SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity
Files
4c7e8d09008ea4e46dd09dccfbd518bb2b792e75
HuggingFace_transformer/tests/models
History
Younes Belkada 163ac3d3ee Add Switch transformers (#19323) 
* first commit

* add more comments

* add router v1

* clean up

- remove `tf` modeling files

* clean up

- remove `tf` modeling files

* clean up

* v0 routers

* added more router

- Implemented `ExpertsChooseMaskedRouter`

- added tests
- 2 more routers to implement

* last router

* improved docstring

- completed the docstring in `router.py`
- added more args in the config

* v0 sparse mlp

* replace wrong naming

* forward pass run

* update MOE layer

* small router update

* fixup

* consistency

* remove scatter router

* remove abstract layer

* update test and model for integration testing

* v1 conversion

* update

* hardcode hack

* all keys match

* add gin conversion, without additional libraries

* update conversion sctipy

* delete router file

* update tests wrt router deletion

* fix router issues

* update expert code

* update, logits match, code needsREFACTORING

* Refactor code

Co-authored-by: Younes Belkada <younesbelkada@users.noreply.github.com>

* add generate tests

Co-authored-by: younesbelkada <younesbelkada@gmail.com>

* add support for router loss

Co-authored-by: Younes Belkada <younesbelkada@users.noreply.github.com>

* fix forward error

* refactor a bit

* remove `FlaxSwitchTransformers` modules

* more tests pass

* Update code

Co-authored-by: Younes Belkada <younesbelkada@users.noreply.github.com>

* fixup

* fix tests

* fix doc

* fix doc + tokenization

* fix tokenizer test

* fix test

* fix loss output

* update code for backward pass

* add loss support

* update documentation

* fix documentation, clean tokenizer

* more doc fix, cleanup example_switch

* fix failing test

* fix test

* fix test

* fix loss issue

* move layer

* update doc and fix router capacity usage

* fixup

* add sparse mlp index for documentation on hub

* fixup

* test sparse mix architecture

* Apply suggestions from code review

* Update docs/source/en/model_doc/switch_transformers.mdx

* fixup on update

* fix tests

* fix another test

* attempt fix

* Update src/transformers/models/switch_transformers/configuration_switch_transformers.py

Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* Update src/transformers/models/switch_transformers/convert_switch_transformers_original_flax_checkpoint_to_pytorch.py

Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* try

* all tests pass

* fix jitter noise

* Apply suggestions from code review

* doc tests pass

* Update src/transformers/models/switch_transformers/modeling_switch_transformers.py

Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* Update src/transformers/models/switch_transformers/modeling_switch_transformers.py

Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* remove assert

* change config order

* fix readme japanese

* Apply suggestions from code review

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* remove parallelizable tests + add one liners

* remove ONNX config

* fix nits

- add `T5Tokenizer` in auto mapping
- remove `Switch Transformers` from ONNX supported models

* remove `_get_router`

* remove asserts

* add check in test for `router_dtype`

* add `SwitchTransformersConfig` in `run_pipeline_test`

* Update tests/pipelines/test_pipelines_summarization.py

* add huge model conversion script

* fix slow tests

- add better casting for `Linear8bitLt`
- remove `torchscript` tests

* add make dir

* style on new script

* fix nits

- doctest
- remove `_keys_to_ignore_on_load_unexpected`

* Update src/transformers/models/switch_transformers/configuration_switch_transformers.py

