Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

test #2

Open
wants to merge 774 commits into
base: v3
Choose a base branch
from
Open
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
774 commits
Select commit Hold shift + click to select a range
98ebe4f
Fix bugs in MuST-C preprocessing (#3887)
gegallego Dec 22, 2021
ca36b43
add METEOR scorer; fix chrF scorer config
kahne Dec 22, 2021
0b128f4
update ignore_prefix_size in label_smoothed_cross_entropy
kahne Dec 24, 2021
87d0ede
fix evaluation tokenizer for sacrebleu >= 2.0.0
kahne Dec 24, 2021
5cd7a21
S2ST oss (#2756)
an918tw Dec 28, 2021
7fddb9d
lint fixes (#2834)
Dec 29, 2021
2762a1c
Add regularization for multihead attention module and ffn module
Dec 30, 2021
6a183f3
add strict option to checkpoint_utils. load_pretrained_component_from…
kahne Dec 31, 2021
75f8fa2
update xm_transformer
kahne Dec 31, 2021
59b3ada
fix SacrebleuScorer.score()
kahne Dec 31, 2021
ee177fc
add xm_transformer test; refactor speech tests
kahne Dec 31, 2021
43defa1
add hub interface for TTS
kahne Jan 4, 2022
1d5da6d
add hub interface for S2T
kahne Jan 4, 2022
c9a8bea
Formatting fix: get CI green (#2860)
Jan 7, 2022
e69f1fa
speech integration tests for jointly trained models
yuntang Jan 7, 2022
40ff55a
conformer (#2859)
sravyapopuri388 Jan 11, 2022
da48518
Benchmarking OSS (#2852)
sravyapopuri388 Jan 11, 2022
b3fa510
Add mha prune to fairseq
Jan 11, 2022
2797962
Preprocess Split (#2738)
Mortimerp9 Jan 11, 2022
fa7663c
Update commands, checkpoints and contact info for shuffled word order…
shagunsodhani Jan 13, 2022
cf8ff8c
Add unittests for jitting EMA model
Jan 13, 2022
1575f30
Add ffn prune to fairseq
Jan 15, 2022
40eb731
Code cleanup
hygong-fb Jan 19, 2022
a075481
Decode using EMA model in IPL recipe
Jan 19, 2022
a59cea5
attn head selection
hygong-fb Jan 19, 2022
995c204
Data2vec prelim (#2929)
alexeib Jan 20, 2022
c71870f
Data2vec (#2936)
alexeib Jan 20, 2022
fc758bb
fix readme (#2939)
alexeib Jan 20, 2022
5fd38e3
Fix breakage from D33649708
tonybruguier-fb Jan 21, 2022
509e83e
Add XGLM downstream task evaluation examples (#4154)
todpole3 Jan 24, 2022
4a7835b
Hubert unit test (#2766)
Jan 25, 2022
1b61bba
Fix broken EMA in fairseq
Jan 27, 2022
d839d84
Miscellaneous S2T & S2 bug fixes
kahne Feb 1, 2022
790f3be
Add XGLM pre-training data format explaination (#4158)
todpole3 Feb 1, 2022
6b77013
Add citation details and other wording fixes to model card (#4172)
todpole3 Feb 1, 2022
5d2be95
add defaults again after importing user_module (#3007)
Feb 2, 2022
f591cc9
upgrade black for lints (#3004)
Mortimerp9 Feb 2, 2022
272c4c5
Fix hubert (#3019)
Feb 3, 2022
53cc55c
add BERTScore scorer
kahne Feb 3, 2022
6b7a7d6
Fix EMA GPU test
Feb 4, 2022
11b2830
Refactor speech tests and add missing regression tests (#3001)
sravyapopuri388 Feb 4, 2022
327cff2
Create a separate EMA implementation for in-model tracking (#3036)
alexeib Feb 7, 2022
8b02f00
fix s2s test - disable multitasking by setting multitask_config_yaml …
sravyapopuri388 Feb 9, 2022
5551a19
Change ParameterList and ParameterDict to be able to contain any kind…
albanD Feb 9, 2022
5f2515e
Fix failing test (#3065)
dianaml0 Feb 10, 2022
cfc4d84
add missing transformer arch and update PadDataset (#4212)
todpole3 Feb 16, 2022
67eaecd
Add regression test for SimulConvTransformerModel (#3031)
sravyapopuri388 Feb 16, 2022
5b87224
Open source conformer models and update documentation
sravyapopuri388 Feb 17, 2022
420136a
fix failing convtransformer test (#3107)
Feb 22, 2022
0c0ef06
Prosody-aware Generative Spoken Language Modelling (#3063)
eugene-kharitonov Feb 23, 2022
5175fd5
update readme for conformer based models (#3104)
Feb 23, 2022
4184752
Best metric is now only logged for the first of all the validation su…
Feb 25, 2022
