Datasets:

renwei2024
/

nemo-github-issues

Dataset card Data Studio Files Files and versions Community

Split (1)

train · 10k rows

url stringlengths 52 53	repository_url stringclasses 1 value	labels_url stringlengths 66 67	comments_url stringlengths 61 62	events_url stringlengths 59 60	html_url stringlengths 40 43	id int64 719M 2.8B	node_id stringlengths 18 32	number int64 1.28k 11.9k	title stringlengths 1 350	user dict	labels listlengths 0 10	state stringclasses 2 values	locked bool 2 classes	assignee dict	assignees listlengths 0 3	milestone null	comments sequencelengths 0 30 ⌀	created_at timestamp[s]	updated_at timestamp[s]	closed_at timestamp[s]	author_association stringclasses 4 values	sub_issues_summary dict	active_lock_reason stringclasses 1 value	draft bool 2 classes	pull_request dict	body stringlengths 0 73.4k ⌀	closed_by dict	reactions dict	timeline_url stringlengths 61 62	performed_via_github_app null	state_reason stringclasses 3 values	is_pull_request bool 2 classes
https://api.github.com/repos/NVIDIA/NeMo/issues/7196	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7196/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7196/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7196/events	https://github.com/NVIDIA/NeMo/pull/7196	1,843,924,402	PR_kwDOC_bI7s5Xkooe	7,196	Fix T5 using FA	{ "login": "hsiehjackson", "id": 37269846, "node_id": "MDQ6VXNlcjM3MjY5ODQ2", "avatar_url": "https://avatars.githubusercontent.com/u/37269846?v=4", "gravatar_id": "", "url": "https://api.github.com/users/hsiehjackson", "html_url": "https://github.com/hsiehjackson", "followers_url": "https://api.github.com/users/hsiehjackson/followers", "following_url": "https://api.github.com/users/hsiehjackson/following{/other_user}", "gists_url": "https://api.github.com/users/hsiehjackson/gists{/gist_id}", "starred_url": "https://api.github.com/users/hsiehjackson/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/hsiehjackson/subscriptions", "organizations_url": "https://api.github.com/users/hsiehjackson/orgs", "repos_url": "https://api.github.com/users/hsiehjackson/repos", "events_url": "https://api.github.com/users/hsiehjackson/events{/privacy}", "received_events_url": "https://api.github.com/users/hsiehjackson/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-08-09T20:02:08	2023-08-10T19:12:55	2023-08-10T19:12:54	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7196", "html_url": "https://github.com/NVIDIA/NeMo/pull/7196", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7196.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7196.patch", "merged_at": "2023-08-10T19:12:54" }	# What does this PR do ? * Fix T5 mask when using FA * Add additional configs 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 ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [x] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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 - [x] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "hsiehjackson", "id": 37269846, "node_id": "MDQ6VXNlcjM3MjY5ODQ2", "avatar_url": "https://avatars.githubusercontent.com/u/37269846?v=4", "gravatar_id": "", "url": "https://api.github.com/users/hsiehjackson", "html_url": "https://github.com/hsiehjackson", "followers_url": "https://api.github.com/users/hsiehjackson/followers", "following_url": "https://api.github.com/users/hsiehjackson/following{/other_user}", "gists_url": "https://api.github.com/users/hsiehjackson/gists{/gist_id}", "starred_url": "https://api.github.com/users/hsiehjackson/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/hsiehjackson/subscriptions", "organizations_url": "https://api.github.com/users/hsiehjackson/orgs", "repos_url": "https://api.github.com/users/hsiehjackson/repos", "events_url": "https://api.github.com/users/hsiehjackson/events{/privacy}", "received_events_url": "https://api.github.com/users/hsiehjackson/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7196/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7196/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7195	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7195/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7195/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7195/events	https://github.com/NVIDIA/NeMo/pull/7195	1,843,732,058	PR_kwDOC_bI7s5Xj_XJ	7,195	minor fix for conformer subsampling docstring.	{ "login": "XuesongYang", "id": 1646669, "node_id": "MDQ6VXNlcjE2NDY2Njk=", "avatar_url": "https://avatars.githubusercontent.com/u/1646669?v=4", "gravatar_id": "", "url": "https://api.github.com/users/XuesongYang", "html_url": "https://github.com/XuesongYang", "followers_url": "https://api.github.com/users/XuesongYang/followers", "following_url": "https://api.github.com/users/XuesongYang/following{/other_user}", "gists_url": "https://api.github.com/users/XuesongYang/gists{/gist_id}", "starred_url": "https://api.github.com/users/XuesongYang/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/XuesongYang/subscriptions", "organizations_url": "https://api.github.com/users/XuesongYang/orgs", "repos_url": "https://api.github.com/users/XuesongYang/repos", "events_url": "https://api.github.com/users/XuesongYang/events{/privacy}", "received_events_url": "https://api.github.com/users/XuesongYang/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4811884691, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR", "name": "ASR", "color": "F3B33E", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-08-09T17:40:56	2023-08-09T19:48:31	2023-08-09T19:48:30	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7195", "html_url": "https://github.com/NVIDIA/NeMo/pull/7195", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7195.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7195.patch", "merged_at": "2023-08-09T19:48:30" }	# 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 ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [x] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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 - [x] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "XuesongYang", "id": 1646669, "node_id": "MDQ6VXNlcjE2NDY2Njk=", "avatar_url": "https://avatars.githubusercontent.com/u/1646669?v=4", "gravatar_id": "", "url": "https://api.github.com/users/XuesongYang", "html_url": "https://github.com/XuesongYang", "followers_url": "https://api.github.com/users/XuesongYang/followers", "following_url": "https://api.github.com/users/XuesongYang/following{/other_user}", "gists_url": "https://api.github.com/users/XuesongYang/gists{/gist_id}", "starred_url": "https://api.github.com/users/XuesongYang/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/XuesongYang/subscriptions", "organizations_url": "https://api.github.com/users/XuesongYang/orgs", "repos_url": "https://api.github.com/users/XuesongYang/repos", "events_url": "https://api.github.com/users/XuesongYang/events{/privacy}", "received_events_url": "https://api.github.com/users/XuesongYang/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7195/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7195/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7193	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7193/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7193/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7193/events	https://github.com/NVIDIA/NeMo/pull/7193	1,843,665,141	PR_kwDOC_bI7s5Xjw5G	7,193	deprecation warning	{ "login": "arendu", "id": 108822655, "node_id": "U_kgDOBnyAfw", "avatar_url": "https://avatars.githubusercontent.com/u/108822655?v=4", "gravatar_id": "", "url": "https://api.github.com/users/arendu", "html_url": "https://github.com/arendu", "followers_url": "https://api.github.com/users/arendu/followers", "following_url": "https://api.github.com/users/arendu/following{/other_user}", "gists_url": "https://api.github.com/users/arendu/gists{/gist_id}", "starred_url": "https://api.github.com/users/arendu/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/arendu/subscriptions", "organizations_url": "https://api.github.com/users/arendu/orgs", "repos_url": "https://api.github.com/users/arendu/repos", "events_url": "https://api.github.com/users/arendu/events{/privacy}", "received_events_url": "https://api.github.com/users/arendu/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-08-09T16:55:21	2023-08-14T18:09:43	2023-08-14T18:09:42	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7193", "html_url": "https://github.com/NVIDIA/NeMo/pull/7193", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7193.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7193.patch", "merged_at": "2023-08-14T18:09:42" }	# What does this PR do ? adds a warning for multiple peft scripts. Collection: [NLP] # Changelog - Add specific line by line info of high level changes in this PR. # Usage * You can potentially add a usage example below ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "arendu", "id": 108822655, "node_id": "U_kgDOBnyAfw", "avatar_url": "https://avatars.githubusercontent.com/u/108822655?v=4", "gravatar_id": "", "url": "https://api.github.com/users/arendu", "html_url": "https://github.com/arendu", "followers_url": "https://api.github.com/users/arendu/followers", "following_url": "https://api.github.com/users/arendu/following{/other_user}", "gists_url": "https://api.github.com/users/arendu/gists{/gist_id}", "starred_url": "https://api.github.com/users/arendu/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/arendu/subscriptions", "organizations_url": "https://api.github.com/users/arendu/orgs", "repos_url": "https://api.github.com/users/arendu/repos", "events_url": "https://api.github.com/users/arendu/events{/privacy}", "received_events_url": "https://api.github.com/users/arendu/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7193/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7193/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7192	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7192/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7192/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7192/events	https://github.com/NVIDIA/NeMo/issues/7192	1,843,431,044	I_kwDOC_bI7s5t4IaE	7,192	List index out of range when transcribing with RNNT conformer large	{ "login": "robertveenhoven", "id": 10922796, "node_id": "MDQ6VXNlcjEwOTIyNzk2", "avatar_url": "https://avatars.githubusercontent.com/u/10922796?v=4", "gravatar_id": "", "url": "https://api.github.com/users/robertveenhoven", "html_url": "https://github.com/robertveenhoven", "followers_url": "https://api.github.com/users/robertveenhoven/followers", "following_url": "https://api.github.com/users/robertveenhoven/following{/other_user}", "gists_url": "https://api.github.com/users/robertveenhoven/gists{/gist_id}", "starred_url": "https://api.github.com/users/robertveenhoven/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/robertveenhoven/subscriptions", "organizations_url": "https://api.github.com/users/robertveenhoven/orgs", "repos_url": "https://api.github.com/users/robertveenhoven/repos", "events_url": "https://api.github.com/users/robertveenhoven/events{/privacy}", "received_events_url": "https://api.github.com/users/robertveenhoven/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 1485815557, "node_id": "MDU6TGFiZWwxNDg1ODE1NTU3", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/bug", "name": "bug", "color": "d73a4a", "default": true, "description": "Something isn't working" } ]	closed	false	null	[]	null	null	2023-08-09T14:46:24	2023-09-20T18:27:11	2023-09-19T15:53:16	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	Describe the bug A clear and concise description of what the bug is. Hello, we are running a Dutch trained version of the conformer RNNT large model. However, in a small amount of cases we receive the following list index out of range error. ![image](https://github.com/NVIDIA/NeMo/assets/10922796/5341394d-a158-43d5-a724-52261052f0bb) I saw that in [https://github.com/NVIDIA/NeMo/issues/4981 ](url) the issue seemed solved for the word timings, but maybe this issue still exists for the confidence scores? We use the