Commit graph

3936 commits

Author SHA1 Message Date
Eric Harper 7ce63be27b
Remove PTL 1.4 upper bound (#2600)
* update exp_manager

Signed-off-by: ericharper <complex451@gmail.com>

* update ptl trainer dataclass

Signed-off-by: ericharper <complex451@gmail.com>

* update check for ranks

Signed-off-by: ericharper <complex451@gmail.com>

* style

Signed-off-by: ericharper <complex451@gmail.com>

* remove returned from val epoch end

Signed-off-by: ericharper <complex451@gmail.com>

* revert

Signed-off-by: ericharper <complex451@gmail.com>

* Try to fix multi validation logging

Signed-off-by: smajumdar <titu1994@gmail.com>

* Try to fix multi validation logging

Signed-off-by: smajumdar <titu1994@gmail.com>

* Try to fix multi validation logging

Signed-off-by: smajumdar <titu1994@gmail.com>

* update for glue with PTL1.4 (#2634)

Signed-off-by: ekmb <ebakhturina@nvidia.com>

* add rank_zero_only to self.log

Signed-off-by: ericharper <complex451@gmail.com>

Co-authored-by: Somshubra Majumdar <titu1994@gmail.com>
Co-authored-by: Evelina <10428420+ekmb@users.noreply.github.com>
2021-08-10 12:50:22 -07:00
Yang Zhang 58c78869bd
Tn class eval (#2614)
* addde comments

Signed-off-by: Yang Zhang <yangzhang@nvidia.com>

* add class level evaluation for duplex text norm

Signed-off-by: Yang Zhang <yangzhang@nvidia.com>

* added comments

Signed-off-by: Yang Zhang <yangzhang@nvidia.com>

* fix doc string

Signed-off-by: Yang Zhang <yangzhang@nvidia.com>

* revert previous change

Signed-off-by: Yang Zhang <yangzhang@nvidia.com>

* fix class based eval to work with input with punctuation

Signed-off-by: Yang Zhang <yangzhang@nvidia.com>

* fix eval for itn

Signed-off-by: Yang Zhang <yangzhang@nvidia.com>

* add counts

Signed-off-by: Yang Zhang <yangzhang@nvidia.com>
2021-08-10 11:46:43 -07:00
Somshubra Majumdar 4af3986326
Move ASR Webapp (#2632)
Signed-off-by: smajumdar <titu1994@gmail.com>
2021-08-10 10:44:49 -06:00
Ryan Leary 2be5853cdb
Add basic grpc MT server (#1807)
* Add basic grpc MT server

Add readme, server updates

Signed-off-by: Ryan Leary <rleary@nvidia.com>

* style fix

Signed-off-by: Oleksii Kuchaiev <okuchaiev@nvidia.com>

* fixing license headers

Signed-off-by: Oleksii Kuchaiev <okuchaiev@nvidia.com>

* Add punctuation model into NMT service

Signed-off-by: MaximumEntropy <sandeep.subramanian.1@umontreal.ca>

* Fix merge conflicts

Signed-off-by: MaximumEntropy <sandeep.subramanian.1@umontreal.ca>

* style fixes to unblock CI

Signed-off-by: Oleksii Kuchaiev <okuchaiev@nvidia.com>

* Add a Jarvis ASR + NeMo NMT client

Signed-off-by: MaximumEntropy <sandeep.subramanian.1@umontreal.ca>

* style fixes

Signed-off-by: MaximumEntropy <sandeep.subramanian.1@umontreal.ca>

* Refactor gRPC service

Signed-off-by: MaximumEntropy <sandeep.subramanian.1@umontreal.ca>

* Update license headers

Signed-off-by: MaximumEntropy <sandeep.subramanian.1@umontreal.ca>

* Update one more license header

Signed-off-by: MaximumEntropy <sandeep.subramanian.1@umontreal.ca>

* Whitepsace in header

Signed-off-by: MaximumEntropy <sandeep.subramanian.1@umontreal.ca>

* Style fixes

Signed-off-by: MaximumEntropy <sandeep.subramanian.1@umontreal.ca>

* Fix grpc requirement

Signed-off-by: MaximumEntropy <sandeep.subramanian.1@umontreal.ca>

* Update license headers

Signed-off-by: MaximumEntropy <sandeep.subramanian.1@umontreal.ca>

* Add option to specify src/tgt lang and import fixes

Signed-off-by: MaximumEntropy <sandeep.subramanian.1@umontreal.ca>

* Style fixes

Signed-off-by: MaximumEntropy <sandeep.subramanian.1@umontreal.ca>

* Fix unused imports

Signed-off-by: MaximumEntropy <sandeep.subramanian.1@umontreal.ca>

* Renaming variables

Signed-off-by: MaximumEntropy <sandeep.subramanian.1@umontreal.ca>

Co-authored-by: Ryan Leary <rleary@nvidia.com>
Co-authored-by: Oleksii Kuchaiev <okuchaiev@users.noreply.github.com>
Co-authored-by: Oleksii Kuchaiev <okuchaiev@nvidia.com>
Co-authored-by: MaximumEntropy <sandeep.subramanian.1@umontreal.ca>
2021-08-09 16:31:41 -06:00
Tuan Manh Lai 7e6197d33d
Fixed an error related to the task indicator of the tagger during inference. (#2627)
* A minor fix to _infer() of DuplexTaggerModel
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

