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

Sync vfm branch with main branch #11288

Merged
merged 24 commits into from
Nov 14, 2024
Merged

Sync vfm branch with main branch #11288

merged 24 commits into from
Nov 14, 2024

Conversation

zpx01
Copy link
Collaborator

@zpx01 zpx01 commented Nov 14, 2024

What does this PR do ?

Add a one line overview of what this PR aims to accomplish.

Collection: [Note which collection this PR will affect]

Changelog

  • Add specific line by line info of high level changes in this PR.

Usage

  • You can potentially add a usage example below
# Add a code snippet demonstrating how to use this 

GitHub Actions CI

The Jenkins CI system has been replaced by GitHub Actions self-hosted runners.

The GitHub Actions CI will run automatically when the "Run CICD" label is added to the PR.
To re-run CI remove and add the label again.
To run CI on an untrusted fork, a NeMo user with write access must first click "Approve and run".

Before your PR is "Ready for review"

Pre checks:

  • Make sure you read and followed Contributor guidelines
  • Did you write any new necessary tests?
  • Did you add or update any necessary documentation?
  • Does the PR affect components that are optional to install? (Ex: Numba, Pynini, Apex etc)
    • Reviewer: Does the PR have correct import guards for all optional libraries?

PR Type:

  • New Feature
  • Bugfix
  • Documentation

If you haven't finished some of the above items you can still open "Draft" PR.

Who can review?

Anyone in the community is free to review the PR once the checks have passed.
Contributor guidelines contains specific people who can review PRs to various areas.

Additional Information

  • Related to # (issue)

nithinraok and others added 24 commits November 10, 2024 15:01
* inital version

Signed-off-by: Nithin Rao Koluguri <nithinraok>

* Support for RNNT, TDT, Hybrid Models

Signed-off-by: Nithin Rao Koluguri <nithinraok>

* move change of decoder stratery from mixin to individual model class

Signed-off-by: Nithin Rao Koluguri <nithinraok>

* Apply isort and black reformatting

Signed-off-by: nithinraok <[email protected]>

* update transcribe_speech.py

Signed-off-by: Nithin Rao Koluguri <nithinraok>

* uncomment

Signed-off-by: Nithin Rao Koluguri <nithinraok>

* Apply isort and black reformatting

Signed-off-by: nithinraok <[email protected]>

* add docs

Signed-off-by: Nithin Rao Koluguri <nithinraok>

* fix docs

Signed-off-by: Nithin Rao Koluguri <nithinraok>

* Apply isort and black reformatting

Signed-off-by: nithinraok <[email protected]>

* codeql fixes

Signed-off-by: Nithin Rao Koluguri <nithinraok>

* unit tests

Signed-off-by: Nithin Rao Koluguri <nithinraok>

* minor rebase fix

Signed-off-by: Nithin Rao Koluguri <nithinraok>

* Apply isort and black reformatting

Signed-off-by: nithinraok <[email protected]>

* add None case to restore the state set outside using decoding_stratergy()

Signed-off-by: Nithin Rao Koluguri <nithinraok>

* Apply isort and black reformatting

Signed-off-by: nithinraok <[email protected]>

* remove ipdb traces

Signed-off-by: Nithin Rao Koluguri <nithinraok>

* updates doc for transcription.py

Signed-off-by: Nithin Rao Koluguri <nithinraok>

* remove preserve alignment for AED models as it doesn;t support it

Signed-off-by: Nithin Rao Koluguri <nithinraok>

* lint warnings

Signed-off-by: Nithin Rao Koluguri <nithinraok>

* Apply isort and black reformatting

Signed-off-by: nithinraok <[email protected]>

---------

Signed-off-by: Nithin Rao Koluguri <nithinraok>
Signed-off-by: nithinraok <[email protected]>
Co-authored-by: Nithin Rao Koluguri <nithinraok>
Co-authored-by: nithinraok <[email protected]>
Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
* Handling tokenizer in PTQ for Nemo 2.0

