Integrated circuit chip apparatus
AssigneeCambricon Technologies Corporation Limited
InventorsShaoli Liu, Xinkai Song, Bingrui Wang, Yao Zhang, Shuai Hu
Life of this patent
Estimating the term…
Abstract
Provided are an integrated circuit chip apparatus and a related product, the integrated circuit chip apparatus being used for executing a multiplication operation, a convolution operation or a training operation of a neural network. The present technical solution has the advantages of a small amount of calculation and low power consumption.
Claims20
An integrated circuit chip apparatus, comprising:
- a main processing circuit;
- and
- a plurality of basic processing circuits comprising data type conversion circuits configured to convert data between a floating data type and a fixed point data type,
- wherein the main processing circuit is configured to transfer data and a computation instruction to the plurality of basic processing circuits,
- wherein the plurality of basic processing circuits are configured to, when convert the data transferred to the basic processing circuits is in the floating point data type:
- convert the data into the fixed point data type;
- perform a first set of neural network computations in parallel in the fixed point data type on the data to obtain computation results in the fixed point data type;
- convert the computation results in the fixed point data type into the floating point data type;
- and
- transfer the computation results to the main processing circuit in the floating point data type.
A processing system, comprising:
- a neural network computing apparatus;
- a general interconnection interface;
- and
- a general-purpose processing apparatus connected to the neural network computing apparatus via the general interconnection interface,
- wherein the neural network computing apparatus further comprises:
- a main processing circuit;
- and
- a plurality of basic processing circuits comprising data type conversion circuits configured to convert data between a floating data type and a fixed point data type,
- wherein the main processing circuit is configured to transfer data and a computation instruction to the plurality of basic processing circuits,
- wherein the plurality of basic processing circuits are configured to, when the data transferred to the basic processing circuits is in the floating point data type:
- convert the data into the fixed point data type;
- perform a first set of neural network computations in parallel in the fixed point data type on the data to obtain computation results in the fixed point data type;
- convert the computation results in the fixed point data type into the floating point data type:
- and
- transfer the computation results to the main processing circuit in the floating point data type.
A method for performing neural network operations using an integrated circuit chip apparatus comprising a main processing circuit, and a plurality of basic processing circuits comprising data type conversion circuits configured to convert data between a floating data type and a fixed point data type, the method comprising:
- transferring data and a computation instruction from the main processing circuit to the plurality of basic processing circuits;
- when the data transferred to the basic processing circuits is in the floating point data type, converting, by the plurality of basic processing circuits, the data into the fixed point data type;
- performing, by the plurality of basic processing circuits, a first set of neural network computations in parallel in the fixed point data type on the data to obtain computation results in the fixed point data type;
- convert the computation results in the fixed point data type into the floating point data type;
- transferring the computation results to the main processing circuit in the floating point data type;
- and
- performing, by the main processing circuit, a second set of neural network computations in series on the computation results.
Dependencies are inferred from the claim text.