ignoreInput = false;
 }
 
 /**
  * Check for Standard scaler type.
  */
-BOOST_AUTO_TEST_CASE(StandardScalerTest)
+TEST_CASE_METHOD(PreprocessScaleTestFixture, "StandardScalerBindingTest",
+                 "PreprocessScaleMainTest")
 {
   std::string method = "standard_scaler";
   // Input custom data points.
   SetInputParam("input", dataset);
   SetInputParam("scaler_method", method);
-  BOOST_REQUIRE_NO_THROW(mlpackMain());
+  REQUIRE_NOTHROW(mlpackMain());
   SetInputParam("scaler_method", std::move(method));
   SetInputParam("input", dataset);
   SetInputParam("input_model",
                 IO::GetParam<ScalingModel*>("output_model"));
   SetInputParam("inverse_scaling", true);
-  BOOST_REQUIRE_NO_THROW(mlpackMain());
+  REQUIRE_NOTHROW(mlpackMain());
 }
 
 /**
  * Check for MaxAbs scaler type.
  */
-BOOST_AUTO_TEST_CASE(MaxAbsScalerTest)
+TEST_CASE_METHOD(PreprocessScaleTestFixture, "MaxAbsScalerBindingTest",
+                 "PreprocessScaleMainTest")
 {
   std::string method = "max_abs_scaler";
   // Input custom data points.
   SetInputParam("input", dataset);
   SetInputParam("scaler_method", method);
-  BOOST_REQUIRE_NO_THROW(mlpackMain());
+  REQUIRE_NOTHROW(mlpackMain());
   SetInputParam("scaler_method", std::move(method));
   SetInputParam("input", dataset);
   SetInputParam("input_model",
                 IO::GetParam<ScalingModel*>("output_model"));
   SetInputParam("inverse_scaling", true);
-  BOOST_REQUIRE_NO_THROW(mlpackMain());
+  REQUIRE_NOTHROW(mlpackMain());
 }
 
 /**
  * Check for MinMax scaler type.
  */
-BOOST_AUTO_TEST_CASE(MinMaxScalerTest)
+TEST_CASE_METHOD(PreprocessScaleTestFixture, "MinMaxScalerBindingTest",
+                 "PreprocessScaleMainTest")
 {
   std::string method = "min_max_scaler";
   // Input custom data points.
@@ -288,69 +298,70 @@ BOOST_AUTO_TEST_CASE(MinMaxScalerTest)
   SetInputParam("scaler_method", method);
   SetInputParam("min_value", 2);
   SetInputParam("max_value", 4);
-  BOOST_REQUIRE_NO_THROW(mlpackMain());
+  REQUIRE_NOTHROW(mlpackMain());
   SetInputParam("scaler_method", std::move(method));
   SetInputParam("input", dataset);
   SetInputParam("input_model",
                 IO::GetParam<ScalingModel*>("output_model"));
   SetInputParam("inverse_scaling", true);
-  BOOST_REQUIRE_NO_THROW(mlpackMain());
+  REQUIRE_NOTHROW(mlpackMain());
 }
 
 /**
  * Check for PCA scaler type.
  */
-BOOST_AUTO_TEST_CASE(PCAScalerTest)
+TEST_CASE_METHOD(PreprocessScaleTestFixture, "PCAScalerBindingTest",
+                 "PreprocessScaleMainTest")
 {
   std::string method = "pca_whitening";
   // Input custom data points.
   SetInputParam("input", dataset);
   SetInputParam("scaler_method", method);
   SetInputParam("epsilon", 1.0);
-  BOOST_REQUIRE_NO_THROW(mlpackMain());
+  REQUIRE_NOTHROW(mlpackMain());
   SetInputParam("scaler_method", std::move(method));
   SetInputParam("input", dataset);
   SetInputParam("input_model",
                 IO::GetParam<ScalingModel*>("output_model"));
   SetInputParam("inverse_scaling", true);
-  BOOST_REQUIRE_NO_THROW(mlpackMain());
+  REQUIRE_NOTHROW(mlpackMain());
 }
 