Signed-off-by: Jan Lasek <[email protected]>

* Print log msg and enable overriding

Signed-off-by: Jan Lasek <[email protected]>

* Warning for legacy tokenizer config

Signed-off-by: Jan Lasek <[email protected]>

* Save HF tokenizer to make tokenizer_config.yaml (almost) redundant

Signed-off-by: Jan Lasek <[email protected]>

* Handle tokenizer in a unified way

Signed-off-by: Jan Lasek <[email protected]>

* Move saving context within export

Signed-off-by: Jan Lasek <[email protected]>

* Fix typo in get_tokenzier

Signed-off-by: Jan Lasek <[email protected]>

* Reduce diff

Signed-off-by: Jan Lasek <[email protected]>

* Drop unused import

Signed-off-by: Jan Lasek <[email protected]>

---------

Signed-off-by: Jan Lasek <[email protected]>
* fix datamodule resume

Signed-off-by: Chen Cui <[email protected]>

* Apply isort and black reformatting

Signed-off-by: cuichenx <[email protected]>

* fix subclass

Signed-off-by: Chen Cui <[email protected]>

* docstrings and formats

Signed-off-by: Chen Cui <[email protected]>

* Apply isort and black reformatting

Signed-off-by: cuichenx <[email protected]>

---------

Signed-off-by: Chen Cui <[email protected]>
Signed-off-by: cuichenx <[email protected]>
Co-authored-by: cuichenx <[email protected]>
* ci: Move `bump mcore` to templates

Signed-off-by: Oliver Koenig <[email protected]>

* fix

Signed-off-by: Oliver Koenig <[email protected]>

* fix

Signed-off-by: Oliver Koenig <[email protected]>

* fix

Signed-off-by: Oliver Koenig <[email protected]>

* final

Signed-off-by: Oliver Koenig <[email protected]>

---------

Signed-off-by: Oliver Koenig <[email protected]>
Signed-off-by: Oliver Koenig <[email protected]>
Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
…11045)

* rename eval_beamsearch_ngram.py to eval_beamsearch_ngram_ctc.py in docs

Signed-off-by: Elena Rastorgueva <[email protected]>

* replace out of date installation instructions with pointer to NeMo README installation section

Signed-off-by: Elena Rastorgueva <[email protected]>

* point to user guide instead of readme

Signed-off-by: Elena Rastorgueva <[email protected]>

* some link updates

Signed-off-by: Elena Rastorgueva <[email protected]>

* update more links

Signed-off-by: Elena Rastorgueva <[email protected]>

---------

Signed-off-by: Elena Rastorgueva <[email protected]>
Signed-off-by: Elena Rastorgueva <[email protected]>
…or Aligner (#10863)

* fix(export): update API for disabling device reassignment in TRTLLM for Aligner

[feat] Upgrade nemo-export path for aligner to TRTLLM-v12 and use python runtime

Signed-off-by: Terry Kong <[email protected]>

fix: forgot to always set _disable_torch_cuda_device_set

Signed-off-by: Terry Kong <[email protected]>

Signed-off-by: Terry Kong <[email protected]>

Apply isort and black reformatting

Signed-off-by: terrykong <[email protected]>

invert torch device set

Signed-off-by: Terry Kong <[email protected]>

* remove comment

Signed-off-by: Terry Kong <[email protected]>

---------

Signed-off-by: Terry Kong <[email protected]>
Signed-off-by: Zeeshan Patel <[email protected]>
Co-authored-by: Zeeshan Patel <[email protected]>
* Update pruning and distillation tutorial notebooks