* Remove tagger data augmentation
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

* Minor fixes to cache path
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

* Style fix
Signed-off-by: Tuan Lai <tuanl@nvidia.com>
2021-08-09 14:32:11 -07:00
Ryan Leary 241855cc2a
Force fastpitch output to fp32 (#2629)
Signed-off-by: Ryan Leary <rleary@nvidia.com>

Co-authored-by: Ryan Leary <rleary@nvidia.com>
2021-08-09 12:11:01 -04:00
Tuan Manh Lai 18b51accbd
Evaluate the performance of the decoder for each semiotic class (#2625)
* Add class_based_decoding_evaluation.py
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

* Remove unused imports
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

* Add evaluation for tagger to class_based_eval
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

Co-authored-by: Yang Zhang <yzhang123@users.noreply.github.com>
2021-08-07 16:01:32 -07:00
Tuan Manh Lai 0d41f3a8fe
Allow using covering grammars for neural English TN model (#2602)
* Compute probability of each sequence
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

* Use CGs when the model is not confident
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

* Add script for visualization
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

* Add comments on how to generate visualizations
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

* Remove spaces in URLs when using CGs
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

* Allow setting n_tagged
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

* Add docstring
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

* if there is any exception, fall back to the input
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

* Minor changes to URL processing
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

* Style fix
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

* Add docstrings and comments
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

* PYNINI_AVAILABLE check
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

Co-authored-by: Yang Zhang <yzhang123@users.noreply.github.com>
2021-08-07 15:19:57 -07:00
Somshubra Majumdar fa7d95583e
Fix return config path for new SaveRestoreConnector (#2626)
* Fix return config path for new SaveRestore connector

Signed-off-by: smajumdar <titu1994@gmail.com>

* Fix return config path for new SaveRestore connector

Signed-off-by: smajumdar <titu1994@gmail.com>
2021-08-06 15:28:24 -07:00
Tuan Manh Lai 2150fbd2c8
Allowed setting the train set size of duplex TN training (#2605)
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

Co-authored-by: Yang Zhang <yzhang123@users.noreply.github.com>
2021-08-06 12:34:52 -07:00
Eric Harper ac73f9117c
Add save restore connector to ModelPT (#2592)
* add save restore connector

