Systems and methods for processing a stream of data values using data value subset groups

A method of processing a stream of data values in a data processing system, the method comprising:

• processing, in each of a plurality of iterations, a respective particular data value of the stream by operating on a respective particular subset of data values of the stream, by, in each of the plurality of iterations:
• retrieving group indication data which indicates at least one group defined for a previous, different subset of data values stored in a previous iteration, wherein the particular subset of data values comprises data values from the previous subset of data values;
• using the retrieved group indication data to define, for the particular subset of data values in the current iteration, a set of groups into which data values within the particular subset can be grouped;
• classifying each of the data values within the particular subset into one of the groups of the set of groups;
• processing the particular data value using one or more of the data values of the particular subset in dependence on the classification of the data values of the particular subset into the groups;
• and
• storing group indication data for at least one group of the set of groups defined for the particular subset of data values in the current iteration, for use in a subsequent iteration;
• wherein the method further comprises outputting the processed data values.

A non-transitory computer readable storage medium having stored thereon a computer readable description of an integrated circuit that, when processed in an integrated circuit manufacturing system, causes the integrated circuit manufacturing system to manufacture a data processing system configured to process a stream of data values, the data processing system comprising:

• a store configured to store group indication data which indicates one or more groups into which data values can be grouped;
• and
• a processing module configured to process, in each of a plurality of iterations, a respective particular data value of the stream by operating on a respective particular subset of data values of the stream, by, in each of the plurality of iterations:
• retrieving, from the store, group indication data for at least one group defined for a previous, different subset of data values stored in a previous iteration, wherein the particular subset of data values comprises data values from the previous subset of data values;
• using the retrieved group indication data to define, for the particular subset of data values in the current iteration, a set of groups into which data values within the particular subset can be grouped;
• classifying each of the data values within the particular subset into one of the groups of the set of groups;
• processing the particular data value using one or more of the data values of the particular subset in dependence on the classification of the data values of the particular subset into the groups;
• and
• storing, in the store, group indication data for at least one group of the set of groups defined for the particular subset of data values in the current iteration, for use in a subsequent iteration;
• wherein the processing module is configured to output the processed data values.