Skip to content
Closed
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
335 commits
Select commit Hold shift + click to select a range
205bc41
Fix GPT-NeoX-20B past handling, attention computation (#17811)
zphang Jun 30, 2022
4f8361a
Unifying training argument type annotations (#17934)
jannisborn Jun 30, 2022
e4d2588
[Pipelines] Add revision tag to all default pipelines (#17667)
patrickvonplaten Jun 30, 2022
f25457b
Fix number of examples for iterable dataset in distributed training (…
sgugger Jun 30, 2022
91e1f24
CLI: convert sharded PT models (#17959)
gante Jun 30, 2022
fe14046
skip some ipex tests until it works with torch 1.12 (#17964)
ydshieh Jun 30, 2022
9cb7cef
Add ONNX support for LayoutLMv3 (#17953)
regisss Jun 30, 2022
49cd736
feat: add pipeline registry abstraction (#17905)
aarnphm Jun 30, 2022
14fb8a6
skip some gpt_neox tests that require 80G RAM (#17923)
ydshieh Jul 1, 2022
cb42502
Fix typo in perf_train_gpu_one.mdx (#17983)
aliencaocao Jul 1, 2022
569b679
Update expected values in CodeGen tests (#17888)
ydshieh Jul 1, 2022
3a064bd
fix `bias` keyword argument in TFDebertaEmbeddings (#17940)
WissamAntoun Jul 1, 2022
462b7f3
fixing fsdp autowrap functionality (#17922)
pacman100 Jul 1, 2022
b68d408
add ONNX support for BLOOM (#17961)
NouamaneTazi Jul 1, 2022
8bb2c38
Fix FlaxBigBirdEmbeddings (#17842)
ydshieh Jul 1, 2022
664688b
higher atol to avoid flaky trainer test failure (#17979)
ydshieh Jul 1, 2022
485bbe7
[Flax] Add remat (gradient checkpointing) (#17843)
sanchit-gandhi Jul 1, 2022
d6cec45
XLA train step fixes (#17973)
Rocketknight1 Jul 1, 2022
009171d
Ensure PT model is in evaluation mode and lightweight forward pass do…
amyeroberts Jul 1, 2022
6f0723a
Restore original task in test_warning_logs (#17985)
ydshieh Jul 1, 2022
6890d19
Shifting labels for causal LM when using label smoother (#17987)
seungeunrho Jul 1, 2022
49c8c67
Exclude Databricks from notebook env only if the runtime is below 11.…
davidheryanto Jul 1, 2022
a045cbd
only a stupid typo, but it can lead to confusion (#17930)
Dobatymo Jul 4, 2022
7b18702
Add link to existing documentation (#17931)
LysandreJik Jul 4, 2022
77ea513
Add TF ResNet model (#17427)
amyeroberts Jul 4, 2022
cf2578a
Refactor to inherit from nn.Module instead of nn.ModuleList (#17501)
amyeroberts Jul 4, 2022
3cfdefa
Fix typo in error message in generation_utils (#18000)
regisss Jul 4, 2022
7498db0
Replace BloomTokenizer by BloomTokenizerFast in doc (#18005)
regisss Jul 4, 2022
6cb1954
sort list of models (#18011)
hollance Jul 4, 2022
96d833b
Return scalar losses instead of per-sample means (#18013)
Rocketknight1 Jul 4, 2022
e3139ad
fixed calculation of ctc loss in TFWav2Vec2ForCTC (#18014)
Sreyan88 Jul 4, 2022
f098268
TF: T5 can now handle a padded past (i.e. XLA generation) (#17969)
gante Jul 4, 2022
97db5b4
Update expected values in DecisionTransformerModelIntegrationTest (#1…
ydshieh Jul 5, 2022
ec07ecc
[Flax] Bump to v0.4.1 (#17966)
sanchit-gandhi Jul 5, 2022
5ae087c
Fix T5/mT5 tests (#18029)
Rocketknight1 Jul 5, 2022
f681437
Enable Past CI (#17919)
ydshieh Jul 5, 2022
22edb68
Squash commits (#17981)
NielsRogge Jul 6, 2022
bf37e5c
Fix T5 incorrect weight decay in Trainer and official summarization e…
ADAning Jul 6, 2022
360719a
TF: GPT-J compatible with XLA generation (#17986)
gante Jul 6, 2022
be79cd7
Protect `TFGenerationMixin.seed_generator` so it's not created at imp…
Rocketknight1 Jul 6, 2022
2e90c3d
Doc to dataset (#18037)
sgugger Jul 6, 2022
870ff9e
Skip failing test until @gante fix it.
sgugger Jul 6, 2022
1b5ea74
Place inputs on device when include_inputs_for_metrics is True (#18046)
sgugger Jul 7, 2022
1b749a7
Sort doc toc (#18034)
sgugger Jul 7, 2022
91c4a3a
Added Command for windows VENV activation in installation docs (#18008)
darthvader2 Jul 7, 2022
2544c14
[Generate Tests] Make sure no tokens are force-generated (#18053)
patrickvonplaten Jul 7, 2022
de46cde
Drop columns after loading samples in prepare_tf_dataset (#17967)
Rocketknight1 Jul 7, 2022
9bd3968
Fix slow CI by pinning resampy (#18077)
sgugger Jul 8, 2022
94ca7d2
