A processing unit and a method for multiplying at least two multiplicands. The multiplicands are present in an exponential notation, that is, each multiplicand is assigned an exponent and a base. The processing unit is configured to carry out a multiplication of the multiplicands and includes at least one bitshift unit, the bitshift unit shifting a binary number a specified number of places, in particular, to the left; an arithmetic unit, which carries out an addition of two input variables and a subtraction of two input variables; and a storage device. A computer program, which is configured to execute the method, and a machine-readable storage element, in which the computer program is stored, are also described.

A processing unit and a method for multiplying at least two multiplicands. The multiplicands are present in an exponential notation, that is, each multiplicand is assigned an exponent and a base. The processing unit is configured to carry out a multiplication of the multiplicands and includes at least one bitshift unit, the bitshift unit shifting a binary number a specified number of places, in particular, to the left; an arithmetic unit, which carries out an addition of two input variables and a subtraction of two input variables; and a storage device. A computer program, which is configured to execute the method, and a machine-readable storage element, in which the computer program is stored, are also described.

A neural network processor is disclosed that includes a combined convolution and pooling circuit that can perform both convolution and pooling operations. The circuit can perform a convolution operation by a multiply circuit determining products of corresponding input feature map and convolution kernel weight values, and an add circuit accumulating the products determined by the multiply circuit in storage. The circuit can perform an average pooling operation by the add circuit accumulating input feature map data values in the storage, a divisor circuit determining a divisor value, and a division circuit dividing the data value accumulated in the storage by the determined divisor value. The circuit can perform a maximum pooling operation by a maximum circuit determining a maximum value of input feature map data values, and storing the determined maximum value in the storage.

A neural network processor is disclosed that includes a combined convolution and pooling circuit that can perform both convolution and pooling operations. The circuit can perform a convolution operation by a multiply circuit determining products of corresponding input feature map and convolution kernel weight values, and an add circuit accumulating the products determined by the multiply circuit in storage. The circuit can perform an average pooling operation by the add circuit accumulating input feature map data values in the storage, a divisor circuit determining a divisor value, and a division circuit dividing the data value accumulated in the storage by the determined divisor value. The circuit can perform a maximum pooling operation by a maximum circuit determining a maximum value of input feature map data values, and storing the determined maximum value in the storage.

This disclosure relates to a data processing method, a data processing apparatus, and related products. The products include a control unit. The control unit includes: an instruction caching unit, an instruction processing unit, and a storage queue unit. The instruction caching unit is used for storing a calculation instruction associated with an artificial neural network computation; the instruction processing unit is used for parsing the calculation instruction to obtain a plurality of computation instructions; and the storage queue unit is used for storing an instruction queue, where the instruction queue includes the plurality of computation instructions or calculation instructions to be executed according to a front-back sequence of a queue. Through the above method of this disclosure, computation efficiency of the related products during a neural network model computation may be improved.

This disclosure relates to a data processing method, a data processing apparatus, and related products. The products include a control unit. The control unit includes: an instruction caching unit, an instruction processing unit, and a storage queue unit. The instruction caching unit is used for storing a calculation instruction associated with an artificial neural network computation; the instruction processing unit is used for parsing the calculation instruction to obtain a plurality of computation instructions; and the storage queue unit is used for storing an instruction queue, where the instruction queue includes the plurality of computation instructions or calculation instructions to be executed according to a front-back sequence of a queue. Through the above method of this disclosure, computation efficiency of the related products during a neural network model computation may be improved.

A neural network accelerator is disclosed that includes a multiplication unit, an adder-tree unit and an accumulator unit. The multiplication unit and the adder tree unit are configured to perform lattice-multiplication operations. The accumulator unit is coupled to an output of the adder tree to form dot-product values from the lattice-multiplication operations performed by the multiplication unit and the adder tree unit. The multiplication unit includes n multiplier units that perform lattice-multiplication-based operations and output product values. Each multiplier unit includes a plurality of multipliers. Each multiplier unit receives first and second multiplicands that each include a most significant nibble (MSN) and a least significant nibble (LSN). The multipliers in each multiplier unit receive different combinations of the MSNs and the LSNs of the multiplicands. The multiplication unit and the adder can provide mixed-precision dot-product computations.

A neural network accelerator is disclosed that includes a multiplication unit, an adder-tree unit and an accumulator unit. The multiplication unit and the adder tree unit are configured to perform lattice-multiplication operations. The accumulator unit is coupled to an output of the adder tree to form dot-product values from the lattice-multiplication operations performed by the multiplication unit and the adder tree unit. The multiplication unit includes n multiplier units that perform lattice-multiplication-based operations and output product values. Each multiplier unit includes a plurality of multipliers. Each multiplier unit receives first and second multiplicands that each include a most significant nibble (MSN) and a least significant nibble (LSN). The multipliers in each multiplier unit receive different combinations of the MSNs and the LSNs of the multiplicands. The multiplication unit and the adder can provide mixed-precision dot-product computations.

A combined function specified by an instruction is performed. The combined function includes a plurality of operations performed as part of one invocation of the combined function. The performing the combined function includes performing a matrix multiplication of a first tensor and a second tensor to obtain one or more intermediate results. The second tensor includes an adjusted weight tensor created using a multiplier. Values of a bias tensor are added to the one or more intermediate results to obtain one or more results for the combined function. The one or more results are at least a part of an output tensor.

A combined function specified by an instruction is performed. The combined function includes a plurality of operations performed as part of one invocation of the combined function. The performing the combined function includes performing a matrix multiplication of a first tensor and a second tensor to obtain one or more intermediate results. The second tensor includes an adjusted weight tensor created using a multiplier. Values of a bias tensor are added to the one or more intermediate results to obtain one or more results for the combined function. The one or more results are at least a part of an output tensor.