COG Datalyer by aasseman · Pull Request #2 · tkornuta-nvidia/NeMo

aasseman · 2020-06-02T23:54:31Z

Ported COG dataset from https://github.com/IBM/mi-prometheus as a NeMo datalayer.
Also made lots of cleanup of the original code. Tried to organize the commits in a clean and logical way, so consulting the commits one by one should help with tracking the modifications.

To run test, in the root of the NeMo python module dir:

python -m nemo.collections.visual_reasoning.modules.data_layers.cog.datalayer

If X is available, the test will show some samples using matplotlib.

update repo

…VIDIA-NeMo#693) * update sgd numbers after fix in seen services and slot request loss Signed-off-by: Yang Zhang <yangzhang@nvidia.com> * fix table Signed-off-by: Yang Zhang <yangzhang@nvidia.com> * add more information to documentation Signed-off-by: Yang Zhang <yangzhang@nvidia.com> * fix doc Signed-off-by: Yang Zhang <yangzhang@nvidia.com> * fix doc Signed-off-by: Yang Zhang <yangzhang@nvidia.com> * fix doc Signed-off-by: Yang Zhang <yangzhang@nvidia.com>

* Added user sys tag to TRADE. Signed-off-by: Vahid Noroozi <VahidooX@users.noreply.github.com>

…-NeMo#695) * megatron glue numbers added, default amp level reverted to O0 Signed-off-by: Evelina Bakhturina <ebakhturina@nvidia.com> * table reformatted Signed-off-by: Evelina Bakhturina <ebakhturina@nvidia.com>

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

editdistance package for fast WER calculation

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

…IA-NeMo#673) * add VAD Signed-off-by: fayejf <fayejf07@gmail.com> * update with PR comments Signed-off-by: fayejf <fayejf07@gmail.com> * revert change on asr notebook 4&5 Signed-off-by: fayejf <fayejf07@gmail.com> * update with PR comments Signed-off-by: fayejf <fayejf07@gmail.com> * fix typos Signed-off-by: fayejf <fayejf07@gmail.com> * upload docs and resolve parts of PR comments Signed-off-by: fayejf <fayejf07@gmail.com> * fix doc bib issue Signed-off-by: fayejf <fayejf07@gmail.com> * fix jenksin doc issue Signed-off-by: fayejf <fayejf07@gmail.com> * fix some warning/typo Signed-off-by: fayejf <fayejf07@gmail.com> * update notebook#6, improve data process scripts, and some fix Signed-off-by: fayejf <fayejf07@gmail.com> * some minor changes Signed-off-by: fayejf <fayejf07@gmail.com> * fix bib issue Signed-off-by: fayejf <fayejf07@gmail.com> * little fix to avoid misunderstanding Signed-off-by: fayejf <fayejf07@gmail.com>

* update an4 notebook Signed-off-by: Jason <jasoli@nvidia.com> * colab bugfix Signed-off-by: Jason <jasoli@nvidia.com> * update script Signed-off-by: Jason <jasoli@nvidia.com> * fix notebooks Signed-off-by: Jason <jasoli@nvidia.com> * fix notebooks Signed-off-by: Jason <jasoli@nvidia.com>