Fix type issue in using bucketing with Trainer (#18051)
seopbo Jul 8, 2022
7c046c5
Update localized READMES when template is filled. (#18062)
sgugger Jul 8, 2022
8b332a6
Make predict() close progress bars after finishing (#17952) (#18078)
neverix Jul 8, 2022
ad28ca2
[bloom] fix alibi device placement (#18087)
stas00 Jul 10, 2022
95113d1
Fix some typos. (#17560)
Yulv-git Jul 11, 2022
ac98a88
Fix torchscript tests for GPT-NeoX (#18012)
ydshieh Jul 11, 2022
1e8140c
Fix RESOURCE_EXHAUSTED error when dealing with large datasets in Flax…
duongna21 Jul 11, 2022
b0520f5
Skip failing tests
sgugger Jul 11, 2022
6c8017a
Fix image segmentation and object detection pipeline tests (#18100)
sgugger Jul 11, 2022
b1b8222
Add filename to info diaplyed when downloading things in from_pretrai…
sgugger Jul 11, 2022
5ff6f85
Mark slow test as such
sgugger Jul 11, 2022
a462fc9
Bloom Optimize operations (#17866)
younesbelkada Jul 11, 2022
b7d8bd3
Enhance IPEX integration in Trainer (#18072)
jianan-gu Jul 12, 2022
d4ebd4e
speed up test (#18106)
sijunhe Jul 12, 2022
f5221c0
Report value for a step instead of epoch. (#18095)
zhawe01 Jul 12, 2022
981714e
CLI: reenable `pt_to_tf` test (#18108)
gante Jul 12, 2022
bc34c21
Fix BLOOM dtype (#17995)
Muennighoff Jul 12, 2022
fcefa20
TF: remove graph mode distinction when processing boolean options (#1…
gante Jul 12, 2022
20509ab
TF: unpack_inputs decorator independent from main_input_name (#18110)
gante Jul 13, 2022
1951333
supported python versions reference (#18116)
CakeCrusher Jul 13, 2022
fde22c7
Add summarization name mapping for MultiNews (#18117)
JohnGiorgi Jul 13, 2022
56e6487
add dataset split and config to model-index in TrainingSummary.from_t…
loicmagne Jul 13, 2022
3564c65
Make sharded checkpoints work in offline mode
sgugger Jul 13, 2022
0a21a48
Revert "Make sharded checkpoints work in offline mode"
sgugger Jul 13, 2022
37aeb57
Make sharded checkpoints work in offline mode (#18125)
sgugger Jul 13, 2022
7ea6ccc
Enable torchdynamo with torch_tensorrt(fx path) (#17765)
Jul 13, 2022
8581a79
Add TF DeiT implementation (#17806)
amyeroberts Jul 13, 2022
fca66ec
Fixing a hard to trigger bug for `text-generation` pipeline. (#18131)
Narsil Jul 15, 2022
ccc0897
Adding support for `device_map` directly in `pipeline(..)` function. …
Narsil Jul 15, 2022
c46d39f
Fix check for falsey inputs in run_summarization (#18155)
JohnGiorgi Jul 18, 2022
a4f97e6
Fix incorrect type hint for lang (#18161)
JohnGiorgi Jul 18, 2022
c1c79b0
NLLB tokenizer (#18126)
LysandreJik Jul 18, 2022
8c14b34
add ONNX support for LeVit (#18154)
gcheron Jul 18, 2022
7417f3a
[HPO] update to sigopt new experiment api (#18147)
sywangyi Jul 18, 2022
cb19c2a
Fix expected loss values in some (m)T5 tests (#18177)
ydshieh Jul 18, 2022
6561fbc
Update TF(Vision)EncoderDecoderModel PT/TF equivalence tests (#18073)
ydshieh Jul 18, 2022
4525581
FIX: Typo (#18156)
ayansengupta17 Jul 18, 2022
aeeab1f
Fix template for new models in README (#18182)
sgugger Jul 18, 2022
edadfc5
Better default for offload_state_dict in from_pretrained (#18183)
sgugger Jul 18, 2022
ced1f1f
fix typo inside bloom documentation (#18187)
SaulLu Jul 18, 2022
0a5b61d
Added preprocessing.mdx italian translation (#17600)
nickprock Jul 18, 2022
c4cc894
Translation italian: multilingual.mdx (#17768)
nickprock Jul 18, 2022
6a1b1bf
BLOOM minor fixes small test (#18175)
younesbelkada Jul 18, 2022
8e445ca
Translation/training: italian translation training.mdx (#17662)
nickprock Jul 18, 2022
bc8e30b
FSDP integration enhancements and fixes (#18134)
pacman100 Jul 18, 2022
29fd471
Use smaller variant of BLOOM for doc to fix tests
sgugger Jul 18, 2022
4bea658
Remove use_auth_token from the from_config method (#18192)
duongna21 Jul 19, 2022
e630dad
Add vision example to README (#18194)
sgugger Jul 19, 2022
7983844
bugfix: div-->dim (#18135)
orgoro Jul 19, 2022
ce01528
Update docs README with instructions on locally previewing docs (#18196)
snehankekre Jul 19, 2022
3bb6356
[From pretrained] Allow download from subfolder inside model repo (#1…
patrickvonplaten Jul 19, 2022
dc9147f
Custom pipeline (#18079)
