US 12,141,579 B2
Inference processing of decision tree models using vector instructions
Jan Van Lunteren, Ruschlikon (CH)
Assigned to INTERNATIONAL BUSINESS MACHINES CORPORATION, Armonk, NY (US)
Filed by INTERNATIONAL BUSINESS MACHINES CORPORATION, Armonk, NY (US)
Filed on Mar. 14, 2023, as Appl. No. 18/183,187.
Prior Publication US 2024/0311148 A1, Sep. 19, 2024
Int. Cl. G06F 9/30 (2018.01); G06F 16/901 (2019.01); G06N 5/01 (2023.01); G06N 5/04 (2023.01); G06N 20/00 (2019.01)
CPC G06F 9/30036 (2013.01) [G06F 16/9027 (2019.01); G06N 5/01 (2023.01); G06N 5/04 (2013.01); G06N 20/00 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A method for inference processing of a decision tree model in processing apparatus which executes vector instructions to perform inference computations on vectors of operands stored in vector registers of the processing apparatus, the method comprising:
for each decision tree of the decision tree model, indexing nodes of the decision tree by consecutive node indexes which are assigned to nodes in a breadth-first order and increase with node-depth in the decision tree model;
during the inference processing, storing a vector of N node indexes, corresponding to a set of nodes for which N inference computations will be performed in parallel, in a vector register of the processing apparatus; and
adaptively selecting a granularity N of the vector of node indexes, in dependence on the node-depth of nodes in the set of nodes, to accelerate inference processing of the decision tree model.