ENH: Ridge Regression CPU SPMD support by DDJHB · Pull Request #3506 · uxlfoundation/oneDAL

DDJHB · 2026-02-09T21:54:56Z

Description

This PR adds distributed (SPMD) training support for Ridge Regression on CPU, following Linear Regression changes.

Checklist:

Completeness and readability

I have commented my code, particularly in hard-to-understand areas.
Git commit message contains an appropriate signed-off-by string (see CONTRIBUTING.md for details).
I have resolved any merge conflicts that might occur with the base branch.

Testing

I have run it locally and tested the changes extensively.
All CI jobs are green or I have provided justification why they aren't.
No new failures.
I have extended testing suite if new functionality was introduced in this PR.

Performance

Not applicable

DDJHB · 2026-02-09T21:56:40Z

/intelci: run

david-cortes-intel · 2026-02-10T07:26:25Z

 }

+TEMPLATE_LIST_TEST_M(lr_spmd_test, "RR common flow", "[rr][spmd]", lr_types) {
+    SKIP_IF(this->not_float64_friendly());


Why would it need to skip this one if it's on CPU?

This test is common for CPU and GPU and this check is necessary just for GPU without native float64 support

@DDJHB @Alexandr-Solovev it would be better to add if_gpu condition then.

DDJHB · 2026-03-29T13:15:26Z

/intelci: run

Copilot

Pull request overview

Adds CPU distributed (SPMD) Ridge Regression training support to the oneAPI flow, mirroring existing distributed Linear Regression patterns and extending the SPMD test suite accordingly.

Changes:

Added new distributed Ridge Regression C++ samples for MPI and CCL backends.
Extended linear_regression SPMD tests with a Ridge Regression (alpha-regularized) common-flow test.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 5 comments.

File	Description
samples/oneapi/cpp/mpi/sources/ridge_regression_distr_mpi.cpp	New MPI-based distributed Ridge Regression sample using SPMD `preview::train` / `preview::infer` with `alpha` regularization.
samples/oneapi/cpp/ccl/sources/ridge_regression_distr_ccl.cpp	New CCL-based distributed Ridge Regression sample using SPMD communicator and `alpha` regularization.
cpp/oneapi/dal/algo/linear_regression/test/spmd.cpp	Adds an RR-tagged SPMD test invoking `run_and_check_ridge()` to validate distributed ridge flow.

Copilot · 2026-04-14T11:13:40Z

+#include <iomanip>
+#include <iostream>
+
+#include "oneapi/dal/algo/linear_regression.hpp"
+#include "oneapi/dal/io/csv.hpp"
+#include "oneapi/dal/spmd/mpi/communicator.hpp"
+


std::runtime_error is used in main() but <stdexcept> is not included. Relying on transitive includes is non-portable and may break builds with stricter standard library headers. Include <stdexcept> directly in this sample.

Valid, but this should be changed in all the samples - a task for another PR.

Copilot · 2026-04-14T11:13:40Z

+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" };
+    }


If run() throws (e.g., CSV read / train / infer exceptions), MPI_Finalize() will be skipped, which can leave MPI in an undefined state and cause hangs/aborts on some runtimes. Consider wrapping run() in a try/catch that always calls MPI_Finalize() (or introduce an RAII guard that finalizes in its destructor).

This code is aligned with other samples. Need to make those changes in all samples - a task for another PR.

Copilot · 2026-04-14T11:13:40Z

+#include <iomanip>
+#include <iostream>
+
+#include "oneapi/dal/algo/linear_regression.hpp"
+#include "oneapi/dal/io/csv.hpp"
+#include "oneapi/dal/spmd/ccl/communicator.hpp"
+


std::runtime_error is used in main() but <stdexcept> is not included. Relying on transitive includes is non-portable and may break builds with stricter standard library headers. Include <stdexcept> directly in this sample.

Copilot · 2026-04-14T11:13:40Z

+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" };
+    }


If run() throws (e.g., CSV read / train / infer exceptions), MPI_Finalize() will be skipped, which can leave MPI in an undefined state and cause hangs/aborts on some runtimes. Consider wrapping run() in a try/catch that always calls MPI_Finalize() (or introduce an RAII guard that finalizes in its destructor).

Vika-F · 2026-04-15T08:21:00Z

/intelci: run

ethanglaser · 2026-04-30T18:57:51Z

These samples (oneapi/cpp) do not appear to be running anywhere in our private CI - is this expected? Or if they are, please point to where

feat: cpu spmd ridge

f97bfbc

DDJHB added the enhancement label Feb 9, 2026

david-cortes-intel reviewed Feb 10, 2026

View reviewed changes

DDJHB added 4 commits February 11, 2026 01:57

refactor: minor stylistic updates

5ec7f36

Merge branch 'main' into dev_ridge_cpu_spmd

8eb0879

merge with main cleanup

1e00670

refactor: clang format

e786cba

Vika-F reviewed Mar 30, 2026

View reviewed changes

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

Vika-F and others added 2 commits April 14, 2026 13:09

Skip float64 testing only on CPU

9d7deff

Merge branch 'main' into dev_ridge_cpu_spmd

a31bc4d

Vika-F requested a review from Copilot April 14, 2026 11:09

Copilot started reviewing on behalf of Vika-F April 14, 2026 11:10 View session

Copilot AI reviewed Apr 14, 2026

View reviewed changes

Vika-F reviewed Apr 14, 2026

View reviewed changes

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

Skip float64 testing only on GPU

c3055b0

Vika-F marked this pull request as ready for review April 16, 2026 08:31

Vika-F requested review from avolkov-intel, ethanglaser and icfaust as code owners April 16, 2026 08:31

Vika-F approved these changes Apr 16, 2026

View reviewed changes

Conversation

DDJHB commented Feb 9, 2026 • edited by Vika-F Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Description

Uh oh!

DDJHB commented Feb 9, 2026

Uh oh!

david-cortes-intel Feb 10, 2026

Choose a reason for hiding this comment

Uh oh!

Alexandr-Solovev Feb 10, 2026

Choose a reason for hiding this comment

Uh oh!

Vika-F Feb 11, 2026

Choose a reason for hiding this comment

Uh oh!

DDJHB commented Mar 29, 2026

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Uh oh!

Copilot AI Apr 14, 2026

Choose a reason for hiding this comment

Uh oh!

Vika-F Apr 15, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 14, 2026

Choose a reason for hiding this comment

Uh oh!

Vika-F Apr 15, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 14, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 14, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Vika-F commented Apr 15, 2026

Uh oh!

ethanglaser commented Apr 30, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants

DDJHB commented Feb 9, 2026 •

edited by Vika-F

Loading