Fix to prefix tuning to fit transformers#2096

Merged

BenjaminBossan merged 31 commits intohuggingface:mainfrom

BenjaminBossan:fix-prefix-tuning-dynamic-cache

Oct 24, 2024

Commits on Sep 25, 2024

[WIP] Fix to prefix tuning to fit transformers
BenjaminBossan
committed

Commits on Oct 22, 2024

Update src/peft/peft_model.py

BenjaminBossan
and
zucchini-nlp
authored
FIX: Change check if past_key_values is empty (#2106)
BenjaminBossan
committed
DOC Update source install instruction (#2110)

Salehbigdeli
authored and
BenjaminBossan
committed
FIX Refactor OFT, small changes to BOFT (#1996)

zqiu24
authored and
BenjaminBossan
committed
ENH: Improved attribute access for modules_to_save (#2117)
BenjaminBossan
committed
FIX low_cpu_mem_usage consolidates devices (#2113)
BenjaminBossan
committed
TST Mark flaky X-LoRA test as xfail (#2114)
BenjaminBossan
committed
ENH: Warn when from_pretrained misses PEFT keys (#2118)
BenjaminBossan
committed
FEAT: Adding exclude modules param(#2044) (#2102)

JINO-ROHIT
authored and
BenjaminBossan
committed
FIX BC breaking change to boft conv2d scaling variable (#2127)

zqiu24
authored and
BenjaminBossan
committed
FEAT: VeRA quantization using bitsandbytes (#2070) (#2076)

ZiadHelal
authored and
BenjaminBossan
committed
Bump version to 0.13.2.dev0 (#2137)
BenjaminBossan
committed
FEAT: Support torchao (#2062)

BenjaminBossan
and
SunMarc
committed
FIX: PiSSA now works with Conv1D layers (#2103) (#2104)

authored and
BenjaminBossan
committed
FIX Type annoations in vera/bnb.py (#2139)
BenjaminBossan
committed
ENH Make PEFT configs forward compatible (#2038)
BenjaminBossan
committed
FIX Raise mixed adapter infer with missing adapter (#2090)
BenjaminBossan
committed
FIX Prompt learning with latest transformers error (#2140)
BenjaminBossan
committed
ENH LoRA notebook for NER task (#2126)

JINO-ROHIT
authored and
BenjaminBossan
committed
FIX TST NaN issue with HQQ GPU test (#2143)
BenjaminBossan
committed
FIX Bug in target module optimization if suffix (#2144)
BenjaminBossan
committed
Bump version to 0.13.2.dev0 (#2145)
BenjaminBossan
committed
FIX Don't assume past_key_valus for encoder models (#2149)
BenjaminBossan
committed
FIX Use `SFTConfig` instead of `SFTTrainer` keyword args (#2150)

qgallouedec
authored and
BenjaminBossan
committed
make style
BenjaminBossan
committed
Merge branch 'main' into fix-prefix-tuning-dynamic-cache
BenjaminBossan
committed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Fix to prefix tuning to fit transformers#2096

Fix to prefix tuning to fit transformers#2096
BenjaminBossan merged 31 commits intohuggingface:mainfrom
BenjaminBossan:fix-prefix-tuning-dynamic-cache

Commits on Sep 25, 2024

[WIP] Fix to prefix tuning to fit transformers

Commits on Oct 22, 2024

Update src/peft/peft_model.py

FIX: Change check if past_key_values is empty (#2106)

DOC Update source install instruction (#2110)

FIX Refactor OFT, small changes to BOFT (#1996)

ENH: Improved attribute access for modules_to_save (#2117)

FIX low_cpu_mem_usage consolidates devices (#2113)

TST Mark flaky X-LoRA test as xfail (#2114)

ENH: Warn when from_pretrained misses PEFT keys (#2118)

FEAT: Adding exclude modules param(#2044) (#2102)

FIX BC breaking change to boft conv2d scaling variable (#2127)

FEAT: VeRA quantization using bitsandbytes (#2070) (#2076)

Bump version to 0.13.2.dev0 (#2137)

FEAT: Support torchao (#2062)

FIX: PiSSA now works with Conv1D layers (#2103) (#2104)

FIX Type annoations in vera/bnb.py (#2139)

ENH Make PEFT configs forward compatible (#2038)

FIX Raise mixed adapter infer with missing adapter (#2090)

FIX Prompt learning with latest transformers error (#2140)

ENH LoRA notebook for NER task (#2126)

FIX TST NaN issue with HQQ GPU test (#2143)

FIX Bug in target module optimization if suffix (#2144)

Bump version to 0.13.2.dev0 (#2145)

FIX Don't assume past_key_valus for encoder models (#2149)

FIX Use `SFTConfig` instead of `SFTTrainer` keyword args (#2150)

make style

Merge branch 'main' into fix-prefix-tuning-dynamic-cache

Commits on Oct 23, 2024

Some further fixes

Fixes for seq2seq model

Mistral test requires protobuf (???)

Also sentencepiece for mistral -_-

Fix to prefix tuning to fit transformers#2096BenjaminBossan merged 31 commits intohuggingface:mainhuggingface/peft:mainfrom BenjaminBossan:fix-prefix-tuning-dynamic-cacheBenjaminBossan/peft:fix-prefix-tuning-dynamic-cacheCopy head branch name to clipboard

Commits

Commits on Sep 25, 2024

Commits on Oct 22, 2024

Commits on Oct 23, 2024

Fix to prefix tuning to fit transformers#2096
BenjaminBossan merged 31 commits intohuggingface:mainfrom
BenjaminBossan:fix-prefix-tuning-dynamic-cache