d421749
Add s2s_conformer model to support conformer encoder in S2UT model (#…
sravyapopuri388 Feb 28, 2022
a24fdf2
Fixing and error related to Floor Division (#4221)
choprahetarth Mar 1, 2022
e55e094
AddTargetDataset now first adds EOS then pads target sequences (#4243)
Mar 1, 2022
1479d31
update s2st vocoder training instructions (#3156)
an918tw Mar 2, 2022
2e0b961
fix import_user_module (#3144)
Mar 3, 2022
592c122
docs: add social button in support of Ukraine (#4249)
dmitryvinn Mar 5, 2022
0f078de
Re-land D34058196 [sacrebleu==2.0.0] buckification [Back out D34503161]
theweiho Mar 8, 2022
d03f4e7
Minor fixes (#3198)
sravyapopuri388 Mar 11, 2022
f9d07a9
support tgt_lang_tag in speech-to-speech (#3187)
hygong-fb Mar 14, 2022
b554f5e
replace "prepend-tgt-lang-tag" with "prepend-tgt-lang-tag-as-bos" to …
hygong-fb Mar 16, 2022
f71c03f
Don't fsdp_wrap transformer encoder & decoder
theweiho Mar 24, 2022
7b9118b
Open source code for "Enhanced Direct Speech-to-Speech Translation Us…
sravyapopuri388 Mar 25, 2022
e9b8952
Fix an indentation issue for decoder sweep config
vineelpratap Mar 25, 2022
5265840
add CTC auxiliary loss to S2T Transformer
kahne Mar 26, 2022
c8a8e2c
release pre-trained models (#3245)
an918tw Mar 28, 2022
fef5006
formatting fix (#4310)
Mar 28, 2022
54ea689
adding readme (#4314)
huihuifan Mar 29, 2022
7d72f28
formatting fix (#4313)
Mar 29, 2022
06c65c8
Add Aim support for logging (#4311)
gorarakelyan Mar 29, 2022
47e2798
add large speech model and link to data2vec vision (#3264)
alexeib Apr 1, 2022
97076f9
Add codeowners file for UST related codebase (#3269)
sravyapopuri388 Apr 5, 2022
8fce12d
Support decimal label rate (#3937)
freewym Apr 5, 2022
31d94f5
Update XGLM README with model FFN dimension (#4331)
todpole3 Apr 6, 2022
8ca4813
Optimize ngram blocking in SequenceGenerator
anchit Apr 11, 2022
7e75884
Fix documentation errors in Enhanced Direct Speech-to-Speech Translat…
sravyapopuri388 Apr 12, 2022
edb25c6
Update README.md (#4349)
patrickvonplaten Apr 13, 2022
2e3c810
Fix typo in exception value (#4334)
mklimasz Apr 14, 2022
f862ff5
fix a bug in compute_mask_indices() for wav2vec (#3266)
freewym Apr 16, 2022
355ffbe
add masked_lm test (#4344)
azzhipa Apr 18, 2022
7356784
fix missing links for vocabulary and config in the Enhanced direct S2…
sravyapopuri388 Apr 19, 2022
806855b
Fast Beamable enc-dec attention
anchit Apr 20, 2022
caac187
fix formatting (#3346)
Apr 27, 2022
72d3408
Pull out some code into separate methods (#3068)
Apr 27, 2022
ab98e94
fairseq[-hydra]-train torchrun compatibility: default device_id set t…
colinclement Apr 28, 2022
0b54d9f
fix formatting (#3350)
Apr 28, 2022
51478ad
xformer integration (#2263)
dianaml0 May 4, 2022
e71c4d0
fix broken build and docs (#3362)
dianaml0 May 5, 2022
c8d6fb1
README for using xFormers in FairSeq (#4402)
dianaml0 May 6, 2022
4368ede
Remove code used for blocksparse masking (#4401)
dianaml0 May 10, 2022
993129d
Merge STPT: Step 3
yuntang May 11, 2022
1ab7a75
support appending lang tags in S2T task (#3353)
sravyapopuri388 May 11, 2022
17fc870
Release code and models for X-MOD (#3333)
artetxem May 12, 2022
f7303c3
format fix (#4418)
May 12, 2022
eb2d786
fix ema memory leak (#3384)
alexeib May 13, 2022
b5a039c
Add X-MOD paper citation (#3386)
artetxem May 13, 2022
b5e7b25
Missing `f` prefix on f-strings fix (#4380)
code-review-doctor May 23, 2022
e0884db
don't use half precision in test_ema on CPU (#3408)
jspark1105 May 27, 2022
86f7065
CircleCI deprecating Ubuntu 16.04-based machine images (#4218)
dianaml0 May 31, 2022
e35c593
xm transformer decoder arg loading & optional weight norm remove (#3400)
uralik Jun 2, 2022
f97cdf7
Working Build Wheels for Fairseq (#4206)
Jun 3, 2022
97b2d81
Auto release (#4455)
Jun 8, 2022
44a8660