sgugger Jul 19, 2022
9f12ec7
Typo in readme (#18195)
flozi00 Jul 19, 2022
05ed569
Use next-gen CircleCI convenience images (#18197)
ydshieh Jul 19, 2022
ec6cd76
TF: Add missing cast to GPT-J (#18201)
gante Jul 19, 2022
8a61fe0
Reduce console spam when using the KerasMetricCallback (#18202)
Rocketknight1 Jul 19, 2022
4b1ed79
update cache to v0.5 (#18203)
ydshieh Jul 20, 2022
0ed4d0d
Fix `LayoutXLM` docstrings (#17038)
qqaatw Jul 20, 2022
dcec4c4
Adding OPTForSeqClassification class (#18123)
oneraghavan Jul 20, 2022
bc6fe6f
Change to FlavaProcessor in PROCESSOR_MAPPING_NAMES (#18213)
ydshieh Jul 20, 2022
9edff45
skip some test_multi_gpu_data_parallel_forward (#18188)
ydshieh Jul 20, 2022
ff56b8f
Add custom config to quicktour (#18115)
stevhliu Jul 20, 2022
dbfeffd
Update add_new_pipeline.mdx (#18224)
zh-zheng Jul 21, 2022
d23cf5b
Add support for Sagemaker Model Parallel >= 1.10 new checkpoint API (…
viclzhu Jul 21, 2022
5e2f2d7
Better messaging and fix for incorrect shape when collating data. (#1…
CakeCrusher Jul 21, 2022
9f787ce
Translation/debugging (#18230)
nickprock Jul 21, 2022
43a5375
Fix TrainingArguments help section (#18232)
sgugger Jul 21, 2022
ba552dd
Make errors for loss-less models more user-friendly (#18233)
sgugger Jul 21, 2022
0d971be
Italian translation of run_scripts.mdx gh-17459 (#17642)
lorenzobalzani Jul 21, 2022
8881e58
Italian/model sharing (#17828)
mfumanelli Jul 21, 2022
07575e8
Italian/accelerate (#17698)
mfumanelli Jul 21, 2022
2c5747e
Update notification service (#17921)
ydshieh Jul 21, 2022
561b9a8
[SegFormer] TensorFlow port (#17910)
sayakpaul Jul 21, 2022
99eb9b5
Fix `no_trainer` CI (#18242)
muellerzr Jul 21, 2022
12d66b4
Add OWL-ViT model for zero-shot object detection (#17938)
alaradirik Jul 22, 2022
06d98e2
Fix OwlViT tests (#18253)
sgugger Jul 22, 2022
4935409
Add Italian translation of create_model.mdx and serialization.mdx (…
F02934 Jul 22, 2022
1fc4b2a
TF: use the correct config with `(...)EncoderDecoder` models (#18097)
gante Jul 22, 2022
7cb4da1
change bloom parameters to 176B (#18235)
muhammad-ahmed-ghani Jul 22, 2022
d95a32c
Fix torch version check in Vilt (#18260)
ydshieh Jul 22, 2022
0750535
Change how `take_along_axis` is computed in DeBERTa to stop confusing…
Rocketknight1 Jul 22, 2022
8e83846
Update serving code to enable `saved_model=True` (#18153)
amyeroberts Jul 22, 2022
7e44226
Generate: deprecate default `max_length` (#18018)
gante Jul 23, 2022
bbb62f2
Skip passes report for `--make-reports` (#18250)
ydshieh Jul 25, 2022
f4e1727
Allows `KerasMetricCallback` to use XLA generation (#18265)
Rocketknight1 Jul 25, 2022
45a1475
Fix TF bad words filter with XLA (#18286)
Rocketknight1 Jul 25, 2022
bd87480
Fix command of doc tests for local testing (#18236)
oneraghavan Jul 26, 2022
f65307e
Fix dtype of input_features in docstring (#18258)
ydshieh Jul 26, 2022
d32558c
Good difficult issue override for the stalebot (#18094)
LysandreJik Jul 26, 2022
002915a
Owlvit docs test (#18257)
alaradirik Jul 26, 2022
2b09650
Add ViltForTokenClassification e.g. for Named-Entity-Recognition (NER…
gilad19 Jul 26, 2022
2844c5d
Fix ORTTrainer failure on gpt2 fp16 training (#18017)
JingyaHuang Jul 26, 2022
c8ed1b8
Replace false parameter by a buffer (#18259)
sgugger Jul 26, 2022
5bb211b
Adding type hints of TF:CTRL (#18264)
Mathews-Tom Jul 26, 2022
f374d39
Adding type hints of TF:OpenAIGPT (#18263)
Mathews-Tom Jul 26, 2022
b516952
Add TFAutoModelForImageClassification to pipelines.py (#18292)
ydshieh Jul 26, 2022
f58b9c0
Update translation.mdx (#18169)
gorkemozkaya Jul 26, 2022
5e0ffd9
[ create_a_model.mdx ] translate to pt (#18098)
Fellip15 Jul 26, 2022
a649de5
Raise a TF-specific error when importing Torch classes (#18280)
Rocketknight1 Jul 26, 2022
bbc2810
Add Italian translation of converting_tensorflow_models.mdx (#18283)
Xpiri Jul 26, 2022
c4c6b4d
Add PyTorch 1.11 to past CI (#18302)
ydshieh Jul 26, 2022
7ea7eba
Add Italian translation of sharing_custom_models.mdx (#17631)
Xpiri Jul 26, 2022
a5d5048
Add Spanish translation of custom_models.mdx (#17807)
donelianc Jul 26, 2022