Signed-off-by: ericharper <complex451@gmail.com>

* add save restore connector

Signed-off-by: ericharper <complex451@gmail.com>

* add save restore connector property

Signed-off-by: ericharper <complex451@gmail.com>

* add _default_save_to

Signed-off-by: ericharper <complex451@gmail.com>

* moving globals to app_state

Signed-off-by: ericharper <complex451@gmail.com>

* moving globals to app_state

Signed-off-by: ericharper <complex451@gmail.com>

* add model attribute to connector

Signed-off-by: ericharper <complex451@gmail.com>

* add model attribute to connector

Signed-off-by: ericharper <complex451@gmail.com>

* fix tabs

Signed-off-by: ericharper <complex451@gmail.com>

* fix tabs

Signed-off-by: ericharper <complex451@gmail.com>

* remove ModelPT import

Signed-off-by: ericharper <complex451@gmail.com>

* add default restore

Signed-off-by: ericharper <complex451@gmail.com>

* add default restore

Signed-off-by: ericharper <complex451@gmail.com>

* remove eff globals

Signed-off-by: ericharper <complex451@gmail.com>

* style

Signed-off-by: ericharper <complex451@gmail.com>

* fix tabs

Signed-off-by: ericharper <complex451@gmail.com>

* style

Signed-off-by: ericharper <complex451@gmail.com>

* update globals, remove save restore property

Signed-off-by: ericharper <complex451@gmail.com>

* fix typo

Signed-off-by: ericharper <complex451@gmail.com>

* add setter

Signed-off-by: ericharper <complex451@gmail.com>

* add setter

Signed-off-by: ericharper <complex451@gmail.com>

* fix app_state restore flag

Signed-off-by: ericharper <complex451@gmail.com>

* typo

Signed-off-by: ericharper <complex451@gmail.com>

* update paths

Signed-off-by: ericharper <complex451@gmail.com>

* add connector arg to from_pretrained

Signed-off-by: ericharper <complex451@gmail.com>

* update save restore connector after instantiating

Signed-off-by: ericharper <complex451@gmail.com>

* use connector

Signed-off-by: ericharper <complex451@gmail.com>

* get class from config in .nemo

Signed-off-by: ericharper <complex451@gmail.com>

* add TODO

Signed-off-by: ericharper <complex451@gmail.com>

* move extract_state_dict to connector

Signed-off-by: ericharper <complex451@gmail.com>

* add methods for toch save and torch load

Signed-off-by: ericharper <complex451@gmail.com>

* typo

Signed-off-by: ericharper <complex451@gmail.com>

* typo

Signed-off-by: ericharper <complex451@gmail.com>

* typo

Signed-off-by: ericharper <complex451@gmail.com>

* update mock model conf

Signed-off-by: ericharper <complex451@gmail.com>

* revert

Signed-off-by: ericharper <complex451@gmail.com>

* move mock model to common collection

Signed-off-by: ericharper <complex451@gmail.com>

* update NLPModel

Signed-off-by: ericharper <complex451@gmail.com>

* update test to use connector

Signed-off-by: ericharper <complex451@gmail.com>

* move artifacts to save restore connector

Signed-off-by: ericharper <complex451@gmail.com>

* add save_restore_connector arg to register_artifact

Signed-off-by: ericharper <complex451@gmail.com>

* style

Signed-off-by: ericharper <complex451@gmail.com>

* default save_restore_connector arg to None

Signed-off-by: ericharper <complex451@gmail.com>

* default save_restore_connector arg to None

Signed-off-by: ericharper <complex451@gmail.com>

* clean commented line

Signed-off-by: ericharper <complex451@gmail.com>

* default save_restore_connector arg to None

Signed-off-by: ericharper <complex451@gmail.com>

* move MockModel

Signed-off-by: ericharper <complex451@gmail.com>

* fix docstrings, remove underscores, default from connector

Signed-off-by: ericharper <complex451@gmail.com>

* update docstring

Signed-off-by: ericharper <complex451@gmail.com>

* update docstring

Signed-off-by: ericharper <complex451@gmail.com>

* change name to is_model_being_restored

Signed-off-by: ericharper <complex451@gmail.com>

* move constants from AppState to SaveRestoreConnector

Signed-off-by: ericharper <complex451@gmail.com>

* encapsulate logic for model parallel checkpoint

Signed-off-by: ericharper <complex451@gmail.com>

* style

Signed-off-by: ericharper <complex451@gmail.com>

* update mock config

Signed-off-by: ericharper <complex451@gmail.com>

* remove unused import

Signed-off-by: ericharper <complex451@gmail.com>

* add init_subclass, remove connector arg from register_artifact, move MockModel to tests

Signed-off-by: ericharper <complex451@gmail.com>

* remove old import

Signed-off-by: ericharper <complex451@gmail.com>

* Add tests

Signed-off-by: smajumdar <titu1994@gmail.com>

* Finalize tests

Signed-off-by: smajumdar <titu1994@gmail.com>

* Finalize tests

Signed-off-by: smajumdar <titu1994@gmail.com>

* fixing lgtm

Signed-off-by: ericharper <complex451@gmail.com>

* fix lgtm

Signed-off-by: ericharper <complex451@gmail.com>

* update NLPModel.restore_from

Signed-off-by: ericharper <complex451@gmail.com>

* Fix classpath resolution

Signed-off-by: smajumdar <titu1994@gmail.com>

Co-authored-by: Somshubra Majumdar <titu1994@gmail.com>
2021-08-05 17:45:57 -07:00
Boris Fomitchev 17b68d73c9
CitriNet export fix (#2620)
* Fixing CitriNet ONNX export

Signed-off-by: Boris Fomitchev <bfomitchev@nvidia.com>

* Reverting context_window back to int

Signed-off-by: Boris Fomitchev <bfomitchev@nvidia.com>

* Cleaning up extra script

Signed-off-by: Boris Fomitchev <bfomitchev@nvidia.com>
2021-08-05 16:50:48 -07:00
Tuan Manh Lai b0c9afb4e8
Allow training a multilingual duplex TN model (#2583)
* Allow training a multilingual duplex TN model
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

* Add copyright header to combine_processed_datasets.py
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

* Minor Fix
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

Co-authored-by: Yang Zhang <yzhang123@users.noreply.github.com>
2021-08-05 15:52:39 -07:00
Tuan Manh Lai 3412ee0b44
Remove unused arg in TextNormalizationTestDataset (#2613)
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

Co-authored-by: Yang Zhang <yzhang123@users.noreply.github.com>
2021-08-05 11:27:59 -07:00
QIliang 7426a9d60b
resolved MisconfigurationException (#2616)
end of epoch will throw execption:

pytorch_lightning.utilities.exceptions.MisconfigurationException: ModelCheckpoint(monitor='val_loss') not found in the returned metrics: ['loss', 'l_mle', 'l_length', 'logdet']. HINT: Did you call self.log('val_loss', value) in the LightningModule?


