1670 Commits (release-1.5)
 

Author SHA1 Message Date
  Megvii Engine Team 809d5056cd feat(mge/distributed): enable pt shm allreduce 4 years ago
  Megvii Engine Team 0245594156 test(autograd): test jvp emulated by 2nd grad 3 years ago
  Megvii Engine Team 8480302da8 fix(autograd): make higher order grad experimental 4 years ago
  Megvii Engine Team 72531f2b4a test(autograd): add more tests for higher order grad 4 years ago
  Megvii Engine Team 522e556b83 feat(autodiff): support higher order grad 4 years ago
  Megvii Engine Team 5198b7834c fix(mge/functional): fix expand_dims for scalar 4 years ago
  Megvii Engine Team 88898e63a5 fix(mgb): replace if_constexpr with runtime function to avoid potential 4 years ago
  Megvii Engine Team 25932352e9 refactor(mgb/dnn): rocm pooling rebase algochooser 4 years ago
  Megvii Engine Team 1cfdbc565c feat(dnn): add deterministic max pooling 4 years ago
  Megvii Engine Team 20ab82d00c fix(tee): fix tee crash 4 years ago
  Megvii Engine Team 933dd9a497 feat(mge/distributed): add cuda env check before forked thread 4 years ago
  Megvii Engine Team 2a54196117 fix(tee): fix tee link 4 years ago
  Megvii Engine Team a5060a2bfe feat(mgb/opr): add check_has_inf kernel and opr 4 years ago
  Megvii Engine Team 3597a6dbd7 feat(dnn/arm): nchw_nchw44 conv support 1x1s1 4 years ago
  Megvii Engine Team c64b1c94cd feat(imperative/functional): add roll in functional 4 years ago
  Megvii Engine Team d915c5a3fd refactor(mgb): make convolution3D handle noncontiguous tensors 4 years ago
  Megvii Engine Team d04cd67faf refactor(mgb): make conv-backward-filter handle noncontiguous tensors 4 years ago
  Megvii Engine Team 44376f702a refactor(mgb): make conv-backward-data handle noncontiguous tensors 4 years ago
  Megvii Engine Team 7b2a76d1ee refactor(mgb): make conv handle noncontiguous tensors 4 years ago
  Megvii Engine Team ca2828ddcb fix(dnn/x86): fix x86 int8 matmul ldc bug 4 years ago
  Megvii Engine Team aa4e847692 fix(interpreter): release gil when interpreter blocking 4 years ago
  Megvii Engine Team bd62a0a647 fix(interpreter): remove notice flag of produce_tensor 4 years ago
  Megvii Engine Team c78a7848a1 test(profiler): simple tests for profiler 4 years ago
  Megvii Engine Team d6079db1f6 feat(profiler): add memory flow format for profiler 4 years ago
  Megvii Engine Team 4286bb9f51 feat(profiler): add launch script for profiler 4 years ago
  Megvii Engine Team 9d47c3babd feat(profiler): imperative profiler support tracing 4 years ago
  Megvii Engine Team cdcb46ba95 feat(profiler): add chrome timeline format for profiler 4 years ago
  Megvii Engine Team 1d64792b41 refactor(profiler): detach profiler from interpreter 4 years ago
  Megvii Engine Team f2027b8d17 refactor(interpreter): recompute with do_apply_op 4 years ago
  Megvii Engine Team 5cb35c1bed refactor(profiler): add state structs to replay recorded events 4 years ago
  Megvii Engine Team 84fc5c927d refactor(profiler): remove dump logic for old profiler 4 years ago
  Megvii Engine Team 5b5a826166 refactor(profiler): use macro to simplify event recording/definition 4 years ago
  Megvii Engine Team 1ce40b5bf7 refactor(interpreter): wrap accesses to channel/worker state 4 years ago
  Megvii Engine Team 90d3905799 feat(mge): add mge._exit 4 years ago
  Megvii Engine Team 846642a881 docs(misc): add some notes 4 years ago
  Megvii Engine Team fb8f1534ca fix(cmake/debug): fix asan do not work at windows env 4 years ago
  Megvii Engine Team 40085acbae fix(mgb): remove unnecessary cudnn8 warning 4 years ago
  Megvii Engine Team 54a4d70eb5 feat(src/serialization): add support of serializing metadata 4 years ago
  Megvii Engine Team 721091faf0 fix(core): fix thread local is not supported in ios 4 years ago
  Megvii Engine Team 1d86528162 ci(config): change pip mirror to pypi cache proxy 4 years ago
  Megvii Engine Team 82b95a3aa2 fix(lar): fix build error 4 years ago
  Megvii Engine Team 8cfe3c3937 fix(lar): update --verbose to --model-info 4 years ago
  Megvii Engine Team 2e5a473de1 feat(lar): support --verbose 4 years ago
  Megvii Engine Team b7e596b4a1 perf(autograd): copy inputs before capture in backward_graph_grad_rule 4 years ago
  Megvii Engine Team 2ac3c9dc8b fix(trace): constants in backward graph treat as ImmutableTensor corectlly 4 years ago
  Megvii Engine Team 7eea1fc65d feat(lite/opencl): fix CMake OpenCL and megenginelite py import 4 years ago
  Megvii Engine Team 62bd6c823b feat(cmake/debug): misc for build 4 years ago
  Megvii Engine Team c7a5c21a7a chore(bazel/cmake): fix lite bazel/cmake symbols 4 years ago
  Megvii Engine Team 3e4e4c4604 feat(mgb/jit): add graph_opt_config and jit_config interfaces 4 years ago
  Megvii Engine Team 1c7d0802ab fix(cuda): remove cuda driver version check and runtime minor version 4 years ago