* add google as authors

* fix year

* remove last `assert` statements

* standardize vertical spaces

* fix failing import

* fix another failing test

* Remove strange àuthorized_keys`

* removing todo and padding that is never used

Co-authored-by: Arthur Zucker <arthur.zucker@gmail.com>
Co-authored-by: ybelkada <younes@huggingface.co>
Co-authored-by: Younes Belkada <younesbelkada@users.noreply.github.com>
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
Co-authored-by: Arthur Zucker <arthur@huggingface.co>
2022-11-15 13:06:45 +01:00
..
albert
Improve model tester (#19984)
2022-11-02 17:38:44 +01:00
auto
Fix tapas scatter (#20149)
2022-11-14 01:04:26 -05:00
bart
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
barthez
Move test model folders (#17034)
2022-05-03 14:42:02 +02:00
bartpho
Move test model folders (#17034)
2022-05-03 14:42:02 +02:00
beit
Add Image Processors (#19796)
2022-11-02 11:57:36 +00:00
bert
update relative positional embedding (#20203)
2022-11-15 10:46:34 +01:00
bert_generation
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
bert_japanese
Add sentencepiece to BertJapaneseTokenizer (#19769)
2022-10-21 10:04:49 -04:00
bertweet
Move test model folders (#17034)
2022-05-03 14:42:02 +02:00
big_bird
wrap forward passes with torch.no_grad() (#19273)
2022-10-04 16:13:22 +02:00
bigbird_pegasus
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
blenderbot
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
blenderbot_small
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
bloom
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
bort
Move test model folders (#17034)
2022-05-03 14:42:02 +02:00
byt5
Replace as_target context managers by direct calls (#18325)
2022-07-29 08:09:09 -04:00
camembert
Move test model folders (#17034)
2022-05-03 14:42:02 +02:00
canine
Improve model tester (#19984)
2022-11-02 17:38:44 +01:00
clip
[processor] Add 'model input names' property (#20117)
2022-11-10 19:29:20 +00:00
clipseg
Update CLIPSegModelTester (#20134)
2022-11-09 15:21:52 +01:00
codegen
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
conditional_detr
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
convbert
wrap forward passes with torch.no_grad() (#19274)
2022-10-04 16:12:03 +02:00
convnext
Add Image Processors (#19796)
2022-11-02 11:57:36 +00:00
cpm
Move test model folders (#17034)
2022-05-03 14:42:02 +02:00
ctrl
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
cvt
[CvT] Tensorflow implementation (#18597)
2022-10-11 18:16:52 +01:00
data2vec
update relative positional embedding (#20203)
2022-11-15 10:46:34 +01:00
deberta
fix train_new_from_iterator in the case of byte-level tokenizers (#17549)
2022-06-08 15:30:41 +02:00
deberta_v2
🚨 🚨 🚨 Fix Issue 15003: SentencePiece Tokenizers Not Adding Special Tokens in convert_tokens_to_string (#15775)
2022-11-02 15:45:38 -04:00
decision_transformer
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
deformable_detr
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
deit
Add accelerate support for ViT family (#20174)
2022-11-15 11:06:01 +01:00
detr
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
distilbert
Move test model folders (#17034)
2022-05-03 14:42:02 +02:00
dit
Move test model folders (#17034)
2022-05-03 14:42:02 +02:00
donut
remvoe _create_and_check_torch_fx_tracing in specific test files (#18667)
2022-09-07 16:22:09 +02:00
dpr
Replace as_target context managers by direct calls (#18325)
2022-07-29 08:09:09 -04:00
dpt
Add Image Processors (#19796)
2022-11-02 11:57:36 +00:00
electra
Improve model tester (#19984)
2022-11-02 17:38:44 +01:00
encoder_decoder
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
ernie
update relative positional embedding (#20203)
2022-11-15 10:46:34 +01:00
esm
Fix ESM LM head test (#20045)
2022-11-04 12:45:34 +00:00
flaubert
Improve model tester (#19984)
2022-11-02 17:38:44 +01:00
flava
Fix type - update any PIL.Image.Resampling (#20172)
2022-11-11 16:55:59 +00:00
fnet
wrap forward passes with torch.no_grad() (#19413)
2022-10-10 15:03:46 -04:00
fsmt
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
funnel
Update serving code to enable saved_model=True (#18153)