Fix typo in release.yml (#4471)
dianaml0 Jun 9, 2022
3864ffa
Document Release Process (#4466)
dianaml0 Jun 9, 2022
10b797a
“new_criterions_for_end-point_detection”
Jun 9, 2022
c39fefc
Do not append commit hash to version (#4472)
dianaml0 Jun 9, 2022
3a72168
Add CircleCI status badge (#4473)
dianaml0 Jun 10, 2022
c42d11c
0.12.0 release (#4476)
dianaml0 Jun 10, 2022
fea865c
Update Inference example in speech text joint training
yuntang Jun 10, 2022
73b42f0
Refactor release.yml (#4475)
Jun 11, 2022
4bcab98
Add command to release workflow (#4483)
dianaml0 Jun 13, 2022
4b98d93
v0.12.1 release (#4486)
dianaml0 Jun 13, 2022
a0ceabc
include wav2vec-u 2.0 (#2826)
Alexander-H-Liu Jun 15, 2022
d9c661b
don't allow language tokens in output for 1:many decoding
jhcross Jun 15, 2022
3a757d7
BT enablement on fairseq - fairseq change (#4480)
Jun 16, 2022
08fe884
reuse dataloader and workers (#3442)
alexeib Jun 16, 2022
956fcf4
Back out "BT enablement on fairseq - fairseq change"
theweiho Jun 22, 2022
ae9b76a
add check for OC version in fairseq
jieru-hu Jun 22, 2022
5c8aaaf
Fix for dynamic quant decoder embedding
Jun 23, 2022
fe06598
Explicitly convert sequence length to integer
Jun 24, 2022
de26c45
Add fix and back compat for changed private omegaconf api (#4440)
voznesenskym Jun 24, 2022
5528b6a
add reading from zip audio to hubert dataset and scripts (#3403)
uralik Jun 24, 2022
d364fdb
Reland BT enablement on fairseq - fairseq change (#4513)
Jun 25, 2022
d83f376
Fix sdist install error (#4511)
dianaml0 Jun 27, 2022
4ee361c
fix format (#4515)
dianaml0 Jun 27, 2022
58c8041
v0.12.2 release (#4516)
dianaml0 Jun 27, 2022
fe56de4
Hot_reload_debug (#4519)
nxphi47 Jun 28, 2022
a6a6327
switch denoising and multilingual_denoising tasks to OmegaConf (#4447)
Jun 28, 2022
5d8d067
[docs] Update Flashlight Bindings Docs (#4522)
jacobkahn Jun 29, 2022
ba415c9
add span_masked_lm task (#4366)
Jun 29, 2022
5307a0e
turn persistent workers off by default (#4524)
alexeib Jun 30, 2022
71a21df
closes #4549 (#4550)
st-vincent1 Jul 7, 2022
570c942
fix multilingual training with tags only on target side (#4543)
mgaido91 Jul 7, 2022
cba35cd
fix multi modality dataset error with resampling datasets after first…
mgaido91 Jul 7, 2022
c82b7c5
STOP Dataset Release and Experiment Reproduction (#4578)
padentomasello Jul 18, 2022
e83bd93
OSS ckpts for Interspeech 2022 paper (#4588)
sravyapopuri388 Jul 19, 2022
8097fbe
Update xformers install, other small fixes (#4539)
dianaml0 Jul 20, 2022
8e804cb
Support direct S2ST models in HF demo (#4593)
sravyapopuri388 Jul 21, 2022
a546702
Merge source and target embedding tables into one table (#4594)
rasto2211 Jul 26, 2022
ad3bec5
Fix Linting Errors (#4611)
dianaml0 Jul 26, 2022
0c5731f
Support En->Hk model in fairseq hub. (#4614)
sravyapopuri388 Jul 28, 2022
4fe8583
padding fix in the adaptor layer (#4613)
uralik Jul 29, 2022
42360db
Fix uninitialized bias parameters in RelPositionMultiHeadedAttention …
sanchit-gandhi Aug 1, 2022
acd9a53
update isort (#4568)
Aug 1, 2022
f826615
fix padding of s2t_wav_transformer (#4629)
mgaido91 Aug 24, 2022
eda7037
Fix under generation issue for speech to speech translation models by…
sravyapopuri388 Aug 24, 2022
d81fac8
Fix edit_dist.cu header file directives (#4667)
cbalioglu Aug 29, 2022
c0c326c
Fix for #4689. (#4690)
BrightXiaoHan Sep 1, 2022
0bad0ce
Revert "Fix for #4689. (#4690)" (#4698)
cbalioglu Sep 8, 2022
9a00e03
Update CODEOWNERS (#4699)
cbalioglu Sep 8, 2022
eba8a50
[feat][ust] Noise and data augmentation suite (#4692)
aghatare Sep 8, 2022
dec1e6b
[fairseq] Guard call to `shape_as_tensor` with `is_in_onnx_export()` …
suo Sep 9, 2022
5ec3a27
add an option to fetch datapoints within a batch in an async manner, …
000Justin000 Sep 9, 2022
0cc475b
Reformat speech translation modules (#4707)
