ml_types.h File Reference

Strongly typed enumerations describing the ML backend, tensor element type and inference task. More...

This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Namespaces
namespace	MLLIB
	Tensors, model abstraction, ONNX Runtime inference and Python training drivers used across mne-cpp.

Enumerations
enum class	MLLIB::MlBackend { MLLIB::OnnxRuntime , MLLIB::BuiltIn }
enum class	MLLIB::MlDataType { MLLIB::Float32 , MLLIB::Float64 , MLLIB::Int64 }
enum class	MLLIB::MlTaskType { MLLIB::Classification , MLLIB::Regression , MLLIB::FeatureExtraction }

Detailed Description

Strongly typed enumerations describing the ML backend, tensor element type and inference task.

SPDX-License-Identifier: BSD-3-Clause Copyright (c) 2026 MNE-CPP Authors

Author

Christoph Dinh chris.nosp@m.toph.nosp@m..dinh.nosp@m.@mne.nosp@m.-cpp..nosp@m.org

Since

2.2.0

Date

April 2026

The three enum class definitions in this header are the shared vocabulary used across every MLLIB interface: MLLIB::MlBackend selects which runtime executes a model (currently ONNX Runtime, with a placeholder BuiltIn entry reserved for native fallbacks), MLLIB::MlDataType names the scalar element of MLLIB::MlTensor, and MLLIB::MlTaskType labels what a loaded model is meant to do so calling code can route the output tensor without inspecting the graph (classification logits, regression scalars, embedding vectors).

Keeping these as scoped enums avoids the usual implicit-int hazards and lets the header stay dependency-free so any other MLLIB header can pull it in without dragging Eigen or ONNX along.

Definition in file ml_types.h.