intel/auto-round

Pull Requests Commits

disable quantizing lm-head with tied weights as a workaround (#102)

wenhuach21 committed 2 years ago

Verified 1fe6aae5

disable quantizing lm-head with tied weights as a workaround (#101)

wenhuach21 committed 2 years ago

Verified c839e825

update readme of calibration dataset and lm-head usage (#98)

wenhuach21 committed 2 years ago

Verified f95b8c7d

fix critic bug for gradient_accumulate_steps!=1 and reduce cpu memory of lm-head tuning (#97)

WeiweiZhang1 committed 2 years ago

Verified 7dd02eb4

handle invalid layername in weight_config (#93)

WeiweiZhang1 committed 2 years ago

Verified 89562261

yintong-lu committed 2 years ago

Verified 9b08de48

deprecate use_quant_inp arg (#90)

yintong-lu committed 2 years ago

Verified 1bf1b486

Add acc data (#89)

pursure-D committed 2 years ago

Verified 8a3da144

fix old eval bug (#86)

WeiweiZhang1 committed 2 years ago

Verified 23d35e32

Update lm-head quantization readme

wenhuach21 committed 2 years ago

Verified 511a5385

add Yi-6b-chat results (#85)

yintong-lu committed 2 years ago

Verified d38a4a56

fix old eval tasks order (#78)

WeiweiZhang1 committed 2 years ago

Verified 257c5be2

Update llama3 acc (#84)

wenhuach21 committed 2 years ago

Verified 849cf9fe

support lm head quantizaiton and export to Intel cpu (#76)

wenhuach21 committed 2 years ago

Verified 16f9b7bd

fix bloom issue

pursure-D committed 2 years ago

Verified bd2fcc9f

update W2g32 accuracy (#74)

pursure-D committed 2 years ago

Verified 16d830b4

Add baichuan-7b chat recipe (#73)

wenhuach21 committed 2 years ago

Verified b68eaf72

fix eval bug of autogptq model (#72)

yintong-lu committed 2 years ago

Verified b22f6aad

fix filter func issue (#71)

wenhuach21 committed 2 years ago

Verified 900154bf

support combination of calibration datasets (#70)

wenhuach21 committed 2 years ago

Verified e8fb5da1

[pre-commit.ci] pre-commit autoupdate (#69)

pre-commit-ci[bot] committed 2 years ago

Verified 9f7e8b81

fix Baichuan2-13B issue (#37)

WeiweiZhang1 committed 2 years ago

Verified fd3298b9

wenhuach21 committed 2 years ago

Verified 7ff627b8

Update README.md

wenhuach21 committed 2 years ago

Verified 62995a97

Update cpu inference code.md

wenhuach21 committed 2 years ago

Verified 3aa4b1a7

qwen1.5 acc update (#67)

yintong-lu committed 2 years ago

Verified 3cd0615e

Fix hpu autocast (#64)

wenhuach21 committed 2 years ago

Verified c7d2fa84

Add model recipe (#63)

wenhuach21 committed 2 years ago

Verified f39638d9

update omniquant accuracy (#62)

pursure-D committed 2 years ago

Verified bf97215b

add Omniquant (#59)

pursure-D committed 2 years ago

Verified d2ecc75e

Older