wangwei
/
MegEngine
Not watched
1
0
Fork 0
Code
Releases 31 Wiki Activity
Issues 0 Pull Requests 0 Datasets Model Cloudbrain
2512 Commits
28 Branches
25 MB
895 Download
Tree: bde2efa3b5
Branches Tags
${ item.name }
Create branch ${ searchTerm }
from 'bde2efa3b5'
${ noResults }
Commit Graph

2512 Commits (bde2efa3b58412b52b8be9d05be35f11b74ae906)
 

Author SHA1 Message Date
  Megvii Engine Team bde2efa3b5 feat(lite/load_and_run): support put and get model redis cache 4 years ago
  Megvii Engine Team 48526abb79 fix(mgb): fix concat cd4 tensor check size invalid 4 years ago
  Megvii Engine Team c87d998e59 feat(mgb): add interface to support opencl IO zero copy when inference 4 years ago
  Megvii Engine Team af6cdb2004 feat(fallback): fix ci 4 years ago
  Megvii Engine Team e4cc85e52c feat(fallback): move arm_common f32 convbias to fallback gi 4 years ago
  Megvii Engine Team 0f1afb0935 feat(fallback): imp gi matmul AlgoF32GiMK4_4x8 algo, 4 years ago
  Megvii Engine Team 410dcb6c69 feat(fallback): add more gi api for conv, and add gi API test 4 years ago
  Megvii Engine Team a0e531180d fix(src/comp_node): fix calling cuda driver api 4 years ago
  Megvii Engine Team ccea0e2386 fix(dnn/rdnn): add warmup before profile 4 years ago
  Megvii Engine Team 58df717e37 fix(mge/autodiff): fix attaching tensor already in gradient path 4 years ago
  Megvii Engine Team 05186e7bd9 fix(midout): fix elemwise crash after midout 4 years ago
  Megvii Engine Team 9be8de6025 fix(midout): formatting midout tools 4 years ago
  Megvii Engine Team 8182af6eb6 fix(mgb): fix strategy of grad_op and opr_attr 4 years ago
  Megvii Engine Team 70209667e8 fix(dnn/test): fix some bug when force_deduce_layout is off 4 years ago
  Megvii Engine Team 597a1e791b refactor(imperative): add interface to clear algorithm cache 4 years ago
  Megvii Engine Team e2f5156b69 refactor(megbrain): save fastrun result to algorithm cache 4 years ago
  Megvii Engine Team f902ba2433 docs(megbrain): add notes for fastrun 4 years ago
  Megvii Engine Team d968942fe3 perf(cuda): speedup direct large kernel conv 4 years ago
  Megvii Engine Team b2cffdde3e fix(lite): fix lite cpu default not work 4 years ago
  Megvii Engine Team dbce6526d6 fix(mge/functional): fix return dtype of comparison function 4 years ago
  Megvii Engine Team 7dc347697a feat(dnn/cuda): add typecvt uint16 4 years ago
  Megvii Engine Team b92866d2c2 fix(build): fix build depends dirty file issue 4 years ago
  megvii-mge 9d8983f323
Merge pull request #457 from kagome1007/updatenewlogo 4 years ago
  “wenjuan” e3f591cadf fix(mge): update logo 4 years ago
  Megvii Engine Team 4b27e861f4 fix(ops): implement from_op_node for reshape 4 years ago
  Megvii Engine Team 4fb3d886b8 perf(misc): use reinterpret_cast to convert valueshape 4 years ago
  Megvii Engine Team 951ed476d6 fix(minigraph): supports varnode forwarding 4 years ago
  Megvii Engine Team 27d4c4b36c refactor(stats): use static inline variable declaration 4 years ago
  Megvii Engine Team 787a22a9d6 perf(tensor): implement __new__ in cpp 4 years ago
  Megvii Engine Team 99df4a7996 fix(dtype): dtype scalar set_retain_dtype supports bool 4 years ago
  Megvii Engine Team 53d48ca399 fix(fastrun): persistent cache add prefix 4 years ago
  Megvii Engine Team 7005b7d691 fix(mge): fix fastpath check 4 years ago
  Megvii Engine Team 7bf5b0ee1e test(imperative): check env values after each pytest 4 years ago
  Megvii Engine Team 811579b0e7 fix(imperative): restrict value converts to symbolvar 4 years ago
  Megvii Engine Team b3f79966fd fix(mgb): fix "TRT_ERROR: INVALID_ARGUMENT: Get binding data type failed." 4 years ago
  Megvii Engine Team 99a85c4079 fix(mge): fix advanced indexing grad 4 years ago
  Megvii Engine Team 409c988163 fix(imperative): add matmul apply_on_varnode 4 years ago
  Megvii Engine Team d52ba79d89 fix(lite): support set data by copy on device tensor 4 years ago
  Megvii Engine Team 275f12c98a fix(mge): fix dimshuffle shape infer 4 years ago
  Megvii Engine Team e59b6e13a3 fix(imperative/src): fix empty_tensor bug of convbwd&rng 4 years ago
  Megvii Engine Team 115c4592c0 fix(dnn/opencl): fix opencl elemwise tuning issue 4 years ago
  Megvii Engine Team b9cbc10120 feat(lite): add pack model 4 years ago
  Megvii Engine Team 7927e98fd6 perf(mge): speed up PixelShuffle 4 years ago
  Megvii Engine Team 43e5f41c4b docs(mgb/docs): add alternative interface for deprecated api 4 years ago
  Megvii Engine Team ba90a02834 feat(mge): opt third_party prepare 2/2 4 years ago
  Megvii Engine Team ff9e612152 fix(lite): fix lar LITE_WITH_CUDA and LITE_WITH_OPENCL invalid for bazel 4 years ago
  Megvii Engine Team 73112558d0 feat(mge/dnn): support checknonfinite for fp16 4 years ago
  Megvii Engine Team f7e10ea8d7 perf(imperative): improve matmul/batch_matmul 4 years ago
  Megvii Engine Team 1c2a323e78 feat(mge): add warning message when mismatched cuda sm is detected 4 years ago
  Megvii Engine Team 877bda4180 perf(mge): improve cross stream memory borrowing 4 years ago