* Durations extraction with script draft. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Durations extraction notebooks. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Finished bulk part of durations predictor. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Add tensorboard logging. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Add general-style train logger. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Change LibriSpeech parts order and move train logger callback to core. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Add one big file durs saving. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Big batch params change. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Add full pad option to data loader as default. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Complete durs pipeline with evaluation. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Rename durs ngc script. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Adjust duration main script default for ngc run. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Fix problem with torch.bool dist eval. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Add LibriTTS processing. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Add FasterSpeech full pipeline reaching about 0.4 MSE for LibriTTS. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Add QN retrain NGC pipeline, new dur XE steps loss and mel Griffin-Lim sampling. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Add train logging for mel with audio sampling and super sampler. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Add length sampler. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Set SSS as default and introduce local shuffling. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * New defaults. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * W&B Support, new speaker system and some refactoring Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Add simple durs aug. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * New baseline (1) Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Fix trim bug and make default O2. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Add pad16. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Fix dist eval error and add variable steps. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Add WaveGlow inference and fix pad16 bug. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Generalize mel loss. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Generalize pad op. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Move pad16 logic to loss. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Add fmin/fmax to griffin-lim vocoding. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Move model params to config. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Add denoiser argument to WaveGlow inference. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Add new durs with all 1s by default. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * New baseline (3) Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * New baseline (4) Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * New baseline (5) Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Refactor durs predictor script. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Durs predictor baseline Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Adjusted durs scirpt for NGC. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * New durs lj baseline Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Update durs baseline params. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Add durs/blanks acc metrics. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Fix durs baseline. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Current state Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Update NGC scripts and implement shake_all aug. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Update augmentations implementations. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Bunch of things Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Change name to TalkNet. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Latest notebooks changes Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Working scripts with latest master changes Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Finished trimming durs predictor code. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Trimmed mels part. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Delete dev folder. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Fix style errors. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Fix LGTM errors. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Revert simple logging changes. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Fix problems. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Fix problems. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com> * Remove WG inference and add type hints for data layer. Signed-off-by: Stanislav Beliaev <stasbelyaev96@gmail.com>

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

…s) (NVIDIA-NeMo#675) * git history clean up Signed-off-by: Evelina Bakhturina <ebakhturina@nvidia.com> * nlp references to the tutotials Signed-off-by: Evelina Bakhturina <ebakhturina@nvidia.com> * sphinx fix Signed-off-by: Evelina Bakhturina <ebakhturina@nvidia.com> * review feedback Signed-off-by: Evelina Bakhturina <ebakhturina@nvidia.com>

Signed-off-by: Jocelyn Huang <jocelynh@nvidia.com>

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

Signed-off-by: Jocelyn Huang <jocelynh@nvidia.com>

* initial commit of callback documentation Signed-off-by: Jason <jasoli@nvidia.com> * some syntax fixes Signed-off-by: Jason <jasoli@nvidia.com> * add old callbacks file Signed-off-by: Jason <jasoli@nvidia.com> * finalize docs; change train to action Signed-off-by: Jason <jasoli@nvidia.com> * style Signed-off-by: Jason <jasoli@nvidia.com> * update sphinx style Signed-off-by: Jason <jasoli@nvidia.com> * update sphinx warnings Signed-off-by: Jason <jasoli@nvidia.com> * train->action rename bug Signed-off-by: Jason <jasoli@nvidia.com> * address comments Signed-off-by: Jason <jasoli@nvidia.com> * comments Signed-off-by: Jason <jasoli@nvidia.com>

Update README (pretrained ASR model information)