following to enable the confidence scores: ` decoding_config = copy.deepcopy(conformer.cfg.decoding) ` ` confidence_cfg = ConfidenceConfig(preserve_word_confidence=True, preserve_token_confidence=False) ` ` with open_dict(decoding_config): ` ` decoding_config.preserve_alignments = True, ` ` decoding_config.confidence_cfg = confidence_cfg ` ` conformer.change_decoding_strategy(decoding_config) ` Expected behavior Transcription succeeding without errors Environment overview (please complete the following information) Docker nvcr.io/nvidia/pytorch:23.04-py3 NeMo installed from main with git clone and ./reinstall.sh ( cloned today ) Additional context Ran on a 4090 GPU	{ "login": "titu1994", "id": 3048602, "node_id": "MDQ6VXNlcjMwNDg2MDI=", "avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4", "gravatar_id": "", "url": "https://api.github.com/users/titu1994", "html_url": "https://github.com/titu1994", "followers_url": "https://api.github.com/users/titu1994/followers", "following_url": "https://api.github.com/users/titu1994/following{/other_user}", "gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}", "starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/titu1994/subscriptions", "organizations_url": "https://api.github.com/users/titu1994/orgs", "repos_url": "https://api.github.com/users/titu1994/repos", "events_url": "https://api.github.com/users/titu1994/events{/privacy}", "received_events_url": "https://api.github.com/users/titu1994/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7192/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7192/timeline	null	completed	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7191	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7191/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7191/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7191/events	https://github.com/NVIDIA/NeMo/pull/7191	1,843,251,364	PR_kwDOC_bI7s5XiXL6	7,191	Fix ZeroDivisionError in DistributedBucketSampler	{ "login": "yonas-g", "id": 23583963, "node_id": "MDQ6VXNlcjIzNTgzOTYz", "avatar_url": "https://avatars.githubusercontent.com/u/23583963?v=4", "gravatar_id": "", "url": "https://api.github.com/users/yonas-g", "html_url": "https://github.com/yonas-g", "followers_url": "https://api.github.com/users/yonas-g/followers", "following_url": "https://api.github.com/users/yonas-g/following{/other_user}", "gists_url": "https://api.github.com/users/yonas-g/gists{/gist_id}", "starred_url": "https://api.github.com/users/yonas-g/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/yonas-g/subscriptions", "organizations_url": "https://api.github.com/users/yonas-g/orgs", "repos_url": "https://api.github.com/users/yonas-g/repos", "events_url": "https://api.github.com/users/yonas-g/events{/privacy}", "received_events_url": "https://api.github.com/users/yonas-g/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4418253970, "node_id": "LA_kwDOC_bI7s8AAAABB1k0kg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/TTS", "name": "TTS", "color": "49AF28", "default": false, "description": "" }, { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null } ]	closed	false	null	[]	null	null	2023-08-09T13:32:33	2023-09-01T01:45:22	2023-09-01T01:45:22	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7191", "html_url": "https://github.com/NVIDIA/NeMo/pull/7191", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7191.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7191.patch", "merged_at": null }	In the `_create_buckets` method of the `DistributedBucketSampler` class, a ZeroDivisionError was being raised when `len_bucket` was zero. This fix adds a conditional check to skip the problematic division if `len_bucket` is zero, thus avoiding the error and ensuring that the rest of the code functions as expected. # What does this PR do ? Fixes a ZeroDivisionError in the DistributedBucketSampler class to ensure proper functionality. Collection: TTS Collection # Changelog - Added a conditional check in the `_create_buckets` method to prevent division by zero when `len_bucket` is zero. # Usage ```python from nemo.collections.tts.data.dataset import DistributedBucketSampler # Initialize the sampler with your dataset, batch_size, and boundaries sampler = DistributedBucketSampler(dataset, batch_size, boundaries) # Use the sampler in your training loop for batch in sampler: # Process the batch ``` # Before your PR is "Ready for review" Pre checks: - [x] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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) - [x] Reviewer: Does the PR have correct import guards for all optional libraries? PR Type: - [ ] New Feature - [x] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas.	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7191/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7191/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7190	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7190/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7190/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7190/events	https://github.com/NVIDIA/NeMo/pull/7190	1,842,274,673	PR_kwDOC_bI7s5XfDzz	7,190	[Fix] AudioCodecModel training fails with PTL 2.0	{ "login": "anteju", "id": 108555623, "node_id": "U_kgDOBnhtZw", "avatar_url": "https://avatars.githubusercontent.com/u/108555623?v=4", "gravatar_id": "", "url": "https://api.github.com/users/anteju", "html_url": "https://github.com/anteju", "followers_url": "https://api.github.com/users/anteju/followers", "following_url": "https://api.github.com/users/anteju/following{/other_user}", "gists_url": "https://api.github.com/users/anteju/gists{/gist_id}", "starred_url": "https://api.github.com/users/anteju/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/anteju/subscriptions", "organizations_url": "https://api.github.com/users/anteju/orgs", "repos_url": "https://api.github.com/users/anteju/repos", "events_url": "https://api.github.com/users/anteju/events{/privacy}", "received_events_url": "https://api.github.com/users/anteju/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4418253970, "node_id": "LA_kwDOC_bI7s8AAAABB1k0kg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/TTS", "name": "TTS", "color": "49AF28", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-08-09T00:34:36	2023-08-09T05:17:11	2023-08-09T05:17:10	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7190", "html_url": "https://github.com/NVIDIA/NeMo/pull/7190", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7190.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7190.patch", "merged_at": null }	Currently, `AudioCodecModel` training fails on `main` due to changes in PTL 2.0. # What does this PR do ? This PR is fixing two issues when using PTL 2.0. Collection: TTS # Changelog - Support for `training_epoch_end` has been removed in v2.0.0, and we use `on_train_epoch_end` - Reference: https://github.com/Lightning-AI/lightning/pull/16520 - Enable the detection of unused parameters in DDP, either by setting the string value `strategy='ddp_find_unused_parameters_true'` - Reference: https://github.com/Lightning-AI/lightning/pull/16520 # Usage N/A # Before your PR is "Ready for review" Pre checks: - [x] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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 - [x] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to #6433	{ "login": "anteju", "id": 108555623, "node_id": "U_kgDOBnhtZw", "avatar_url": "https://avatars.githubusercontent.com/u/108555623?v=4", "gravatar_id": "", "url": "https://api.github.com/users/anteju", "html_url": "https://github.com/anteju", "followers_url": "https://api.github.com/users/anteju/followers", "following_url": "https://api.github.com/users/anteju/following{/other_user}", "gists_url": "https://api.github.com/users/anteju/gists{/gist_id}", "starred_url": "https://api.github.com/users/anteju/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/anteju/subscriptions", "organizations_url": "https://api.github.com/users/anteju/orgs", "repos_url": "https://api.github.com/users/anteju/repos", "events_url": "https://api.github.com/users/anteju/events{/privacy}", "received_events_url": "https://api.github.com/users/anteju/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7190/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7190/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7189	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7189/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7189/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7189/events	https://github.com/NVIDIA/NeMo/pull/7189	1,842,244,122	PR_kwDOC_bI7s5Xe9Sr	7,189	Make NFA subtitles stay until end of video	{ "login": "erastorgueva-nv", "id": 80532067, "node_id": "MDQ6VXNlcjgwNTMyMDY3", "avatar_url": "https://avatars.githubusercontent.com/u/80532067?v=4", "gravatar_id": "", "url": "https://api.github.com/users/erastorgueva-nv", "html_url": "https://github.com/erastorgueva-nv", "followers_url": "https://api.github.com/users/erastorgueva-nv/followers", "following_url": "https://api.github.com/users/erastorgueva-nv/following{/other_user}", "gists_url": "https://api.github.com/users/erastorgueva-nv/gists{/gist_id}", "starred_url": "https://api.github.com/users/erastorgueva-nv/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/erastorgueva-nv/subscriptions", "organizations_url": "https://api.github.com/users/erastorgueva-nv/orgs", "repos_url": "https://api.github.com/users/erastorgueva-nv/repos", "events_url": "https://api.github.com/users/erastorgueva-nv/events{/privacy}", "received_events_url": "https://api.github.com/users/erastorgueva-nv/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[]	closed	false	null	[]	null	null	2023-08-08T23:45:11	2023-08-09T04:32:22	2023-08-09T04:32:21	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7189", "html_url": "https://github.com/NVIDIA/NeMo/pull/7189", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7189.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7189.patch", "merged_at": "2023-08-09T04:32:21" }	# What does this PR do ? Write an extra line at the end of the ASS files which will ensure the subtitles of the final segment are visible until the end of the video. Without this, it means that if we create an MP4 file from an audio file by writing the subtitles on a plain background, at the end of the video we are always left with a blank screen, which is not a nice UX as the text feels to disappear very suddenly. Collection: `tools/nemo_forced_aligner.py` # Changelog - extra code to get the duration of the audio file and write the extra lines at the end of the ASS files. # Usage No change in usage. # Before your PR is "Ready for review" Pre checks: - [x] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ N/A ] Did you write any new necessary tests? - [ N/A ] Did you add or update any necessary documentation? - [ N/A ] 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 - [x] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "titu1994", "id": 3048602, "node_id": "MDQ6VXNlcjMwNDg2MDI=", "avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4", "gravatar_id": "", "url": "https://api.github.com/users/titu1994", "html_url": "https://github.com/titu1994", "followers_url": "https://api.github.com/users/titu1994/followers", "following_url": "https://api.github.com/users/titu1994/following{/other_user}", "gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}", "starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/titu1994/subscriptions", "organizations_url": "https://api.github.com/users/titu1994/orgs", "repos_url": "https://api.github.com/users/titu1994/repos", "events_url": "https://api.github.com/users/titu1994/events{/privacy}", "received_events_url": "https://api.github.com/users/titu1994/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7189/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7189/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7188	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7188/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7188/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7188/events	https://github.com/NVIDIA/NeMo/pull/7188	1,842,202,641	PR_kwDOC_bI7s5Xe0Uc	7,188	[TTS] Fix TTS recipes with PTL 2.0	{ "login": "rlangman", "id": 9942053, "node_id": "MDQ6VXNlcjk5NDIwNTM=", "avatar_url": "https://avatars.githubusercontent.com/u/9942053?v=4", "gravatar_id": "", "url": "https://api.github.com/users/rlangman", "html_url": "https://github.com/rlangman", "followers_url": "https://api.github.com/users/rlangman/followers", "following_url": "https://api.github.com/users/rlangman/following{/other_user}", "gists_url": "https://api.github.com/users/rlangman/gists{/gist_id}", "starred_url": "https://api.github.com/users/rlangman/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/rlangman/subscriptions", "organizations_url": "https://api.github.com/users/rlangman/orgs", "repos_url": "https://api.github.com/users/rlangman/repos", "events_url": "https://api.github.com/users/rlangman/events{/privacy}", "received_events_url": "https://api.github.com/users/rlangman/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4418253970, "node_id": "LA_kwDOC_bI7s8AAAABB1k0kg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/TTS", "name": "TTS", "color": "49AF28", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-08-08T22:44:13	2023-08-09T05:10:10	2023-08-09T05:10:09	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7188", "html_url": "https://github.com/NVIDIA/NeMo/pull/7188", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7188.