6649133
Add PYTEST_TIMEOUT for CircleCI test jobs (#18251)
ydshieh Jul 26, 2022
68097dc
Fix Sylvain's nits on the original KerasMetricCallback PR (#18300)
Rocketknight1 Jul 26, 2022
ee67e7a
patch for smddp import (#18244)
carolynwang Jul 26, 2022
47c2af0
[DETR] Improve code examples (#18262)
NielsRogge Jul 27, 2022
5dfec70
Remove duplicated line (#18310)
manuelciosici Jul 27, 2022
ccd4180
[EncoderDecoder] Improve docs (#18271)
NielsRogge Jul 27, 2022
e318cda
Apply type correction to `TFSwinModelOutput` (#18295)
harrydrippin Jul 27, 2022
d5610b5
[XLA] Improve t5 model performance (#18288)
ymwangg Jul 27, 2022
1d71ad8
Update CodeParrot readme to include training in Megatron (#17798)
loubnabnl Jul 27, 2022
70e7d1d
Fixes torch jit tracing for LayoutLMv2 model (re-open) (#18313)
mikkeldenker Jul 27, 2022
7996ef7
fix module order (#18312)
younesbelkada Jul 27, 2022
83d2d74
fix loading from pretrained for sharded model with `torch_dtype="auto…
NouamaneTazi Jul 27, 2022
170fcaa
Generalize decay_mask_fn to apply mask to all LayerNorm params (#18273)
duongna21 Jul 27, 2022
cf32b2e
Remove all uses of six (#18318)
sgugger Jul 27, 2022
5c5676c
sentencepiece shouldn't be required for the fast LayoutXLM tokenizer …
LysandreJik Jul 27, 2022
0077360
Fix sacremoses sof dependency for Transformers XL (#18321)
sgugger Jul 27, 2022
9caf68a
Owlvit test fixes (#18303)
alaradirik Jul 27, 2022
7490a97
[Flax] Fix incomplete batches in example scripts (#17863)
sanchit-gandhi Jul 27, 2022
c89a592
Dev version
LysandreJik Jul 27, 2022
e87ac9d
Add swin transformer v2 (#17469)
nandwalritik Jul 27, 2022
2b81f72
start from 1.12, torch_ccl is renamed as oneccl_bindings_for_pytorch …
sywangyi Jul 27, 2022
96be1b7
Update feature extractor docs (#18324)
stevhliu Jul 27, 2022
1e380c7
fixed typo (#18331)
banda-larga Jul 28, 2022
a8e2795
updated translation (#18333)
banda-larga Jul 28, 2022
985c7e3
Updated _toctree.yml (#18337)
nickprock Jul 28, 2022
5d1fed0
Update automatic_speech_recognition.py (#18339)
bofenghuang Jul 28, 2022
286a18f
Fix codeparrot deduplication - ignore whitespaces (#18023)
loubnabnl Jul 28, 2022
b53dab6
Remove Flax OPT from doctest for now (#18338)
ydshieh Jul 28, 2022
9c33665
Include tensorflow-aarch64 as a candidate (#18345)
ankrgyl Jul 28, 2022
7b09087
[BLOOM] Deprecate `position_ids` (#18342)
thomasw21 Jul 28, 2022
a258679
Migrate metric to Evaluate library for tensorflow examples (#18327)
VijayKalmath Jul 28, 2022
da503ea
Migrate metrics used in flax examples to Evaluate (#18348)
VijayKalmath Jul 28, 2022
a4ee463
[Docs] Fix Speech Encoder Decoder doc sample (#18346)
sanchit-gandhi Jul 29, 2022
a64bcb5
Fix OwlViT torchscript tests (#18347)
ydshieh Jul 29, 2022
986526a
Replace `as_target` context managers by direct calls (#18325)
sgugger Jul 29, 2022
1763770
Fix some doctests (#18359)
ydshieh Jul 29, 2022
4e2f4a9
[FX] Symbolic trace for Bloom (#18356)
michaelbenayoun Jul 29, 2022
51227e2
Fix TFSegformerForSemanticSegmentation doctest (#18362)
ydshieh Jul 29, 2022
b2e4b09
fix FSDP ShardedGradScaler (#18358)
pacman100 Jul 30, 2022
a736038
Bump mistune in /examples/research_projects/visual_bert (#18371)
dependabot[bot] Aug 1, 2022
25ec12e
Bump mistune from 0.8.4 to 2.0.3 in /examples/research_projects/lxmer…
dependabot[bot] Aug 1, 2022
1f84399
Migrate metric to Evaluate in Pytorch examples (#18369)
atturaioe Aug 1, 2022
679d68a
Correct the spelling of bleu metric (#18375)
ToluClassics Aug 1, 2022
96b5d7d
Remove pt-like calls on tf tensor (#18393)
amyeroberts Aug 1, 2022
1cd7c6f
Fix from_pretrained kwargs passing (#18387)
YouJiacheng Aug 1, 2022
bd6d1b4
Add a check regarding the number of occurrences of ``` (#18389)
ydshieh Aug 1, 2022
af1e6b4
Add evaluate to test dependencies (#18396)
sgugger Aug 1, 2022
1141371
Fix OPT doc tests (#18365)
ArthurZucker Aug 1, 2022
39e76d7
Fix doc tests (#18397)
NielsRogge Aug 1, 2022
e0bc4c7
Add balanced strategies for device_map in from_pretrained (#18349)
sgugger Aug 1, 2022
7b9e995
Fix docs (#18399)
NielsRogge Aug 1, 2022
62098b9
Adding fine-tuning models to LUKE (#18353)
ikuyamada Aug 1, 2022
