Floating-point conversion with denormalization

A computer system for floating-point conversion with denormalization in a single clock cycle, the computer system comprising:

• one or more computer-readable non-transitory storage media configured to store computer program code;
• and
• one or more computer processors configured to access said computer program code and operate as instructed by said computer program code, said computer program code including:
• receiving code configured to cause the one or more computer processors to receive an input floating-point number corresponding to an input data type;
• extracting code configured to cause the one or more computer processors to extract an exponent value and a fraction value from the received input floating-point number;
• determining code configured to cause the one or more computer processors to determine a biasing constant associated with converting the received input floating-point number from the input data type to an output data type;
• biasing code configured to cause the one or more computer processors to bias the exponent value based on the biasing constant;
• converting code configured to cause the one or more computer processors to convert the fraction value to the output data type based on a denormalization constant associated with the extracted exponent value and the determined biasing constant, wherein biasing the exponent value and converting the fraction value to the output data type occurs in a single clock cycle based on the converting the fraction value occurring in parallel with biasing the exponent value;
• and
• outputting code configured to cause the one or more computer processors to output a floating-point number in the output data type corresponding to the converted fraction value and the biased exponent value, wherein the output data type is configured to be used with an artificial intelligence system.

A non-transitory computer readable medium having stored thereon a computer program for floating-point conversion with denormalization in a single clock cycle, the computer program configured to cause one or more computer processors to:

• receive an input floating-point number corresponding to an input data type;
• extract an exponent value and a fraction value from the received input floating-point number;
• determine a biasing constant associated with converting the received input floating-point number from the input data type to an output data type;
• bias the exponent value based on the biasing constant;
• convert the fraction value to the output data type based on a denormalization constant associated with the extracted exponent value and the determined biasing constant, wherein biasing the exponent value and converting the fraction value to the output data type occurs in a single clock cycle based on the converting the fraction value occurring in parallel with biasing the exponent value;
• and
• output a floating-point number in the output data type corresponding to the converted fraction value and the biased exponent value, wherein the output data type is configured to be used with an artificial intelligence system.