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7188.patch", "merged_at": "2023-08-09T05:10:09" }	# What does this PR do ? Update TTS configs so they run with PTL 2.0 Collection: [TTS] # Changelog - Replace _ddp_ with _ddp_find_unused_parameters_true_ for all vocoder recipes, otherwise it fails because there are unused parameters during the alternating generator and discriminator training steps. - Update training_epoch_end() for AudioCodec to be compatible with new PTL interface. # Before your PR is "Ready for review" Pre checks: - [x] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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 - [x] Bugfix - [ ] Documentation	{ "login": "rlangman", "id": 9942053, "node_id": "MDQ6VXNlcjk5NDIwNTM=", "avatar_url": "https://avatars.githubusercontent.com/u/9942053?v=4", "gravatar_id": "", "url": "https://api.github.com/users/rlangman", "html_url": "https://github.com/rlangman", "followers_url": "https://api.github.com/users/rlangman/followers", "following_url": "https://api.github.com/users/rlangman/following{/other_user}", "gists_url": "https://api.github.com/users/rlangman/gists{/gist_id}", "starred_url": "https://api.github.com/users/rlangman/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/rlangman/subscriptions", "organizations_url": "https://api.github.com/users/rlangman/orgs", "repos_url": "https://api.github.com/users/rlangman/repos", "events_url": "https://api.github.com/users/rlangman/events{/privacy}", "received_events_url": "https://api.github.com/users/rlangman/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7188/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7188/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7187	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7187/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7187/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7187/events	https://github.com/NVIDIA/NeMo/issues/7187	1,842,086,783	I_kwDOC_bI7s5tzAN_	7,187	Concerns about the implementation of `end_of_generation_condition()`	{ "login": "odelalleau", "id": 507137, "node_id": "MDQ6VXNlcjUwNzEzNw==", "avatar_url": "https://avatars.githubusercontent.com/u/507137?v=4", "gravatar_id": "", "url": "https://api.github.com/users/odelalleau", "html_url": "https://github.com/odelalleau", "followers_url": "https://api.github.com/users/odelalleau/followers", "following_url": "https://api.github.com/users/odelalleau/following{/other_user}", "gists_url": "https://api.github.com/users/odelalleau/gists{/gist_id}", "starred_url": "https://api.github.com/users/odelalleau/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/odelalleau/subscriptions", "organizations_url": "https://api.github.com/users/odelalleau/orgs", "repos_url": "https://api.github.com/users/odelalleau/repos", "events_url": "https://api.github.com/users/odelalleau/events{/privacy}", "received_events_url": "https://api.github.com/users/odelalleau/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 1485815557, "node_id": "MDU6TGFiZWwxNDg1ODE1NTU3", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/bug", "name": "bug", "color": "d73a4a", "default": true, "description": "Something isn't working" } ]	closed	false	null	[]	null	null	2023-08-08T20:54:23	2023-08-28T18:29:06	2023-08-28T18:29:06	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	The code of `end_of_generation_condition()` https://github.com/NVIDIA/NeMo/blob/13791d27ec8b322da11e8324ff07124aa4e6c55f/nemo/collections/nlp/modules/common/text_generation_strategy.py#L138 seems weird to me => opening an issue to clarify whether it is working as intended (cc @gshennvm @yidong72) I have two concerns: 1. I find it weird to see `<extra_id_1>` hardcoded in this function -- I thought that the goal of providing `end_strings` as argument would be to avoid having to hardcode special strings 2. I am confused by something that is best explained through an example. Let's say I want to use `end_strings = ["abc"]`, which is tokenized into [1, 2, 3] (tokens for "a", "b", "c"), while `<extra_id_1>` is tokenized into [0]. It seems to me that this code will lead to `ids_1 = tokenizer.text_to_ids(f'<extra_id_1>abc') = [0, 1, 2, 3]`, `ids_2 = [0]`, and `token_id = ids_1[len(ids_2) :][0] = 1`. As a result, as soon as `p` (the last token generated) is equal to 1 (i.e., as soon as an `"a"` is generated), we consider that the generation has ended. Am I missing something?	{ "login": "yidong72", "id": 43824965, "node_id": "MDQ6VXNlcjQzODI0OTY1", "avatar_url": "https://avatars.githubusercontent.com/u/43824965?v=4", "gravatar_id": "", "url": "https://api.github.com/users/yidong72", "html_url": "https://github.com/yidong72", "followers_url": "https://api.github.com/users/yidong72/followers", "following_url": "https://api.github.com/users/yidong72/following{/other_user}", "gists_url": "https://api.github.com/users/yidong72/gists{/gist_id}", "starred_url": "https://api.github.com/users/yidong72/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/yidong72/subscriptions", "organizations_url": "https://api.github.com/users/yidong72/orgs", "repos_url": "https://api.github.com/users/yidong72/repos", "events_url": "https://api.github.com/users/yidong72/events{/privacy}", "received_events_url": "https://api.github.com/users/yidong72/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7187/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7187/timeline	null	completed	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7186	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7186/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7186/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7186/events	https://github.com/NVIDIA/NeMo/pull/7186	1,841,689,274	PR_kwDOC_bI7s5XdCRj	7,186	Speaker emb -> speaker encoder	{ "login": "styagi130", "id": 26252097, "node_id": "MDQ6VXNlcjI2MjUyMDk3", "avatar_url": "https://avatars.githubusercontent.com/u/26252097?v=4", "gravatar_id": "", "url": "https://api.github.com/users/styagi130", "html_url": "https://github.com/styagi130", "followers_url": "https://api.github.com/users/styagi130/followers", "following_url": "https://api.github.com/users/styagi130/following{/other_user}", "gists_url": "https://api.github.com/users/styagi130/gists{/gist_id}", "starred_url": "https://api.github.com/users/styagi130/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/styagi130/subscriptions", "organizations_url": "https://api.github.com/users/styagi130/orgs", "repos_url": "https://api.github.com/users/styagi130/repos", "events_url": "https://api.github.com/users/styagi130/events{/privacy}", "received_events_url": "https://api.github.com/users/styagi130/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4418253970, "node_id": "LA_kwDOC_bI7s8AAAABB1k0kg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/TTS", "name": "TTS", "color": "49AF28", "default": false, "description": "" }, { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null } ]	closed	false	null	[]	null	null	2023-08-08T17:03:18	2023-08-30T01:44:00	2023-08-30T01:44:00	CONTRIBUTOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	true	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7186", "html_url": "https://github.com/NVIDIA/NeMo/pull/7186", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7186.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7186.patch", "merged_at": null }	# What does this PR do ? Rename spk_emb api to speaker_encoder 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 ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7186/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7186/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7185	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7185/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7185/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7185/events	https://github.com/NVIDIA/NeMo/pull/7185	1,841,613,956	PR_kwDOC_bI7s5Xcxua	7,185	Fix token_ids list truncation to specified max_len	{ "login": "ssh-meister", "id": 117230141, "node_id": "U_kgDOBvzKPQ", "avatar_url": "https://avatars.githubusercontent.com/u/117230141?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ssh-meister", "html_url": "https://github.com/ssh-meister", "followers_url": "https://api.github.com/users/ssh-meister/followers", "following_url": "https://api.github.com/users/ssh-meister/following{/other_user}", "gists_url": "https://api.github.com/users/ssh-meister/gists{/gist_id}", "starred_url": "https://api.github.com/users/ssh-meister/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ssh-meister/subscriptions", "organizations_url": "https://api.github.com/users/ssh-meister/orgs", "repos_url": "https://api.github.com/users/ssh-meister/repos", "events_url": "https://api.github.com/users/ssh-meister/events{/privacy}", "received_events_url": "https://api.github.com/users/ssh-meister/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null }, { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" }, { "id": 4847373924, "node_id": "LA_kwDOC_bI7s8AAAABIO0OZA", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/common", "name": "common", "color": "c5def5", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-08-08T16:13:58	2023-08-24T09:55:00	2023-08-24T09:55:00	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7185", "html_url": "https://github.com/NVIDIA/NeMo/pull/7185", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7185.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7185.patch", "merged_at": null }	# What does this PR do? Fixed samples processing: - removed multiple conversions from text_to_ids and ids_to_tokens which causes the difference in the token amounts between these operations - changed the prompt template processing: tokens inside the prompt template (eg "<extra_token_1") are converted to token_ids and after merging with "input" and "output" token_ids lists - due to some tokenizers have no specified space tokens (like " " and "\n") the prompt_template is mandatory (or it will be set by default as "{input} <extra_id_1> {output}" 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 ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "hsiehjackson", "id": 37269846, "node_id": "MDQ6VXNlcjM3MjY5ODQ2", "avatar_url": "https://avatars.githubusercontent.com/u/37269846?v=4", "gravatar_id": "", "url": "https://api.github.com/users/hsiehjackson", "html_url": "https://github.com/hsiehjackson", "followers_url": "https://api.github.com/users/hsiehjackson/followers", "following_url": "https://api.github.com/users/hsiehjackson/following{/other_user}", "gists_url": "https://api.github.com/users/hsiehjackson/gists{/gist_id}", "starred_url": "https://api.github.com/users/hsiehjackson/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/hsiehjackson/subscriptions", "organizations_url": "https://api.github.com/users/hsiehjackson/orgs", "repos_url": "https://api.github.com/users/hsiehjackson/repos", "events_url": "https://api.github.com/users/hsiehjackson/events{/privacy}", "received_events_url": "https://api.github.com/users/hsiehjackson/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7185/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7185/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7184	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7184/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7184/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7184/events	https://github.com/NVIDIA/NeMo/pull/7184	1,841,452,804	PR_kwDOC_bI7s5XcOvR	7,184	allreduce layernorm params when enable sequence_parallel	{ "login": "Agoniii", "id": 18502794, "node_id": "MDQ6VXNlcjE4NTAyNzk0", "avatar_url": "https://avatars.githubusercontent.com/u/18502794?