 /**
  * Check for ZCA scaler type.
  */
-BOOST_AUTO_TEST_CASE(ZCAScalerTest)
+TEST_CASE_METHOD(PreprocessScaleTestFixture, "ZCAScalerBindingTest",
+                 "PreprocessScaleMainTest")
 {
   std::string method = "zca_whitening";
   // Input custom data points.
   SetInputParam("input", dataset);
   SetInputParam("scaler_method", method);
   SetInputParam("epsilon", 1.0);
-  BOOST_REQUIRE_NO_THROW(mlpackMain());
+  REQUIRE_NOTHROW(mlpackMain());
   SetInputParam("scaler_method", std::move(method));
   SetInputParam("input", dataset);
   SetInputParam("input_model",
                 IO::GetParam<ScalingModel*>("output_model"));
   SetInputParam("inverse_scaling", true);
-  BOOST_REQUIRE_NO_THROW(mlpackMain());
+  REQUIRE_NOTHROW(mlpackMain());
 }
 
 /**
  * Check for Mean Normalization scaler type.
  */
-BOOST_AUTO_TEST_CASE(MeanNormalizationTest)
+TEST_CASE_METHOD(PreprocessScaleTestFixture, "MeanNormalizationBindingTest",
+                 "PreprocessScaleMainTest")
 {
   std::string method = "mean_normalization";
   // Input custom data points.
   SetInputParam("input", dataset);
   SetInputParam("scaler_method", method);
-  BOOST_REQUIRE_NO_THROW(mlpackMain());
+  REQUIRE_NOTHROW(mlpackMain());
   SetInputParam("scaler_method", std::move(method));
   SetInputParam("input", dataset);
   SetInputParam("input_model",
                 IO::GetParam<ScalingModel*>("output_model"));
   SetInputParam("inverse_scaling", true);
-  BOOST_REQUIRE_NO_THROW(mlpackMain());
+  REQUIRE_NOTHROW(mlpackMain());
 }
-
-BOOST_AUTO_TEST_SUITE_END();
diff --git a/src/mlpack/tests/main_tests/preprocess_split_test.cpp b/src/mlpack/tests/main_tests/preprocess_split_test.cpp
index b9032af752..375af31f6e 100644
--- a/src/mlpack/tests/main_tests/preprocess_split_test.cpp
+++ b/src/mlpack/tests/main_tests/preprocess_split_test.cpp
@@ -18,8 +18,8 @@ static const std::string testName = "PreprocessSplit";
 #include <mlpack/methods/preprocess/preprocess_split_main.cpp>
 
 #include "test_helper.hpp"
-#include <boost/test/unit_test.hpp>
-#include "../test_tools.hpp"
+#include "../test_catch_tools.hpp"
+#include "../catch.hpp"
 
 #include <cmath>
 
@@ -42,14 +42,12 @@ struct PreprocessSplitTestFixture
   }
 };
 
-BOOST_FIXTURE_TEST_SUITE(PreprocessSplitMainTest,
-                         PreprocessSplitTestFixture);
-
 /**
  * Check that desired output dimensions are received for both input data and
  * labels.
  */
-BOOST_AUTO_TEST_CASE(PreprocessSplitDimensionTest)
+TEST_CASE_METHOD(PreprocessSplitTestFixture, "PreprocessSplitDimensionTest",
+                 "PreprocessSplitMainTest")
 {
   // Load custom dataset.
   arma::mat inputData;
@@ -71,15 +69,15 @@ BOOST_AUTO_TEST_CASE(PreprocessSplitDimensionTest)
   mlpackMain();
 
   // Now check that the output has desired dimensions.
-  BOOST_REQUIRE_EQUAL(IO::GetParam<arma::mat>("training").n_cols,
-                      std::ceil(0.9 * inputSize));
-  BOOST_REQUIRE_EQUAL(IO::GetParam<arma::mat>("test").n_cols,
-                      std::floor(0.1 * inputSize));
+  REQUIRE(IO::GetParam<arma::mat>("training").n_cols ==
+      std::ceil(0.9 * inputSize));
+  REQUIRE(IO::GetParam<arma::mat>("test").n_cols ==
+      std::floor(0.1 * inputSize));
 
