Zhejiang Lab

Lingfang Zeng

Hongsheng Wang

Guang Chen

Hujun Bao

Yong Li

Hongcai Cheng

Huanbo Zheng

Aimin Pan

Jian Zhu

The present disclosure discloses a method for execution of a computational graph in a neural network model and an apparatus thereof, including: creating task execution bodies on a native machine according to a physical computational graph compiled and generated by a deep learning framework, and designing a solution for allocating a plurality of idle memory blocks to each task execution body, so that the entire computational graph participates in deep learning training tasks of different batches of data in a pipelining and parallelizing manner.

Method for execution of computational graph in neural network model and apparatus thereof

Disclosed are an intermediate representation method and apparatus for parallel execution of graph computation. The method includes the following steps: S: compiling a neural network into a computational graph on a computer; S: defining branch states of tensor variables in the computational graph; S: defining a data dependency relationship of the tensor variables in the computational graph; S: defining a control dependency relationship of the tensor variables in the computational graph; S: building a data dependency relationship graph of the tensor variables in the computational graph; S: building a control dependency relationship graph of the tensor variables in the computational graph; and S: transforming control dependencies into data dependencies. The present application derives, based on the dependency relationship, a parallel computing method that can execute the branch threads in parallel in the global computational graph, and optimizes the compilation efficiency of the computational graph.

Intermediate representation method and apparatus for parallel execution of graph computation

Growing community of inventors

Hangzhou, China

Lingfang Zeng