Megvii Engine Team
|
6667100638
|
feat(mge): use weakref for GradManger.attach
GitOrigin-RevId: 6df336c3c1
|
4 years ago |
Megvii Engine Team
|
b9918c329d
|
feat(mge/distributed): support distributed key-value store
GitOrigin-RevId: b4abe80014
|
4 years ago |
Megvii Engine Team
|
495472954d
|
fix(trace): link io-op to avoid deadlock
GitOrigin-RevId: 872cb6b715
|
4 years ago |
Megvii Engine Team
|
094601e834
|
feat(mge/distributed): allow remote grad by using grad manager
GitOrigin-RevId: a890c206a5
|
4 years ago |
Megvii Engine Team
|
02438ee635
|
refactor(mge/distributed): use thread.Threading to create Server
GitOrigin-RevId: fc994411bf
|
4 years ago |
Megvii Engine Team
|
a1efd4d56d
|
fix(mge/parampack): fix param pack when no param left
GitOrigin-RevId: b88b876064
|
4 years ago |
Megvii Engine Team
|
b309890c66
|
docs(mge): pytest for sphinx docstring
GitOrigin-RevId: 8bed12562a
|
4 years ago |
Megvii Engine Team
|
09241a1ff7
|
feat(mge): remove param_pack_* from functional
GitOrigin-RevId: a5fe25be8c
|
4 years ago |
Megvii Engine Team
|
f5f86a05c4
|
docs(mge/distributed): add distributed.server docs
GitOrigin-RevId: 929d6adfcc
|
4 years ago |
Megvii Engine Team
|
6c5cf25f4d
|
docs(mge/distributed): add distributed.helper docs
GitOrigin-RevId: 37c14aa11f
|
4 years ago |
Megvii Engine Team
|
026af62042
|
docs(mge): docs typo fix
GitOrigin-RevId: 851f6de02f
|
4 years ago |
Megvii Engine Team
|
36a4fb5611
|
ci(image): update docker image install python package
GitOrigin-RevId: 171e95b3d9
|
4 years ago |
Megvii Engine Team
|
1a24fb29c1
|
perf(mge/allreduce): put allreduce on another cuda stream
GitOrigin-RevId: 2e778dfa04
|
4 years ago |
Megvii Engine Team
|
aac8de554e
|
fix(mge/parampacksplit): fix parampacksplit refcnt error
GitOrigin-RevId: c964465596
|
4 years ago |
Megvii Engine Team
|
e507228e74
|
feat(mge/examples): add distributed training examples using launcher
GitOrigin-RevId: 5db26f58eb
|
4 years ago |
Megvii Engine Team
|
8d02d10483
|
refactor(mge/distributed): change bcast_params_ to bcast_list_
GitOrigin-RevId: 26b452a6b7
|
4 years ago |
Megvii Engine Team
|
c7acba41fc
|
refactor(mge/optimizer): refine gradmanager api, record = __enter__
GitOrigin-RevId: 5376177237
|
4 years ago |
Megvii Engine Team
|
8c482b6709
|
fix(mge/grad): make register_after_backward_callback private
GitOrigin-RevId: 8eb6c0e628
|
4 years ago |
Megvii Engine Team
|
66b6daf777
|
test(mge/optimizer): fix test for new optimizer api
GitOrigin-RevId: 482ee62652
|
4 years ago |
Megvii Engine Team
|
e9104ef157
|
fix(mge/parampack): fix copy stream, import cycle
GitOrigin-RevId: 673e11c5b6
|
4 years ago |
Megvii Engine Team
|
e283663a02
|
fix(mge/imperative): update tests to new optimizer api
GitOrigin-RevId: 3d06e3db3c
|
4 years ago |
Megvii Engine Team
|
b5016b9d29
|
feat(mge/parampack): add parampack in allreduce callback
GitOrigin-RevId: 73d53eeba1
|
4 years ago |
Megvii Engine Team
|
6070266766
|
refactor(mge/grad_manager): refactor gradmanager, add allreduce callback
GitOrigin-RevId: 086e2871e8
|
4 years ago |
Megvii Engine Team
|
3f2eac2fe1
|
fix(mge/imperative): move functional/distributed.py to distributed/functional.py
GitOrigin-RevId: 30cf2f514b
|
4 years ago |
Megvii Engine Team
|
e1fba6ece7
|
test(mge/distributed): add get_device_count_by_fork to fix distributed test skip
GitOrigin-RevId: 9ffd8a6149
|
4 years ago |
Megvii Engine Team
|
6b380e8965
|
feat(mge/imperative): run oss test and restore cmake list build items
GitOrigin-RevId: 11411b6964
|
4 years ago |