Signed-off-by: Gomathy Venkata Krishnan <[email protected]>

* Update README

Signed-off-by: Gomathy Venkata Krishnan <[email protected]>

* Update batch size in width pruning script

Signed-off-by: Gomathy Venkata Krishnan <[email protected]>

* Update README

Signed-off-by: Gomathy Venkata Krishnan <[email protected]>

---------

Signed-off-by: Gomathy Venkata Krishnan <[email protected]>
* initial commit

Signed-off-by: lilithgrigoryan <[email protected]>

* add: default beam search implementation

Signed-off-by: lilithgrigoryan <[email protected]>

* fix: changed to removing duplicate hypothesis in separate function

Signed-off-by: lilithgrigoryan <[email protected]>

* fix: changed to cartesian product in choosing best hyp

Signed-off-by: lilithgrigoryan <[email protected]>

* fix: minor fixes in comments

Signed-off-by: lilithgrigoryan <[email protected]>

* add: maes decoding strategy

Signed-off-by: lilithgrigoryan <[email protected]>

* add: durations filtering in maes, lm fusion in progress

Signed-off-by: lilithgrigoryan <[email protected]>

* fix: refactored, added comments, command line args, finalized

Signed-off-by: lilithgrigoryan <[email protected]>

* fix: removed prints

Signed-off-by: lilithgrigoryan <[email protected]>

* add: docs

Signed-off-by: lilithgrigoryan <[email protected]>

* Apply isort and black reformatting

Signed-off-by: lilithgrigoryan <[email protected]>

* fix: minor fix

Signed-off-by: lilithgrigoryan <[email protected]>

* fix: rm beam_size=1 exception, rm duplicates check, fix error handling

Signed-off-by: lilithgrigoryan <[email protected]>

* fix: error handling

Signed-off-by: lilithgrigoryan <[email protected]>

* Apply isort and black reformatting

Signed-off-by: lilithgrigoryan <[email protected]>

* fix: removed evaluations file

Signed-off-by: lilithgrigoryan <[email protected]>

* rn: blank scoring

Signed-off-by: lilithgrigoryan <[email protected]>

* clean up

Signed-off-by: lilithgrigoryan <[email protected]>

* rm: blank scoring and duration beam size

Signed-off-by: lilithgrigoryan <[email protected]>

* Apply isort and black reformatting

Signed-off-by: lilithgrigoryan <[email protected]>

* fix: removed durations_beam_size from default beam search

Signed-off-by: lilithgrigoryan <[email protected]>

* add: logaddexp

Signed-off-by: lilithgrigoryan <[email protected]>

* rm: prefix search

Signed-off-by: lilithgrigoryan <[email protected]>

* rn: nested loop over extensions

Signed-off-by: lilithgrigoryan <[email protected]>

* fix: bug with caching

Signed-off-by: lilithgrigoryan <[email protected]>

* rm: topk on durations

Signed-off-by: lilithgrigoryan <[email protected]>

* add: restored prefix search

Signed-off-by: lilithgrigoryan <[email protected]>

* Apply isort and black reformatting

Signed-off-by: lilithgrigoryan <[email protected]>

* clean up