941d233
Fix ROUGE add example check and update README (#18398)
sgugger Aug 1, 2022
3909d7f
Add Flax BART pretraining script (#18297)
duongna21 Aug 1, 2022
01db72a
Rewrite push_to_hub to use upload_files (#18366)
sgugger Aug 1, 2022
151a2aa
Split model list on modality (#18328)
stevhliu Aug 1, 2022
24845ae
Layoutlmv2 tesseractconfig (#17733)
kelvinAI Aug 1, 2022
df5e423
fix: create a copy for tokenizer object (#18408)
YBooks Aug 1, 2022
68a894a
Fix uninitialized parameter in conformer relative attention. (#18368)
Aug 2, 2022
dd21fb3
Fix the hub user name in a longformer doctest checkpoint (#18418)
ydshieh Aug 2, 2022
0b8c1b6
Change audio kwarg to images in TROCR processor (#18421)
ydshieh Aug 2, 2022
8ae7784
update maskformer docs (#18423)
alaradirik Aug 2, 2022
2959d09
Fix `test_load_default_pipelines_tf` test error (#18422)
ydshieh Aug 2, 2022
5546fb6
fix run_clip README (#18332)
ydshieh Aug 2, 2022
dbd9641
Improve `generate` docstring (#18198)
JoaoLages Aug 2, 2022
c382ed8
Accept `trust_remote_code` and ignore it in `PreTrainedModel.from_pre…
ydshieh Aug 2, 2022
042f420
Update pipeline word heuristic to work with whitespace in token offse…
davidbenton Aug 2, 2022
5096a65
Add programming languages (#18434)
cakiki Aug 2, 2022
22a0dd2
fixing error when using sharded ddp (#18435)
pacman100 Aug 3, 2022
92915eb
Update _toctree.yml (#18440)
stevhliu Aug 3, 2022
9d7b70b
support ONNX export of XDropout in deberta{,_v2} and sew_d (#17502)
garymm Aug 3, 2022
10e1ec9
Add Spanish translation of run_scripts.mdx (#18415)
donelianc Aug 3, 2022
3db4378
Update no trainer scripts for language modeling and image classificat…
nandwalritik Aug 3, 2022
a507908
Update pinned hhub version (#18448)
osanseviero Aug 3, 2022
8fb7c90
Fix failing tests for XLA generation in TF (#18298)
dsuess Aug 3, 2022
fc54633
add zero-shot obj detection notebook to docs (#18453)
alaradirik Aug 3, 2022
be41eaf
fix: keras fit tests for segformer tf and minor refactors. (#18412)
sayakpaul Aug 3, 2022
02b176c
Fix torch version comparisons (#18460)
LSinev Aug 3, 2022
b69a62d
[BLOOM] Clean modeling code (#18344)
thomasw21 Aug 4, 2022
fc1d841
change shape to support dynamic batch input in tf.function XLA genera…
nlpcat Aug 4, 2022
c74befc
HFTracer.trace can now take callables and torch.nn.Module (#18457)
michaelbenayoun Aug 4, 2022
330247e
Update no trainer scripts for multiple-choice (#18468)
kiansierra Aug 4, 2022
df28de0
Fix load of model checkpoints in the Trainer (#18470)
sgugger Aug 4, 2022
672b662
Add FX support for torch.baddbmm andd torch.Tensor.baddbmm (#18363)
thomasw21 Aug 4, 2022
f9a0008
Add VideoMAE (#17821)
NielsRogge Aug 4, 2022
d2704c4
Add machine type in the artifact of Examples directory job (#18459)
ydshieh Aug 4, 2022
0bf1e1a
Update no trainer examples for QA and Semantic Segmentation (#18474)
kiansierra Aug 4, 2022
1492892
Add `TF_MODEL_FOR_SEMANTIC_SEGMENTATION_MAPPING` (#18469)
ydshieh Aug 4, 2022
586dcf6
Fixing issue where generic model types wouldn't load properly with th…
Narsil Aug 5, 2022
575aa6e
Fix TFSwinSelfAttention to have relative position index as non-traina…
harrydrippin Aug 5, 2022
bf174f9
Refactor `TFSwinLayer` to increase serving compatibility (#18352)
harrydrippin Aug 5, 2022
893122f
Add TF prefix to TF-Res test class (#18481)
ydshieh Aug 5, 2022
f5cdb14
Update methods to optionally rescale
amyeroberts Aug 5, 2022
134e7a7
Cast images to numpy arrays in call to enable consistent behaviour wi…
amyeroberts Aug 5, 2022
94515a2
Cast frames to numpy arrays in call to enable consistent behaviour wi…
amyeroberts Aug 5, 2022
1ace93b
Remove accidental clip changes
amyeroberts Aug 5, 2022
6e21b7d
Remove accidental clip changes
amyeroberts Aug 5, 2022
51e4959
Merge branch 'type-cast-before-normalize-update-methods' into type-ca…
amyeroberts Aug 5, 2022
2b967ed
Make sure defaults are the same as before
amyeroberts Aug 11, 2022
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
The table of contents is too big for display.
Diff view
Diff view
  •  
  •  
  •  
247 changes: 137 additions & 110 deletions .circleci/config.yml