Sep 15, 2022
34ac732
Fix conformer encoder (#4721)
Sep 15, 2022
56e7fb4
Fix eval mode for S2ST models during training (#4722)
Sep 16, 2022
59e0027
update vocoder to support cpu (#4727)
lpw0 Sep 19, 2022
adebac6
Add Conv2dSubsampler (#4724)
Sep 20, 2022
03f3a73
Support multi-task learning for speech-to-text task (#4732)
Sep 21, 2022
699ab19
run all tests (#4733)
gwenzek Sep 23, 2022
25451b2
Add Rdrop (#4730)
Sep 28, 2022
dd0079b
use src_lengths from net_input if possible if src_tokens in the input…
uralik Sep 30, 2022
0e9a425
drop apex/megatron in CI (not used) (#4757)
gwenzek Oct 6, 2022
9273676
Support multitask for XMTransformer (#4763)
Oct 6, 2022
21865d5
Add a missing method to XMTransformer (#4764)
Oct 6, 2022
6d90f79
Refactor S2ST (#4747)
Oct 7, 2022
b400118
UnitY implementation (#4670)
Oct 7, 2022
a3bd672
make a scriptable dynamicconv (#4772)
moslehpour Oct 10, 2022
c20ba1f
make Multihead_attention scriptable (#4773)
moslehpour Oct 11, 2022
411d365
make lightconv scriptable (#4775)
moslehpour Oct 11, 2022
16538a0
fix dynamicconv test (#4779)
moslehpour Oct 12, 2022
8df77ea
Fix error in workflow (#4790)
dianaml0 Oct 14, 2022
2727d17
Remove support for cpython 3.6 (#4791)
dianaml0 Oct 14, 2022
144e7c5
Remove cpython support as well (#4792)
dianaml0 Oct 14, 2022
8adff21
Update release.yml (#4793)
dianaml0 Oct 14, 2022
d55ac5d
Lower required numpy (#4794)
dianaml0 Oct 14, 2022
4fa4e1f
Revert "Update release.yml (#4793)" (#4795)
dianaml0 Oct 15, 2022
e40c435
Update release.yml (#4796)
dianaml0 Oct 15, 2022
fdcfd79
Revert "Update release.yml (#4796)" (#4797)
dianaml0 Oct 15, 2022
b7b7928
Lower numpy version requirement (#4798)
dianaml0 Oct 15, 2022
f342939
Fix GLIBC 2.14 not found error (#4802)
dianaml0 Oct 17, 2022
66d713b
Update release.yml (#4804)
dianaml0 Oct 17, 2022
05625e3
fix #4708 (#4805)
moslehpour Oct 17, 2022
7a71b84
Revert changes to sinusoidal_positional_embedding.py (#4808)
moslehpour Oct 18, 2022
f2fefe4
Update README.md (#4809)
androstj Oct 18, 2022
0272196
Add LID generate & eval script (#4810)
androstj Oct 18, 2022
c8e4ab7
Fix UnitY (#4830)
Oct 26, 2022
59d966a
Make representation computation branchless in TransformerEncoderBase …
zhxchen17 Nov 2, 2022
b8ac3fa
Fix Torchscript typing in transformer_encoder.py (#4847)
zhxchen17 Nov 8, 2022
f131336
Add Generative Spoken Dialogue Language Modeling (#4879)
tuanh208 Nov 29, 2022
1f53315
Update deprecated torch.qr in glow.py example (#4685)
kit1980 Dec 10, 2022
0f33ccf
Emotion Conversion Paper Open Source (#4895)
felixkreuk Dec 12, 2022
d871f61
data2vec v2.0 (#4903)
alexeib Dec 12, 2022
902f4aa
remove missing config entries when loading task from checkpoint (#4905)
alexeib Dec 13, 2022
c19aed8
make apex optional (#4906)
alexeib Dec 13, 2022
6c57de3
Add file to generate manifests for stop dataset. (#4891)
padentomasello Dec 14, 2022
6188342
Update STOP dataset README to include proper link. (#4892)
padentomasello Dec 14, 2022
9ab5f32
Update README.md (#4893)
padentomasello Dec 14, 2022
3c1abb5
using foreach to reduce kernel (#4904)
000Justin000 Dec 17, 2022
58cc6cc
Update README.md to add data2vec blog post (#4913)
arbabu123 Dec 20, 2022
7d050ad
Update config to fix circleci failure (#4949)
dianaml0 Jan 19, 2023
e7f6596
Generative Spoken Dialogue Language Modeling Paper Open Source (#4957)
tuanh208 Jan 26, 2023
83c4e49
wav2vec2_laser (#4968)
Mortimerp9 Feb 2, 2023
214c0cb
ASR BLEU tool copied from ust branch into main (#4914)
uralik Feb 2, 2023
ad0e69c
Add transcript option for asr-bleu (#4981)
xutaima Feb 9, 2023
0338cdc
fix imports referencing moved metrics.py file (#4840)
franznowak Feb 23, 2023
176cd93
add data2vec2 nlp model dictionary (#5045)
wnhsu Mar 27, 2023
3f6ba43
Fix DiverseBeamSearch so that no diversity groups will be dropped. (#…
