ENH: Ridge Regression CPU SPMD support

cpp/oneapi/dal/algo/linear_regression/test/spmd.cpp

@@ -27,4 +27,13 @@ TEMPLATE_LIST_TEST_M(lr_spmd_test, "LR common flow", "[lr][spmd]", lr_types) {
     this->run_and_check_linear();
 }
 
+TEMPLATE_LIST_TEST_M(lr_spmd_test, "RR common flow", "[rr][spmd]", lr_types) {
+    SKIP_IF(this->not_float64_friendly());
+
+    this->generate(777);
+    this->set_rank_count(GENERATE(2, 3));
+
+    this->run_and_check_ridge();
+}
+
 } // namespace oneapi::dal::linear_regression::test

samples/oneapi/cpp/ccl/sources/ridge_regression_distr_ccl.cpp

@@ -0,0 +1,82 @@
+/*******************************************************************************
+* Copyright contributors to the oneDAL project
+*
+* Licensed under the Apache License, Version 2.0 (the "License");
+* you may not use this file except in compliance with the License.
+* You may obtain a copy of the License at
+*
+*     http://www.apache.org/licenses/LICENSE-2.0
+*
+* Unless required by applicable law or agreed to in writing, software
+* distributed under the License is distributed on an "AS IS" BASIS,
+* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+* See the License for the specific language governing permissions and
+* limitations under the License.
+*******************************************************************************/
+
+#include <iomanip>
+#include <iostream>
+
+#include "oneapi/dal/algo/linear_regression.hpp"
+#include "oneapi/dal/io/csv.hpp"
+#include "oneapi/dal/spmd/ccl/communicator.hpp"
+
+#include "utils.hpp"
+
+namespace dal = oneapi::dal;
+namespace lr = dal::linear_regression;
+
+void run() {
+    const auto train_data_file_name = get_data_path("data/linear_regression_train_data.csv");
+    const auto train_response_file_name =
+        get_data_path("data/linear_regression_train_responses.csv");
+    const auto test_data_file_name = get_data_path("data/linear_regression_test_data.csv");
+    const auto test_response_file_name = get_data_path("data/linear_regression_test_responses.csv");
+
+    const auto x_train = dal::read<dal::table>(dal::csv::data_source{ train_data_file_name });
+    const auto y_train = dal::read<dal::table>(dal::csv::data_source{ train_response_file_name });
+    const auto x_test = dal::read<dal::table>(dal::csv::data_source{ test_data_file_name });
+    const auto y_test = dal::read<dal::table>(dal::csv::data_source{ test_response_file_name });
+
+    auto comm = dal::preview::spmd::make_communicator<dal::preview::spmd::backend::ccl>();
+    auto rank_id = comm.get_rank();
+    auto rank_count = comm.get_rank_count();
+
+    auto x_train_vec = split_table_by_rows<float>(x_train, rank_count);
+    auto y_train_vec = split_table_by_rows<float>(y_train, rank_count);
+    auto x_test_vec = split_table_by_rows<float>(x_test, rank_count);
+    auto y_test_vec = split_table_by_rows<float>(y_test, rank_count);
+
+    const double alpha = 1.0;
+    const auto rr_desc = lr::descriptor<float>{ true, alpha };
+
+    const auto result_train =
+        dal::preview::train(comm, rr_desc, x_train_vec.at(rank_id), y_train_vec.at(rank_id));
+
+    const auto result_infer =
+        dal::preview::infer(comm, rr_desc, x_test_vec.at(rank_id), result_train.get_model());
+
+    if (comm.get_rank() == 0) {
+        std::cout << "Ridge regression alpha: " << alpha << std::endl;
+
+        std::cout << "Prediction results:\n" << result_infer.get_responses() << std::endl;
+
+        std::cout << "Ground truth:\n" << y_test_vec.at(rank_id) << std::endl;
+    }
+}
+
+int main(int argc, char const *argv[]) {
+    ccl::init();
+    int status = MPI_Init(nullptr, nullptr);
+    if (status != MPI_SUCCESS) {
+        throw std::runtime_error{ "Problem occurred during MPI init" };
+    }
+
+    run();
+
+    status = MPI_Finalize();
+    if (status != MPI_SUCCESS) {
+        throw std::runtime_error{ "Problem occurred during MPI finalize" };
+    }
+    return 0;
+}

