auto-round

72f5ce80 - support autoround hpu format (#182)

support autoround hpu format (#182) * qlinear_hpu Signed-off-by: sys-lpot-val <sys_lpot_val@intel.com> * remove comments Signed-off-by: sys-lpot-val <sys_lpot_val@intel.com> * code change for post_init Signed-off-by: sys-lpot-val <sys_lpot_val@intel.com> * remove unnecessary log and add device arg for evaluation Signed-off-by: sys-lpot-val <sys_lpot_val@intel.com> * update readme Signed-off-by: sys-lpot-val <sys_lpot_val@intel.com> * update readme doc Signed-off-by: sys-lpot-val <sys_lpot_val@intel.com> * restore evaluation code Signed-off-by: sys-lpot-val <sys_lpot_val@intel.com> * fix zp plus_1 Signed-off-by: sys-lpot-val <sys_lpot_val@intel.com> * enable hpu inference of gptq models Signed-off-by: yintong-lu <yintong.lu@intel.com> * change auto_quantizer logic Signed-off-by: yintong-lu <yintong.lu@intel.com> * add ut and add qlinear_hpu_gptq Signed-off-by: Heng Guo <henggou@habana.ai> * add hqt version check and refine doc Signed-off-by: Heng Guo <henggou@habana.ai> * comment qlinear_hpu Signed-off-by: Heng Guo <henggou@habana.ai> * fix coverage issue Signed-off-by: yintong-lu <yintong.lu@intel.com> * fix coverage Signed-off-by: yintong-lu <yintong.lu@intel.com> * fix coverage Signed-off-by: yintong-lu <yintong.lu@intel.com> * fix coverage Signed-off-by: yintong-lu <yintong.lu@intel.com> --------- Signed-off-by: sys-lpot-val <sys_lpot_val@intel.com> Signed-off-by: yintong-lu <yintong.lu@intel.com> Signed-off-by: Heng Guo <henggou@habana.ai> Co-authored-by: sys-lpot-val <sys_lpot_val@intel.com> Co-authored-by: Heng Guo <henggou@habana.ai>