transformers
🚨🚨🚨 [`Quantization`] Store the original dtype in the config as a private attribute 🚨🚨🚨
#26761

Merged

🚨🚨🚨 [`Quantization`] Store the original dtype in the config as a private attribute 🚨🚨🚨 #26761

younesbelkada merged 8 commits into huggingface:main from younesbelkada:add-orig-dtype

First step

2797129d

fix

b7e797f3

younesbelkada requested a review from

LysandreJik 2 years ago

younesbelkada commented on 2023-10-12

ArthurZucker commented on 2023-10-13

add adjustements for gptq

73d31090

change to `_pre_quantization_dtype`

28d2b27a

Update src/transformers/modeling_utils.py

316f7766

fix serialization

948de7ed

younesbelkada requested a review from

ArthurZucker 2 years ago

ArthurZucker approved these changes on 2023-10-13

Apply suggestions from code review

06700c73

fixup

1ebe674d

younesbelkada changed the title ~~[`Quantization`] Store the original dtype in the config as a private attribute~~ 🚨🚨🚨 [`Quantization`] Store the original dtype in the config as a private attribute 🚨🚨🚨 2 years ago

younesbelkada merged fd6a0ade into main 2 years ago

younesbelkada deleted the add-orig-dtype branch 2 years ago

Reviewers

ArthurZucker

LysandreJik

Assignees

No one assigned

Labels

None yet

Milestone

No milestone

transformers 🚨🚨🚨 [`Quantization`] Store the original dtype in the config as a private attribute 🚨🚨🚨 #26761 Merged

🚨🚨🚨 [`Quantization`] Store the original dtype in the config as a private attribute 🚨🚨🚨 #26761

transformers
🚨🚨🚨 [`Quantization`] Store the original dtype in the config as a private attribute 🚨🚨🚨
#26761

Merged