 justheuristic
|
c440e31232
use manual python calls instead of pytest
|
%!s(int64=4) %!d(string=hai) anos |
 Aleksandr Borzunov
|
cb22719ac4
Unskip test_allreduce_grid()
|
%!s(int64=4) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
f917b9d578
Skip test_allreduce_grid()
|
%!s(int64=4) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
2bb98a94c5
Blackify
|
%!s(int64=4) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
356cff0acc
Fix test_allreduce_grid(), skip test_overcrowded()
|
%!s(int64=4) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
6c20858d2e
Return timeout=5
|
%!s(int64=4) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
df26bfca33
Try removing timeout
|
%!s(int64=4) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
95574522ef
Make diff smaller
|
%!s(int64=4) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
2e51140cc0
Remove `endpoint` parameter of GroupKeyManager
|
%!s(int64=4) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
2ae476fe96
Speed up DHT swarm creation
|
%!s(int64=4) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
36282f81cc
Continue fix test_averaging.py
|
%!s(int64=4) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
7eb91a3b25
Move launch_dht_instances() to test_utils.py
|
%!s(int64=4) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
4f5acb5eee
Fix test_averaging.py
|
%!s(int64=4) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
0384737569
Fix test_allreduce_once
|
%!s(int64=4) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
a8fcb0a609
Convert AllReduceRunner, Matchmaking, and GroupKeyManager to libp2p backend
|
%!s(int64=4) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
ea3b56d479
Start converting averager to libp2p backend
|
%!s(int64=4) %!d(string=hai) anos |
|
justheuristic
|
11db5fd56f
Refactor for v0.9.10 and fix example (#319)
|
%!s(int64=4) %!d(string=hai) anos |
 Max Ryabinin
|
2f07a556e6
Reformat code with Black (#274)
|
%!s(int64=4) %!d(string=hai) anos |
|
Alexander Borzunov
|
0be1512c74
Convert DHT to libp2p backend (#296)
|
%!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
|
e9956b84f6
Add graceful shutdown to DHT and Averager (#301)
|
%!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 |
 foksly
|
e58f65db33
Support auxiliary participants in AllReduceProtocol (#260)
|
%!s(int64=4) %!d(string=hai) anos |
 Roman Zhytar
|
8c3bd93e87
Statistics averaging (#229)
|
%!s(int64=4) %!d(string=hai) anos |
|
justheuristic
|
053c7c7d13
Disentangle DecentralizedAverager components, add weights (#217)
|
%!s(int64=4) %!d(string=hai) anos |
|
foksly
|
edf9327e45
Support client-only participants in AllReduceProtocol (#176)
|
%!s(int64=4) %!d(string=hai) anos |
|
justheuristic
|
25fdf0d94f
Move Averager metadata serialization out of user scope (#168)
|
%!s(int64=4) %!d(string=hai) anos |
|
justheuristic
|
690c9dc32b
Fix the remaining tests for py37 (#166)
|
%!s(int64=4) %!d(string=hai) anos |
|
justheuristic
|
ca4b8c5df6
Remove pickle.loads in averager (#160)
|
%!s(int64=4) %!d(string=hai) anos |