Signed-off-by: lilithgrigoryan <[email protected]>

* fix: fixed comments

Signed-off-by: lilithgrigoryan <[email protected]>

* refactored duplicate merging

Signed-off-by: lilithgrigoryan <[email protected]>

* changes batch scoring

Signed-off-by: lilithgrigoryan <[email protected]>

* refactored rnnt batch scoring

Signed-off-by: lilithgrigoryan <[email protected]>

* alsd first working

Signed-off-by: lilithgrigoryan <[email protected]>

* refactored

Signed-off-by: lilithgrigoryan <[email protected]>

* clean up

Signed-off-by: lilithgrigoryan <[email protected]>

* remove stacking operations

Signed-off-by: lilithgrigoryan <[email protected]>

* fixes im base class

Signed-off-by: lilithgrigoryan <[email protected]>

* clean up

Signed-off-by: lilithgrigoryan <[email protected]>

* Apply isort and black reformatting

Signed-off-by: lilithgrigoryan <[email protected]>

* remove potentially uninitialized local variable

Signed-off-by: lilithgrigoryan <[email protected]>

* default beam search minor fixes

Signed-off-by: lilithgrigoryan <[email protected]>

* add test, fix maes timesteps

Signed-off-by: lilithgrigoryan <[email protected]>

* rm file

Signed-off-by: lilithgrigoryan <[email protected]>

* rm file

Signed-off-by: lilithgrigoryan <[email protected]>

* clean up

Signed-off-by: lilithgrigoryan <[email protected]>

* Apply isort and black reformatting

Signed-off-by: lilithgrigoryan <[email protected]>

* clean up

Signed-off-by: lilithgrigoryan <[email protected]>

* fix comments

Signed-off-by: lilithgrigoryan <[email protected]>

* add ngram lm test

Signed-off-by: lilithgrigoryan <[email protected]>

* Apply isort and black reformatting

Signed-off-by: lilithgrigoryan <[email protected]>

* fix maes_num_steps=1

Signed-off-by: lilithgrigoryan <[email protected]>

* fix kenlm model path

Signed-off-by: lilithgrigoryan <[email protected]>

* fix kenlm model full path

Signed-off-by: lilithgrigoryan <[email protected]>

* Apply isort and black reformatting

Signed-off-by: lilithgrigoryan <[email protected]>

* made requested changes

Signed-off-by: lilithgrigoryan <[email protected]>

* merge after isort

Signed-off-by: lilithgrigoryan <[email protected]>

* add prints to test

Signed-off-by: lilithgrigoryan <[email protected]>

* Apply isort and black reformatting

Signed-off-by: lilithgrigoryan <[email protected]>

* add Kenlm to asr requirements

Signed-off-by: lilithgrigoryan <[email protected]>

* remove prints in tests

Signed-off-by: lilithgrigoryan <[email protected]>

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* add kenlm to test requirements

Signed-off-by: lilithgrigoryan <[email protected]>

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* rm kenlm from link, add package-name

Signed-off-by: lilithgrigoryan <[email protected]>

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* rm second kenlm installation

