Skip to content

PTQ-Level Efficiency, QAT-Level Accuracy | 训练后量化的效率,量化感知训练的精度

Commits on main

change to new api in readme (#507)
Wenhua Cheng committed 1 year ago
1d63c7b
fix bug (#506)
Heng Guo committed 1 year ago
9f5d01c
use --tasks to control evaluation enabling (#505)
Wenhua Cheng committed 1 year ago
ef33b49
9a6f325
refine inference step 2 (#498)
Wenhua Cheng committed 1 year ago
37341f5
refine inference backend/code step 1 (#486)
Wenhua Cheng committed 1 year ago
ea4d843
709b6d3
[pre-commit.ci] pre-commit autoupdate (#494)
pre-commit-ci[bot] committed 1 year ago
23447b5
fix bug of mix dataset (#492)
Heng Guo committed 1 year ago
943070e
update 2bits acc results (#491)
WeiweiZhang1 committed 1 year ago
226d2bf
fix llama 8b time cost (#490)
WeiweiZhang1 committed 1 year ago
eb79348
fix dataset combine bug (#489)
Wenhua Cheng committed 1 year ago
0d0e1cb
bump into v0.4.7 (#487)
Sun, Xuehao committed 1 year ago
f55b6de
add light ut, fixtypos (#483)
WeiweiZhang1 committed 1 year ago
7b5a2f1
support xpu in tuning and inference (#481)
Wenhua Cheng committed 1 year ago
1603573
af56508
d2cbed9
rm redundant line break (#475)
WeiweiZhang1 committed 1 year ago
050594f
34336d7
support to export gemma3 gguf format (#470)
Heng Guo committed 1 year ago
43629a7
support for mistral3 (#472)
Heng Guo committed 1 year ago
4e83d78
fix incorrect recipe data (#471)
WeiweiZhang1 committed 1 year ago
002ca16
12539ef
upload_auto-round-light results (#454)
WeiweiZhang1 committed 1 year ago
eec34cd
support for gemma3 (#468)
Heng Guo committed 1 year ago
451c029
5ed7e57
W4Afp8 export (#378)
Wenhua Cheng committed 1 year ago
7f6e5d4
fixl eval bug (#461)
Heng Guo committed 1 year ago
5b95a8e