-  BOOST_REQUIRE_EQUAL(
-      IO::GetParam<arma::Mat<size_t>>("training_labels").n_cols,
+  REQUIRE(
+      IO::GetParam<arma::Mat<size_t>>("training_labels").n_cols ==
       std::ceil(0.9 * labelSize));
-  BOOST_REQUIRE_EQUAL(IO::GetParam<arma::Mat<size_t>>("test_labels").n_cols,
+  REQUIRE(IO::GetParam<arma::Mat<size_t>>("test_labels").n_cols ==
       std::floor(0.1 * labelSize));
 }
 
@@ -87,7 +85,9 @@ BOOST_AUTO_TEST_CASE(PreprocessSplitDimensionTest)
  * Check that desired output dimensions are received for the input data when
  * labels are not provided.
  */
-BOOST_AUTO_TEST_CASE(PreprocessSplitLabelLessDimensionTest)
+TEST_CASE_METHOD(
+    PreprocessSplitTestFixture,
+    "PreprocessSplitLabelLessDimensionTest", "PreprocessSplitMainTest")
 {
   // Load custom dataset.
   arma::mat inputData;
@@ -105,16 +105,17 @@ BOOST_AUTO_TEST_CASE(PreprocessSplitLabelLessDimensionTest)
   mlpackMain();
 
   // Now check that the output has desired dimensions.
-  BOOST_REQUIRE_EQUAL(IO::GetParam<arma::mat>("training").n_cols,
+  REQUIRE(IO::GetParam<arma::mat>("training").n_cols ==
       std::ceil(0.9 * inputSize));
-  BOOST_REQUIRE_EQUAL(IO::GetParam<arma::mat>("test").n_cols,
+  REQUIRE(IO::GetParam<arma::mat>("test").n_cols ==
       std::floor(0.1 * inputSize));
 }
 
 /**
  * Ensure that test ratio is always a non-negative number.
  */
-BOOST_AUTO_TEST_CASE(PreprocessSplitTestRatioTest)
+TEST_CASE_METHOD(PreprocessSplitTestFixture, "PreprocessSplitTestRatioTest",
+                 "PreprocessSplitMainTest")
 {
   // Load custom dataset.
   arma::mat inputData;
@@ -129,14 +130,16 @@ BOOST_AUTO_TEST_CASE(PreprocessSplitTestRatioTest)
   SetInputParam("test_ratio", (double) -0.2);
 
   Log::Fatal.ignoreInput = true;
-  BOOST_REQUIRE_THROW(mlpackMain(), std::runtime_error);
+  REQUIRE_THROWS_AS(mlpackMain(), std::runtime_error);
   Log::Fatal.ignoreInput = false;
 }
 
 /**
  * Check that if test size is 0 then train consist of whole input data.
  */
-BOOST_AUTO_TEST_CASE(PreprocessSplitZeroTestRatioTest)
+TEST_CASE_METHOD(
+    PreprocessSplitTestFixture, "PreprocessSplitZeroTestRatioTest",
+    "PreprocessSplitMainTest")
 {
   // Load custom dataset.
   arma::mat inputData;
@@ -157,19 +160,19 @@ BOOST_AUTO_TEST_CASE(PreprocessSplitZeroTestRatioTest)
   mlpackMain();
 
   // Now check that the output has desired dimensions.
-  BOOST_REQUIRE_EQUAL(IO::GetParam<arma::mat>("training").n_cols, inputSize);
-  BOOST_REQUIRE_EQUAL(IO::GetParam<arma::mat>("test").n_cols, 0);
+  REQUIRE(IO::GetParam<arma::mat>("training").n_cols == inputSize);
+  REQUIRE(IO::GetParam<arma::mat>("test").n_cols == 0);
 
