transformers
[cache] make all classes cache compatible finally
#38635
Merged

[cache] make all classes cache compatible finally #38635

zucchini-nlp
zucchini-nlp dump
5d14a877
zucchini-nlp push other models
5c5825bc
zucchini-nlp fix simple greedy generation
051fe7f8
zucchini-nlp xmod
b04ddbc6
HuggingFaceDocBuilderDev
zucchini-nlp add fmst and clean up some mentions of old cache format
6a289a7c
zucchini-nlp gpt-bigcode now follows standards
b3be72b4
zucchini-nlp delete tuple cache reference in generation
85061bc1
zucchini-nlp fix some models
1424600b
zucchini-nlp fix some models
02fb0d2b
zucchini-nlp fix mambas and support cache in tapas
f7494bcf
zucchini-nlp fix some more tests
576fb7b6
zucchini-nlp fix copies
8757e846
zucchini-nlp delete `_reorder_cache`
bcf0cc7c
zucchini-nlp another fix copies
91d92f1d
zucchini-nlp fix typos and delete unnecessary test
edf5f6e0
zucchini-nlp fix rag generate, needs special cache reordering
b236e90a
zucchini-nlp fix tapas and superglue
1893f8a8
zucchini-nlp reformer create special cache
46e50b5a
zucchini-nlp recurrent gemma `reorder_cache` was a no-op, delete
204ed55d
zucchini-nlp fix-copies
7b61dfda
zucchini-nlp fix blio and musicgen pipeline tests
69c20ae2
zucchini-nlp Merge branch 'main' into cache-class-finalize
d281a6c5
zucchini-nlp fix reformer
b5088140
zucchini-nlp
zucchini-nlp commented on 2025-06-11
zucchini-nlp fix reformer, again...
b7deae60
zucchini-nlp delete `_supports_cache_class`
ae88ecc8
zucchini-nlp delete `supports_quantized_cache`
f1ec0ba9
zucchini-nlp fix failing tests
8f5d8a03
zucchini-nlp fix copies
08ad1b07
zucchini-nlp some minor clean up
dfdf50bd
zucchini-nlp zucchini-nlp requested a review from gante gante 345 days ago
zucchini-nlp zucchini-nlp requested a review from ArthurZucker ArthurZucker 345 days ago
zucchini-nlp
zucchini-nlp style
e9a281f0
gante
gante commented on 2025-06-20
zucchini-nlp merge main, so many conflicts
e735b877
zucchini-nlp style
e1a3fc4e
zucchini-nlp fix copies
3190a9e1
gante
zucchini-nlp fix tests
2f942f86
zucchini-nlp merge main
0fc01593
zucchini-nlp fix copies
ccdd784f
zucchini-nlp create causal mask now needs positions?
63f1bd3f
zucchini-nlp
zucchini-nlp zucchini-nlp requested a review from gante gante 319 days ago
zucchini-nlp
zucchini-nlp fixc copies
2dc3b01a
gante
gante approved these changes on 2025-07-08
zucchini-nlp merge main
f050810f
zucchini-nlp style
e945d2f2
zucchini-nlp
zucchini-nlp Update tests/test_modeling_common.py
8a7a05bd
zucchini-nlp clean-up of non-generative model after merging main
ce665c0d
zucchini-nlp check `is_decoder` for cache
d0f68d0b
zucchini-nlp delete transpose for scores
72bc51a6
zucchini-nlp remove tuple cache from docs everywhere
ac739595
zucchini-nlp fix tests
fd84e679
zucchini-nlp fix copies
7c2d5b67
zucchini-nlp fix copies once more
122564eb
zucchini-nlp properly deprecate `encoder_attention_mask` in Bert-like models
b2652873
zucchini-nlp import `deprecate_kwarg` where needed
8218c5bc
zucchini-nlp fix copies again
bb1866ca
zucchini-nlp
zucchini-nlp zucchini-nlp requested a review from Cyrilvallez Cyrilvallez 316 days ago
zucchini-nlp Merge branch 'main' into cache-class-finalize
b9fe72c3
zucchini-nlp fix copies
b67a4c3f
Cyrilvallez
Cyrilvallez commented on 2025-07-14
zucchini-nlp delete `nex_decoder_cache`
5eeeeb32
zucchini-nlp fix copies asks to update for PLM
5231ed5f
zucchini-nlp merge main
27e95392
zucchini-nlp fix copies
5a745092
zucchini-nlp rebasing had a few new models, fix them and merge asap!
011ee196
zucchini-nlp
zucchini-nlp fix copies once more
91f80729
zucchini-nlp
github-actions
ArthurZucker
ArthurZucker approved these changes on 2025-07-15
zucchini-nlp fix slow tests
ec311c3a
zucchini-nlp Merge branch 'main' into cache-class-finalize
78d44b49
zucchini-nlp fix tests and updare PLM checkpoint
f9592f62
zucchini-nlp add read token and revert accidentally removed line
abccbee8
zucchini-nlp oh com -on, style
327b9e10
zucchini-nlp just skip it, read token has no access to PLM yet
651febb5
github-actions
zucchini-nlp
zucchini-nlp zucchini-nlp merged c8524aeb into main 310 days ago

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone