US 12,242,973 B2
Graph context-based operator checks to improve graph clustering and execution in AI accelerator framework integration
Chandrakant Khandelwal, Bengaluru (IN); Ritesh Kumar Rajore, Bengaluru (IN); Laxmi Ganesan, Bengaluru (IN); Sai Jayanthi, Bengaluru (IN); and Yamini Nimmagadda, Portland, OR (US)
Assigned to Intel Corporation, Santa Clara, CA (US)
Filed by Intel Corporation, Santa Clara, CA (US)
Filed on Aug. 13, 2021, as Appl. No. 17/402,114.
Prior Publication US 2021/0374554 A1, Dec. 2, 2021
Int. Cl. G06N 3/10 (2006.01); G06F 8/41 (2018.01); G06F 9/445 (2018.01); G06F 16/901 (2019.01)
CPC G06N 3/10 (2013.01) [G06F 8/427 (2013.01); G06F 8/443 (2013.01); G06F 9/44542 (2013.01); G06F 16/9024 (2019.01)] 16 Claims
OG exemplary drawing
 
1. A computing system comprising:
a network controller;
a processor coupled to the network controller; and,
a memory coupled to the processor, the memory including a plurality of executable instructions which, when executed by the processor, cause the processor to:
parse a deep learning graph in topological order to identify a plurality of nodes,
mark a first set of nodes in the plurality of nodes as unsupported by target hardware, and
mark a second set of nodes in the plurality of nodes as supported by the target hardware, wherein the first set of nodes and the second set of nodes are marked based on one or more attributes defining operation functionality, wherein the one or more attributes include one or more of an input node parameter, a dimension, or a shape, and wherein one or more nodes in the second set of nodes are to be associated with a node that has one or more of a constant operator input node, an input node that has a shape attribute with no zero dimension values, or the one or more attributes supported on the target hardware.