Large diffs are not rendered by default.

2 changes: 0 additions & 2 deletions .github/ISSUE_TEMPLATE/bug-report.yml
Original file line number Diff line number Diff line change
Expand Up @@ -7,7 +7,6 @@ body:
attributes:
label: System Info
description: Please share your system info with us. You can run the command `transformers-cli env` and copy-paste its output below.
render: shell
placeholder: transformers version, platform, python version, ...
validations:
required: true
Expand Down Expand Up @@ -118,4 +117,3 @@ body:
attributes:
label: Expected behavior
description: "A clear and concise description of what you would expect to happen."
render: shell
4 changes: 2 additions & 2 deletions .github/conda/meta.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -25,7 +25,7 @@ requirements:
- sacremoses
- regex !=2019.12.17
- protobuf
- tokenizers >=0.10.1,<0.11.0
- tokenizers >=0.11.1,!=0.11.3,<0.13
- pyyaml >=5.1
run:
- python
Expand All @@ -40,7 +40,7 @@ requirements:
- sacremoses
- regex !=2019.12.17
- protobuf
- tokenizers >=0.10.1,<0.11.0
- tokenizers >=0.11.1,!=0.11.3,<0.13
- pyyaml >=5.1

test:
Expand Down
2 changes: 1 addition & 1 deletion .github/workflows/add-model-like.yml
Original file line number Diff line number Diff line change
Expand Up @@ -27,7 +27,7 @@ jobs:
id: cache
with:
path: ~/venv/
key: v3-tests_model_like-${{ hashFiles('setup.py') }}
key: v4-tests_model_like-${{ hashFiles('setup.py') }}

