transformers
pass module to Params4bit.from_prequantized to ensure quant_state
#32524
Merged
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Overview
Commits
5
Changes
View On
GitHub
pass module to Params4bit.from_prequantized to ensure quant_state
#32524
ArthurZucker
merged 5 commits into
huggingface:main
from
winglian:bnb-load-quant-state-fix
SunMarc
commented on 2024-08-08
SunMarc
requested a review
from
Titus-von-Koeller
1 year ago
SunMarc
requested a review
from
matthewdouglas
1 year ago
matthewdouglas
requested changes on 2024-08-08
matthewdouglas
added
Quantization
matthewdouglas
added
PyTorch FSDP
matthewdouglas
approved these changes on 2024-08-08
SunMarc
approved these changes on 2024-08-09
SunMarc
requested a review
from
ArthurZucker
1 year ago
pass module to Params4bit.from_prequantized to ensure quant_state
1c5d3c43
make sure to check bnb version
6e545501
revert min bnb version and use inspect on method instead
f5521cb4
use version instead of inspect to prevent performance hit
009ba7ca
make the property name readable
7a8c1f58
winglian
force pushed
to
7a8c1f58
1 year ago
muellerzr
approved these changes on 2024-08-28
ArthurZucker
approved these changes on 2024-08-29
ArthurZucker
merged
39bfb2f5
into main
1 year ago
winglian
deleted the bnb-load-quant-state-fix branch
1 year ago
Login to write a write a comment.
Login via GitHub
Reviewers
ArthurZucker
matthewdouglas
SunMarc
muellerzr
Titus-von-Koeller
Assignees
No one assigned
Labels
Quantization
PyTorch FSDP
Milestone
No milestone
Login to write a write a comment.
Login via GitHub