-  BOOST_REQUIRE_EQUAL(
-      IO::GetParam<arma::Mat<size_t>>("training_labels").n_cols, labelSize);
-  BOOST_REQUIRE_EQUAL(
-      IO::GetParam<arma::Mat<size_t>>("test_labels").n_cols, 0);
+  REQUIRE(IO::GetParam<arma::Mat<size_t>>("training_labels").n_cols == labelSize);
+  REQUIRE(IO::GetParam<arma::Mat<size_t>>("test_labels").n_cols == 0);
 }
 
 /**
  * Check that if test size is 1 then test consist of whole input data.
  */
-BOOST_AUTO_TEST_CASE(PreprocessSplitUnityTestRatioTest)
+TEST_CASE_METHOD(
+    PreprocessSplitTestFixture, "PreprocessSplitUnityTestRatioTest",
+    "PreprocessSplitMainTest")
 {
   // Load custom dataset.
   arma::mat inputData;
@@ -190,19 +193,19 @@ BOOST_AUTO_TEST_CASE(PreprocessSplitUnityTestRatioTest)
   mlpackMain();
 
   // Now check that the output has desired dimensions.
-  BOOST_REQUIRE_EQUAL(IO::GetParam<arma::mat>("training").n_cols, 0);
-  BOOST_REQUIRE_EQUAL(IO::GetParam<arma::mat>("test").n_cols, inputSize);
+  REQUIRE(IO::GetParam<arma::mat>("training").n_cols == 0);
+  REQUIRE(IO::GetParam<arma::mat>("test").n_cols == inputSize);
 
-  BOOST_REQUIRE_EQUAL(
-      IO::GetParam<arma::Mat<size_t>>("training_labels").n_cols, 0);
-  BOOST_REQUIRE_EQUAL(IO::GetParam<arma::Mat<size_t>>("test_labels").n_cols,
-      labelSize);
+  REQUIRE(IO::GetParam<arma::Mat<size_t>>("training_labels").n_cols == 0);
+  REQUIRE(IO::GetParam<arma::Mat<size_t>>("test_labels").n_cols == labelSize);
 }
 
 /**
  * Check shuffle_data flag is working as expected.
  */
-BOOST_AUTO_TEST_CASE(PreprocessSplitLabelShuffleDataTest)
+TEST_CASE_METHOD(
+    PreprocessSplitTestFixture, "PreprocessSplitLabelShuffleDataTest",
+    "PreprocessSplitMainTest")
 {
   // Load custom dataset.
   arma::mat inputData;
@@ -220,14 +223,12 @@ BOOST_AUTO_TEST_CASE(PreprocessSplitLabelShuffleDataTest)
   mlpackMain();
 
   // Now check that the output has desired dimensions.
-  BOOST_REQUIRE_EQUAL(IO::GetParam<arma::mat>("training").n_cols,
+  REQUIRE(IO::GetParam<arma::mat>("training").n_cols ==
       std::ceil(0.9 * inputSize));
-  BOOST_REQUIRE_EQUAL(IO::GetParam<arma::mat>("test").n_cols,
+  REQUIRE(IO::GetParam<arma::mat>("test").n_cols ==
       std::floor(0.1 * inputSize));
 
   arma::mat concat = arma::join_rows(IO::GetParam<arma::mat>("training"),
       IO::GetParam<arma::mat>("test"));
   CheckMatrices(inputData, concat);
 }
-
-BOOST_AUTO_TEST_SUITE_END();
diff --git a/src/mlpack/tests/scaling_test.cpp b/src/mlpack/tests/scaling_test.cpp
index 07a343ce74..dd3f425556 100644
--- a/src/mlpack/tests/scaling_test.cpp
+++ b/src/mlpack/tests/scaling_test.cpp
@@ -16,15 +16,13 @@
 #include <mlpack/core/data/scaler_methods/standard_scaler.hpp>
 #include <mlpack/core/data/scaler_methods/mean_normalization.hpp>
 