Signed-off-by: lilithgrigoryan <[email protected]>

* rm kenlm from dependencies make test optional

Signed-off-by: lilithgrigoryan <[email protected]>

* Apply isort and black reformatting

Signed-off-by: lilithgrigoryan <[email protected]>

* fix in test

Signed-off-by: lilithgrigoryan <[email protected]>

* fix in test

Signed-off-by: lilithgrigoryan <[email protected]>

* Apply isort and black reformatting

Signed-off-by: lilithgrigoryan <[email protected]>

* fix comments

Signed-off-by: lilithgrigoryan <[email protected]>

* Apply isort and black reformatting

Signed-off-by: lilithgrigoryan <[email protected]>

* add comments

Signed-off-by: lilithgrigoryan <[email protected]>

* add comments

Signed-off-by: lilithgrigoryan <[email protected]>

* splitted docstrings

Signed-off-by: lilithgrigoryan <[email protected]>

* Apply isort and black reformatting

Signed-off-by: lilithgrigoryan <[email protected]>

* add comments

Signed-off-by: lilithgrigoryan <[email protected]>

* splitted docstrings

Signed-off-by: lilithgrigoryan <[email protected]>

* Apply isort and black reformatting

Signed-off-by: lilithgrigoryan <[email protected]>

* add comments

Signed-off-by: lilithgrigoryan <[email protected]>

* Apply isort and black reformatting

Signed-off-by: lilithgrigoryan <[email protected]>

* fixes to python3 type annotations

Signed-off-by: lilithgrigoryan <[email protected]>

* Apply isort and black reformatting

Signed-off-by: lilithgrigoryan <[email protected]>

* merging

Signed-off-by: lilithgrigoryan <[email protected]>

* merging

Signed-off-by: lilithgrigoryan <[email protected]>

* fix in return type

Signed-off-by: lilithgrigoryan <[email protected]>

* Apply isort and black reformatting

Signed-off-by: lilithgrigoryan <[email protected]>

* fix test

Signed-off-by: lilithgrigoryan <[email protected]>

* Apply isort and black reformatting

Signed-off-by: lilithgrigoryan <[email protected]>

* rm time_idx

Signed-off-by: lilithgrigoryan <[email protected]>

* fix comments to python3 style

Signed-off-by: lilithgrigoryan <[email protected]>

---------

Signed-off-by: lilithgrigoryan <[email protected]>
Signed-off-by: lilithgrigoryan <[email protected]>
Co-authored-by: lilithgrigoryan <[email protected]>
Co-authored-by: lilithgrigoryan <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* update nemo1->2 conversion according to changes in main

Signed-off-by: Huiying Li <[email protected]>

* Apply isort and black reformatting

Signed-off-by: HuiyingLi <[email protected]>

* format fix

Signed-off-by: Huiying Li <[email protected]>

* add docstrings

Signed-off-by: Huiying Li <[email protected]>

---------

Signed-off-by: Huiying Li <[email protected]>
Signed-off-by: HuiyingLi <[email protected]>
Co-authored-by: HuiyingLi <[email protected]>
* add llama 3.1 recipes

Signed-off-by: Chen Cui <[email protected]>

* Apply isort and black reformatting

Signed-off-by: cuichenx <[email protected]>

* fix pylint

Signed-off-by: Chen Cui <[email protected]>

* Fix llama3.1 wrong config in io.json

---------

Signed-off-by: Chen Cui <[email protected]>
Signed-off-by: cuichenx <[email protected]>
Co-authored-by: cuichenx <[email protected]>
Co-authored-by: Ao Tang <[email protected]>
* Fix Starcoder_15 SFT recipe

* Fix PP type SFT recipe

* Fix PP type SFT recipe

* Fix Gemma2b SFT TP=1

* Fix more sft recipe

* Fix more sft recipe

* Fix more sft recipe

* Fix more sft recipe

* Fix more sft recipe

* Fix more sft recipe

* Fix more sft recipe

* Fix more sft recipe

* Fix more sft recipe

* remove pp dtype

* remove pp dtype
* Configure no restart validation loop in nl.Trainer

Signed-off-by: Hemil Desai <[email protected]>

* fix

Signed-off-by: Hemil Desai <[email protected]>

* Skip validation whenever restarting=True

Signed-off-by: Hemil Desai <[email protected]>

* PR feedback

Signed-off-by: Hemil Desai <[email protected]>

* Apply isort and black reformatting

Signed-off-by: hemildesai <[email protected]>

---------

Signed-off-by: Hemil Desai <[email protected]>
Signed-off-by: hemildesai <[email protected]>
Co-authored-by: hemildesai <[email protected]>
* Use MegatronDataSampler in HfDataset

Signed-off-by: Alexandros Koumparoulis <[email protected]>

* Apply isort and black reformatting

Signed-off-by: akoumpa <[email protected]>

---------

Signed-off-by: Alexandros Koumparoulis <[email protected]>
Signed-off-by: akoumpa <[email protected]>
Co-authored-by: akoumpa <[email protected]>
Signed-off-by: Dong Hyuk Chang <[email protected]>

video_latent = torch.ones(self.seq_length, c * p**2, dtype=torch.bfloat16) * 0.5
text_embedding = torch.randn(self.text_seqlen, self.text_dim, dtype=torch.bfloat16)
pos_emb = pos_id_3d.get_pos_id_3d(t=t, h=h // p, w=w // p).reshape(-1, 3)

Check notice

Code scanning / CodeQL

Unused local variable Note

Variable pos_emb is not used.
Comment on lines +540 to +546
torch.cuda.memory._record_memory_history(
True,
# Keep 100,000 alloc/free events from before the snapshot
trace_alloc_max_entries=100000,
# Record stack information for the trace events
trace_alloc_record_context=True,
)

Check warning

Code scanning / CodeQL

Unreachable code Warning

This statement is unreachable.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.