Abstract:
A system receives a neural network model that includes asymmetric operations. Each asymmetric operation includes one or more fixed-point operands that are asymmetrically-quantized from corresponding floating-point operands. The system compiles a given asymmetric operation of the neural network model into a symmetric operation that includes a combined bias value. A compiler computes the combined bias value is a constant by merging at least zero points of input and output of the given asymmetric operation. The system then generates a symmetric neural network model including the symmetric operation for inference hardware to execute in fixed-point arithmetic.
Abstract:
A digital compute-in-memory (DCIM) system includes a first DCIM macro. The first DCIM macro includes a first memory cell array and a first arithmetic logic unit (ALU). The first memory cell array has N rows that are configured to store weight data of a neural network in a single weight data download session, wherein N is a positive integer not smaller than two. The first ALU is configured to receive a first activation input, and perform convolution operations upon the first activation input and a single row of weight data selected from the N rows of the first memory cell array to generate first convolution outputs.
Abstract:
A digital compute-in-memory (DCIM) macro includes a memory cell array and an arithmetic logic unit (ALU). The memory cell array stores weight data of a neural network. The ALU receives parallel bits of a same input channel in an activation input, and generates a convolution computation output of the parallel bits and target weight data in the memory cell array.
Abstract:
Techniques pertaining to dynamic enablement, disablement and adjustment of offset of a virtual periodic timing control signal based on one or more predefined events are described. A method may determine whether a first predefined event is beginning. The method may also enable an offset of the virtual periodic timing control signal for synchronizing one or more first system modules in response to a determination that the first predefined event is beginning. The one or more first system modules may be configured to control one or more operations of one or more second system modules. The one or more second system modules may be configured to process one or more image frames. The method may further determine whether the first predefined event is ending. The method may additionally disable the offset in response to a determination that the first predefined event is ending.
Abstract:
Techniques pertaining to dynamic enablement, disablement and adjustment of offset of a virtual periodic timing control signal based on one or more predefined events are described. A method may determine whether a first predefined event is beginning. The method may also enable an offset of the virtual periodic timing control signal for synchronizing one or more first system modules in response to a determination that the first predefined event is beginning. The one or more first system modules may be configured to control one or more operations of one or more second system modules. The one or more second system modules may be configured to process one or more image frames. The method may further determine whether the first predefined event is ending. The method may additionally disable the offset in response to a determination that the first predefined event is ending.