-#include <boost/test/unit_test.hpp>
-#include "test_tools.hpp"
+#include "test_catch_tools.hpp"
+#include "catch.hpp"
 
 using namespace mlpack;
 using namespace mlpack::data;
 using namespace std;
 
-BOOST_AUTO_TEST_SUITE(ScalingTest);
-
 arma::mat dataset = "-1 -0.5 0 1;"
                     "2 6 10 18;";
 arma::mat scaleddataset;
@@ -33,7 +31,7 @@ arma::mat temp;
 /**
  * Test For MinMax Scaler Class.
  */
-BOOST_AUTO_TEST_CASE(MinMaxScalerTest)
+TEST_CASE("MinMaxScalerTest", "[ScalingTest]")
 {
   arma::mat scaled = "0 0.2500 0.5000 1.000;"
                      "0 0.2500 0.5000 1.000;";
@@ -48,7 +46,7 @@ BOOST_AUTO_TEST_CASE(MinMaxScalerTest)
 /**
  * Test For MaxAbs Scaler Class.
  */
-BOOST_AUTO_TEST_CASE(MaxAbsScalerTest)
+TEST_CASE("MaxAbsScalerTest", "[ScalingTest]")
 {
   arma::mat scaled = "-1 -0.5 0 1;"
                      "0.1111111111 0.3333333333 0.55555556 1.0000;";
@@ -63,7 +61,7 @@ BOOST_AUTO_TEST_CASE(MaxAbsScalerTest)
 /**
  * Test For Standard Scaler Class.
  */
-BOOST_AUTO_TEST_CASE(StandardScalerTest)
+TEST_CASE("StandardScalerTest", "[ScalingTest]")
 {
   arma::mat scaled = "-1.18321596 -0.50709255  0.16903085 1.52127766;"
                      "-1.18321596 -0.50709255  0.16903085 1.52127766;";
@@ -78,7 +76,7 @@ BOOST_AUTO_TEST_CASE(StandardScalerTest)
 /**
  * Test For MeanNormalization Scaler Class.
  */
-BOOST_AUTO_TEST_CASE(MeanNormalizationTest)
+TEST_CASE("MeanNormalizationTest", "[ScalingTest]")
 {
   arma::mat scaled = "-0.43750000000 -0.187500000 0.062500000 0.562500000;"
                      "-0.43750000000 -0.187500000 0.062500000 0.562500000;";
@@ -93,7 +91,7 @@ BOOST_AUTO_TEST_CASE(MeanNormalizationTest)
 /**
  * Test to pass same matrix as input and output
  */
-BOOST_AUTO_TEST_CASE(SameInputOutputTest)
+TEST_CASE("SameInputOutputTest", "[ScalingTest]")
 {
   temp = dataset;
   arma::mat scaled = "-0.43750000000 -0.187500000 0.062500000 0.562500000;"
@@ -109,7 +107,7 @@ BOOST_AUTO_TEST_CASE(SameInputOutputTest)
 /**
  * Test for Zero Matrix.
  */
-BOOST_AUTO_TEST_CASE(ZeroMatrixTest)
+TEST_CASE("ZeroMatrixTest", "[ScalingTest]")
 {
   arma::mat input(2, 4, arma::fill::zeros);
   data::MeanNormalization scale;
@@ -123,7 +121,7 @@ BOOST_AUTO_TEST_CASE(ZeroMatrixTest)
 /**
  * Test for Zero Scale.
  */
-BOOST_AUTO_TEST_CASE(ZeroScaleTest)
+TEST_CASE("ZeroScaleTest", "[ScalingTest]")
 {
   dataset = "1 1 1 1;"
             "2 6 10 18;";
@@ -140,7 +138,7 @@ BOOST_AUTO_TEST_CASE(ZeroScaleTest)
 /**
  * Test for PCA whitening Scale.
  */
-BOOST_AUTO_TEST_CASE(PCAWhiteningTest)
+TEST_CASE("PCAWhiteningTest", "[ScalingTest]")
 {
   data::PCAWhitening scale;
   arma::mat output;
@@ -151,7 +149,7 @@ BOOST_AUTO_TEST_CASE(PCAWhiteningTest)
   double ccovsum = 0.0;
   for (size_t i = 0; i < diagonals.n_elem; ++i)
     ccovsum += diagonals(i);
-  BOOST_REQUIRE_CLOSE(ccovsum, 1.0, 1e-3);
+  REQUIRE(ccovsum == Approx(1.0).epsilon(1e-3 / 100));
   scale.InverseTransform(output, temp);
   CheckMatrices(dataset, temp);
 }
@@ -159,7 +157,7 @@ BOOST_AUTO_TEST_CASE(PCAWhiteningTest)
 /**
  * Test for ZCA whitening Scale.
  */
-BOOST_AUTO_TEST_CASE(ZCAWhiteningTest)
+TEST_CASE("ZCAWhiteningTest", "[ScalingTest]")
 {
   data::ZCAWhitening scale;
   arma::mat output;
@@ -170,9 +168,7 @@ BOOST_AUTO_TEST_CASE(ZCAWhiteningTest)
   double ccovsum = 0.0;
   for (size_t i = 0; i < diagonals.n_elem; ++i)
     ccovsum += diagonals(i);
-  BOOST_REQUIRE_CLOSE(ccovsum, 1.0, 1e-3);
+  REQUIRE(ccovsum == Approx(1.0).epsilon(1e-3 / 100));
   scale.InverseTransform(output, temp);
   CheckMatrices(dataset, temp);
 }
-
-BOOST_AUTO_TEST_SUITE_END();
diff --git a/src/mlpack/tests/split_data_test.cpp b/src/mlpack/tests/split_data_test.cpp
index 6e5beaac21..788e80e372 100644
--- a/src/mlpack/tests/split_data_test.cpp
+++ b/src/mlpack/tests/split_data_test.cpp
@@ -12,15 +12,13 @@
 #include <mlpack/core.hpp>
 #include <mlpack/core/data/split_data.hpp>
 
