Methods and apparatus for configuring a texture filter pipeline for deep learning operation

An apparatus for a deep learning operation, the apparatus comprising:

• a memory;
• and
• at least one processor coupled to the memory and configured to:
• map one or more inputs of the deep learning operation to a respective input of at least one texture filtering logic unit in a graphics pipeline, wherein the one or more inputs are based on one or more batch weight features and one or more activation features;
• generate, by the at least one texture filtering logic unit, at least one output for the deep learning operation based on the one or more inputs mapped to the at least one texture filtering logic unit;
• and
• communicate the at least one output to a programmable shader.

An apparatus for a deep learning operation, the apparatus comprising:

• means for mapping one or more inputs of the deep learning operation to a respective input of at least one texture filtering logic unit in a graphics pipeline, wherein the one or more inputs are based on one or more batch weight features and one or more activation features;
• means for generating, by the at least one texture filtering logic unit, at least one output for the deep learning operation based on the one or more inputs mapped to the at least one texture filtering logic unit;
• and
• means for communicating the at least one output to a programmable shader.

A non-transitory computer-readable medium storing computer executable code for a deep learning operation, the code when executed by a processor causes the processor to:

• map one or more inputs of the deep learning operation to a respective input of at least one texture filtering logic unit in a graphics pipeline, wherein the one or more inputs are based on one or more batch weight features and one or more activation features;
• generate, by the at least one texture filtering logic unit, at least one output for the deep learning operation based on the one or more inputs mapped to the at least one texture filtering logic unit;
• and
• communicate the at least one output to a programmable shader.