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

Add setseed in autoround (#201)
WeiweiZhang1 committed 2 years ago
9718d20
0daad3f
fix memory issue (#193)
wenhuach21 committed 2 years ago
e24b907
add check of mixed device (#190)
n1ck-guo committed 2 years ago
61cf9ee
remove force fp16 dtype export (#192)
WeiweiZhang1 committed 2 years ago
d9f7cee
Fix multimodal and moe issue (#191)
WeiweiZhang1 committed 2 years ago
ca59d36
low_cpu_mem refinement (#186)
n1ck-guo committed 2 years ago
7b9611e
support autoround hpu format (#182)
yintong-lu committed 2 years ago
72f5ce8
add initial support of mxfp4 (#187)
wenhuach21 committed 2 years ago
d48d040
[pre-commit.ci] pre-commit autoupdate (#167)
pre-commit-ci[bot] committed 2 years ago
9226a89
fix typos (#185)
wenhuach21 committed 2 years ago
24b2e74
Add layer wise mode to save memory (#136)
n1ck-guo committed 2 years ago
e2e33f0
aeb9e40
Fix UT coverage report (#180)
Sun, Xuehao committed 2 years ago
8d08400
Fix autoround format accuracy issue (#179)
wenhuach21 committed 2 years ago
0126180
Add unit test (#173)
Sun, Xuehao committed 2 years ago
8162409
5f67048
speedup the tuning a little (#175)
wenhuach21 committed 2 years ago
473f474
add chat template in calib tokenization (#171)
yintong-lu committed 2 years ago
735dfc9
support marlin in auto_round format (#172)
wenhuach21 committed 2 years ago
2b1448d
5947e9c
fix typos, update overview img (#166)
WeiweiZhang1 committed 2 years ago
8d5765a
bugfix (#160)
WeiweiZhang1 committed 2 years ago
31c566c
fix bug and limit numpy version (#159)
yintong-lu committed 2 years ago
77320b0
support calibration dataset concat (#147)
yintong-lu committed 2 years ago
75e3fde
remove gpt ppl eval from lm-0.4.2 (#158)
wenhuach21 committed 2 years ago
77d6a88
edcec56