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

new mllm eval (#317)
Heng Guo committed 2 years ago
ee816b3
fix merge error (#325)
wenhuach21 committed 2 years ago
dadb95a
add warning for no exllamav2 kernel (#324)
wenhuach21 committed 2 years ago
f2ccd6a
fix multiple device bug (#321)
wenhuach21 committed 2 years ago
8918df2
fix eval device issue (#319)
wenhuach21 committed 2 years ago
6aa3d27
update the hpu check (#318)
Yi Liu committed 2 years ago
e1791cd
fix merge error (#316)
Heng Guo committed 2 years ago
606a5c3
add torch compile arg (#314)
wenhuach21 committed 2 years ago
8a9eed0
patch 1 for mllm (#298)
n1ck-guo committed 2 years ago
73861bb
HPU only release binary (#302)
Yi Liu committed 2 years ago
5cd255e
multiple gpu evaluation/calibration refine (#312)
wenhuach21 committed 2 years ago
c5687ba
qwen2 vision quantization bugfix (#313)
WeiweiZhang1 committed 2 years ago
97bbb9e
fix typo (#311)
wenhuach21 committed 2 years ago
b3f0287
Update autogptq exporting (#310)
wenhuach21 committed 2 years ago
bb07dc4
Enable torch.compile on HPU (#307)
Yi Liu committed 2 years ago
d6447fb
fix glm4-9b batch dim issue (#304)
wenhuach21 committed 2 years ago
3376edd
Porting Numba-based packing from INC (#301)
Yi Liu committed 2 years ago
fe19011
mllm eval bug fix (#297)
n1ck-guo committed 2 years ago
0bb70a6
eval for MLLMs (#296)
n1ck-guo committed 2 years ago
4384914
refine forward hook (#290)
WeiweiZhang1 committed 2 years ago
25d977b
fix bug of backend (#294)
wenhuach21 committed 2 years ago
4f22871
fix ipex tqdm mismatch issue (#293)
wenhuach21 committed 2 years ago
487abd6
Add ipex support for intel cpu (#292)
wenhuach21 committed 2 years ago
168a1f6
Refine code (#291)
wenhuach21 committed 2 years ago
f41094a
update readme (#287)
wenhuach21 committed 2 years ago
8efff6f