shuminghu Apr 12, 2023
b35e8ef
XGLM paper camera-ready: add XStoryCloze data opensource (#4820)
todpole3 May 8, 2023
5ecbbf5
update XStoryCloze dataset description (#5100)
todpole3 May 8, 2023
bfd9dc6
fix dict order (#5109)
vineelpratap May 19, 2023
728b947
Mms release (#3948) (#5110)
vineelpratap May 22, 2023
aec128c
Update blog post link for MMS (#5114)
vineelpratap May 22, 2023
c7bfa9b
Update MMS README (#5115)
vineelpratap May 22, 2023
af12c9c
Update README.md (#5118)
vineelpratap May 22, 2023
87d3005
[MMS] Fix concatenation of emissions (#5133)
vineelpratap May 23, 2023
bc8e8b1
MMS: Fix forced alignment API usage (#5138)
vineelpratap May 23, 2023
1082b61
[MMS] Update README.md (#5137)
vineelpratap May 23, 2023
fea3361
[MMS] TTS text uromanization + cpu inference (#5140)
chevalierNoir May 24, 2023
b50b649
Fix the MMS doc about LID manifest (#5144)
kirill-fedyanin May 24, 2023
25c20e6
Fix wrong input-output ASR input utts order (#5149)
androstj May 24, 2023
478787b
Create MMS_ASR_Inference_Colab.ipynb (#5151)
androstj May 25, 2023
68c52f5
[MMS] Create Colab notebook for LID inference (#5157)
vineelpratap May 25, 2023
35641fb
[MMS] Add a tutorial on CC LM decoding for ASR model (#5160)
vineelpratap May 26, 2023
b309803
Feature: Implement UTF-8 readfile support in MMS for TTS example (#5148)
serkandyck May 26, 2023
ae59bd6
MMS TTS Colab Notebook (#5165)
chevalierNoir May 26, 2023
533644c
MMS TTS Romanian char fix + MPS support + full checkpoint (#5168)
chevalierNoir May 30, 2023
456ffcf
fix missing extra args in ConformerLayer (#5176)
androstj May 31, 2023
b2d5b78
Add transformers MMS checkpoints to docs (#5186)
patrickvonplaten Jun 4, 2023
91c364b
Update MMS README.md (#5202)
vineelpratap Jun 14, 2023
8deb43a
add new instructions on how to get manifest *.tsv file (#5207)
androstj Jun 17, 2023
a29952c
Update README.md (#5211)
patrickvonplaten Jun 20, 2023
31fba01
Register `weights` as a non-persistent buffer of `SinusoidalPositiona…
MaigoAkisame Jun 23, 2023
100cd91
Make RotaryPositionalEmbedding jit-compatible (#5237)
EgorLakomkin Jul 7, 2023
4db2649
Add batchnorm option to hubert/wav2vec2 positional convolution layer …
tuanh208 Aug 18, 2023
b5d89cd
Update align_and_segment.py (#5317)
vineelpratap Sep 7, 2023
e29f53b
initial revision (#5328)
piyush-kansal Sep 15, 2023
7409af7
Keep task level checkpoint key name generic (#5330)
piyush-kansal Sep 15, 2023
c7c478b
fix iterator when loading from checkpoint (#5344)
000Justin000 Oct 9, 2023
da8fb63
Change Meta AI to FAIR (#5346)
cbalioglu Oct 10, 2023
fad2c4d
Update README.md (#5407)
ylacombe Jan 8, 2024
3f0f20f
MMS alignment README fixes (#5432)
raphaelmerx Jan 24, 2024
34973a9
Multires hubert (#5363)
ftshijt Feb 26, 2024
bedb259
Delete .circleci directory (#5458)
bigfootjon Mar 13, 2024
d9a6270
Create depreview.yml (#5501)
bigfootjon May 30, 2024
920a548
Create README.md (#5529)
vineelpratap Jul 22, 2024
c214511
Add LID rerank for MMS (#5545)
brianyan918 Sep 27, 2024
018621f
update paper link + bug fix (#5547)
brianyan918 Oct 3, 2024
ecbf110
Add an `auto_expand` option to `SinusoidalPositionalEmbedding` (#5555)
MaigoAkisame Oct 18, 2024
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
  •  
  •  
  •  
21 changes: 21 additions & 0 deletions .github/CODEOWNERS
Validating CODEOWNERS rules …
Original file line number Diff line number Diff line change
@@ -0,0 +1,21 @@
# Setting up CODEOWNERS for UST related codebase
# Documentation for open sourced models relevant to UST
examples/speech_to_text @kahne @sravyapopuri388 @jmp84
examples/speech_to_speech @an918tw @sravyapopuri388 @jmp84
examples/speech_synthesis @kahne @jmp84
examples/simultaneous_translation @kahne @jmp84
examples/speech_text_joint_to_text @yuntang @jmp84

# Speech related models relevant to UST
fairseq/models/speech_to_speech @sravyapopuri388 @jmp84
fairseq/models/speech_to_text @kahne @sravyapopuri388 @jmp84
fairseq/models/text_to_speech @kahne @jmp84

# CONFORMER IMPLEMENTATION
fairseq/modules/conformer_layer.py @sravyapopuri388 @jmp84
fairseq/modules/espnet_multihead_attention.py @sravyapopuri388 @jmp84
fairseq/modules/rotary_positional_embedding.py @sravyapopuri388 @jmp84
fairseq/modules/positional_encoding.py @sravyapopuri388 @jmp84

# Machine Translation/NLLB
fairseq/tasks/translation.py @gwenzek
4 changes: 2 additions & 2 deletions .github/ISSUE_TEMPLATE/bug_report.md
Original file line number Diff line number Diff line change
Expand Up @@ -19,7 +19,7 @@ Steps to reproduce the behavior (**always include the command you ran**):


#### Code sample
<!-- Ideally attach a minimal code sample to reproduce the decried issue.
<!-- Ideally attach a minimal code sample to reproduce the decried issue.
Minimal means having the shortest code but still preserving the bug. -->

### Expected behavior
Expand All @@ -28,7 +28,7 @@ Minimal means having the shortest code but still preserving the bug. -->

### Environment

- fairseq Version (e.g., 1.0 or master):
- fairseq Version (e.g., 1.0 or main):
- PyTorch Version (e.g., 1.0)
- OS (e.g., Linux):
- How you installed fairseq (`pip`, source):
Expand Down
10 changes: 5 additions & 5 deletions .github/ISSUE_TEMPLATE/how-to-question.md
Original file line number Diff line number Diff line change
Expand Up @@ -6,23 +6,23 @@ labels: 'question, needs triage'

## ❓ Questions and Help

### Before asking:
1. search the issues.
2. search the docs.
### Before asking:
1. search the issues.
2. search the docs.

<!-- If you still can't find what you need: -->

#### What is your question?

#### Code

<!-- Please paste a code snippet if your question requires it! -->
<!-- Please paste a code snippet if your question requires it! -->

#### What have you tried?

#### What's your environment?

- fairseq Version (e.g., 1.0 or master):
- fairseq Version (e.g., 1.0 or main):
- PyTorch Version (e.g., 1.0)
- OS (e.g., Linux):
- How you installed fairseq (`pip`, source):
Expand Down
10 changes: 5 additions & 5 deletions .github/PULL_REQUEST_TEMPLATE.md
Original file line number Diff line number Diff line change
@@ -1,15 +1,15 @@
# Before submitting

- [ ] Was this discussed/approved via a Github issue? (no need for typos, doc improvements)
- [ ] Did you read the [contributor guideline](https://github.com/pytorch/fairseq/blob/master/CONTRIBUTING.md)?
- [ ] Did you make sure to update the docs?
- [ ] Did you write any new necessary tests?
- [ ] Did you read the [contributor guideline](https://github.com/pytorch/fairseq/blob/main/CONTRIBUTING.md)?
- [ ] Did you make sure to update the docs?
- [ ] Did you write any new necessary tests?

## What does this PR do?
Fixes # (issue).

## PR review
Anyone in the community is free to review the PR once the tests have passed.
## PR review
Anyone in the community is free to review the PR once the tests have passed.
If we didn't discuss your PR in Github issues there's a high chance it will not be merged.

