US Patent Application 18214101. NODE FUSION METHOD FOR COMPUTATIONAL GRAPH AND DEVICE simplified abstract
Contents
NODE FUSION METHOD FOR COMPUTATIONAL GRAPH AND DEVICE
Organization Name
Inventor(s)
Zhaochuang Zhang of Shenzhen (CN)
NODE FUSION METHOD FOR COMPUTATIONAL GRAPH AND DEVICE - A simplified explanation of the abstract
- This abstract for appeared for US patent application number 18214101 Titled 'NODE FUSION METHOD FOR COMPUTATIONAL GRAPH AND DEVICE'
Simplified Explanation
This application presents a method for combining nodes in a computational graph. The method involves converting a neural network into a computational graph and identifying groups of nodes that can be processed in parallel. These groups are determined based on the relationships between the nodes, such as having a common parent or child node, or having no parent or child node. Finally, the method fuses multiple nodes from different sub-branches within each parallelizable group to create a new computational graph.
Original Abstract Submitted
Embodiments of this application disclose a node fusion method for a computational graph and a device. The method includes: converting a neural network into a computational graph; extracting one or more parallelizable branch groups from the computational graph based on a dependency relationship between nodes in the computational graph, where the dependency relationship indicates at least one of the following relationships: the parallelizable branch group has a common parent node, the parallelizable branch group has a common child node, the parallelizable branch group has no parent node, and the parallelizable branch group has no child node; and finally, fusing a plurality of nodes in any parallelizable branch group that respectively belong to different sub-branches to obtain a new computational graph.