-#include <boost/test/unit_test.hpp>
-#include "test_tools.hpp"
+#include "test_catch_tools.hpp"
+#include "catch.hpp"
 
 using namespace mlpack;
 using namespace arma;
 using namespace mlpack::data;
 
-BOOST_AUTO_TEST_SUITE(SplitDataTest);
-
 /**
  * Compare the data after train test split.  This assumes that the labels
  * correspond to each column, so that we can easily check each point against its
@@ -42,9 +40,9 @@ void CompareData(const mat& inputData,
     for (size_t j = 0; j != lhsCol.n_rows; ++j)
     {
       if (std::abs(rhsCol(j)) < 1e-5)
-        BOOST_REQUIRE_SMALL(lhsCol(j), 1e-5);
+        REQUIRE(lhsCol(j) == Approx(0.0).margin(1e-5));
       else
-        BOOST_REQUIRE_CLOSE(lhsCol(j), rhsCol(j), 1e-5);
+        REQUIRE(lhsCol(j) == Approx(rhsCol(j)).epsilon(1e-5 / 100));
     }
   }
 }
@@ -61,9 +59,9 @@ void CheckMatEqual(const mat& inputData,
     for (size_t j = 0; j < lhsCol.n_rows; ++j)
     {
       if (std::abs(rhsCol(j)) < 1e-5)
-        BOOST_REQUIRE_SMALL(lhsCol(j), 1e-5);
+        REQUIRE(lhsCol(j) == Approx(0.0).margin(1e-5));
       else
-        BOOST_REQUIRE_CLOSE(lhsCol(j), rhsCol(j), 1e-5);
+        REQUIRE(lhsCol(j) == Approx(rhsCol(j)).epsilon(1e-5 / 100));
     }
   }
 }
@@ -80,80 +78,80 @@ void CheckDuplication(const Row<size_t>& trainLabels,
 
   for (size_t i = 0; i < trainLabels.n_elem; ++i)
   {
-    BOOST_REQUIRE_LT(trainLabels[i], counts.n_elem);
+    REQUIRE(trainLabels[i] < counts.n_elem);
     counts[trainLabels[i]]++;
   }
   for (size_t i = 0; i < testLabels.n_elem; ++i)
   {
-    BOOST_REQUIRE_LT(testLabels[i], counts.n_elem);
+    REQUIRE(testLabels[i] < counts.n_elem);
     counts[testLabels[i]]++;
   }
 