Signed-off-by: Qiliang <xiaoqlster@gmail.com>
2021-08-05 10:07:29 -04:00
Somshubra Majumdar f092c7f656
Make ITN tests optional (run only on change) (#2611)
* Make ITN tests optional run only on change

Signed-off-by: smajumdar <titu1994@gmail.com>

* Revert Jenkinsfile for RNNT test

Signed-off-by: smajumdar <titu1994@gmail.com>
2021-08-04 10:48:33 -07:00
Somshubra Majumdar 7051487c7e
Update contextnet configs (#2601)
Signed-off-by: smajumdar <titu1994@gmail.com>
2021-08-03 15:01:04 -07:00
Evelina 97e67f7790
import fix (#2597)
* import fix

Signed-off-by: ekmb <ebakhturina@nvidia.com>

* remove unused import

Signed-off-by: ekmb <ebakhturina@nvidia.com>

* example import fix

Signed-off-by: ekmb <ebakhturina@nvidia.com>
2021-08-03 12:03:14 -07:00
Somshubra Majumdar d04c7e9b4e
Integrate NVIDIA DALI 1.4 to NeMo ASR (#2567)
* Initial prototype of ASR DALI integration with DALI 1.4

Signed-off-by: smajumdar <titu1994@gmail.com>

* Update dali support to 1.4

Signed-off-by: smajumdar <titu1994@gmail.com>

* Fix docs

Signed-off-by: smajumdar <titu1994@gmail.com>

* Address comments

Signed-off-by: smajumdar <titu1994@gmail.com>

* Apply suggestions from code review

Co-authored-by: Janusz Lisiecki <39967756+JanuszL@users.noreply.github.com>

* Address comments

Signed-off-by: smajumdar <titu1994@gmail.com>

* Correct module utils

Signed-off-by: smajumdar <titu1994@gmail.com>

Co-authored-by: Janusz Lisiecki <39967756+JanuszL@users.noreply.github.com>
2021-08-03 11:01:51 -07:00
Yang Zhang dfb5bf5b74
asr gemran (#2590)
Signed-off-by: Yang Zhang <yangzhang@nvidia.com>
2021-08-02 10:49:54 -07:00
Tuan Manh Lai 2133e7e834
Fixes for neural TN (#2581)
* Preprocessed Google TN data not need basic tokenization
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

* Minor Fix
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

* Error fix
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

* Added test after train
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

* Minor Fix
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

* Allow data caching for Tagger Dataset
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

* Allow data caching for Decoder dataset
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

* Minor fixes
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

Co-authored-by: Yang Zhang <yzhang123@users.noreply.github.com>
2021-07-30 14:52:58 -07:00
Eric Harper 4a7c3e3df4
Merge 1.2 bugfixes into main (#2588)
* update jenkinsfile

Signed-off-by: ericharper <complex451@gmail.com>

* update BRANCH

Signed-off-by: ericharper <complex451@gmail.com>

* Fix onnx for ASR notebook (#2542)

* Update onnx version

Signed-off-by: smajumdar <titu1994@gmail.com>

* Fix onnx

Signed-off-by: smajumdar <titu1994@gmail.com>

* Fix onnx

Signed-off-by: smajumdar <titu1994@gmail.com>

* Fix typos and MeCab import (#2541)

Signed-off-by: MaximumEntropy <sandeep.subramanian.1@umontreal.ca>

* Fix branch for ASR notebooks (#2549)

Signed-off-by: smajumdar <titu1994@gmail.com>

* rmtok (#2559)

Signed-off-by: Abhinav Khattar <aklife97@gmail.com>

* Add xxhash dependency (#2564)

Signed-off-by: MaximumEntropy <sandeep.subramanian.1@umontreal.ca>

* fix (#2566)

* fix

Signed-off-by: nithinraok <nithinrao.koluguri@gmail.com>

* doc add

Signed-off-by: nithinraok <nithinrao.koluguri@gmail.com>

* style fix

Signed-off-by: nithinraok <nithinrao.koluguri@gmail.com>

* Fix moses path issue (#2573)

Signed-off-by: MaximumEntropy <sandeep.subramanian.1@umontreal.ca>

* More moses data path fixes (#2575)

Signed-off-by: MaximumEntropy <sandeep.subramanian.1@umontreal.ca>

* Path fixes (#2580)

Signed-off-by: MaximumEntropy <sandeep.subramanian.1@umontreal.ca>

* Upper bound transformers for 1.2 (#2584)

* upper bound transformers and name change jarvis to riva

Signed-off-by: ericharper <complex451@gmail.com>

* upper bound transformers and name change jarvis to riva

Signed-off-by: ericharper <complex451@gmail.com>

* update jenkinsfile

Signed-off-by: ericharper <complex451@gmail.com>

* update notebooks branch

Signed-off-by: ericharper <complex451@gmail.com>

* update notebooks branch

Signed-off-by: ericharper <complex451@gmail.com>

* update notebooks branch

Signed-off-by: ericharper <complex451@gmail.com>

Co-authored-by: Somshubra Majumdar <titu1994@gmail.com>
Co-authored-by: Sandeep Subramanian <sandeep.subramanian.1@umontreal.ca>
Co-authored-by: Abhinav Khattar <aklife97@gmail.com>
Co-authored-by: Nithin Rao <nithinrao.koluguri@gmail.com>
2021-07-30 11:53:07 -06:00
PeganovAnton b5b29a69cc
Fix manifest name (#2546)
Signed-off-by: PeganovAnton <peganoff2@mail.ru>