Bugfix to output ports of Kaldi data layer

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

fixes to the asr model

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

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

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

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

* pm+nlg for multiwoz init Signed-off-by: Evelina Bakhturina <ebakhturina@nvidia.com> * pipeline is working, init clean up Signed-off-by: Evelina Bakhturina <ebakhturina@nvidia.com> * headers added Signed-off-by: Evelina Bakhturina <ebakhturina@nvidia.com> * fixed invalid .json file, added db files to multiwoz preprocessing Signed-off-by: Evelina Bakhturina <ebakhturina@nvidia.com> * code clean up Signed-off-by: Evelina Bakhturina <ebakhturina@nvidia.com> * lgtm fixes Signed-off-by: Evelina Bakhturina <ebakhturina@nvidia.com> * docs for TRADE update, jenkins for ruled_based example Signed-off-by: Evelina Bakhturina <ebakhturina@nvidia.com> * jenkins fix Signed-off-by: Evelina Bakhturina <ebakhturina@nvidia.com> * ports refactor wip Signed-off-by: Evelina Bakhturina <ebakhturina@nvidia.com> * ports refactor wip Signed-off-by: Evelina Bakhturina <ebakhturina@nvidia.com> * wip works Signed-off-by: Evelina Bakhturina <ebakhturina@nvidia.com> * neural types refactored Signed-off-by: Evelina Bakhturina <ebakhturina@nvidia.com> * remove unused Signed-off-by: Evelina Bakhturina <ebakhturina@nvidia.com> * lgtm fixes Signed-off-by: Evelina Bakhturina <ebakhturina@nvidia.com> * typo Signed-off-by: Evelina Bakhturina <ebakhturina@nvidia.com> * state dict splited Signed-off-by: Evelina Bakhturina <ebakhturina@nvidia.com> * lgtm fixes Signed-off-by: Evelina Bakhturina <ebakhturina@nvidia.com> * fixing the process script, moved multiwoz_mapping.pair to multiwoz, enabled utilization of relative paths Signed-off-by: Tomasz Kornuta <tkornuta@nvidia.com> * formatting fix Signed-off-by: Tomasz Kornuta <tkornuta@nvidia.com> * reformatted the code, ready for definition of NG by connecting the modules - and fixing the definitions Signed-off-by: Tomasz Kornuta <tkornuta@nvidia.com> * work in progress-ess, not working, internet issues Signed-off-by: Tomasz Kornuta <tkornuta@nvidia.com> * UtteranceEncoder neural types wip Signed-off-by: nvidia <tkornuta@nvidia.com> * utterance encoder neural types Signed-off-by: nvidia <tkornuta@nvidia.com> * updating trade outputs Signed-off-by: nvidia <tkornuta@nvidia.com> * updating trade outputs Signed-off-by: nvidia <tkornuta@nvidia.com> * fightihg with belief state Signed-off-by: nvidia <tkornuta@nvidia.com> * Cannot make second named tuple work Signed-off-by: nvidia <tkornuta@nvidia.com> * reorganized files, whole pipeline handshaking works Signed-off-by: nvidia <tkornuta@nvidia.com> * reorganized files, whole pipeline handshaking works Signed-off-by: nvidia <tkornuta@nvidia.com> * polish Signed-off-by: nvidia <tkornuta@nvidia.com> * Fix of my dummy error Signed-off-by: nvidia <tkornuta@nvidia.com> * new examples Signed-off-by: nvidia <tkornuta@nvidia.com> * style fix Signed-off-by: Evelina Bakhturina <ebakhturina@nvidia.com> * fixed TRADE training Signed-off-by: Evelina Bakhturina <ebakhturina@nvidia.com> * Added module responsible for sys uttr dialog history update Signed-off-by: nvidia <tkornuta@nvidia.com> * LGTM fix Signed-off-by: nvidia <tkornuta@nvidia.com> * moved dialog specific axesc andctypes to nlp/neural_types.py, refactored the modules Signed-off-by: nvidia <tkornuta@nvidia.com> * style fix Signed-off-by: nvidia <tkornuta@nvidia.com> Co-authored-by: Tomasz Kornuta <tkornuta@nvidia.com>

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

Callchain fix and update logging

* make test better Signed-off-by: Jason <jasoli@nvidia.com> * fix rename error during topological sort Signed-off-by: Jason <jasoli@nvidia.com> * test fix Signed-off-by: Jason <jasoli@nvidia.com>

aasseman · 2020-06-11T00:54:42Z

FYI, I just rebased to a more recent upstream master.