## Did you have fun?
Expand Down
30 changes: 30 additions & 0 deletions .github/stale.yml
Original file line number Diff line number Diff line change
@@ -0,0 +1,30 @@
# Configuration for probot-stale - https://github.com/probot/stale
# Mostly copied from github.com/facebook/react/blob/master/.github/stale.yml
# Number of days of inactivity before an issue becomes stale
daysUntilStale: 90
# Number of days of inactivity before a stale issue is closed
daysUntilClose: 7
# Issues with these labels will never be considered stale
exemptLabels:
- bug
# Label to use when marking an issue as stale
staleLabel: stale
issues:
# Comment to post when marking an issue as stale.
markComment: >
This issue has been automatically marked as stale.
**If this issue is still affecting you, please leave any comment** (for example, "bump"), and we'll keep it open.
We are sorry that we haven't been able to prioritize it yet. If you have any new additional information, please include it with your comment!
# Comment to post when closing a stale issue.
closeComment: >
Closing this issue after a prolonged period of inactivity. If this issue is still present in the latest release, please create a new issue with up-to-date information. Thank you!
pulls:
# Comment to post when marking a pull request as stale.
markComment: >
This pull request has been automatically marked as stale.
**If this pull request is still relevant, please leave any comment** (for example, "bump"), and we'll keep it open.
We are sorry that we haven't been able to prioritize reviewing it yet. Your contribution is very much appreciated.
# Comment to post when closing a stale pull request.
closeComment: >
Closing this pull request after a prolonged period of inactivity. If this issue is still present in the latest release, please ask for this pull request to be reopened. Thank you!

57 changes: 47 additions & 10 deletions .github/workflows/build.yml
Original file line number Diff line number Diff line change
@@ -1,10 +1,10 @@
name: build

on:
# Trigger the workflow on push to master or any pull request
# Trigger the workflow on push to main or any pull request
push:
branches:
- master
- main
pull_request:

jobs:
Expand All @@ -14,31 +14,68 @@ jobs:
max-parallel: 4
matrix:
platform: [ubuntu-latest, macos-latest]
python-version: [3.6, 3.7]
python-version: [3.8, 3.9]

runs-on: ${{ matrix.platform }}

steps:
- uses: actions/checkout@v1
- uses: actions/checkout@v2

- name: Set up Python ${{ matrix.python-version }}
uses: actions/setup-python@v1
uses: actions/setup-python@v2
with:
python-version: ${{ matrix.python-version }}

- name: Conditionally install pytorch
if: matrix.platform == 'windows-latest'
run: pip3 install torch -f https://download.pytorch.org/whl/torch_stable.html

- name: Install locally
run: |
python -m pip install --upgrade pip
python setup.py build_ext --inplace
python -m pip install --editable .
git submodule update --init --recursive
python -m pip install .

- name: Check installation
working-directory: /tmp
run: python $GITHUB_WORKSPACE/scripts/check_installation.py

- name: Install optional test requirements
run: |
python -m pip install '.[dev,docs]'
python -m pip install iopath transformers pyarrow
python -m pip install git+https://github.com/facebookresearch/fairscale.git@main
python -m pip install pygit2 pgzip

- name: Install xformers for Macos
if: matrix.platform == 'macos-latest'
run: |
brew install llvm libomp
CC=/usr/local/opt/llvm/bin/clang CXX=clang++ pip install git+https://github.com/facebookresearch/xformers.git@main

- name: Install xformers for non-MacOS
if: matrix.platform != 'macos-latest'
run: |
python -m pip install --progress-bar off git+https://github.com/facebookresearch/xformers.git@main

- name: Lint with black
run: black --check --diff .

- name: Lint with flake8
run: |
pip install flake8
# stop the build if there are Python syntax errors or undefined names
flake8 . --count --select=E9,F63,F7,F82 --show-source --statistics
# exit-zero treats all errors as warnings. The GitHub editor is 127 chars wide
flake8 . --count --exit-zero --max-complexity=10 --max-line-length=127 --statistics

- name: Build doc
run: make singlehtml
working-directory: docs/

- name: Run tests
run: |
python setup.py test
# When installing in non-editable mode, the .so files will be generated in 'site-packages/fairseq'.
# But by default, pytest import machinery will load local fairseq, and won't see the .so.
# Use --import-mode=append to favorize the 'site-packages/fairseq'.
# https://docs.pytest.org/en/7.1.x/explanation/pythonpath.html
run: pytest --import-mode=append -vvv tests/

14 changes: 14 additions & 0 deletions .github/workflows/depreview.yml
Original file line number Diff line number Diff line change
@@ -0,0 +1,14 @@
name: 'Dependency Review'
on: [pull_request]

permissions:
contents: read

jobs:
dependency-review:
runs-on: ubuntu-latest
steps:
- name: 'Checkout Repository'
uses: actions/checkout@v4
- name: Dependency Review
uses: actions/dependency-review-action@v4
161 changes: 161 additions & 0 deletions .github/workflows/release.yml
Original file line number Diff line number Diff line change
@@ -0,0 +1,161 @@
name: Fairseq Release

on:
workflow_dispatch:
inputs:
name:
description: 'Release Type'
default: 'patch'
required: true

jobs:

get_next_version:
runs-on: ubuntu-latest
steps:
- name: checkout-repo-content
uses: actions/checkout@v2