2022-07-22 18:05:38 +01:00
glpn
Move test model folders (#17034)
2022-05-03 14:42:02 +02:00
gpt2
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
gpt_neo
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
gpt_neox
skip some gpt_neox tests that require 80G RAM (#17923)
2022-07-01 09:04:38 -04:00
gpt_neox_japanese
Add support for Japanese GPT-NeoX-based model by ABEJA, Inc. (#18814)
2022-09-14 10:17:40 -04:00
gptj
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
groupvit
Allow passing arguments to model testers for CLIP-like models (#20044)
2022-11-04 18:01:41 +01:00
herbert
Move test model folders (#17034)
2022-05-03 14:42:02 +02:00
hubert
Fix train_step, test_step and tests for CLIP (#18684)
2022-09-09 20:01:02 +01:00
ibert
Improve model tester (#19984)
2022-11-02 17:38:44 +01:00
imagegpt
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
jukebox
Add Jukebox model (replaces #16875) (#17826)
2022-11-10 21:05:27 +01:00
layoutlm
TF: TF 2.10 unpin + related onnx test skips (#18995)
2022-09-12 19:30:27 +01:00
layoutlmv2
[processor] Add 'model input names' property (#20117)
2022-11-10 19:29:20 +00:00
layoutlmv3
[processor] Add 'model input names' property (#20117)
2022-11-10 19:29:20 +00:00
layoutxlm
Fix tapas scatter (#20149)
2022-11-14 01:04:26 -05:00
led
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
levit
Add Image Processors (#19796)
2022-11-02 11:57:36 +00:00
lilt
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
longformer
Improve model tester (#19984)
2022-11-02 17:38:44 +01:00
longt5
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
luke
Adding fine-tuning models to LUKE (#18353)
2022-08-01 11:09:47 -04:00
lxmert
Update serving code to enable saved_model=True (#18153)
2022-07-22 18:05:38 +01:00
m2m_100
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
marian
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
markuplm
[processor] Add 'model input names' property (#20117)
2022-11-10 19:29:20 +00:00
maskformer
Fix image segmentation pipeline errors, resolve backward compatibility issues (#19768)
2022-10-21 18:09:58 +03:00
mbart
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
mbart50
Replace as_target context managers by direct calls (#18325)
2022-07-29 08:09:09 -04:00
mctct
[processor] Add 'model input names' property (#20117)
2022-11-10 19:29:20 +00:00
megatron_bert
Move test model folders (#17034)
2022-05-03 14:42:02 +02:00
megatron_gpt2
Move test model folders (#17034)
2022-05-03 14:42:02 +02:00
mluke
Black preview (#17217)
2022-05-12 16:25:55 -04:00
mobilebert
Fix train_step, test_step and tests for CLIP (#18684)
2022-09-09 20:01:02 +01:00
mobilenet_v2
add MobileNetV2 model (#17845)
2022-11-14 01:00:10 -05:00
mobilevit
Add Image Processors (#19796)
2022-11-02 11:57:36 +00:00
mpnet
Move test model folders (#17034)
2022-05-03 14:42:02 +02:00
mt5
Fix expected loss values in some (m)T5 tests (#18177)
2022-07-18 15:26:21 +02:00
mvp
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
nezha
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
nllb
Replace as_target context managers by direct calls (#18325)
2022-07-29 08:09:09 -04:00
nystromformer
Move test model folders (#17034)
2022-05-03 14:42:02 +02:00
openai
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
opt
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
owlvit
[OWL-ViT] Make model consistent with CLIP (#20144)
2022-11-11 11:36:17 +01:00
pegasus
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
pegasus_x
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
perceiver
Update defaults and logic to match old FE (#20065)
2022-11-04 19:14:56 +00:00
phobert
Move test model folders (#17034)
2022-05-03 14:42:02 +02:00
plbart
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
poolformer
Add Image Processors (#19796)
2022-11-02 11:57:36 +00:00
prophetnet
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
qdqbert
Move test model folders (#17034)
2022-05-03 14:42:02 +02:00
rag
Avoid GPU OOM for a TF Rag test (#17638)
2022-06-10 18:50:29 +02:00
realm
Black preview (#17217)
2022-05-12 16:25:55 -04:00
reformer
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
regnet
Run tests if skip condition not met (#18764)