Co-authored-by: fayejf <36722593+fayejf@users.noreply.github.com>
Co-authored-by: Oleksii Kuchaiev <okuchaiev@users.noreply.github.com>
2021-07-27 11:00:58 -07:00
Yang Zhang eaa2b1707b
add seconds to time for english tn (#2550)
Signed-off-by: Yang Zhang <yangzhang@nvidia.com>

Co-authored-by: Oleksii Kuchaiev <okuchaiev@users.noreply.github.com>
2021-07-27 07:34:10 -07:00
Aleksei Kalinov 92c79faf5f
Remove deprecated import from source file. (#2481)
Signed-off-by: Aleksei Kalinov <alekseia@nvidia.com>

Co-authored-by: Oleksii Kuchaiev <okuchaiev@users.noreply.github.com>
Co-authored-by: Somshubra Majumdar <titu1994@gmail.com>
2021-07-26 22:51:26 -07:00
Yang Zhang 222411fb71
fix bug (#2527)
Signed-off-by: Yang Zhang <yangzhang@nvidia.com>

Co-authored-by: Oleksii Kuchaiev <okuchaiev@users.noreply.github.com>
2021-07-26 21:22:26 -07:00
Jason cd6a691607
Style Patch (#2556)
* Update audio_preprocessing.py

* Update audio_preprocessing.py

Signed-off-by: Jason <jasoli@nvidia.com>

* style

Signed-off-by: Jason <jasoli@nvidia.com>
2021-07-26 10:51:43 -04:00
Jason f2910cb01d
Update audio_preprocessing.py (#2538)
* Update audio_preprocessing.py

* Update audio_preprocessing.py

Signed-off-by: Jason <jasoli@nvidia.com>
2021-07-26 09:03:55 -04:00
Ghasem bb7a9304fa
Adding Non English Tutor (#2532)
* Adding Non English Tutor

Signed-off-by: Ghasem Pasandi <gpasandi@nvidia.com>

* applying Evelina's comments

Signed-off-by: Ghasem Pasandi <gpasandi@nvidia.com>

Co-authored-by: Ghasem Pasandi <gpasandi@nvidia.com>
2021-07-23 09:02:12 -07:00
Eric Harper 0349415662
Merge bugfixes and doc updates from 1.2.0 to main (#2533)
* update jenkinsfile

Signed-off-by: ericharper <complex451@gmail.com>

* update BRANCH

Signed-off-by: ericharper <complex451@gmail.com>

* update package_info.py

Signed-off-by: ericharper <complex451@gmail.com>

* Update Dockerfile numba install for 21.06 (#2515)

* update Dockerfile

Signed-off-by: ericharper <complex451@gmail.com>

* update Dockerfile

Signed-off-by: ericharper <complex451@gmail.com>

* upper bound ptl 1.4 (#2517)

Signed-off-by: ericharper <complex451@gmail.com>

* Typo correction in asr streaming tutorial (#2520)

* Corrected typos

Signed-off-by: jbalam <jbalam@nvidia.com>

* datalayer->data layer

Signed-off-by: jbalam <jbalam@nvidia.com>

* Jarvis to Riva changes for doc 1.2.0 (#2521)

* Change Jarvis to Riva in export.rst (#2529)

Signed-off-by: Herb Kelly <hkelly@nvidia.com>

* update branch

Signed-off-by: ericharper <complex451@gmail.com>

* update version

Signed-off-by: ericharper <complex451@gmail.com>

Co-authored-by: Jagadeesh Balam <4916480+jbalam-nv@users.noreply.github.com>
Co-authored-by: hkelly33 <58792115+hkelly33@users.noreply.github.com>
2021-07-22 09:31:38 -06:00
Aleksandr Laptev a367249872
Convert kaldi data folder to manifest.json (#2447)
* kaldi data folder to manifest converting

Signed-off-by: GNroy <laptevsasha12@gmail.com>

* pipe and offset support

Signed-off-by: GNroy <laptevsasha12@gmail.com>

* logging added

Signed-off-by: GNroy <laptevsasha12@gmail.com>

* apply style fix

Signed-off-by: Aleksandr Laptev <alaptev@nvidia.com>

Co-authored-by: Oleksii Kuchaiev <okuchaiev@users.noreply.github.com>
Co-authored-by: Aleksandr Laptev <alaptev@nvidia.com>
2021-07-21 17:04:36 -07:00
Nithin Rao 9fad1dc929
Fix time stamps (#2522)
* time stamps done