   // Now make sure each point has been used once.
   for (size_t i = 0; i < counts.n_elem; ++i)
-    BOOST_REQUIRE_EQUAL(counts[i], 1);
+    REQUIRE(counts[i] == 1);
 }
 
-BOOST_AUTO_TEST_CASE(SplitShuffleDataResultMat)
+TEST_CASE("SplitShuffleDataResultMat", "[SplitDataTest]")
 {
   mat input(2, 10);
   size_t count = 0; // Counter for unique sequential values.
   input.imbue([&count] () { return ++count; });
 
   const auto value = Split(input, 0.2);
-  BOOST_REQUIRE_EQUAL(std::get<0>(value).n_cols, 8); // Train data.
-  BOOST_REQUIRE_EQUAL(std::get<1>(value).n_cols, 2); // Test data.
+  REQUIRE(std::get<0>(value).n_cols == 8); // Train data.
+  REQUIRE(std::get<1>(value).n_cols == 2); // Test data.
 
   mat concat = arma::join_rows(std::get<0>(value), std::get<1>(value));
   CheckMatEqual(input, concat);
 }
 
-BOOST_AUTO_TEST_CASE(SplitDataResultMat)
+TEST_CASE("SplitDataResultMat", "[SplitDataTest]")
 {
   mat input(2, 10);
   size_t count = 0; // Counter for unique sequential values.
   input.imbue([&count] () { return ++count; });
 
   const auto value = Split(input, 0.2, false);
-  BOOST_REQUIRE_EQUAL(std::get<0>(value).n_cols, 8); // Train data.
-  BOOST_REQUIRE_EQUAL(std::get<1>(value).n_cols, 2); // Test data.
+  REQUIRE(std::get<0>(value).n_cols == 8); // Train data.
+  REQUIRE(std::get<1>(value).n_cols == 2); // Test data.
 
   mat concat = arma::join_rows(std::get<0>(value), std::get<1>(value));
   // Order matters here.
   CheckMatrices(input, concat);
 }
 
-BOOST_AUTO_TEST_CASE(ZeroRatioSplitData)
+TEST_CASE("ZeroRatioSplitData", "[SplitDataTest]")
 {
   mat input(2, 10);
   size_t count = 0; // Counter for unique sequential values.
   input.imbue([&count] () { return ++count; });
 
   const auto value = Split(input, 0, false);
-  BOOST_REQUIRE_EQUAL(std::get<0>(value).n_cols, 10); // Train data.
-  BOOST_REQUIRE_EQUAL(std::get<1>(value).n_cols, 0); // Test data.
+  REQUIRE(std::get<0>(value).n_cols == 10); // Train data.
+  REQUIRE(std::get<1>(value).n_cols == 0); // Test data.
 
   mat concat = arma::join_rows(std::get<0>(value), std::get<1>(value));
   // Order matters here.
   CheckMatrices(input, concat);
 }
 
-BOOST_AUTO_TEST_CASE(TotalRatioSplitData)
+TEST_CASE("TotalRatioSplitData", "[SplitDataTest]")
 {
   mat input(2, 10);
   size_t count = 0; // Counter for unique sequential values.
   input.imbue([&count] () { return ++count; });
 
   const auto value = Split(input, 1, false);
-  BOOST_REQUIRE_EQUAL(std::get<0>(value).n_cols, 0); // Train data.
-  BOOST_REQUIRE_EQUAL(std::get<1>(value).n_cols, 10); // Test data.
+  REQUIRE(std::get<0>(value).n_cols == 0); // Train data.
+  REQUIRE(std::get<1>(value).n_cols == 10); // Test data.
 