- name: setup-python
uses: actions/setup-python@v2
with:
python-version: 3.8

- name: get next version and tag
id: get-next-version-and-tag
run: |
output=$(python3 release_utils.py --release-type ${{ github.event.inputs.name }})
echo $output
new_version=$(echo $output | awk '{print $1}')
new_tag=$(echo $output | awk '{print $2}')
echo "new version is $new_version"
echo "new tag is $new_tag"
echo ::set-output name=version::$new_version
echo ::set-output name=tag::$new_tag
echo ::set-output name=branch_name::$new_version-release
echo "NEW_TAG=$new_tag" >> $GITHUB_ENV
echo "NEW_BRANCH=$new_version-release" >> $GITHUB_ENV


# update the version number in version.txt
- name: update version
id: update-version
run : |
echo "current folder = $PWD"
echo "current branch = $(git branch --show-current)"
output=$(python3 release_utils.py --release-type ${{ github.event.inputs.name }} --update-version)

- name: add and commit
uses: EndBug/add-and-commit@v9
with:
author_name: ${{ secrets.AUTHOR_NAME }}
author_email: ${{ secrets.AUTHOR_EMAIL }}

# TODO: change this to main once shipit is disabled.
new_branch: '${{ env.NEW_BRANCH }}'
default_author: github_actor
message: '${{ env.NEW_TAG }} release'
pathspec_error_handling: exitAtEnd

# Arguments for the git pull command. Use NO-PULL to avoid the action pulling at all.
# pull: 'NO-PULL'
tag: '${{ env.NEW_TAG }}'

outputs:
new_version: ${{ steps.get-next-version-and-tag.outputs.version }}
new_tag: ${{ steps.get-next-version-and-tag.outputs.tag }}
branch_name: ${{ steps.get-next-version-and-tag.outputs.branch_name }}

create_sdist:
runs-on: ubuntu-latest
name: Create Source Distribution
needs: get_next_version
steps:
- uses: actions/checkout@v3
with:
ref: ${{ needs.get_next_version.outputs.branch_name }}

- name: Install Python
uses: actions/setup-python@v2
with:
python-version: '3.8'

- name: Upgrade pip
run: |
python3 -m pip install --upgrade pip

- name: Create Source Distribution
run: |
python3 -m pip install setuptools wheel twine torch
python3 setup.py sdist

- uses: actions/upload-artifact@v2
with:
path: dist/*.tar.gz

build_wheels:
name: Build wheels on ${{ matrix.os }}
runs-on: ${{ matrix.os }}
needs: get_next_version
strategy:
matrix:
os: [ubuntu-latest, macos-latest]

steps:
- uses: actions/checkout@v3
with:
ref: ${{ needs.get_next_version.outputs.branch_name }}

- name: Install Python
uses: actions/setup-python@v2
with:
python-version: '3.8'

- name: Upgrade pip
run: |
python3 -m pip install --upgrade pip

- name: Install cibuildwheel
run: |
python3 -m pip install cibuildwheel

- name: Build wheels for CPython
run: |
python3 -m cibuildwheel --output-dir dist
env:
CIBW_BUILD: "cp38-*64"
CIBW_MANYLINUX_X86_64_IMAGE: manylinux1
CIBW_BEFORE_BUILD: git submodule update --init --recursive && pip install .
# Install system library
CIBW_BEFORE_BUILD_LINUX: (yum install -y libffi-devel || apt-get install -y libffi-devel || apk add --update --no-cache libffi-devel || true) && (yum install -y libc6 || apt-get install -y libc6 || apk add --update --no-cache libc6 || true)
CIBW_ENVIRONMENT: "PIP_ONLY_BINARY=numpy"
CIBW_SKIP: "*musllinux*"

- uses: actions/upload-artifact@v2
with:
path: dist

upload:
name: Upload to PyPi and create release
runs-on: ubuntu-latest
needs: [build_wheels, create_sdist, get_next_version]
steps:
- uses: actions/download-artifact@v2
with:
name: artifact
path: dist

# build the PyPI package and upload it
- name: upload
env:
TWINE_USERNAME: ${{ secrets.PYPI_USERNAME }}
TWINE_PASSWORD: ${{ secrets.PYPI_PASSWORD }}
run: |
pip install setuptools wheel twine
python3 -m twine upload --repository pypi dist/*

# create the release on github
- name: create release on github
uses: ncipollo/release-action@v1
with:
tag: '${{ needs.get_next_version.outputs.new_tag }}'
8 changes: 8 additions & 0 deletions .gitignore
Original file line number Diff line number Diff line change
Expand Up @@ -131,3 +131,11 @@ data-bin/

# Experimental Folder
experimental/*

# Weights and Biases logs
wandb/

# Hydra artifacts
nohup.out
multirun
outputs
Loading