- name: Create virtual environment on cache miss
if: steps.cache.outputs.cache-hit != 'true'
Expand Down
1 change: 1 addition & 0 deletions .github/workflows/build-docker-images.yml
Original file line number Diff line number Diff line change
Expand Up @@ -5,6 +5,7 @@ on:
branches:
- docker-image*
repository_dispatch:
workflow_call:
schedule:
- cron: "0 1 * * *"

Expand Down
108 changes: 108 additions & 0 deletions .github/workflows/build-past-ci-docker-images.yml
Original file line number Diff line number Diff line change
@@ -0,0 +1,108 @@
name: Build docker images (Past CI)

on:
push:
branches:
- past-ci-docker-image*

concurrency:
group: docker-images-builds
cancel-in-progress: false

jobs:
past-pytorch-docker:
name: "Past PyTorch Docker"
strategy:
fail-fast: false
matrix:
version: ["1.11", "1.10", "1.9", "1.8", "1.7", "1.6", "1.5", "1.4"]
runs-on: ubuntu-latest
steps:
-
name: Set up Docker Buildx
uses: docker/setup-buildx-action@v1
-
name: Check out code
uses: actions/checkout@v2
-
name: Login to DockerHub
uses: docker/login-action@v1
with:
username: ${{ secrets.DOCKERHUB_USERNAME }}
password: ${{ secrets.DOCKERHUB_PASSWORD }}
-
name: Build and push
uses: docker/build-push-action@v2
with:
context: ./docker/transformers-past-gpu
build-args: |
REF=main
FRAMEWORK=pytorch
VERSION=${{ matrix.version }}
push: true
tags: huggingface/transformers-pytorch-past-${{ matrix.version }}-gpu

past-tensorflow-docker:
name: "Past TensorFlow Docker"
strategy:
fail-fast: false
matrix:
version: ["2.8", "2.7", "2.6", "2.5"]
runs-on: ubuntu-latest
steps:
-
name: Set up Docker Buildx
uses: docker/setup-buildx-action@v1
-
name: Check out code
uses: actions/checkout@v2
-
name: Login to DockerHub
uses: docker/login-action@v1
with:
username: ${{ secrets.DOCKERHUB_USERNAME }}
password: ${{ secrets.DOCKERHUB_PASSWORD }}
-
name: Build and push
uses: docker/build-push-action@v2
with:
context: ./docker/transformers-past-gpu
build-args: |
REF=main
FRAMEWORK=tensorflow
VERSION=${{ matrix.version }}
push: true
tags: huggingface/transformers-tensorflow-past-${{ matrix.version }}-gpu