Signed-off-by: nithinraok <nithinrao.koluguri@gmail.com>

* multi batch_size support

Signed-off-by: nithinraok <nithinrao.koluguri@gmail.com>

* add doc strings

Signed-off-by: nithinraok <nithinrao.koluguri@gmail.com>

* spelling fix

Signed-off-by: nithinraok <nithinrao.koluguri@gmail.com>

* bs default

Signed-off-by: nithinraok <nithinrao.koluguri@gmail.com>

* test jenkins remove file

Signed-off-by: nithinraok <nithinrao.koluguri@gmail.com>

* revert file not found jenkins fix

Signed-off-by: nithinraok <nithinrao.koluguri@gmail.com>

* subsegments rename

Signed-off-by: nithinraok <nithinrao.koluguri@gmail.com>

* out_dir check

Signed-off-by: nithinraok <nithinrao.koluguri@gmail.com>
2021-07-21 14:42:29 -07:00
Vahid Noroozi 57406c3975
Support for char-level models. (#2530) 2021-07-21 12:12:58 -07:00
Tuan Manh Lai b472670afa
Extending the neural TN/ITN models for other languages (#2497)
* extending the neural TN/ITN model to handle RU

Signed-off-by: Tuan Lai <tuanl@nvidia.com>

* Support German
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

* Catch AttributeError instead of BaseException
Signed-off-by: Tuan Lai <tuanl@nvidia.com>

* Style fix
Signed-off-by: Tuan Lai <tuanl@nvidia.com>
2021-07-21 09:22:53 -07:00
Somshubra Majumdar dc6960509f
Enable RNNT ONNX Export (#2510)
* Begin export of RNNT models

Signed-off-by: smajumdar <titu1994@gmail.com>

* Prepare export

Signed-off-by: smajumdar <titu1994@gmail.com>

* Refactor Exportable

Signed-off-by: smajumdar <titu1994@gmail.com>

* Enable RNNT export to onnx

Signed-off-by: smajumdar <titu1994@gmail.com>

* RNNT export

Signed-off-by: smajumdar <titu1994@gmail.com>

* Attempt to get stateful inference

Signed-off-by: smajumdar <titu1994@gmail.com>

* Revert changes attempt state fix

Signed-off-by: smajumdar <titu1994@gmail.com>

* Hack together forward pass <FIX LATER>

Signed-off-by: smajumdar <titu1994@gmail.com>

* Return length

Signed-off-by: smajumdar <titu1994@gmail.com>

* Correct signature

Signed-off-by: smajumdar <titu1994@gmail.com>

* Prepare stateful

Signed-off-by: smajumdar <titu1994@gmail.com>

* Continie checks for states

Signed-off-by: smajumdar <titu1994@gmail.com>

* Correct naming

Signed-off-by: smajumdar <titu1994@gmail.com>

* Correct names

Signed-off-by: smajumdar <titu1994@gmail.com>

* Pass trace checks

Signed-off-by: smajumdar <titu1994@gmail.com>

* Update test to be dynamic

Signed-off-by: smajumdar <titu1994@gmail.com>

* Update naming semantic

Signed-off-by: smajumdar <titu1994@gmail.com>

* Initial prototype of RNNT greedy decoding

Signed-off-by: smajumdar <titu1994@gmail.com>

* Fix tracer warnings

Signed-off-by: smajumdar <titu1994@gmail.com>

* Corrected names of outputs for RNNT encoder

Signed-off-by: smajumdar <titu1994@gmail.com>

* Fix export rnnt script

Signed-off-by: smajumdar <titu1994@gmail.com>

* Prototype RNNT forward inference

Signed-off-by: smajumdar <titu1994@gmail.com>

* Update asr export

Signed-off-by: smajumdar <titu1994@gmail.com>

* Runnable inference

Signed-off-by: smajumdar <titu1994@gmail.com>

* Correct log softmax for joint

Signed-off-by: smajumdar <titu1994@gmail.com>

* Finish RNNT export to onnx

Signed-off-by: smajumdar <titu1994@gmail.com>

* Fix onnx export

Signed-off-by: smajumdar <titu1994@gmail.com>

* Create parallel version of ConvASREncoder (#2456)

* Add parallel block that supports Jasper-like blocks.

Signed-off-by: Aleksei Kalinov <alekseia@nvidia.com>

* Add Encoder with parallel blocks.

Signed-off-by: Aleksei Kalinov <alekseia@nvidia.com>

* Mark test as unit test.

Signed-off-by: Aleksei Kalinov <alekseia@nvidia.com>

* Support pointwise residual connections to connect layers with different fitlers.

Signed-off-by: Aleksei Kalinov <alekseia@nvidia.com>

* Format source code

Signed-off-by: Aleksei Kalinov <alekseia@nvidia.com>

* Switch to 1D tensor for dropout weights.

Signed-off-by: Aleksei Kalinov <alekseia@nvidia.com>

* Format source code.

