 Michael Diskin
|
aa6d65a818
Fix "too many open files" issue (#444)
|
%!s(int64=3) %!d(string=hai) anos |
 justheuristic
|
ed4204009f
move PerformanceEMA to utils, TrainingAverager to optim, update utils (#405)
|
%!s(int64=3) %!d(string=hai) anos |
|
justheuristic
|
025e095d55
backport PerformanceEMA from server_side_averaging (#397)
|
%!s(int64=3) %!d(string=hai) anos |
|
justheuristic
|
09985d843b
Implement simplified all-reduce for asymmetric TCP connections (#385)
|
%!s(int64=3) %!d(string=hai) anos |
 Denis Mazur
|
b44236972a
Fix pickle vulnerability (#386)
|
%!s(int64=3) %!d(string=hai) anos |
|
justheuristic
|
1f54faf82f
Split compression strategies into separate classes (#366)
|
%!s(int64=4) %!d(string=hai) anos |
|
justheuristic
|
2500e3dd58
Fix "Too many open files" and load state freezing (#371)
|
%!s(int64=4) %!d(string=hai) anos |
 Alexander Borzunov
|
7f296a43c3
Improve Matchmaking finalizers (#357)
|
%!s(int64=4) %!d(string=hai) anos |
|
Alexander Borzunov
|
f97b742508
Fix minor asyncio issues in averager (#356)
|
%!s(int64=4) %!d(string=hai) anos |
|
Michael Diskin
|
bedfa6eefb
Reorder imports with isort (#326)
|
%!s(int64=4) %!d(string=hai) anos |
|
justheuristic
|
1b01a8164a
Resolve deadlock in MPFuture (#337)
|
%!s(int64=4) %!d(string=hai) anos |
|
justheuristic
|
197666c2b6
Remove shared memory from MPFuture, fix minor bugs (#317)
|
%!s(int64=4) %!d(string=hai) anos |
 Max Ryabinin
|
2f07a556e6
Reformat code with Black (#274)
|
%!s(int64=4) %!d(string=hai) anos |
|
Max Ryabinin
|
5233b6c085
Split hivemind.client into hivemind.averaging and hivemind.moe (#304)
|
%!s(int64=4) %!d(string=hai) anos |
|
justheuristic
|
200fbecdbf
Refactor MPFuture to use a single pipe/thread per process (#298)
|
%!s(int64=4) %!d(string=hai) anos |
|
justheuristic
|
0a0e290ea3
Add per-tensor compression, make all-reduce faster and more flexible (#272)
|
%!s(int64=4) %!d(string=hai) anos |
 Vsevolod-pl
|
8673071bc0
Fixed nan when compressing the tensor of zeros (#266)
|
%!s(int64=4) %!d(string=hai) anos |
 mponty
|
0080028e25
Add uniform compression (#202)
|
%!s(int64=4) %!d(string=hai) anos |
|
Max Ryabinin
|
916c3db52d
Move compression-related code to hivemind.utils.compression (#213)
|
%!s(int64=4) %!d(string=hai) anos |
|
Max Ryabinin
|
6f8f192150
Improve Runtime exception handling (#207)
|
%!s(int64=4) %!d(string=hai) anos |
|
Vsevolod-pl
|
b9c02ac191
Add quantile compression (#182)
|
%!s(int64=4) %!d(string=hai) anos |
 Alexey Bukhtiyarov
|
bb7c4d5953
Fix scalar deserialization (#190)
|
%!s(int64=4) %!d(string=hai) anos |
|
Alexey Bukhtiyarov
|
06162992fa
Serialize DHTID source with msgpack (#172)
|
%!s(int64=4) %!d(string=hai) anos |
|
justheuristic
|
23e655640c
Handle edge cases in DecentralizedAverager (#171)
|
%!s(int64=4) %!d(string=hai) anos |
 Anton Sinitsin
|
0d7818b2cd
fix problem with NamedTuple inheritance in python3.9 (#142)
|
%!s(int64=4) %!d(string=hai) anos |
|
justheuristic
|
e159605143
Address averaging corner cases, add benchmark_averaging.py, chunk averaged tensors, fix DHTNode get (#134)
|
%!s(int64=4) %!d(string=hai) anos |
|
justheuristic
|
1754792aad
Process-wide channel cache for gRPC+aio (#120)
|
%!s(int64=4) %!d(string=hai) anos |
|
justheuristic
|
2bd481c73f
added torch1.7 support, switch to grpc 1.33, grpc bump, improved tests & logging, (#116)
|
%!s(int64=4) %!d(string=hai) anos |
|
Vsevolod-pl
|
fde83bba02
Implemented float16 compression (#106)
|
%!s(int64=4) %!d(string=hai) anos |
|
Vsevolod-pl
|
d4d9da9d3e
Tensor compression (part1) (#102)
|
%!s(int64=4) %!d(string=hai) anos |
