1. e1651a5 COMPMID-2414 Add add_2D_tensor_argument_if for OpenCL by Michalis Spyrou · 5 years ago
  2. 2d7e683 COMPMID-1694: Fuse offset contribution with the output stage when we use NEGEMMLowpMatrixMultiplyCore by George Wort · 5 years ago
  3. 4b90865 COMPMID-1413 - Improve the performance of GEMMLowp with 8 bit dot product on OpenCL by Gian Marco Iodice · 6 years ago