huggingface/peft

Pull Requests Commits

Update tests/test_gpu_examples.py

younesbelkada committed 1 year ago

Verified b3253faa

Update test_gpu_examples.py

younesbelkada committed 1 year ago

Verified 480d935e

FIX Doc error prompt tuning seq len calc (#1686)

jpodivin committed 1 year ago

Verified e19f7bf4

FEAT Show adapter layer and model status (#1663)

BenjaminBossan committed 1 year ago

Verified 250b7eb8

FIX Issues with AdaLora initialization (#1652)

BenjaminBossan committed 1 year ago

Verified f5f7b67d

FIX bf16 dtype issue for IA3 (#1634)

sywangyi committed 1 year ago

Verified 7a22b7da

FIX Init DoRA weights in float32 if float16 used (#1653)

BenjaminBossan committed 1 year ago

Verified e7b47ac0

Update Dockerfile (#1684)

younesbelkada committed 1 year ago

Verified 8bc3c086

Update build_docker_images.yml (#1682)

younesbelkada committed 1 year ago

Verified 383e1fab

FEAT: Add EETQ support in PEFT (#1675)

younesbelkada committed 1 year ago

Verified d0fa70ae

FIX Don't eagerly import bnb for LoftQ (#1683)

BenjaminBossan committed 1 year ago

Verified b1d6c771

FIX Use trl version of tiny random llama (#1681)

BenjaminBossan committed 1 year ago

Verified f0d3c6b8

FIX / Workflow: Fix Mac-OS CI issues (#1680)

younesbelkada committed 1 year ago

Verified 3d9529d1

ENH: Add multi-backend tests for bnb (#1667)

younesbelkada committed 1 year ago

Verified 83518146

DOC DeepSpeed and QLoRA compatibility (#1679)

sanghyuk-choi committed 1 year ago

Verified 3d6520e2

VeRA (Vector Based Random Matrix Adaption) (#1564)

BenjaminBossan committed 1 year ago

Verified 5a4b9cad

ENH Support safetensor in multitask_prompt_tuning (#1662)

sywangyi committed 1 year ago

Verified 144b7345

MNT Remove dreambooth git submodule (#1660)

charliermarsh committed 1 year ago

Verified bdb85678

FIX Bug with handling of active adapters (#1659)

BenjaminBossan committed 1 year ago

Verified ed865e28

ENH Float fmt in print_trainable_parameters (#1648)

stas00 committed 1 year ago

Verified 56773b9a

DOC Update figure assets of BOFT (#1642)

YuliangXiu committed 1 year ago

Verified c8974c58

FIX Errors in the transformers integration docs (#1629)

BenjaminBossan committed 1 year ago

Verified 76719262

Don't use deprecated `Repository` anymore (#1641)

Wauplin committed 1 year ago

Verified 9f0cfc99

BOFT: Orthogonal Finetuning via Butterfly Factorization (#1326)

yfeng95 committed 1 year ago

Verified 81116993

add deepspeed support for adalora finetune (#1625)

sywangyi committed 1 year ago

Verified b0f1bb46

FEAT Allow load_adapter to use different device (#1631)

yhZhai committed 1 year ago

Verified 31c884e9

FIX Correctly call element_size (#1635)

BenjaminBossan committed 1 year ago

Verified 88875f1c

FIX Multiple adapters and modules_to_save (#1615)

BenjaminBossan committed 1 year ago

Verified 0d283ae0

itemsize is torch>=2.1, use element_size() (#1630)

winglian committed 1 year ago

Verified e07095a6

FEAT: Allow ignoring mismatched sizes when loading (#1620)

BenjaminBossan committed 1 year ago

Verified 5b60ec02

Older