Signed-off-by: Aleksei Kalinov <alekseia@nvidia.com>

Co-authored-by: Somshubra Majumdar <titu1994@gmail.com>

* Fix style

Signed-off-by: smajumdar <titu1994@gmail.com>

* Update comments

Signed-off-by: smajumdar <titu1994@gmail.com>

* Freeze the nemo model

Signed-off-by: smajumdar <titu1994@gmail.com>

* Update transcribe script to support RNNT and rnnt model transcriptions

Signed-off-by: smajumdar <titu1994@gmail.com>

* Add threshold

Signed-off-by: smajumdar <titu1994@gmail.com>

* Add warning

Signed-off-by: smajumdar <titu1994@gmail.com>

* Minor fixes

Signed-off-by: smajumdar <titu1994@gmail.com>

* Style fix

Signed-off-by: smajumdar <titu1994@gmail.com>

* Correctly replace forward step

Signed-off-by: smajumdar <titu1994@gmail.com>

* Reset RNNT flag after export is done

Signed-off-by: smajumdar <titu1994@gmail.com>

* Generalize _rnnt_export flag

Signed-off-by: smajumdar <titu1994@gmail.com>

* Generalize _rnnt_export flag

Signed-off-by: smajumdar <titu1994@gmail.com>

Co-authored-by: Aleksei Kalinov <alekseia@nvidia.com>
2021-07-20 20:01:18 -07:00
Evelina c8f9427295
Eng TN update (#2516)
* added url support

Signed-off-by: ekmb <ebakhturina@nvidia.com>

* address added

Signed-off-by: ekmb <ebakhturina@nvidia.com>

* sh test and export update

Signed-off-by: ekmb <ebakhturina@nvidia.com>

* sh test and export update

Signed-off-by: ekmb <ebakhturina@nvidia.com>

* fix fraction for sh

Signed-off-by: ekmb <ebakhturina@nvidia.com>

* telephone with words added

Signed-off-by: ekmb <ebakhturina@nvidia.com>

* remove unused import

Signed-off-by: ekmb <ebakhturina@nvidia.com>

* clean up

Signed-off-by: ekmb <ebakhturina@nvidia.com>

* update

Signed-off-by: ekmb <ebakhturina@nvidia.com>
2021-07-20 14:57:58 -07:00
Yang Zhang a8b6a1a4dd
Itn german (#2486)
* initial german itn

Signed-off-by: Yang Zhang <yangzhang@nvidia.com>

* cardinal now can accept compound, hundred and thousand without any prefix, ordinal verbalization deleted suffix and replaced with dot

Signed-off-by: Yang Zhang <yangzhang@nvidia.com>

* added all date options

Signed-off-by: Yang Zhang <yangzhang@nvidia.com>

* added fraction

Signed-off-by: Yang Zhang <yangzhang@nvidia.com>

* added fraction to measure

Signed-off-by: Yang Zhang <yangzhang@nvidia.com>

* default cent to euro

Signed-off-by: Yang Zhang <yangzhang@nvidia.com>

* added fraction tsv (forgot in the past) and added hour to night to time class

Signed-off-by: Yang Zhang <yangzhang@nvidia.com>

* adjusted docstring to german

Signed-off-by: Yang Zhang <yangzhang@nvidia.com>

* fix lgtm

Signed-off-by: Yang Zhang <yangzhang@nvidia.com>

* delete unnecessary copyright

Signed-off-by: Yang Zhang <yangzhang@nvidia.com>

* fix header and delete wrong spelling

Signed-off-by: Yang Zhang <yangzhang@nvidia.com>

* delete wrong spelling

Signed-off-by: Yang Zhang <yangzhang@nvidia.com>

* added missing data values for time

Signed-off-by: Yang Zhang <yangzhang@nvidia.com>

* delete SH normalization test, cause it doesnt exist

Signed-off-by: Yang Zhang <yangzhang@nvidia.com>

* added all classes to SH test

Signed-off-by: Yang Zhang <yangzhang@nvidia.com>

* deleted redundant files, updated header

Signed-off-by: Yang Zhang <yangzhang@nvidia.com>

* adding back whitelist

Signed-off-by: Yang Zhang <yangzhang@nvidia.com>

Co-authored-by: Oleksii Kuchaiev <okuchaiev@users.noreply.github.com>
2021-07-20 12:41:36 -07:00
Micha Livne d08c7a7b9c
Nmt byte level tokenizer model fix (#2421)
* 1. Updated permissions to include x.

* 1. Fixed byte-level tokenizer when no model for tokenizer is given.

Signed-off-by: Micha Livne <mlivne@nvidia.com>

Co-authored-by: Micha Livne <mlivne@nvidia.com>
Co-authored-by: Oleksii Kuchaiev <okuchaiev@users.noreply.github.com>
2021-07-19 17:12:36 -07:00
Jason 846b150082
Update TTS Docs to recommend fastpitch and hifigan (#2498)
* update docs

Signed-off-by: Jason <jasoli@nvidia.com>

* update

Signed-off-by: Jason <jasoli@nvidia.com>
2021-07-19 16:30:35 -07:00
David ffb80e1bbd
minor updates (#2465)
Signed-off-by: David Mosallanezhad <amosalla@asu.edu>

