Methods and apparatuses for high performance and accuracy fixed-point scale implementation

1. A data processing method, comprising:

receiving floating-point input data over a channel of a standalone floating-point scale layer in a neural network, and converting the floating-point input data into fixed-point input data of the standalone floating-point scale layer;

obtaining fixed-point quantization parameters in each channel based on input data and two floating-point parameters γ i , β i in each channel, wherein the standalone floating-point scale layer comprises a plurality of channels, and the fixed-point quantization parameters are generated separately for each of the plurality of channels;

converting the standalone floating-point scale layer based on the fixed-point quantization parameters into a fixed-point scale layer for processing the fixed-point input data to generate fixed-point output data; and

mapping the fixed-point scale layer to a fixed-point convolution layer, wherein computation of convolution is done by matrix multiplication executed on a General Matrix Multiplication (GEMM) engine.

2. The data processing method according to claim 1 , wherein processing the fixed-point input data to generate the fixed-point output data further comprises:

for the fixed-point input data in a size of 16-bit, multiplying the fixed-point input data with a first fixed-point quantization parameter in size S16 to receive a first output;

summing up the first output and a second fixed-point quantization parameter in size S47 to receive a second output;

right shifting with rounding the second output with an accumulator shift in size U8 to receive a third output;

clamping the third output to receive a fourth output in size S32;

multiplying the fourth output with an output scale in size U16 to receive a fifth output;

right shifting with rounding the fifth output with an output shift in size U8 to receive a sixth output; and

clamping the sixth output into the fixed-point output data in size S16 or U16.

3. The data processing method according to claim 2 , wherein for the fixed-point input data in the size of 16-bit, a preferred range of the second fixed-point quantization parameter is 30 to 47 bits.

4. The data processing method according to claim 1 , wherein processing the fixed-point input data to generate the fixed-point output data further comprises:

for the fixed-point input data in a size of 8-bit, multiplying the fixed-point input data with a first fixed-point quantization parameter in size S8 or U8 to receive a first output;

summing up the first output and a second fixed-point quantization parameter in size S31 to receive a second output;

right shifting with rounding the second output with an accumulator shift in size U8 to receive a third output;

clamping the third output to receive a fourth output in size S16;

multiplying the fourth output with an output scale in size U16 to receive a fifth output;

right shifting with rounding the fifth output with an output shift in size U8 to receive a sixth output; and

clamping the sixth output into the fixed-point output data in size S8 or U8.

5. The data processing method according to claim 4 , wherein for the fixed-point input data in the size of 8-bit, a preferred range of the second fixed-point quantization parameter is 15 to 31 bits.

6. The data processing method according to claim 1 , wherein mapping the fixed-point scale layer to the fixed-point convolution layer further comprises:

multiplying the fixed-point input data with a filter weight for the channel in the fixed-point scale layer to receive a product; and

summing up the product and a bias for the channel in the fixed-point scale layer.

7. The data processing method according to claim 1 , wherein the matrix multiplication is executed on a GEMM engine or a Multiply-Accumulate (MAC) operations array.

8. An apparatus for implementing a neural network, comprising:

one or more processors; and

a memory configured to store instructions executable by the one or more processors;

wherein the one or more processors, upon execution of the instructions, are configured to:

receive floating-point input data over a channel of a standalone floating-point scale layer in a neural network, and convert the floating-point input data into fixed-point input data of the standalone floating-point scale layer;

obtain fixed-point quantization parameters in each channel based on input data and two floating-point parameters γ i , β i in each channel, wherein the standalone floating-point scale layer comprises a plurality of channels, and the fixed-point quantization parameters are generated separately for each of the plurality of channels;

convert the standalone floating-point scale layer based on the fixed-point quantization parameters into a fixed-point scale layer in the neural network for processing the fixed-point input data to generate fixed-point output data; and

map the fixed-point scale layer to a fixed-point convolution layer, wherein computation of convolution is done by matrix multiplication executed on a GEMM engine.

9. The apparatus of claim 8 , wherein the one or more processors are further configured to:

for the fixed-point input data in a size of 16-bit, multiply the fixed-point input data with a first fixed-point quantization parameter in size S16 to receive a first output;

sum up the first output and a second fixed-point quantization parameter in size S47 to receive a second output;

right shift with rounding the second output with an accumulator shift in size U8 to receive a third output;

clamp the third output to receive a fourth output in size S32;

multiply the fourth output with an output scale in size U16 to receive a fifth output;

right shift with rounding the fifth output with an output shift in size U8 to receive a sixth output; and

clamp the sixth output into the output data in size S16 or U16.