Fixed 2_Online_ASR_Microphone_Demo notebook to support new config

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

…VIDIA-NeMo#724) * Added ability to write audio to tensorboard during Tacotron training Signed-off-by: Polezhaev Sergej <sapolezh@mts.ru> * Removed unused import Signed-off-by: Polezhaev Sergej <sapolezh@mts.ru> Co-authored-by: Sergey Polezhaev <sapolezh@mts.ru>

Signed-off-by: Alexis Asseman <alexis.asseman@ibm.com>

…rent subdirs, as they are not datalayers. Signed-off-by: Alexis Asseman <alexis.asseman@ibm.com>

Signed-off-by: Alexis Asseman <alexis.asseman@ibm.com>

* Integrated Megatron-LM Signed-off-by: Boris Fomitchev <bfomitchev@nvidia.com> * Addressing PR comments, trying NER example Signed-off-by: Boris Fomitchev <bfomitchev@nvidia.com> * manual style fix Signed-off-by: Boris Fomitchev <bfomitchev@nvidia.com> * manual style fix #2 Signed-off-by: Boris Fomitchev <bfomitchev@nvidia.com> * Resolving circular import Signed-off-by: Boris Fomitchev <bfomitchev@nvidia.com> * Static analysys warnings addressed Signed-off-by: Boris Fomitchev <bfomitchev@nvidia.com> * Addressed code review; Jenkins test added Signed-off-by: Boris Fomitchev <bfomitchev@nvidia.com> * Removing parallel feom Megatron Signed-off-by: Boris Fomitchev <bfomitchev@nvidia.com> * Added more info to tokenizer printout, made megatron bert derivative explicit Signed-off-by: Boris Fomitchev <bfomitchev@nvidia.com> * Bumping Megatron-LM version to get APEX fix Signed-off-by: Boris Fomitchev <bfomitchev@nvidia.com>

gkucsko and others added 30 commits June 2, 2020 19:46

Merge pull request NVIDIA-NeMo#1 from NVIDIA/master

79bc099

update repo

Added editdistance package for ASR for fast WER calculation

f9ebb4c

Update the improved state tracker for SGD model (NVIDIA-NeMo#686)

1273d8f

* Added user sys tag to TRADE. Signed-off-by: Vahid Noroozi <VahidooX@users.noreply.github.com>

add links for biobert/biomegatron to nlp documentation (NVIDIA-NeMo#696)

f6f8b47

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

Merge pull request NVIDIA-NeMo#692 from gkucsko/c_lev_package

9a1366c

editdistance package for fast WER calculation

bugfix in asrconvctcmodel

0fbb6f1

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

fixes to the asr model

2efe793

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

Bugfix to output ports of Kaldi data layer

8332d9d

Signed-off-by: Jocelyn Huang <jocelynh@nvidia.com>

bumping version to 0.11.0b3

b183f38

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

mem fix

7fa0df0

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

Update README (pretrained ASR model information)

fb3e233

Signed-off-by: Jocelyn Huang <jocelynh@nvidia.com>

Merge pull request NVIDIA-NeMo#705 from redoctopus/readme_table_update

0b24b1a

Update README (pretrained ASR model information)

Merge pull request NVIDIA-NeMo#704 from redoctopus/kaldi_layer_bugfix

5835bdb

Bugfix to output ports of Kaldi data layer

address review feedback

42855ce

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

Merge pull request NVIDIA-NeMo#703 from NVIDIA/asr_model_fix

42ab7ac

fixes to the asr model

bumping version to 0.11.0b4

b7bfb30

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

revert the previous default value (NVIDIA-NeMo#707)

09c38c9

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

fix biobert qa numbers (NVIDIA-NeMo#708)

cba71d7

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

callchain fix and update logging

a1b7f92

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

add clear state dict after Nan Inf

d115382

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

Merge pull request NVIDIA-NeMo#710 from blisc/mismatched_callchain_fix

76d8385

Callchain fix and update logging

Nmtensor rename bug fix (NVIDIA-NeMo#709)

b4a347d

* make test better Signed-off-by: Jason <jasoli@nvidia.com> * fix rename error during topological sort Signed-off-by: Jason <jasoli@nvidia.com> * test fix Signed-off-by: Jason <jasoli@nvidia.com>

okuchaiev and others added 23 commits June 11, 2020 13:03

Merge pull request NVIDIA-NeMo#723 from NVIDIA/online-asr-nb-fix

fdc712b

Fixed 2_Online_ASR_Microphone_Demo notebook to support new config

bumping version to 0.11.0b6

5e39bcc

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

Copy COG code from https://github.com/IBM/mi-prometheus (Apache-2.0)