Co-authored-by: Sandeep Subramanian <sandeep.subramanian.1@umontreal.ca>
2021-07-19 16:29:52 -07:00
roman-vygon 99557a6f6e
apply fix (#2461)
Signed-off-by: roman-vygon <roman.vygon@gmail.com>

Co-authored-by: fayejf <36722593+fayejf@users.noreply.github.com>
2021-07-19 16:13:45 -07:00
vadam5 e3f6867dd2
Entity linking documentation (#2357)
* Update tutorials.rst

Signed-off-by: Virginia Adams <vadams@nvidia.com>

* Update tutorials.rst

Signed-off-by: Virginia Adams <vadams@nvidia.com>

* Update models.rst

Signed-off-by: Virginia Adams <vadams@nvidia.com>

* Add files via upload

Signed-off-by: Virginia Adams <vadams@nvidia.com>

* Create entity_linking.rst

Signed-off-by: Virginia Adams <vadams@nvidia.com>

* Update README.rst

Signed-off-by: Virginia Adams <vadams@nvidia.com>

* Update entity_linking.rst

Signed-off-by: Virginia Adams <vadams@nvidia.com>

* Update nlp_all.bib

Signed-off-by: Virginia Adams <vadams@nvidia.com>

* Update entity_linking.rst

Signed-off-by: Virginia Adams <vadams@nvidia.com>

* Update entity_linking.rst

Signed-off-by: Virginia Adams <vadams@nvidia.com>

* fixed base typos and doc link

Signed-off-by: Virginia Adams <vadams@nvidia.com>

Co-authored-by: Sandeep Subramanian <sandeep.subramanian.1@umontreal.ca>
Co-authored-by: Oleksii Kuchaiev <okuchaiev@users.noreply.github.com>
2021-07-19 16:10:19 -07:00
khcs 9db4053e07
Update Relation_Extraction-BioMegatron.ipynb (#2389)
fix 'DDP' to 'ddp'

Co-authored-by: Oleksii Kuchaiev <okuchaiev@users.noreply.github.com>
2021-07-19 16:07:33 -07:00
Eric Harper c527e954b6
Update container version to 21.06 (#2431)
* update container version

Signed-off-by: ericharper <complex451@gmail.com>

* remove conda update from reinstall.sh

Signed-off-by: ericharper <complex451@gmail.com>

* pin numba in reinstall

Signed-off-by: ericharper <complex451@gmail.com>
2021-07-16 11:51:53 -07:00
felixmcgregor e5b8570eec
Remove multiplication of sample rate for pydub indexing (#2482)
Signed-off-by: fmcgregor <felix@saigen.co.za>

Co-authored-by: fmcgregor <felix@saigen.co.za>
Co-authored-by: Somshubra Majumdar <titu1994@gmail.com>
2021-07-16 09:11:08 -07:00
Aleksei Kalinov f893721684
Create parallel version of ConvASREncoder (#2456)
* Add parallel block that supports Jasper-like blocks.

Signed-off-by: Aleksei Kalinov <alekseia@nvidia.com>

* Add Encoder with parallel blocks.

Signed-off-by: Aleksei Kalinov <alekseia@nvidia.com>

* Mark test as unit test.

Signed-off-by: Aleksei Kalinov <alekseia@nvidia.com>

* Support pointwise residual connections to connect layers with different fitlers.

Signed-off-by: Aleksei Kalinov <alekseia@nvidia.com>

* Format source code

Signed-off-by: Aleksei Kalinov <alekseia@nvidia.com>

* Switch to 1D tensor for dropout weights.

Signed-off-by: Aleksei Kalinov <alekseia@nvidia.com>

* Format source code.

Signed-off-by: Aleksei Kalinov <alekseia@nvidia.com>

Co-authored-by: Somshubra Majumdar <titu1994@gmail.com>
2021-07-16 16:22:47 +03:00
Yang Zhang 159952d71f
add sgdqa to readme (#2492)
Signed-off-by: Yang Zhang <yangzhang@nvidia.com>

Co-authored-by: Eric Harper <complex451@gmail.com>
2021-07-15 23:14:32 -06:00
Eric Harper 765920cd68
Update README.rst 2021-07-15 18:17:19 -06:00
Sandeep Subramanian 3992946be1
Minor fixes to NMT Data Preprocessing Notebook (#2491)
* Minor fixes to NMT data notebok

Signed-off-by: MaximumEntropy <sandeep.subramanian.1@umontreal.ca>

* One more fix

Signed-off-by: MaximumEntropy <sandeep.subramanian.1@umontreal.ca>

* One more fix

Signed-off-by: MaximumEntropy <sandeep.subramanian.1@umontreal.ca>
2021-07-15 15:01:25 -07:00
Jason d445f44dfd
Add copyright headers check (#2490)
* add cpr hdrs

Signed-off-by: Jason <jasoli@nvidia.com>

* update

Signed-off-by: Jason <jasoli@nvidia.com>

* review

Signed-off-by: Jason <jasoli@nvidia.com>
2021-07-15 15:06:45 -04:00