 Artem Chumachenko
|
1b21dd3217
Add memory cache usage
|
1 year ago |
|
Artem Chumachenko
|
01c3cf8d15
Add first version
|
1 year ago |
 Max Ryabinin
|
1ebd88ae7b
Optimize the Falcon block for inference (#500)
|
1 year ago |
 Alexander Borzunov
|
d40eb6c701
Fix prompt tuning after #464 (#501)
|
1 year ago |
|
Alexander Borzunov
|
dd4a3230bc
Add Falcon support (#499)
|
1 year ago |
|
Alexander Borzunov
|
b4d822afb2
Force use_cache=True in config only (#497)
|
1 year ago |
|
Alexander Borzunov
|
abd547735f
Force use_cache=True (#496)
|
1 year ago |
|
Alexander Borzunov
|
6ef6bf5fa2
Create model index in DHT (#491)
|
2 years ago |
|
Alexander Borzunov
|
6bb3f54e39
Replace dots in repo names when building DHT prefixes (#489)
|
2 years ago |
|
Alexander Borzunov
|
02fc71eb25
Fix race condition in MemoryCache (#487)
|
2 years ago |
|
Alexander Borzunov
|
dc0072fde1
Wait for DHT storing state OFFLINE on shutdown (#486)
|
2 years ago |
|
Alexander Borzunov
|
a26559ff65
Fix `.generate(input_ids=...)` (#485)
|
2 years ago |
|
Alexander Borzunov
|
459933f846
Remove no-op process in PrioritizedTaskPool (#484)
|
2 years ago |
|
Alexander Borzunov
|
26ebbfe8f0
Support macOS (#477)
|
2 years ago |
|
Alexander Borzunov
|
75e516a8c1
Refactor readme (#482)
|
2 years ago |
 justheuristic
|
c08d09c4d3
Rewrite MemoryCache alloc_timeout logic (#434)
|
2 years ago |
|
Alexander Borzunov
|
90840dfea2
Fix requiring transformers>=4.32.0 (#480)
|
2 years ago |
|
Alexander Borzunov
|
915b357740
Require transformers>=4.32.0 (#479)
|
2 years ago |
|
Alexander Borzunov
|
18e93afc73
Don't install cpufeature on non-x86_64 machines (#478)
|
2 years ago |
|
Alexander Borzunov
|
6967904590
Bump version to 2.1.0 (#474)
|
2 years ago |
|
Alexander Borzunov
|
df8ab09ca2
Hide excess key message (#476)
|
2 years ago |
|
Artem Chumachenko
|
a14ae7334d
Update peft to 0.5.0 version (#475)
|
2 years ago |
|
Alexander Borzunov
|
a9b0e9ff1a
Support loading weights from Safetensors on server (#473)
|
2 years ago |
|
justheuristic
|
4f850996bb
Change transformers version assert (#472)
|
2 years ago |
|
justheuristic
|
9250025140
Support transformers 4.32.x (#471)
|
2 years ago |
|
justheuristic
|
adda5f8c20
Temporarily require peft<0.5.0, transformers<4.32.0 (#470)
|
2 years ago |
|
Alexander Borzunov
|
de2475f31c
Make client compatible with transformers' GenerationMixin (#464)
|
2 years ago |
|
Alexander Borzunov
|
063e94b4c8
Move SequenceManagerConfig -> ClientConfig, petals.dht_utils -> petals.utils.dht (#463)
|
2 years ago |
|
Artem Chumachenko
|
568f21dc3b
Add customizable input tensors (#445)
|
2 years ago |
|
Alexander Borzunov
|
329f7d31e8
Add `blocked_servers` argument (#462)
|
2 years ago |
