ELIMINATING MEMORY BOTTLENECKS FOR DEPTHWISE CONVOLUTIONS
Certain aspects of the present disclosure provide techniques for efficient depthwise convolution. A convolution is performed with a compute-in-memory (CIM) array to generate CIM output, and at least a portion of the CIM output corresponding to a first output data channel, of a plurality of output da...
Gespeichert in:
Hauptverfasser: | , , , , , |
---|---|
Format: | Patent |
Sprache: | eng ; fre ; ger |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | Certain aspects of the present disclosure provide techniques for efficient depthwise convolution. A convolution is performed with a compute-in-memory (CIM) array to generate CIM output, and at least a portion of the CIM output corresponding to a first output data channel, of a plurality of output data channels in the CIM output, is written to a digital multiply-accumulate (DMAC) activation buffer. A patch of the CIM output is read from the DMAC activation buffer, and weight data is read from a DMAC weight buffer. Multiply-accumulate (MAC) operations are performed with the patch of CIM output and the weight data to generate a DMAC output. |
---|