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Agoniii", "html_url": "https://github.com/Agoniii", "followers_url": "https://api.github.com/users/Agoniii/followers", "following_url": "https://api.github.com/users/Agoniii/following{/other_user}", "gists_url": "https://api.github.com/users/Agoniii/gists{/gist_id}", "starred_url": "https://api.github.com/users/Agoniii/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Agoniii/subscriptions", "organizations_url": "https://api.github.com/users/Agoniii/orgs", "repos_url": "https://api.github.com/users/Agoniii/repos", "events_url": "https://api.github.com/users/Agoniii/events{/privacy}", "received_events_url": "https://api.github.com/users/Agoniii/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null }, { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-08-08T14:39:06	2023-08-30T01:44:03	2023-08-30T01:44:03	CONTRIBUTOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7184", "html_url": "https://github.com/NVIDIA/NeMo/pull/7184", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7184.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7184.patch", "merged_at": null }	# What does this PR do ? Refer to the [comments](https://github.com/NVIDIA/NeMo/blob/main/nemo/collections/nlp/models/language_modeling/megatron_gpt_model.py#L677), Layernorm parameters need to be all-reduced across model parallel nodes when sequence parallelism is used. NeMo find layernorm parameters through '`sequence_parallel` attribute. Code is [here](https://github.com/NVIDIA/NeMo/blob/main/nemo/collections/nlp/models/language_modeling/megatron_gpt_model.py#L665). But layernorm set the attribute with `sequence_parallel_enabled` in apex. Code is [here](https://github.com/NVIDIA/apex/blob/master/apex/transformer/layers/layer_norm.py#L30). 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 ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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 - [x] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7184/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7184/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7183	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7183/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7183/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7183/events	https://github.com/NVIDIA/NeMo/pull/7183	1,841,428,167	PR_kwDOC_bI7s5XcJZW	7,183	use the MixedFusedLayerNorm which will set sequence_parallel_enabled	{ "login": "Agoniii", "id": 18502794, "node_id": "MDQ6VXNlcjE4NTAyNzk0", "avatar_url": "https://avatars.githubusercontent.com/u/18502794?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Agoniii", "html_url": "https://github.com/Agoniii", "followers_url": "https://api.github.com/users/Agoniii/followers", "following_url": "https://api.github.com/users/Agoniii/following{/other_user}", "gists_url": "https://api.github.com/users/Agoniii/gists{/gist_id}", "starred_url": "https://api.github.com/users/Agoniii/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Agoniii/subscriptions", "organizations_url": "https://api.github.com/users/Agoniii/orgs", "repos_url": "https://api.github.com/users/Agoniii/repos", "events_url": "https://api.github.com/users/Agoniii/events{/privacy}", "received_events_url": "https://api.github.com/users/Agoniii/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null }, { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-08-08T14:25:24	2023-08-30T01:44:06	2023-08-30T01:44:06	CONTRIBUTOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7183", "html_url": "https://github.com/NVIDIA/NeMo/pull/7183", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7183.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7183.patch", "merged_at": null }	# What does this PR do ? NeMo uses `apex.normalization.fused_layer_norm.MixedFusedLayerNorm` when the `persistent_layernorm` is false. But the `apex.normalization.fused_layer_norm.MixedFusedLayerNorm` doesn't set sequence_parallel. And the `apex.transformer.layers.layer_norm.MixedFusedLayerNorm` will set sequence_parallel. This PR use the `apex.transformer.layers.layer_norm.MixedFusedLayerNorm` instead of `apex.normalization.fused_layer_norm.MixedFusedLayerNorm`. 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 ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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 - [x] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7183/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7183/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7182	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7182/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7182/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7182/events	https://github.com/NVIDIA/NeMo/pull/7182	1,841,409,748	PR_kwDOC_bI7s5XcFWE	7,182	Fix t5 trainer test m	{ "login": "jubick1337", "id": 16004982, "node_id": "MDQ6VXNlcjE2MDA0OTgy", "avatar_url": "https://avatars.githubusercontent.com/u/16004982?v=4", "gravatar_id": "", "url": "https://api.github.com/users/jubick1337", "html_url": "https://github.com/jubick1337", "followers_url": "https://api.github.com/users/jubick1337/followers", "following_url": "https://api.github.com/users/jubick1337/following{/other_user}", "gists_url": "https://api.github.com/users/jubick1337/gists{/gist_id}", "starred_url": "https://api.github.com/users/jubick1337/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/jubick1337/subscriptions", "organizations_url": "https://api.github.com/users/jubick1337/orgs", "repos_url": "https://api.github.com/users/jubick1337/repos", "events_url": "https://api.github.com/users/jubick1337/events{/privacy}", "received_events_url": "https://api.github.com/users/jubick1337/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" }, { "id": 4847373924, "node_id": "LA_kwDOC_bI7s8AAAABIO0OZA", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/common", "name": "common", "color": "c5def5", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-08-08T14:15:32	2023-08-10T15:41:44	2023-08-10T15:41:44	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	true	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7182", "html_url": "https://github.com/NVIDIA/NeMo/pull/7182", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7182.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7182.patch", "merged_at": null }	# 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 ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "jubick1337", "id": 16004982, "node_id": "MDQ6VXNlcjE2MDA0OTgy", "avatar_url": "https://avatars.githubusercontent.com/u/16004982?v=4", "gravatar_id": "", "url": "https://api.github.com/users/jubick1337", "html_url": "https://github.com/jubick1337", "followers_url": "https://api.github.com/users/jubick1337/followers", "following_url": "https://api.github.com/users/jubick1337/following{/other_user}", "gists_url": "https://api.github.com/users/jubick1337/gists{/gist_id}", "starred_url": "https://api.github.com/users/jubick1337/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/jubick1337/subscriptions", "organizations_url": "https://api.github.com/users/jubick1337/orgs", "repos_url": "https://api.github.com/users/jubick1337/repos", "events_url": "https://api.github.com/users/jubick1337/events{/privacy}", "received_events_url": "https://api.github.com/users/jubick1337/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7182/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7182/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7181	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7181/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7181/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7181/events	https://github.com/NVIDIA/NeMo/issues/7181	1,840,710,035	I_kwDOC_bI7s5ttwGT	7,181	MSDD model share memory.	{ "login": "SagyHarpazGong", "id": 99274232, "node_id": "U_kgDOBerN-A", "avatar_url": "https://avatars.githubusercontent.com/u/99274232?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SagyHarpazGong", "html_url": "https://github.com/SagyHarpazGong", "followers_url": "https://api.github.com/users/SagyHarpazGong/followers", "following_url": "https://api.github.com/users/SagyHarpazGong/following{/other_user}", "gists_url": "https://api.github.com/users/SagyHarpazGong/gists{/gist_id}", "starred_url": "https://api.github.com/users/SagyHarpazGong/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SagyHarpazGong/subscriptions", "organizations_url": "https://api.github.com/users/SagyHarpazGong/orgs", "repos_url": "https://api.github.com/users/SagyHarpazGong/repos", "events_url": "https://api.github.com/users/SagyHarpazGong/events{/privacy}", "received_events_url": "https://api.github.com/users/SagyHarpazGong/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null } ]	closed	false	{ "login": "okuchaiev", "id": 22035961, "node_id": "MDQ6VXNlcjIyMDM1OTYx", "avatar_url": "https://avatars.githubusercontent.com/u/22035961?v=4", "gravatar_id": "", "url": "https://api.github.com/users/okuchaiev", "html_url": "https://github.com/okuchaiev", "followers_url": "https://api.github.com/users/okuchaiev/followers", "following_url": "https://api.github.com/users/okuchaiev/following{/other_user}", "gists_url": "https://api.github.com/users/okuchaiev/gists{/gist_id}", "starred_url": "https://api.github.com/users/okuchaiev/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/okuchaiev/subscriptions", "organizations_url": "https://api.github.com/users/okuchaiev/orgs", "repos_url": "https://api.github.com/users/okuchaiev/repos", "events_url": "https://api.github.com/users/okuchaiev/events{/privacy}", "received_events_url": "https://api.github.com/users/okuchaiev/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "login": "okuchaiev", "id": 22035961, "node_id": "MDQ6VXNlcjIyMDM1OTYx", "avatar_url": "https://avatars.githubusercontent.com/u/22035961?v=4", "gravatar_id": "", "url": "https://api.github.com/users/okuchaiev", "html_url": "https://github.com/okuchaiev", "followers_url": "https://api.github.com/users/okuchaiev/followers", "following_url": "https://api.github.com/users/okuchaiev/following{/other_user}", "gists_url": "https://api.github.com/users/okuchaiev/gists{/gist_id}", "starred_url": "https://api.github.com/users/okuchaiev/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/okuchaiev/subscriptions", "organizations_url": "https://api.github.com/users/okuchaiev/orgs", "repos_url": "https://api.github.com/users/okuchaiev/repos", "events_url": "https://api.github.com/users/okuchaiev/events{/privacy}", "received_events_url": "https://api.github.com/users/okuchaiev/received_events", "type": "User", "user_view_type": "public", "site_admin": false } ]	null	null	2023-08-08T07:15:21	2023-09-15T01:44:34	2023-09-15T01:44:34	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	Hi. I tried to run my MSDD model (using TITANET-large as speaker embedding model) in serving mode, meaning loading the model once and inference it in several processes, and in order to do so I need to use torch share_memory method. when I use share_memory method I get the following error when I start the process: AttributeError: Can't pickle local object 'FilterbankFeatures.