fe5b3fe

Signed-off-by: Alexis Asseman <alexis.asseman@ibm.com>

Style fix

b292fa9

Signed-off-by: Alexis Asseman <alexis.asseman@ibm.com>

Fix __init__.py's, import paths

7ed7cbc

Signed-off-by: Alexis Asseman <alexis.asseman@ibm.com>

cog.py: cleanup redundancy, imports, some python modernization

2d9dc98

Signed-off-by: Alexis Asseman <alexis.asseman@ibm.com>

Removed mpi deps ; more cleaning ; work just enough to run test

462947d

Signed-off-by: Alexis Asseman <alexis.asseman@ibm.com>

COG constants moved to constants.py

fd629bf

Signed-off-by: Alexis Asseman <alexis.asseman@ibm.com>

Fix image RGB scaling

cac9189

Signed-off-by: Alexis Asseman <alexis.asseman@ibm.com>

Moved from MIP's DataDict system to regular Torch dataset system

11bb150

Signed-off-by: Alexis Asseman <alexis.asseman@ibm.com>

Improved COG's show_sample

5cdd0f8

Signed-off-by: Alexis Asseman <alexis.asseman@ibm.com>

Moved COG's loss to LossNM, made COG's stat functions static methods

9f719d9

Signed-off-by: Alexis Asseman <alexis.asseman@ibm.com>

Moved COG stats to separate file

ce03bf8

Signed-off-by: Alexis Asseman <alexis.asseman@ibm.com>

Make COG Dataset a valid DataLayerNM

982f087

Signed-off-by: Alexis Asseman <alexis.asseman@ibm.com>

Cleanup of style, comments

5eabc52

Signed-off-by: Alexis Asseman <alexis.asseman@ibm.com>

Remove unnecessary cog_utils subdir. Move loss and stats into diffe…

ae5d505

…rent subdirs, as they are not datalayers. Signed-off-by: Alexis Asseman <alexis.asseman@ibm.com>

Fix import paths after reshuffling the COG files

2877a34

Signed-off-by: Alexis Asseman <alexis.asseman@ibm.com>

Adjusted the __init__.py's

6acb5ff

Signed-off-by: Alexis Asseman <alexis.asseman@ibm.com>

Added warning comment about error in original COG code

1eb04bd

Signed-off-by: Alexis Asseman <alexis.asseman@ibm.com>

Using NeMo's Tokenizer in COG Datalayer

932d104

Signed-off-by: Alexis Asseman <alexis.asseman@ibm.com>

Various fixes in COG datalayer

36a0d7e

Signed-off-by: Alexis Asseman <alexis.asseman@ibm.com>

COG datalayer multiprocess loading

1d4de8f

Signed-off-by: Alexis Asseman <alexis.asseman@ibm.com>

aasseman force-pushed the feat/cog branch from 92e019f to 4e39d89 Compare June 16, 2020 18:24

Alexis Asseman added 3 commits June 16, 2020 11:26

New COG stats code. Minor fixes to COG.

4e39d89

Signed-off-by: Alexis Asseman <alexis.asseman@ibm.com>

COG Stats simplification

fcbadfd

Signed-off-by: Alexis Asseman <alexis.asseman@ibm.com>

Cleanup

fd286ee

aasseman marked this pull request as ready for review June 19, 2020 17:08

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

COG Datalyer#2

COG Datalyer#2
aasseman wants to merge 70 commits intotkornuta-nvidia:masterfrom
aasseman:feat/cog

aasseman commented Jun 2, 2020 •

edited

Loading

Uh oh!

aasseman commented Jun 11, 2020

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

13 participants

Conversation

aasseman commented Jun 2, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

aasseman commented Jun 11, 2020

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

13 participants

aasseman commented Jun 2, 2020 •

edited

Loading