past-tensorflow-docker-2-4:
name: "Past TensorFlow Docker"
strategy:
fail-fast: false
matrix:
version: ["2.4"]
runs-on: ubuntu-latest
steps:
-
name: Set up Docker Buildx
uses: docker/setup-buildx-action@v1
-
name: Check out code
uses: actions/checkout@v2
-
name: Login to DockerHub
uses: docker/login-action@v1
with:
username: ${{ secrets.DOCKERHUB_USERNAME }}
password: ${{ secrets.DOCKERHUB_PASSWORD }}
-
name: Build and push
uses: docker/build-push-action@v2
with:
context: ./docker/transformers-past-gpu
build-args: |
REF=main
BASE_DOCKER_IMAGE=nvidia/cuda:11.0.3-cudnn8-devel-ubuntu20.04
FRAMEWORK=tensorflow
VERSION=${{ matrix.version }}
push: true
tags: huggingface/transformers-tensorflow-past-${{ matrix.version }}-gpu
2 changes: 1 addition & 1 deletion .github/workflows/model-templates.yml
Original file line number Diff line number Diff line change
Expand Up @@ -21,7 +21,7 @@ jobs:
id: cache
with:
path: ~/venv/
key: v3-tests_templates-${{ hashFiles('setup.py') }}
key: v4-tests_templates-${{ hashFiles('setup.py') }}

- name: Create virtual environment on cache miss
if: steps.cache.outputs.cache-hit != 'true'
Expand Down
136 changes: 136 additions & 0 deletions .github/workflows/self-past-caller.yml
Original file line number Diff line number Diff line change
@@ -0,0 +1,136 @@
name: Self-hosted runner (past-ci-caller)

on:
push:
branches:
- run-past-ci*

jobs:
run_past_ci_pytorch_1-11:
name: PyTorch 1.11
if: always()
uses: ./.github/workflows/self-past.yml
with:
framework: pytorch
version: "1.11"
secrets: inherit

run_past_ci_pytorch_1-10:
name: PyTorch 1.10
if: always()
needs: [run_past_ci_pytorch_1-11]
uses: ./.github/workflows/self-past.yml
with:
framework: pytorch
version: "1.10"
secrets: inherit

run_past_ci_pytorch_1-9:
name: PyTorch 1.9
if: always()
needs: [run_past_ci_pytorch_1-10]
uses: ./.github/workflows/self-past.yml
with:
framework: pytorch
version: "1.9"
secrets: inherit

run_past_ci_pytorch_1-8:
name: PyTorch 1.8
if: always()
needs: [run_past_ci_pytorch_1-9]
uses: ./.github/workflows/self-past.yml
with:
framework: pytorch
version: "1.8"
secrets: inherit

run_past_ci_pytorch_1-7:
name: PyTorch 1.7
if: always()
needs: [run_past_ci_pytorch_1-8]
uses: ./.github/workflows/self-past.yml
with:
framework: pytorch
version: "1.7"
secrets: inherit

run_past_ci_pytorch_1-6:
name: PyTorch 1.6
if: always()
needs: [run_past_ci_pytorch_1-7]
uses: ./.github/workflows/self-past.yml
with:
framework: pytorch
version: "1.6"
secrets: inherit

run_past_ci_pytorch_1-5:
name: PyTorch 1.5
if: always()
needs: [run_past_ci_pytorch_1-6]
uses: ./.github/workflows/self-past.yml
with:
framework: pytorch
version: "1.5"
secrets: inherit

run_past_ci_pytorch_1-4:
name: PyTorch 1.4
if: always()
needs: [run_past_ci_pytorch_1-5]
uses: ./.github/workflows/self-past.yml
with:
framework: pytorch
version: "1.4"
secrets: inherit

run_past_ci_tensorflow_2-8:
name: TensorFlow 2.8
if: always()
needs: [run_past_ci_pytorch_1-4]
uses: ./.github/workflows/self-past.yml
with:
framework: tensorflow
version: "2.8"
secrets: inherit

run_past_ci_tensorflow_2-7:
name: TensorFlow 2.7
if: always()
needs: [run_past_ci_tensorflow_2-8]
uses: ./.github/workflows/self-past.yml
with:
framework: tensorflow
version: "2.7"
secrets: inherit

run_past_ci_tensorflow_2-6:
name: TensorFlow 2.6
if: always()
needs: [run_past_ci_tensorflow_2-7]
uses: ./.github/workflows/self-past.yml
with:
framework: tensorflow
version: "2.6"
secrets: inherit

run_past_ci_tensorflow_2-5:
name: TensorFlow 2.5
if: always()
needs: [run_past_ci_tensorflow_2-6]
uses: ./.github/workflows/self-past.yml
with:
framework: tensorflow
version: "2.5"
secrets: inherit

run_past_ci_tensorflow_2-4:
name: TensorFlow 2.4
if: always()
needs: [run_past_ci_tensorflow_2-5]
uses: ./.github/workflows/self-past.yml
with:
framework: tensorflow
version: "2.4"
secrets: inherit
Loading