PR #29850 Better llava next.

Better llava next. #29850

zucchini-nlp merged 51 commits into huggingface:main from nxphi47:better_llava_next

Better llava next.

febbfd5b

fix bugs in batched generation

d7f98b8a

add tests

b98581b0

fix batch-gen bugs, left-padding positions and incorrect attention mask

da770a96

remove better modeling llava

9b84e7dc

NielsRogge commented on 2024-03-27

NielsRogge commented on 2024-03-30

fix formatting

ce7fadc2

fix test

3b01b22d

fix test

3da4a26c

fix testing

e3543a27

fix test

8c808274

fix formatting

5ee0f0f4

NielsRogge requested a review from

ArthurZucker 2 years ago

sanjari-orb commented on 2024-04-02

amyeroberts commented on 2024-04-02

Update src/transformers/models/llava_next/modeling_llava_next.py

df4ae3fa

Update modeling_llava_next.py

86441796

ArthurZucker removed review request from

ArthurZucker 2 years ago

sanjari-orb commented on 2024-04-07

fix bug modeling_llava_next.py

e000fa3f

update modeling

bc69a599

add pad fn Merge branch 'better_llava_next' of https://github.com/nxp…

30aafbb5

Merge branch 'main' into better_llava_next

bd5e8f6a

fix bugs

edae6e79

Merge branch 'better_llava_next' of https://github.com/nxphi47/transf…

6f433d29

fix format

f6ab4154

fix error

685892cb

NielsRogge requested a review from

zucchini-nlp 2 years ago

zucchini-nlp commented on 2024-04-17

fix new_token_positions

1aa6d825

Update modeling_llava_next.py

66b729c9

update formatting

0706e6b9

add args

4a2a3421

NielsRogge commented on 2024-04-25

removecomments

1e07818f

add slow tests for batched inference

bff40e00

failing tf/flax tests

8f1d029f

this one ic correct

a0102a42

insujang commented on 2024-04-26

Update src/transformers/models/llava_next/modeling_llava_next.py

16ab5ab1

fix docs

4d328fd2

make fixup

c9a8c13a

Merge branch 'main' into better_llava_next

2c01a775

more fixup

2de8e950

zucchini-nlp requested a review from

amyeroberts 2 years ago

amyeroberts commented on 2024-04-26

add test for batch equivalence

c91394fc

Update tests/models/llava_next/test_modeling_llava_next.py

4c2702f0

Update src/transformers/models/llava_next/image_processing_llava_next.py

70ccd03a

Update src/transformers/models/llava_next/image_processing_llava_next.py

473e70b8

Update src/transformers/models/llava_next/modeling_llava_next.py

257f3a96

Update src/transformers/models/llava_next/modeling_llava_next.py

9845a15b

Update src/transformers/models/llava_next/modeling_llava_next.py

daf49c39

Update src/transformers/models/llava_next/modeling_llava_next.py

67a3e6e4

Update src/transformers/models/llava_next/modeling_llava_next.py

fa0e6471

Update src/transformers/models/llava_next/modeling_llava_next.py

96eada0b

pr comments

bc65338b

hardcode padding side for bs=1

e73b5559

Merge branch 'main' into better_llava_next

65bd47b7

amyeroberts commented on 2024-05-09

update

38608ead

zucchini-nlp commented on 2024-05-10

amyeroberts approved these changes on 2024-05-14

[run-slow] llava_next

74bce491

amyeroberts commented on 2024-05-14

[run-slow] llava_next

d11c0349

amyeroberts approved these changes on 2024-05-14

make fix-copies

ef08ca4c

zucchini-nlp merged 5ca085b8 into main 2 years ago

Reviewers

amyeroberts

zucchini-nlp

NielsRogge

insujang

sanjari-orb

Assignees

No one assigned

Labels

None yet

Milestone

No milestone

transformers Better llava next. #29850 Merged

Better llava next. #29850

transformers
Better llava next.
#29850

Merged