samples/oneapi/cpp/mpi/sources/ridge_regression_distr_mpi.cpp

@@ -0,0 +1,81 @@
+/*******************************************************************************
+* Copyright contributors to the oneDAL project
+*
+* Licensed under the Apache License, Version 2.0 (the "License");
+* you may not use this file except in compliance with the License.
+* You may obtain a copy of the License at
+*
+*     http://www.apache.org/licenses/LICENSE-2.0
+*
+* Unless required by applicable law or agreed to in writing, software
+* distributed under the License is distributed on an "AS IS" BASIS,
+* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+* See the License for the specific language governing permissions and
+* limitations under the License.
+*******************************************************************************/
+
+#include <iomanip>
+#include <iostream>
+
+#include "oneapi/dal/algo/linear_regression.hpp"
+#include "oneapi/dal/io/csv.hpp"
+#include "oneapi/dal/spmd/mpi/communicator.hpp"
+
+#include "utils.hpp"
+
+namespace dal = oneapi::dal;
+namespace lr = dal::linear_regression;
+
+void run() {
+    const auto train_data_file_name = get_data_path("data/linear_regression_train_data.csv");
+    const auto train_response_file_name =
+        get_data_path("data/linear_regression_train_responses.csv");
+    const auto test_data_file_name = get_data_path("data/linear_regression_test_data.csv");
+    const auto test_response_file_name = get_data_path("data/linear_regression_test_responses.csv");
+
+    const auto x_train = dal::read<dal::table>(dal::csv::data_source{ train_data_file_name });
+    const auto y_train = dal::read<dal::table>(dal::csv::data_source{ train_response_file_name });
+    const auto x_test = dal::read<dal::table>(dal::csv::data_source{ test_data_file_name });
+    const auto y_test = dal::read<dal::table>(dal::csv::data_source{ test_response_file_name });
+
+    auto comm = dal::preview::spmd::make_communicator<dal::preview::spmd::backend::mpi>();
+    auto rank_id = comm.get_rank();
+    auto rank_count = comm.get_rank_count();
+
+    auto x_train_vec = split_table_by_rows<float>(x_train, rank_count);
+    auto y_train_vec = split_table_by_rows<float>(y_train, rank_count);
+    auto x_test_vec = split_table_by_rows<float>(x_test, rank_count);
+    auto y_test_vec = split_table_by_rows<float>(y_test, rank_count);
+
+    const double alpha = 1.0;
+    const auto rr_desc = lr::descriptor<float>{ true, alpha };
+
+    const auto result_train =
+        dal::preview::train(comm, rr_desc, x_train_vec.at(rank_id), y_train_vec.at(rank_id));
+
+    const auto result_infer =
+        dal::preview::infer(comm, rr_desc, x_test_vec.at(rank_id), result_train.get_model());
+
+    if (comm.get_rank() == 0) {
+        std::cout << "Ridge regression alpha: " << alpha << std::endl;
+
+        std::cout << "Prediction results:\n" << result_infer.get_responses() << std::endl;
+
+        std::cout << "Ground truth:\n" << y_test_vec.at(rank_id) << std::endl;
+    }
+}
+
+int main(int argc, char const *argv[]) {
+    int status = MPI_Init(nullptr, nullptr);
+    if (status != MPI_SUCCESS) {
+        throw std::runtime_error{ "Problem occurred during MPI init" };
+    }
+
+    run();
+
+    status = MPI_Finalize();
+    if (status != MPI_SUCCESS) {
+        throw std::runtime_error{ "Problem occurred during MPI finalize" };
+    }
+    return 0;
+}