__init__.<locals>.<lambda>' Is there any guidelines how to share memory for all MSDD modules or any utils script to do so?	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7181/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7181/timeline	null	not_planned	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7180	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7180/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7180/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7180/events	https://github.com/NVIDIA/NeMo/pull/7180	1,840,649,175	PR_kwDOC_bI7s5XZgVl	7,180	add missing import	{ "login": "XuesongYang", "id": 1646669, "node_id": "MDQ6VXNlcjE2NDY2Njk=", "avatar_url": "https://avatars.githubusercontent.com/u/1646669?v=4", "gravatar_id": "", "url": "https://api.github.com/users/XuesongYang", "html_url": "https://github.com/XuesongYang", "followers_url": "https://api.github.com/users/XuesongYang/followers", "following_url": "https://api.github.com/users/XuesongYang/following{/other_user}", "gists_url": "https://api.github.com/users/XuesongYang/gists{/gist_id}", "starred_url": "https://api.github.com/users/XuesongYang/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/XuesongYang/subscriptions", "organizations_url": "https://api.github.com/users/XuesongYang/orgs", "repos_url": "https://api.github.com/users/XuesongYang/repos", "events_url": "https://api.github.com/users/XuesongYang/events{/privacy}", "received_events_url": "https://api.github.com/users/XuesongYang/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 1923495043, "node_id": "MDU6TGFiZWwxOTIzNDk1MDQz", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/core", "name": "core", "color": "f25e3a", "default": false, "description": "Changes to NeMo Core" }, { "id": 4418253970, "node_id": "LA_kwDOC_bI7s8AAAABB1k0kg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/TTS", "name": "TTS", "color": "49AF28", "default": false, "description": "" }, { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null }, { "id": 4811884691, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR", "name": "ASR", "color": "F3B33E", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-08-08T06:28:55	2024-07-30T17:50:53	2023-08-30T01:44:08	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	true	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7180", "html_url": "https://github.com/NVIDIA/NeMo/pull/7180", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7180.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7180.patch", "merged_at": null }	# What does this PR do ? Not sure why missing lib import did not raise any errors. * added missing import for `Trainer` and `NemoConfig`; * unify `'Trainer'` into `Trainer` as a variable type hint. 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 ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7180/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7180/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7179	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7179/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7179/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7179/events	https://github.com/NVIDIA/NeMo/issues/7179	1,840,635,440	I_kwDOC_bI7s5ttd4w	7,179	[Error] On running this tutorial notebook on Colab	{ "login": "mahimairaja", "id": 81288263, "node_id": "MDQ6VXNlcjgxMjg4MjYz", "avatar_url": "https://avatars.githubusercontent.com/u/81288263?v=4", "gravatar_id": "", "url": "https://api.github.com/users/mahimairaja", "html_url": "https://github.com/mahimairaja", "followers_url": "https://api.github.com/users/mahimairaja/followers", "following_url": "https://api.github.com/users/mahimairaja/following{/other_user}", "gists_url": "https://api.github.com/users/mahimairaja/gists{/gist_id}", "starred_url": "https://api.github.com/users/mahimairaja/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/mahimairaja/subscriptions", "organizations_url": "https://api.github.com/users/mahimairaja/orgs", "repos_url": "https://api.github.com/users/mahimairaja/repos", "events_url": "https://api.github.com/users/mahimairaja/events{/privacy}", "received_events_url": "https://api.github.com/users/mahimairaja/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null } ]	closed	false	null	[]	null	null	2023-08-08T06:16:42	2023-09-15T01:44:36	2023-09-15T01:44:36	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	## ASR Tutorial - Online_Offline_Speech_Commands_Demo.ipynb I tried to execute the notebook in google colab (`https://github.com/NVIDIA/NeMo/blob/stable/tutorials/asr/Online_Offline_Speech_Commands_Demo.ipynb`) While running the code block: ```python wave_file = demo_wave CHANNELS = 1 audio, sample_rate = librosa.load(wave_file, sr=SAMPLE_RATE) dur = librosa.get_duration(audio) print(dur) ``` The below error is produced: ```bash TypeError Traceback (most recent call last) [<ipython-input-20-1dea7463f5dd>](https://al43ym753q8-496ff2e9c6d22116-0-colab.googleusercontent.com/outputframe.html?vrz=colab-20230804-060128-RC00_553764177#) in <cell line: 5>() 3 CHANNELS = 1 4 audio, sample_rate = librosa.load(wave_file, sr=SAMPLE_RATE) ----> 5 dur = librosa.get_duration(audio) 6 print(dur) TypeError: get_duration() takes 0 positional arguments but 1 was given ```	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7179/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7179/timeline	null	not_planned	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7178	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7178/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7178/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7178/events	https://github.com/NVIDIA/NeMo/issues/7178	1,840,595,651	I_kwDOC_bI7s5ttULD	7,178	fine-tuning hifi-gan with libritts	{ "login": "liulfy", "id": 27479464, "node_id": "MDQ6VXNlcjI3NDc5NDY0", "avatar_url": "https://avatars.githubusercontent.com/u/27479464?v=4", "gravatar_id": "", "url": "https://api.github.com/users/liulfy", "html_url": "https://github.com/liulfy", "followers_url": "https://api.github.com/users/liulfy/followers", "following_url": "https://api.github.com/users/liulfy/following{/other_user}", "gists_url": "https://api.github.com/users/liulfy/gists{/gist_id}", "starred_url": "https://api.github.com/users/liulfy/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/liulfy/subscriptions", "organizations_url": "https://api.github.com/users/liulfy/orgs", "repos_url": "https://api.github.com/users/liulfy/repos", "events_url": "https://api.github.com/users/liulfy/events{/privacy}", "received_events_url": "https://api.github.com/users/liulfy/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4418253970, "node_id": "LA_kwDOC_bI7s8AAAABB1k0kg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/TTS", "name": "TTS", "color": "49AF28", "default": false, "description": "" }, { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null } ]	closed	false	{ "login": "XuesongYang", "id": 1646669, "node_id": "MDQ6VXNlcjE2NDY2Njk=", "avatar_url": "https://avatars.githubusercontent.com/u/1646669?v=4", "gravatar_id": "", "url": "https://api.github.com/users/XuesongYang", "html_url": "https://github.com/XuesongYang", "followers_url": "https://api.github.com/users/XuesongYang/followers", "following_url": "https://api.github.com/users/XuesongYang/following{/other_user}", "gists_url": "https://api.github.com/users/XuesongYang/gists{/gist_id}", "starred_url": "https://api.github.com/users/XuesongYang/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/XuesongYang/subscriptions", "organizations_url": "https://api.github.com/users/XuesongYang/orgs", "repos_url": "https://api.github.com/users/XuesongYang/repos", "events_url": "https://api.github.com/users/XuesongYang/events{/privacy}", "received_events_url": "https://api.github.com/users/XuesongYang/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "login": "XuesongYang", "id": 1646669, "node_id": "MDQ6VXNlcjE2NDY2Njk=", "avatar_url": "https://avatars.githubusercontent.com/u/1646669?v=4", "gravatar_id": "", "url": "https://api.github.com/users/XuesongYang", "html_url": "https://github.com/XuesongYang", "followers_url": "https://api.github.com/users/XuesongYang/followers", "following_url": "https://api.github.com/users/XuesongYang/following{/other_user}", "gists_url": "https://api.github.com/users/XuesongYang/gists{/gist_id}", "starred_url": "https://api.github.com/users/XuesongYang/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/XuesongYang/subscriptions", "organizations_url": "https://api.github.com/users/XuesongYang/orgs", "repos_url": "https://api.github.com/users/XuesongYang/repos", "events_url": "https://api.github.com/users/XuesongYang/events{/privacy}", "received_events_url": "https://api.github.com/users/XuesongYang/received_events", "type": "User", "user_view_type": "public", "site_admin": false } ]	null	null	2023-08-08T05:35:06	2023-10-15T01:45:51	2023-10-15T01:45:50	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	I try finetune hifi-gan, which was pre-trained as 'tts_en_hifigan', with libritts, but it failed. I guess I need to modify the libritts data, I have resampled it to 22050kHz, and it did not work. Could you please give me some help, thanks!	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7178/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7178/timeline	null	not_planned	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7177	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7177/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7177/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7177/events	https://github.com/NVIDIA/NeMo/pull/7177	1,840,436,886	PR_kwDOC_bI7s5XYzEk	7,177	Fix logging for Megatron	{ "login": "athitten", "id": 47577437, "node_id": "MDQ6VXNlcjQ3NTc3NDM3", "avatar_url": "https://avatars.githubusercontent.com/u/47577437?v=4", "gravatar_id": "", "url": "https://api.github.com/users/athitten", "html_url": "https://github.com/athitten", "followers_url": "https://api.github.com/users/athitten/followers", "following_url": "https://api.github.com/users/athitten/following{/other_user}", "gists_url": "https://api.github.com/users/athitten/gists{/gist_id}", "starred_url": "https://api.github.com/users/athitten/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/athitten/subscriptions", "organizations_url": "https://api.github.com/users/athitten/orgs", "repos_url": "https://api.github.com/users/athitten/repos", "events_url": "https://api.github.com/users/athitten/events{/privacy}", "received_events_url": "https://api.github.com/users/athitten/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null }, { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-08-08T01:58:33	2023-08-30T01:44:11	2023-08-30T01:44:11	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7177", "html_url": "https://github.com/NVIDIA/NeMo/pull/7177", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7177.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7177.patch", "merged_at": null }	# What does this PR do ? The PR fixes the progress bar stdout in megatron models to remove the `s/it` or `it/s` and instead display the actual cuda synchronized train step timing. 