oneDAL
oneDAL copied to clipboard
moved results_to_compute to [train|infer]_input classes
moved functionality for assignment of results_to_compute to train|infer input classes
/intelci: run
Renaming train_results_to_compute with error_metrics_to_compute. Infer_results_to_compute -> results_to_compute. And returned them back into descriptor (due to only one results_to_compute available now). Also remove std::uint64_t from input and rewrote operators correspondingly. Please see request https://github.com/oneapi-src/oneDAL/pull/799
@makart19 - i assume we don't need this PR?
@makart19 - do we need this PR or it can be removed?