A method of performing operations on a plurality of inputs and a same kernel using a delay time by using a same processor, and a neural network device thereof are provided, the neural network device includes input data including a first input and a second input, and a processor configured to obtain
A method of performing operations on a plurality of inputs and a same kernel using a delay time by using a same processor, and a neural network device thereof are provided, the neural network device includes input data including a first input and a second input, and a processor configured to obtain a first result by performing operations between the first input and a plurality of kernels, to obtain a second result by performing operations between the second input, which is received at a time delayed by a first interval from a time when the first input is received, and the plurality of kernels, and to obtain output data using the first result and the second result. The neural network device may include neuromorphic hardware and may perform convolutional neural network (CNN) mapping.
대표청구항▼
1. A neural network device comprising: input data comprising a first input and a second input; anda processor configured toobtain a first result by performing operations between the first input and a plurality of kernels,obtain a second result by performing operations between the second input, which
1. A neural network device comprising: input data comprising a first input and a second input; anda processor configured toobtain a first result by performing operations between the first input and a plurality of kernels,obtain a second result by performing operations between the second input, which is received at a time delayed by a first interval from a time when the first input is received, and the plurality of kernels, andobtain output data using the first result and the second result. 2. The neural network device of claim 1, wherein the neural network device comprises neuromorphic hardware configured to perform convolution neural network (CNN) mapping using the first input and the second input. 3. The neural network device of claim 1, further comprising a memory storing instructions that, when executed by the processor, cause the processor to drive a neural network by executing the instructions and performing an operation on the input data. 4. The neural network device of claim 1, wherein the input data comprises image data, andwherein the first input comprises data with respect to a first region of the image data and the second input comprises data with respect to a second region of the image data. 5. The neural network device of claim 4, wherein the first region and the second region partially overlap and are adjacent to each other. 6. The neural network device of claim 5, wherein the processor is further configured to obtain the second result by performing operations between the second input and the plurality of kernels, in response to the second input being a valid input. 7. The neural network device of claim 6, wherein the processor is further configured to determine that the second input is the valid input, in response to the second input being pixel data constituting the second region. 8. The neural network device of claim 1, wherein the processor is further configured to receive data streams having different delay times and representing image data from input terminals, to receive the first input from the data streams received from the input terminals, and to receive the second input from the data streams received from the input terminals. 9. The neural network device of claim 8, wherein the first input is received during a first cycle, andwherein the second input is received during a second cycle delayed by the first interval from the first cycle. 10. The neural network device of claim 1, wherein the processor is further configured to obtain the first result by adding operation results between the first input and the plurality of kernels, and to obtain the second result by adding operation results between the second input and the plurality of kernels. 11. The neural network device of claim 1, wherein the processor is further configured to receive a third input included in the input data at a time delayed by a second interval from a time when the second input is received, to obtain a third result by performing operations between the third input and the plurality of kernels, and to obtain the output data by using the first result, the second result, and the third result. 12. A method, performed by a neural network device, of performing an operation on input data comprising a first input and a second input, the method comprising: obtaining a first result by performing operations between the first input and a plurality of kernels using a processor in the neural network device;obtaining a second result by performing operations between the second input received at a time delayed by a first interval from a time when the first input is received and the plurality of kernels using the processor; andobtaining output data using the first result and the second result. 13. The method of claim 12, wherein the neural network device comprises neuromorphic hardware configured to perform convolution neural network (CNN) mapping using the first input and the second input. 14. The method of claim 12, wherein the input data comprises image data, andwherein the first input comprises data with respect to a first region of the image data and the second input comprises data with respect to a second region of the image data. 15. The method of claim 14, wherein the first region and the second region partially overlap and are adjacent to each other. 16. The method of claim 15, wherein the obtaining of the second result comprises: obtaining the second result by performing operations between the second input and the plurality of kernels, in response to determining that the second input is a valid input. 17. The method of claim 16, wherein the determining of whether the second input is the valid input comprises: determining that the second input is the valid input, in response to the second input being pixel data constituting the second region. 18. The method of claim 12, further comprising: receiving data streams having different delay times and representing image data from input terminals; wherein the first input comprises data streams received from the input terminals, andwherein the second input comprises data streams received from the input terminals. 19. The method of claim 18, wherein the first input is received during a first cycle, andwherein the second input is received during a second cycle delayed by the first interval from the first cycle. 20. The method of claim 12, wherein the obtaining of the first result comprises obtaining the first result by adding operation results between the first input and the plurality of kernels, andwherein the obtaining of the second result comprises obtaining the second result by adding operation results between the second input and the plurality of kernels. 21. The method of claim 12, further comprising: receiving a third input included in the input data at a time delayed by a second interval from a time when the second input is received by using the processor; andobtaining a third result by performing operations between the third input and the plurality of kernels,wherein the obtaining of the output data comprises obtaining the output data using the first result, the second result, and the third result. 22. A non-transitory computer-readable recording medium storing instructions that, when executed by a processor, cause the processor to perform the method of claim 12.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.