   mat concat = arma::join_rows(std::get<0>(value), std::get<1>(value));
   // Order matters here.
   CheckMatrices(input, concat);
 }
 
-BOOST_AUTO_TEST_CASE(SplitLabeledDataResultMat)
+TEST_CASE("SplitLabeledDataResultMat", "[SplitDataTest]")
 {
   mat input(2, 10);
   input.randu();
@@ -164,10 +162,10 @@ BOOST_AUTO_TEST_CASE(SplitLabeledDataResultMat)
       input.n_cols);
 
   const auto value = Split(input, labels, 0.2);
-  BOOST_REQUIRE_EQUAL(std::get<0>(value).n_cols, 8);
-  BOOST_REQUIRE_EQUAL(std::get<1>(value).n_cols, 2);
-  BOOST_REQUIRE_EQUAL(std::get<2>(value).n_cols, 8);
-  BOOST_REQUIRE_EQUAL(std::get<3>(value).n_cols, 2);
+  REQUIRE(std::get<0>(value).n_cols == 8);
+  REQUIRE(std::get<1>(value).n_cols == 2);
+  REQUIRE(std::get<2>(value).n_cols == 8);
+  REQUIRE(std::get<3>(value).n_cols == 2);
 
   CompareData(input, std::get<0>(value), std::get<2>(value));
   CompareData(input, std::get<1>(value), std::get<3>(value));
@@ -180,21 +178,21 @@ BOOST_AUTO_TEST_CASE(SplitLabeledDataResultMat)
 /**
  * The same test as above, but on a larger dataset.
  */
-BOOST_AUTO_TEST_CASE(SplitDataLargerTest)
+TEST_CASE("SplitDataLargerTest", "[SplitDataTest]")
 {
   size_t count = 0;
   mat input(10, 497);
   input.imbue([&count] () { return ++count; });
 
   const auto value = Split(input, 0.3);
-  BOOST_REQUIRE_EQUAL(std::get<0>(value).n_cols, 497 - size_t(0.3 * 497));
-  BOOST_REQUIRE_EQUAL(std::get<1>(value).n_cols, size_t(0.3 * 497));
+  REQUIRE(std::get<0>(value).n_cols == 497 - size_t(0.3 * 497));
+  REQUIRE(std::get<1>(value).n_cols == size_t(0.3 * 497));
 
   mat concat = arma::join_rows(std::get<0>(value), std::get<1>(value));
   CheckMatEqual(input, concat);
 }
 
-BOOST_AUTO_TEST_CASE(SplitLabeledDataLargerTest)
+TEST_CASE("SplitLabeledDataLargerTest", "[SplitDataTest]")
 {
   mat input(10, 497);
   input.randu();
@@ -204,15 +202,13 @@ BOOST_AUTO_TEST_CASE(SplitLabeledDataLargerTest)
       input.n_cols);
 
   const auto value = Split(input, labels, 0.3);
-  BOOST_REQUIRE_EQUAL(std::get<0>(value).n_cols, 497 - size_t(0.3 * 497));
-  BOOST_REQUIRE_EQUAL(std::get<1>(value).n_cols, size_t(0.3 * 497));
-  BOOST_REQUIRE_EQUAL(std::get<2>(value).n_cols, 497 - size_t(0.3 * 497));
-  BOOST_REQUIRE_EQUAL(std::get<3>(value).n_cols, size_t(0.3 * 497));
+  REQUIRE(std::get<0>(value).n_cols == 497 - size_t(0.3 * 497));
+  REQUIRE(std::get<1>(value).n_cols == size_t(0.3 * 497));
+  REQUIRE(std::get<2>(value).n_cols == 497 - size_t(0.3 * 497));
+  REQUIRE(std::get<3>(value).n_cols == size_t(0.3 * 497));
 
   CompareData(input, std::get<0>(value), std::get<2>(value));
   CompareData(input, std::get<1>(value), std::get<3>(value));
 
   CheckDuplication(std::get<2>(value), std::get<3>(value));
 }
-
-BOOST_AUTO_TEST_SUITE_END();