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 ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7177/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7177/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7176	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7176/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7176/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7176/events	https://github.com/NVIDIA/NeMo/pull/7176	1,840,264,610	PR_kwDOC_bI7s5XYN1O	7,176	Fixes for lightning 2.0 upgrade	{ "login": "athitten", "id": 47577437, "node_id": "MDQ6VXNlcjQ3NTc3NDM3", "avatar_url": "https://avatars.githubusercontent.com/u/47577437?v=4", "gravatar_id": "", "url": "https://api.github.com/users/athitten", "html_url": "https://github.com/athitten", "followers_url": "https://api.github.com/users/athitten/followers", "following_url": "https://api.github.com/users/athitten/following{/other_user}", "gists_url": "https://api.github.com/users/athitten/gists{/gist_id}", "starred_url": "https://api.github.com/users/athitten/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/athitten/subscriptions", "organizations_url": "https://api.github.com/users/athitten/orgs", "repos_url": "https://api.github.com/users/athitten/repos", "events_url": "https://api.github.com/users/athitten/events{/privacy}", "received_events_url": "https://api.github.com/users/athitten/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" }, { "id": 4840216855, "node_id": "LA_kwDOC_bI7s8AAAABIH_ZFw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/CI", "name": "CI", "color": "1D76DB", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-08-07T22:00:26	2023-08-12T04:51:04	2023-08-12T04:51:03	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7176", "html_url": "https://github.com/NVIDIA/NeMo/pull/7176", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7176.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7176.patch", "merged_at": "2023-08-12T04:51:03" }	# What does this PR do ? Fix the following issues with the previous lightning 2.0 upgrade PR: 1) Decrease `val_check_interval` to 1 while resuming from a ckpt in some CI tests where max_steps is same as that of training and lightning 2.0 errors: [fit_loop.py#L259](https://github.com/Lightning-AI/lightning/blob/2.0.6/src/lightning/pytorch/loops/fit_loop.py#L259). Lightning 2.0 calls `setup_data` right at the beginning of fit_loop.run unlike 1.9 that did it only when `self.skip` for fit was False, hence the error din't show up with 1.9 2) Remove the usage of `trainer._checkpoint_connector` as it's redundant in lightning 2.0 and use `trainer.ckpt_path` directly. Also having `trainer._checkpoint_connector = _CheckpointConnector(trainer)` overrode ckpt_path with None leading to starting training from scratch during resuming from a ckpt. 3) Remove `optimizer_idx` arg in `optimizer_step` of `MegatronHalfPrecisionPlugin` class as the parent function in lightning does not have it anymore. 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 ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "ericharper", "id": 11999610, "node_id": "MDQ6VXNlcjExOTk5NjEw", "avatar_url": "https://avatars.githubusercontent.com/u/11999610?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ericharper", "html_url": "https://github.com/ericharper", "followers_url": "https://api.github.com/users/ericharper/followers", "following_url": "https://api.github.com/users/ericharper/following{/other_user}", "gists_url": "https://api.github.com/users/ericharper/gists{/gist_id}", "starred_url": "https://api.github.com/users/ericharper/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ericharper/subscriptions", "organizations_url": "https://api.github.com/users/ericharper/orgs", "repos_url": "https://api.github.com/users/ericharper/repos", "events_url": "https://api.github.com/users/ericharper/events{/privacy}", "received_events_url": "https://api.github.com/users/ericharper/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7176/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7176/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7175	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7175/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7175/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7175/events	https://github.com/NVIDIA/NeMo/pull/7175	1,838,451,195	PR_kwDOC_bI7s5XSF_p	7,175	Update README.rst	{ "login": "fayejf", "id": 36722593, "node_id": "MDQ6VXNlcjM2NzIyNTkz", "avatar_url": "https://avatars.githubusercontent.com/u/36722593?v=4", "gravatar_id": "", "url": "https://api.github.com/users/fayejf", "html_url": "https://github.com/fayejf", "followers_url": "https://api.github.com/users/fayejf/followers", "following_url": "https://api.github.com/users/fayejf/following{/other_user}", "gists_url": "https://api.github.com/users/fayejf/gists{/gist_id}", "starred_url": "https://api.github.com/users/fayejf/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/fayejf/subscriptions", "organizations_url": "https://api.github.com/users/fayejf/orgs", "repos_url": "https://api.github.com/users/fayejf/repos", "events_url": "https://api.github.com/users/fayejf/events{/privacy}", "received_events_url": "https://api.github.com/users/fayejf/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[]	closed	false	null	[]	null	null	2023-08-06T22:57:08	2023-08-08T20:52:02	2023-08-08T20:52:01	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7175", "html_url": "https://github.com/NVIDIA/NeMo/pull/7175", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7175.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7175.patch", "merged_at": "2023-08-08T20:52:01" }	# What does this PR do ? update readme regarding container version and python version 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 ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "fayejf", "id": 36722593, "node_id": "MDQ6VXNlcjM2NzIyNTkz", "avatar_url": "https://avatars.githubusercontent.com/u/36722593?v=4", "gravatar_id": "", "url": "https://api.github.com/users/fayejf", "html_url": "https://github.com/fayejf", "followers_url": "https://api.github.com/users/fayejf/followers", "following_url": "https://api.github.com/users/fayejf/following{/other_user}", "gists_url": "https://api.github.com/users/fayejf/gists{/gist_id}", "starred_url": "https://api.github.com/users/fayejf/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/fayejf/subscriptions", "organizations_url": "https://api.github.com/users/fayejf/orgs", "repos_url": "https://api.github.com/users/fayejf/repos", "events_url": "https://api.github.com/users/fayejf/events{/privacy}", "received_events_url": "https://api.github.com/users/fayejf/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7175/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7175/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7174	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7174/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7174/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7174/events	https://github.com/NVIDIA/NeMo/pull/7174	1,838,273,748	PR_kwDOC_bI7s5XRgT4	7,174	Allow applications to control the encoder mask type	{ "login": "nzmora-nvidia", "id": 96238833, "node_id": "U_kgDOBbx88Q", "avatar_url": "https://avatars.githubusercontent.com/u/96238833?v=4", "gravatar_id": "", "url": "https://api.github.com/users/nzmora-nvidia", "html_url": "https://github.com/nzmora-nvidia", "followers_url": "https://api.github.com/users/nzmora-nvidia/followers", "following_url": "https://api.github.com/users/nzmora-nvidia/following{/other_user}", "gists_url": "https://api.github.com/users/nzmora-nvidia/gists{/gist_id}", "starred_url": "https://api.github.com/users/nzmora-nvidia/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/nzmora-nvidia/subscriptions", "organizations_url": "https://api.github.com/users/nzmora-nvidia/orgs", "repos_url": "https://api.github.com/users/nzmora-nvidia/repos", "events_url": "https://api.github.com/users/nzmora-nvidia/events{/privacy}", "received_events_url": "https://api.github.com/users/nzmora-nvidia/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null }, { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-08-06T16:48:21	2023-10-03T01:44:40	2023-10-03T01:44:40	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7174", "html_url": "https://github.com/NVIDIA/NeMo/pull/7174", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7174.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7174.patch", "merged_at": null }	# What does this PR do ? When TensorRT optimizes Nemo GPT models it uses a different mask shape in the Context phase (triangular mask, i.e. causal) and in the Generation phase (1D mask, i.e. padding). To support this TensorRT needs to control GPT's encoder mask type (which defaults to causal) via the model configuration. Collection: NLP # Changelog - `MegatronGPTModel::model_provider_func` reads an optional `encoder_attn_mask_type` argument from the config file and passes it to the GPTModel constructor. The default value is a causal mask. - Add an optional `encoder_attn_mask_type` argument to GPTModel constructor. The argument defaults to `AttnMaskType.causal`. - Pass this argument to `get_language_model` (previously hard-coded: `encoder_attn_mask_type=AttnMaskType.causal`) PR Type: - [x] New Feature - [ ] Bugfix - [ ] Documentation ## Who can review? Anyone in the community is free to review the PR once the checks have passed. [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas.	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7174/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7174/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7173	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7173/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7173/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7173/events	https://github.com/NVIDIA/NeMo/issues/7173	1,838,218,698	I_kwDOC_bI7s5tkP3K	7,173	1D conv block in citrinet	{ "login": "bchinnari", "id": 26137195, "node_id": "MDQ6VXNlcjI2MTM3MTk1", "avatar_url": "https://avatars.githubusercontent.com/u/26137195?v=4", "gravatar_id": "", "url": "https://api.github.com/users/bchinnari", "html_url": "https://github.com/bchinnari", "followers_url": "https://api.github.com/users/bchinnari/followers", "following_url": "https://api.github.com/users/bchinnari/following{/other_user}", "gists_url": "https://api.github.com/users/bchinnari/gists{/gist_id}", "starred_url": "https://api.github.com/users/bchinnari/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/bchinnari/subscriptions", "organizations_url": "https://api.github.com/users/bchinnari/orgs", "repos_url": "https://api.github.com/users/bchinnari/repos", "events_url": "https://api.github.com/users/bchinnari/events{/privacy}", "received_events_url": "https://api.github.com/users/bchinnari/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[]	closed	false	null	[]	null	null	2023-08-06T13:54:46	2023-08-07T07:53:27	2023-08-07T07:53:26	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	I am looking at the citrinet paper (https://arxiv.org/pdf/2104.01721.pdf) and I am trying to understand the architecture. Simultaneously I downloaded citrinet model (stt_en_citrinet_1024_gamma_0_24) and printed the model. Part of it looks like follows --------------------- (1): JasperBlock( (mconv): ModuleList( (0): MaskedConv1d( (conv): Conv1d(1024, 1024, kernel_size=(3,), stride=(1,), padding=(1,), groups=1024, bias=False) ) (1): MaskedConv1d( (conv): Conv1d(1024, 1024, kernel_size=(1,), stride=(1,), bias=False) ) (2): BatchNorm1d(1024, eps=0.001, momentum=0.1, affine=True, track_running_stats=True) (3): ReLU(inplace=True) (4): Dropout(p=0.1, inplace=False) (5): MaskedConv1d( (conv): Conv1d(1024, 1024, kernel_size=(3,), stride=(1,), padding=(1,), groups=1024, bias=False) ) (6): MaskedConv1d( (conv): Conv1d(1024, 1024, kernel_size=(1,), stride=(1,), bias=False) ) ........................ --------------------- There are 2 conv1d blocks before batchnorm where as there is only 1 conv1d block as per the paper. Is this observation correct ? I saw few lines of the code , it looks like there is residual module for every sub block in the megablock . Am I correct ? But as per paper, there is only residual connecting from beginning to end of the the megablock. Kindly clarify	{ "login": "bchinnari", "id": 26137195, "node_id": "MDQ6VXNlcjI2MTM3MTk1", "avatar_url": "https://avatars.githubusercontent.com/u/26137195?v=4", "gravatar_id": "", "url": "https://api.github.com/users/bchinnari", "html_url": "https://github.com/bchinnari", "followers_url": "https://api.github.com/users/bchinnari/followers", "following_url": "https://api.github.com/users/bchinnari/following{/other_user}", "gists_url": "https://api.github.com/users/bchinnari/gists{/gist_id}", "starred_url": "https://api.github.com/users/bchinnari/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/bchinnari/subscriptions", "organizations_url": "https://api.github.com/users/bchinnari/orgs", "repos_url": "https://api.github.com/users/bchinnari/repos", "events_url": "https://api.github.com/users/bchinnari/events{/privacy}", "received_events_url": "https://api.github.com/users/bchinnari/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7173/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7173/timeline	null	completed	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7172	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7172/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7172/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7172/events	https://github.com/NVIDIA/NeMo/issues/7172	1,838,112,398	I_kwDOC_bI7s5tj16O	7,172	Upgrading `hydra-core` past 1.3	{ "login": "alllexx88", "id": 6587516, "node_id": "MDQ6VXNlcjY1ODc1MTY=", "avatar_url": "https://avatars.githubusercontent.com/u/6587516?v=4", "gravatar_id": "", "url": "https://api.github.com/users/alllexx88", "html_url": "https://github.com/alllexx88", "followers_url": "https://api.github.com/users/alllexx88/followers", "following_url": "https://api.github.com/users/alllexx88/following{/other_user}", "gists_url": "https://api.github.com/users/alllexx88/gists{/gist_id}", "starred_url": "https://api.github.com/users/alllexx88/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/alllexx88/subscriptions", "organizations_url": "https://api.github.com/users/alllexx88/orgs", "repos_url": "https://api.github.com/users/alllexx88/repos", "events_url": "https://api.github.com/users/alllexx88/events{/privacy}", "received_events_url": "https://api.github.com/users/alllexx88/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[]	closed	false	{ "login": "titu1994", "id": 3048602, "node_id": "MDQ6VXNlcjMwNDg2MDI=", "avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4", "gravatar_id": "", "url": "https://api.github.com/users/titu1994", "html_url": "https://github.com/titu1994", "followers_url": "https://api.github.com/users/titu1994/followers", "following_url": "https://api.github.com/users/titu1994/following{/other_user}", "gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}", "starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/titu1994/subscriptions", "organizations_url": "https://api.github.com/users/titu1994/orgs", "repos_url": "https://api.github.com/users/titu1994/repos", "events_url": "https://api.github.com/users/titu1994/events{/privacy}", "received_events_url": "https://api.github.com/users/titu1994/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "login": "titu1994", "id": 3048602, "node_id": "MDQ6VXNlcjMwNDg2MDI=", "avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4", "gravatar_id": "", "url": "https://api.github.com/users/titu1994", "html_url": "https://github.com/titu1994", "followers_url": "https://api.github.com/users/titu1994/followers", "following_url": "https://api.github.com/users/titu1994/following{/other_user}", "gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}", "starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/titu1994/subscriptions", "organizations_url": "https://api.github.com/users/titu1994/orgs", "repos_url": "https://api.github.com/users/titu1994/repos", "events_url": "https://api.github.com/users/titu1994/events{/privacy}", "received_events_url": "https://api.github.com/users/titu1994/received_events", "type": "User", "user_view_type": "public", "site_admin": false }, { "login": "ericharper", "id": 11999610, "node_id": "MDQ6VXNlcjExOTk5NjEw", "avatar_url": "https://avatars.githubusercontent.com/u/11999610?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ericharper", "html_url": "https://github.com/ericharper", "followers_url": "https://api.github.com/users/ericharper/followers", "following_url": "https://api.github.com/users/ericharper/following{/other_user}", "gists_url": "https://api.github.com/users/ericharper/gists{/gist_id}", "starred_url": "https://api.github.com/users/ericharper/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ericharper/subscriptions", "organizations_url": "https://api.github.com/users/ericharper/orgs", "repos_url": "https://api.github.com/users/ericharper/repos", "events_url": "https://api.github.com/users/ericharper/events{/privacy}", "received_events_url": "https://api.github.com/users/ericharper/received_events", "type": "User", "user_view_type": "public", "site_admin": false }, { "login": "athitten", "id": 47577437, "node_id": "MDQ6VXNlcjQ3NTc3NDM3", "avatar_url": "https://avatars.githubusercontent.com/u/47577437?v=4", "gravatar_id": "", "url": "https://api.github.com/users/athitten", "html_url": "https://github.com/athitten", "followers_url": "https://api.github.com/users/athitten/followers", "following_url": "https://api.github.com/users/athitten/following{/other_user}", "gists_url": "https://api.github.com/users/athitten/gists{/gist_id}", "starred_url": "https://api.github.com/users/athitten/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/athitten/subscriptions", "organizations_url": "https://api.github.com/users/athitten/orgs", "repos_url": "https://api.github.com/users/athitten/repos", "events_url": "https://api.github.com/users/athitten/events{/privacy}", "received_events_url": "https://api.github.com/users/athitten/received_events", "type": "User", "user_view_type": "public", "site_admin": false } ]	null	null	2023-08-06T08:40:07	2023-08-19T12:37:52	2023-08-19T12:37:52	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	Is your feature request related to a problem? Please describe. Current `hydra-core` version (`hydra-core>=1.2.0,<1.3` from `requirements/requirements_lightning.txt`) doesn't work with Python 3.11: see #7166 Describe the solution you'd like Now, after [Upgrade to pytorch lightning 2.0](https://github.com/NVIDIA/NeMo/commit/67de251e980ef8d8aec3b8339e70e96f6c952122) commit, would it be safe to upgrade `hydra-core` past `1.3.0`? `hydra-core==1.3.2` at least imports fine without `1.2.0` `dataclasses` related issues. Describe alternatives you've considered The only alternative I can think of is creating a fork of `1.2.0` version and patching it, but I think it's counter-productive since it has to be maintained separately. Thanks :smiley:	{ "login": "alllexx88", "id": 6587516, "node_id": "MDQ6VXNlcjY1ODc1MTY=", "avatar_url": "https://avatars.githubusercontent.com/u/6587516?v=4", "gravatar_id": "", "url": "https://api.github.com/users/alllexx88", "html_url": "https://github.com/alllexx88", "followers_url": "https://api.github.com/users/alllexx88/followers", "following_url": "https://api.github.com/users/alllexx88/following{/other_user}", "gists_url": "https://api.github.com/users/alllexx88/gists{/gist_id}", "starred_url": "https://api.github.com/users/alllexx88/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/alllexx88/subscriptions", "organizations_url": "https://api.github.com/users/alllexx88/orgs", "repos_url": "https://api.github.com/users/alllexx88/repos", "events_url": "https://api.github.com/users/alllexx88/events{/privacy}", "received_events_url": "https://api.github.com/users/alllexx88/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7172/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7172/timeline	null	completed	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7171	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7171/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7171/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7171/events	https://github.com/NVIDIA/NeMo/pull/7171	1,838,000,642	PR_kwDOC_bI7s5XQttE	7,171	context parallelism	{ "login": "xrennvidia", "id": 103958965, "node_id": "U_kgDOBjJJtQ", "avatar_url": "https://avatars.githubusercontent.com/u/103958965?v=4", "gravatar_id": "", "url": "https://api.github.com/users/xrennvidia", "html_url": "https://github.com/xrennvidia", "followers_url": "https://api.github.com/users/xrennvidia/followers", "following_url": "https://api.github.com/users/xrennvidia/following{/other_user}", "gists_url": "https://api.github.com/users/xrennvidia/gists{/gist_id}", "starred_url": "https://api.github.com/users/xrennvidia/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/xrennvidia/subscriptions", "organizations_url": "https://api.github.com/users/xrennvidia/orgs", "repos_url": "https://api.github.com/users/xrennvidia/repos", "events_url": "https://api.github.com/users/xrennvidia/events{/privacy}", "received_events_url": "https://api.github.com/users/xrennvidia/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 1923495043, "node_id": "MDU6TGFiZWwxOTIzNDk1MDQz", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/core", "name": "core", "color": "f25e3a", "default": false, "description": "Changes to NeMo Core" }, { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" }, { "id": 4840216855, "node_id": "LA_kwDOC_bI7s8AAAABIH_ZFw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/CI", "name": "CI", "color": "1D76DB", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-08-06T01:24:12	2023-10-17T18:13:43	2023-10-17T18:13:02	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7171", "html_url": "https://github.com/NVIDIA/NeMo/pull/7171", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7171.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7171.patch", "merged_at": null }	# What does this PR do ? GPT training with long-context input (e.g., sequence length of 16, 32K, 64K) can easily overflow GPU memory with huge activations. Context parallelism splits long-context input along the dimension of sequence length, parallelizes partitioned sequence segments among multiple GPUs. In this way, each GPU only needs to store activations of a part of sequence length, so that we can avoid memory overflow. # Usage Add a new config in yaml file, e.g., "context_parallel_size: 2". # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "xrennvidia", "id": 103958965, "node_id": "U_kgDOBjJJtQ", "avatar_url": "https://avatars.githubusercontent.com/u/103958965?v=4", "gravatar_id": "", "url": "https://api.github.com/users/xrennvidia", "html_url": "https://github.com/xrennvidia", "followers_url": "https://api.github.com/users/xrennvidia/followers", "following_url": "https://api.github.com/users/xrennvidia/following{/other_user}", "gists_url": "https://api.github.com/users/xrennvidia/gists{/gist_id}", "starred_url": "https://api.github.com/users/xrennvidia/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/xrennvidia/subscriptions", "organizations_url": "https://api.github.com/users/xrennvidia/orgs", "repos_url": "https://api.github.com/users/xrennvidia/repos", "events_url": "https://api.github.com/users/xrennvidia/events{/privacy}", "received_events_url": "https://api.github.com/users/xrennvidia/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7171/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7171/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7168	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7168/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7168/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7168/events	https://github.com/NVIDIA/NeMo/pull/7168	1,837,462,907	PR_kwDOC_bI7s5XPFDl	7,168	initial scripts	{ "login": "nithinraok", "id": 19668129, "node_id": "MDQ6VXNlcjE5NjY4MTI5", "avatar_url": "https://avatars.githubusercontent.com/u/19668129?v=4", "gravatar_id": "", "url": "https://api.github.com/users/nithinraok", "html_url": "https://github.com/nithinraok", "followers_url": "https://api.github.com/users/nithinraok/followers", "following_url": "https://api.github.com/users/nithinraok/following{/other_user}", "gists_url": "https://api.github.com/users/nithinraok/gists{/gist_id}", "starred_url": "https://api.github.com/users/nithinraok/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/nithinraok/subscriptions", "organizations_url": "https://api.github.com/users/nithinraok/orgs", "repos_url": "https://api.github.com/users/nithinraok/repos", "events_url": "https://api.github.com/users/nithinraok/events{/privacy}", "received_events_url": "https://api.github.com/users/nithinraok/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 1923495043, "node_id": "MDU6TGFiZWwxOTIzNDk1MDQz", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/core", "name": "core", "color": "f25e3a", "default": false, "description": "Changes to NeMo Core" }, { "id": 4418253970, "node_id": "LA_kwDOC_bI7s8AAAABB1k0kg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/TTS", "name": "TTS", "color": "49AF28", "default": false, "description": "" }, { "id": 4811884691, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR", "name": "ASR", "color": "F3B33E", "default": false, "description": "" }, { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" }, { "id": 4840216855, "node_id": "LA_kwDOC_bI7s8AAAABIH_ZFw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/CI", "name": "CI", "color": "1D76DB", "default": false, "description": "" }, { "id": 4847373924, "node_id": "LA_kwDOC_bI7s8AAAABIO0OZA", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/common", "name": "common", "color": "c5def5", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-08-05T00:12:56	2023-08-18T00:20:42	2023-08-14T23:33:21	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	true	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7168", "html_url": "https://github.com/NVIDIA/NeMo/pull/7168", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7168.