223 Commits (a64920c46802a3030bfdf8754870ffb1e5c6ba78)

Author SHA1 Message Date
  i-robot a1395344e4 !26848 [MS][LITE] add some gpu fp32 gate models 4 years ago
  i-robot ab26378451 !26900 fix matmul int8 bug 4 years ago
  greatpanc 8681c77d47 test support models 4 years ago
  yeyunpeng2020 a2b2bd516f fix matmul int8 bug && support activation symmetry 4 years ago
  wang_shaocong 2b356dd4a0 [MSLITE] Fix ci smoketest 4 years ago
  i-robot c2af5258cc !26769 optimize the Quantizer base class 4 years ago
  yeyunpeng2020 8aaff90f7e optimize the Quantizer base class and optimize min_quant_weight_channel strategy 4 years ago
  zengxianglong 229b94e8a7 fix npu compatibility bugs and strided_slice-transpose decrease bugs 4 years ago
  i-robot 6b6bdf5d1d !26507 [MSLITE] Add scale fusion. 4 years ago
  i-robot 9d023b6445 !26564 [MS][LITE][develop] fix deconv unstable accuracy 4 years ago
  wang_shaocong ba0d056735 [MSLITE] Add scale_scale fusion 4 years ago
  sunsuodong 6f9330f874 fix deconv unstable accuracy 4 years ago
  i-robot e8226f8f4f !26578 Optimize full quantization code structure 4 years ago
  yeyunpeng2020 730678e7ab Optimize full quantization code structure 4 years ago
  i-robot a148442099 !26509 post training bind thread 4 years ago
  yeyunpeng2020 a107abd292 post training quant bind thread 4 years ago
  i-robot e6fad31e5c !26470 [MS][LITE]modify switch actor to support switch layer 4 years ago
  mengyuanli a091a56ee9 modify switch actor to support switch_layer 4 years ago
  wang_shaocong 70f2024d5a [MSLITE] Complete mul_add fusion 4 years ago
  sunsuodong 3d1066b6a0 add fp16 resize kernel 4 years ago
  wang_shaocong fdd3522f8c [MSLITE] BatchNorm to scale fusion 4 years ago
  i-robot 279a099936 !25138 [MS][LITE][ToD] mindir_fp16 4 years ago
  nizzan 665e2a738e Support mix precision mindir file 4 years ago
  wang_shaocong 1265c27e91 [MSLITE] Fuzz test. 4 years ago
  i-robot 0b7797b413 !25712 gather op、cast op bugfix, and add new cast type op, and new PackNCHWToNHWC4 method 4 years ago
  greatpanc 9d40ceba5e add cast int32 type op, and some model gate 4 years ago
  Liu_Xuu bf5617475d [MSLITE] add max, bias_add, argmax op and modify bugs for ocr models in tensorrt delegate 4 years ago
  i-robot 8ee62721c9 !25296 add broadcast feature for matmul op 4 years ago
  wangyanling 3a0a1c6b72 add broadcast feature for matmul op 4 years ago
  i-robot 324c4f0ca2 !25453 add gpu some gate config 4 years ago
  greatpanc d565704db0 add gpu some gate config 4 years ago
  zengxianglong 06cb50ad6f fix overflow mistakes 4 years ago
  i-robot 1e245f1487 !25433 matmul support per channel 4 years ago
  yeyunpeng2020 ce84d12c49 matmul support per channel 4 years ago
  yangruoqi713 dbd1114e74 [MSLITE][DEVELOP] add channel check for conv 4 years ago
  i-robot 3f2d75f31b !25206 Q888_CV_model_age_gender.pb bugfix, add gate config, ulfgf gaze_corrector 4 years ago
  greatpanc 5b33db5581 add gate config 4 years ago
  i-robot 5a66cff789 !25165 layernorm fusion and int8 div gather bug fix 4 years ago
  i-robot f62f494d75 !23754 [MSLITE] add ci scripts for tensorrt 4 years ago
  zhaozhenlong fc7f495ee1 fix div gather and layernorm fusion 4 years ago
  Liu_Xuu eee4efccd6 [MSLITE] add tensorrt st ci 4 years ago
  zengxianglong 240db233f3 optimize entrance guard 4 years ago
  i-robot 07505278d9 !24745 [MSLITE] add models to ci test. 4 years ago
  wang_shaocong 53fe2effe0 [MSLITE] add models to ci test. 4 years ago
  yoni 097e09d541 mini_yolo bit exact 4 years ago
  i-robot e83df2cdfa !24807 【MS】【LITE】registry clean code 4 years ago
  chenjianping b3353a15c7 clean code 4 years ago
  i-robot 87df077986 !24579 [MS][LITE] simplify npu code and add npu models 4 years ago
  i-robot 47b22be58b !24592 one model onto npu gate 4 years ago
  zengxianglong 55e59b8924 simplify npu code and add npu models 4 years ago