2022-08-30 14:03:28 +02:00
rembert
Move test model folders (#17034)
2022-05-03 14:42:02 +02:00
resnet
FX support for ConvNext, Wav2Vec2 and ResNet (#19053)
2022-09-16 10:57:41 +02:00
retribert
fix retribert's test_torch_encode_plus_sent_to_model (#17231)
2022-05-17 14:33:13 +02:00
roberta
update relative positional embedding (#20203)
2022-11-15 10:46:34 +01:00
roc_bert
update relative positional embedding (#20203)
2022-11-15 10:46:34 +01:00
roformer
wrap forward passes with torch.no_grad() (#19438)
2022-10-10 14:54:54 -04:00
segformer
Add Image Processors (#19796)
2022-11-02 11:57:36 +00:00
sew
Move test model folders (#17034)
2022-05-03 14:42:02 +02:00
sew_d
Move test model folders (#17034)
2022-05-03 14:42:02 +02:00
speech_encoder_decoder
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
speech_to_text
[processor] Add 'model input names' property (#20117)
2022-11-10 19:29:20 +00:00
speech_to_text_2
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
splinter
Fix Splinter test (#17854)
2022-06-24 16:26:14 +02:00
squeezebert
Move test model folders (#17034)
2022-05-03 14:42:02 +02:00
swin
remvoe _create_and_check_torch_fx_tracing in specific test files (#18667)
2022-09-07 16:22:09 +02:00
swinv2
Add swin transformer v2 (#17469)
2022-07-27 11:14:47 -04:00
switch_transformers
Add Switch transformers (#19323)
2022-11-15 13:06:45 +01:00
t5
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
table_transformer
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
tapas
Fix tapas scatter (#20149)
2022-11-14 01:04:26 -05:00
tapex
Replace as_target context managers by direct calls (#18325)
2022-07-29 08:09:09 -04:00
time_series_transformer
Add a decorator for flaky tests (#19498)
2022-10-12 14:00:17 -04:00
trajectory_transformer
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
transfo_xl
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
trocr
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
unispeech
Move test model folders (#17034)
2022-05-03 14:42:02 +02:00
unispeech_sat
Move test model folders (#17034)
2022-05-03 14:42:02 +02:00
van
has_attentions - consistent test skipping logic and tf tests (#17495)
2022-06-09 09:50:03 +02:00
videomae
Add Image Processors (#19796)
2022-11-02 11:57:36 +00:00
vilt
Add Image Processors (#19796)
2022-11-02 11:57:36 +00:00
vision_encoder_decoder
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
vision_text_dual_encoder
[processor] Add 'model input names' property (#20117)
2022-11-10 19:29:20 +00:00
visual_bert
wrap forward passes with torch.no_grad() (#19439)
2022-10-10 14:54:36 -04:00
vit
Add accelerate support for ViT family (#20174)
2022-11-15 11:06:01 +01:00
vit_mae
TF: tests for (de)serializable models with resized tokens (#19013)
2022-09-16 16:38:08 +01:00
vit_msn
Some fixes regarding auto mappings and test class names (#19923)
2022-10-27 14:38:59 +02:00
wav2vec2
[processor] Add 'model input names' property (#20117)
2022-11-10 19:29:20 +00:00
wav2vec2_conformer
[Test] Fix W2V-Conformer integration test (#17303)
2022-05-17 18:20:36 +02:00
wav2vec2_phoneme
Move test model folders (#17034)
2022-05-03 14:42:02 +02:00
wav2vec2_with_lm
[processor] Add 'model input names' property (#20117)
2022-11-10 19:29:20 +00:00
wavlm
Move test model folders (#17034)
2022-05-03 14:42:02 +02:00
whisper
[WHISPER] Update modeling tests (#20162)
2022-11-15 11:04:58 +01:00
x_clip
Allow passing arguments to model testers for CLIP-like models (#20044)
2022-11-04 18:01:41 +01:00
xglm
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
xlm
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
xlm_prophetnet
Black preview (#17217)
2022-05-12 16:25:55 -04:00
xlm_roberta
Black preview (#17217)
2022-05-12 16:25:55 -04:00
xlm_roberta_xl
update relative positional embedding (#20203)
2022-11-15 10:46:34 +01:00
xlnet
Generate: move generation_*.py src files into generation/*.py (#20096)
2022-11-09 15:34:08 +00:00
yolos
[fix] Add DeformableDetrFeatureExtractor (#19140)
2022-09-22 09:45:24 +02:00
yoso
fix train_new_from_iterator in the case of byte-level tokenizers (#17549)
2022-06-08 15:30:41 +02:00
__init__.py
Move test model folders (#17034)
2022-05-03 14:42:02 +02:00