686 Commits (f0088335bbe5e91d3463ca756e7012cfcc2d645a)

Author SHA1 Message Date
  Megvii Engine Team f7b0395976 perf(mgb/compile): improve compile time according the file map of compile time 3 years ago
  Megvii Engine Team 124f38c44d perf(mgb/compile): improve compile time for megbrain 3 years ago
  Megvii Engine Team 0a266d7a1d feat(riscv): speed up bazel build and fix rv64gc without rvv build 3 years ago
  Megvii Engine Team 36ba1d6d39 fix(riscv): fix ci fp16 build and move test GI_TEST_NAIVE by megdnn_gi_api_test 3 years ago
  Megvii Engine Team 698dcef491 feat(gi/x86): fix _mm_slli_si128 build at clang 3 years ago
  Megvii Engine Team 2d806f9c3c feat(gi): make conv_bias apply gi class type 3 years ago
  Megvii Engine Team 19d36fa03c feat(gi): make pooling apply gi class type 3 years ago
  Megvii Engine Team 8546c15d45 feat(gi): make elemwise apply gi class type 3 years ago
  Megvii Engine Team 74fb63db29 feat(gi): make matrix_mul apply gi class type 3 years ago
  Megvii Engine Team 45b26400e7 feat(gi): make resize apply gi class type 3 years ago
  Megvii Engine Team 7d7cc3c8da feat(gi/riscv): add gi support with risc-v 3 years ago
  Megvii Engine Team a32b727720 fix(build): upgrade bazel riscv toolchains 3 years ago
  Megvii Engine Team 24c5c19bf0 fix(imperative): make functional ops support negative axis 3 years ago
  Megvii Engine Team f96429c031 feat(imperative): support empty tensor in roi_align 3 years ago
  Megvii Engine Team 8f17b84ad8 fix(dnn): fix dnn run cd4 on cpu 3 years ago
  Megvii Engine Team 81065cf00e build(mgb/cutlass): merge partial headers 3 years ago
  Megvii Engine Team c2deef1a97 feat(mge): aad atlas710 support 3 years ago
  Megvii Engine Team 4e66e0eb1f feat(megdnn/softmax): add softmax operator in OpenCL 3 years ago
  Megvii Engine Team 6c9b3a58e3 refactor(dnn): remove algorithm cache queries 3 years ago
  Megvii Engine Team 96d90be1c6 feat(dnn): fallback support int4 relayout 3 years ago
  Megvii Engine Team 711b5bf502 fix(dnn/arm_common): fix some load beyond memory 3 years ago
  Megvii Engine Team 3ebb8db01a feat(third_party/cutlass): update to version 2.8 3 years ago
  Megvii Engine Team da91e650a5 refactor(ops/layer_norm): speed up the host speed of layer_norm 3 years ago
  Megvii Engine Team cd26376549 style(imperative/amp): reformat code 3 years ago
  Megvii Engine Team 6f0b582064 chore(imperative/amp): adapt dev 3 years ago
  Megvii Engine Team fc0f454685 fix(dnn/check_non_finite): adjust some details of CheckNonFinite 3 years ago
  Megvii Engine Team 3bd40887b6 feat(mgb/opr): add NHWC support for AdaptivePooling 3 years ago
  Megvii Engine Team 98b5ee78c1 feat(mge/dnn): add lamb optimizer 3 years ago
  Megvii Engine Team 9e0583e13a feat(dnn/arm_common): add arm_common chanwise dot 11x11 3 years ago
  Megvii Engine Team c62ddba238 feat(dnn/opencl): optimize heuristic rule 3 years ago
  Megvii Engine Team c2500cdb7e chore(license): apply change caused by bot forward rebase 3 years ago
  Megvii Engine Team 5f0e7ffb64 feat(fallback): add FB_GI_F32_4x12 benchmark 3 years ago
  Megvii Engine Team f249d387de feat(fallback): imp gi matmul FB_GI_F32_4x12 algo 3 years ago
  Megvii Engine Team 03f78547f7 feat(dnn/arm_common): add 9x9s1s2 dot chanwise kernel 3 years ago
  Megvii Engine Team c2e9860feb chore(license): remove all license in file header 3 years ago
  Megvii Engine Team 4cce2480d5 fix(dnn/opencl): fix some bug for dnn opencl conv bias and relayout format 3 years ago
  Megvii Engine Team e98049d77e feat(fallback): move arm_common resize f32 algo to fallback gi 3 years ago
  Megvii Engine Team 7c8f184723 fix(dnn/x86): fix x86 pooling exec 3 years ago
  Megvii Engine Team 91aaafd587 feat(fallback): move arm_common pooling f32 algo to fallback gi 3 years ago
  Megvii Engine Team 48526abb79 fix(mgb): fix concat cd4 tensor check size invalid 3 years ago
  Megvii Engine Team af6cdb2004 feat(fallback): fix ci 3 years ago
  Megvii Engine Team e4cc85e52c feat(fallback): move arm_common f32 convbias to fallback gi 3 years ago
  Megvii Engine Team 0f1afb0935 feat(fallback): imp gi matmul AlgoF32GiMK4_4x8 algo, 3 years ago
  Megvii Engine Team 410dcb6c69 feat(fallback): add more gi api for conv, and add gi API test 3 years ago
  Megvii Engine Team 05186e7bd9 fix(midout): fix elemwise crash after midout 3 years ago
  Megvii Engine Team 70209667e8 fix(dnn/test): fix some bug when force_deduce_layout is off 3 years ago
  Megvii Engine Team 597a1e791b refactor(imperative): add interface to clear algorithm cache 3 years ago
  Megvii Engine Team e2f5156b69 refactor(megbrain): save fastrun result to algorithm cache 3 years ago
  Megvii Engine Team d968942fe3 perf(cuda): speedup direct large kernel conv 3 years ago
  Megvii Engine Team 7dc347697a feat(dnn/cuda): add typecvt uint16 3 years ago