10. The apparatus of claim 9 , wherein for the fixed-point input data in the size of 16-bit, a preferred range of the second fixed-point quantization parameter is 30 to 47 bits.

11. The apparatus of claim 8 , wherein the one or more processors are further configured to:

for the fixed-point input data in a size of 8-bit, multiply the fixed-point input data with a first fixed-point quantization parameter in size S8 or U8 to receive a first output;

sum up the first output and a second fixed-point quantization parameter in size S31 to receive a second output;

right shift with rounding the second output with an accumulator shift in size U8 to receive a third output;

clamp the third output to receive a fourth output in size S16;

multiply the fourth output with an output scale in size U16 to receive a fifth output;

right shift with rounding the fifth output with a third parameter in size U8 to receive a sixth output; and

clamp the sixth output into the fixed-point output data in size S8 or U8.

12. The apparatus of claim 11 , wherein for the fixed-point input data in the size of 8-bit, a preferred range of the second fixed-point quantization parameter is 15 to 31 bits.

13. The apparatus of claim 8 , the one or more processors are further configured to:

multiply the fixed-point input data with a filter weight for the channel in the fixed-point scale layer to receive a product; and

sum up the product and a bias for the channel in the fixed-point scale layer.

14. The apparatus of claim 8 , wherein the matrix multiplication is executed on a GEMM engine or a MAC operations array.

15. A non-transitory computer readable storage medium, comprising instructions stored therein to implement a neural network, wherein, upon execution of the instructions by one or more processors, the instructions cause the one or more processors to perform acts comprising:

receiving floating-point input data over a channel of a standalone floating-point scale layer in a neural network, and converting the floating-point input data into fixed-point input data of the standalone floating-point scale layer;

obtaining fixed-point quantization parameters in each channel based on input data and two floating-point parameters γ i , β i in each channel, wherein the standalone floating-point scale layer comprises a plurality of channels, and the fixed-point quantization parameters are generated separately for each of the plurality of channels;

converting the standalone floating-point scale layer based on the fixed-point quantization parameters into a fixed-point scale layer in the neural network for processing the fixed-point input data to generate fixed-point output data; and

mapping the fixed-point scale layer to a fixed-point convolution layer, wherein computation of convolution is done by matrix multiplication that executed on a GEMM engine.

16. The non-transitory computer readable storage medium of claim 15 , wherein processing the fixed-point input data to generate the fixed-point output data further comprises:

for the fixed-point input data in a size of 16-bit, multiplying the fixed-point input data with a first fixed-point quantization parameter in size S16 to receive a first output;

summing up the first output and a second fixed-point quantization parameter in size S47 to receive a second output;

right shifting with rounding the second output with an accumulator shift in size U8 to receive a third output;

clamping the third output to receive a fourth output in size S32;

multiplying the fourth output with an output scale in size U16 to receive a fifth output;

right shifting with rounding the fifth output with an output shift in size U8 to receive a sixth output; and

clamping the sixth output into the fixed-point output data in size S16 or U16.

17. The non-transitory computer readable storage medium of claim 16 , wherein for the fixed-point input data in the size of 16-bit, a preferred range of the second fixed-point quantization parameter is 30 to 47 bits.

18. The non-transitory computer readable storage medium of claim 15 , wherein processing the fixed-point input data to generate the fixed-point output data further comprises:

for the fixed-point input data in a size of 8-bit, multiplying the fixed-point input data with a first fixed-point quantization parameter in size S8 or U8 to receive a first output;

summing up the first output and a second fixed-point quantization parameter in size S31 to receive a second output;

right shifting with rounding the second output with an accumulator shift in size U8 to receive a third output;

clamping the third output to receive a fourth output in size S16;

multiplying the fourth output with an output scale in size U16 to receive a fifth output;

right shifting with rounding the fifth output with a third parameter in size U8 to receive a sixth output; and

clamping the sixth output into the fixed-point output data in size S8 or U8.

19. The non-transitory computer readable storage medium of claim 18 , wherein for the fixed-point input data in the size of 8-bit, a preferred range of the second fixed-point quantization parameter is 15 to 31 bits.

20. The non-transitory computer readable storage medium of claim 15 , wherein mapping the fixed-point scale layer to the fixed-point convolution layer further comprises:

multiplying the fixed-point input data with a filter weight for the channel in the fixed-point scale layer to receive a product; and

summing up the product and a bias for the channel in the fixed-point scale layer.

21. The non-transitory computer readable storage medium of claim 15 , wherein the matrix multiplication is executed on GEMM engine or a MAC operations array.