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7168.patch", "merged_at": null }	# 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 ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "nithinraok", "id": 19668129, "node_id": "MDQ6VXNlcjE5NjY4MTI5", "avatar_url": "https://avatars.githubusercontent.com/u/19668129?v=4", "gravatar_id": "", "url": "https://api.github.com/users/nithinraok", "html_url": "https://github.com/nithinraok", "followers_url": "https://api.github.com/users/nithinraok/followers", "following_url": "https://api.github.com/users/nithinraok/following{/other_user}", "gists_url": "https://api.github.com/users/nithinraok/gists{/gist_id}", "starred_url": "https://api.github.com/users/nithinraok/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/nithinraok/subscriptions", "organizations_url": "https://api.github.com/users/nithinraok/orgs", "repos_url": "https://api.github.com/users/nithinraok/repos", "events_url": "https://api.github.com/users/nithinraok/events{/privacy}", "received_events_url": "https://api.github.com/users/nithinraok/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7168/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7168/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7167	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7167/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7167/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7167/events	https://github.com/NVIDIA/NeMo/pull/7167	1,837,330,909	PR_kwDOC_bI7s5XOoag	7,167	Merge release r1.20.0 to main	{ "login": "ericharper", "id": 11999610, "node_id": "MDQ6VXNlcjExOTk5NjEw", "avatar_url": "https://avatars.githubusercontent.com/u/11999610?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ericharper", "html_url": "https://github.com/ericharper", "followers_url": "https://api.github.com/users/ericharper/followers", "following_url": "https://api.github.com/users/ericharper/following{/other_user}", "gists_url": "https://api.github.com/users/ericharper/gists{/gist_id}", "starred_url": "https://api.github.com/users/ericharper/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ericharper/subscriptions", "organizations_url": "https://api.github.com/users/ericharper/orgs", "repos_url": "https://api.github.com/users/ericharper/repos", "events_url": "https://api.github.com/users/ericharper/events{/privacy}", "received_events_url": "https://api.github.com/users/ericharper/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4811884691, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR", "name": "ASR", "color": "F3B33E", "default": false, "description": "" }, { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" }, { "id": 4840216855, "node_id": "LA_kwDOC_bI7s8AAAABIH_ZFw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/CI", "name": "CI", "color": "1D76DB", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-08-04T20:57:50	2023-08-05T05:28:11	2023-08-05T05:28:10	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7167", "html_url": "https://github.com/NVIDIA/NeMo/pull/7167", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7167.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7167.patch", "merged_at": "2023-08-05T05:28:10" }	# What does this PR do ? Make sure all bugfixes and doc updates were properly cherry picked to main. 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 ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "tango4j", "id": 13211483, "node_id": "MDQ6VXNlcjEzMjExNDgz", "avatar_url": "https://avatars.githubusercontent.com/u/13211483?v=4", "gravatar_id": "", "url": "https://api.github.com/users/tango4j", "html_url": "https://github.com/tango4j", "followers_url": "https://api.github.com/users/tango4j/followers", "following_url": "https://api.github.com/users/tango4j/following{/other_user}", "gists_url": "https://api.github.com/users/tango4j/gists{/gist_id}", "starred_url": "https://api.github.com/users/tango4j/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/tango4j/subscriptions", "organizations_url": "https://api.github.com/users/tango4j/orgs", "repos_url": "https://api.github.com/users/tango4j/repos", "events_url": "https://api.github.com/users/tango4j/events{/privacy}", "received_events_url": "https://api.github.com/users/tango4j/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7167/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7167/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7166	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7166/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7166/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7166/events	https://github.com/NVIDIA/NeMo/issues/7166	1,836,660,080	I_kwDOC_bI7s5teTVw	7,166	Ubuntu 22.04 Python 3.11 [asr]: multiple errors `dataclasses ValueError: mutable default * for field * is not allowed: use default_factory`	{ "login": "alllexx88", "id": 6587516, "node_id": "MDQ6VXNlcjY1ODc1MTY=", "avatar_url": "https://avatars.githubusercontent.com/u/6587516?v=4", "gravatar_id": "", "url": "https://api.github.com/users/alllexx88", "html_url": "https://github.com/alllexx88", "followers_url": "https://api.github.com/users/alllexx88/followers", "following_url": "https://api.github.com/users/alllexx88/following{/other_user}", "gists_url": "https://api.github.com/users/alllexx88/gists{/gist_id}", "starred_url": "https://api.github.com/users/alllexx88/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/alllexx88/subscriptions", "organizations_url": "https://api.github.com/users/alllexx88/orgs", "repos_url": "https://api.github.com/users/alllexx88/repos", "events_url": "https://api.github.com/users/alllexx88/events{/privacy}", "received_events_url": "https://api.github.com/users/alllexx88/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 1485815557, "node_id": "MDU6TGFiZWwxNDg1ODE1NTU3", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/bug", "name": "bug", "color": "d73a4a", "default": true, "description": "Something isn't working" } ]	closed	false	{ "login": "titu1994", "id": 3048602, "node_id": "MDQ6VXNlcjMwNDg2MDI=", "avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4", "gravatar_id": "", "url": "https://api.github.com/users/titu1994", "html_url": "https://github.com/titu1994", "followers_url": "https://api.github.com/users/titu1994/followers", "following_url": "https://api.github.com/users/titu1994/following{/other_user}", "gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}", "starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/titu1994/subscriptions", "organizations_url": "https://api.github.com/users/titu1994/orgs", "repos_url": "https://api.github.com/users/titu1994/repos", "events_url": "https://api.github.com/users/titu1994/events{/privacy}", "received_events_url": "https://api.github.com/users/titu1994/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "login": "titu1994", "id": 3048602, "node_id": "MDQ6VXNlcjMwNDg2MDI=", "avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4", "gravatar_id": "", "url": "https://api.github.com/users/titu1994", "html_url": "https://github.com/titu1994", "followers_url": "https://api.github.com/users/titu1994/followers", "following_url": "https://api.github.com/users/titu1994/following{/other_user}", "gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}", "starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/titu1994/subscriptions", "organizations_url": "https://api.github.com/users/titu1994/orgs", "repos_url": "https://api.github.com/users/titu1994/repos", "events_url": "https://api.github.com/users/titu1994/events{/privacy}", "received_events_url": "https://api.github.com/users/titu1994/received_events", "type": "User", "user_view_type": "public", "site_admin": false } ]	null	null	2023-08-04T12:28:02	2024-05-16T23:49:50	2023-10-05T17:18:33	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	Describe the bug After installing latest stable `1.19.1` from pipI, or the latest current commit with `[asr]` extras, I'm getting this error from `hydra-core==1.2.0` when trying to import `nemo.collections.asr`: ``` Traceback (most recent call last): File "/home/alex/T7/src/speaker_verification/nemo_speaker/test.py", line 5, in <module> import nemo.collections.asr as nemo_asr File "/home/alex/T7/src/speaker_verification/nemo_speaker/venv11/lib/python3.11/site-packages/nemo/collections/asr/__init__.py", line 15, in <module> from nemo.collections.asr import data, losses, models, modules File "/home/alex/T7/src/speaker_verification/nemo_speaker/venv11/lib/python3.11/site-packages/nemo/collections/asr/losses/__init__.py", line 15, in <module> from nemo.collections.asr.losses.angularloss import AngularSoftmaxLoss File "/home/alex/T7/src/speaker_verification/nemo_speaker/venv11/lib/python3.11/site-packages/nemo/collections/asr/losses/angularloss.py", line 18, in <module> from nemo.core.classes import Loss, Typing, typecheck File "/home/alex/T7/src/speaker_verification/nemo_speaker/venv11/lib/python3.11/site-packages/nemo/core/__init__.py", line 16, in <module> from nemo.core.classes import * File "/home/alex/T7/src/speaker_verification/nemo_speaker/venv11/lib/python3.11/site-packages/nemo/core/classes/__init__.py", line 16, in <module> import hydra File "/home/alex/T7/src/speaker_verification/nemo_speaker/venv11/lib/python3.11/site-packages/hydra/__init__.py", line 5, in <module> from hydra import utils File "/home/alex/T7/src/speaker_verification/nemo_speaker/venv11/lib/python3.11/site-packages/hydra/utils.py", line 8, in <module> import hydra._internal.instantiate._instantiate2 File "/home/alex/T7/src/speaker_verification/nemo_speaker/venv11/lib/python3.11/site-packages/hydra/_internal/instantiate/_instantiate2.py", line 12, in <module> from hydra._internal.utils import _locate File "/home/alex/T7/src/speaker_verification/nemo_speaker/venv11/lib/python3.11/site-packages/hydra/_internal/utils.py", line 18, in <module> from hydra.core.utils import get_valid_filename, validate_config_path File "/home/alex/T7/src/speaker_verification/nemo_speaker/venv11/lib/python3.11/site-packages/hydra/core/utils.py", line 20, in <module> from hydra.core.hydra_config import HydraConfig File "/home/alex/T7/src/speaker_verification/nemo_speaker/venv11/lib/python3.11/site-packages/hydra/core/hydra_config.py", line 6, in <module> from hydra.conf import HydraConf File "/home/alex/T7/src/speaker_verification/nemo_speaker/venv11/lib/python3.11/site-packages/hydra/conf/__init__.py", line 46, in <module> class JobConf: File "/home/alex/T7/src/speaker_verification/nemo_speaker/venv11/lib/python3.11/site-packages/hydra/conf/__init__.py", line 75, in JobConf @dataclass ^^^^^^^^^ File "/usr/lib/python3.11/dataclasses.py", line 1230, in dataclass return wrap(cls) ^^^^^^^^^ File "/usr/lib/python3.11/dataclasses.py", line 1220, in wrap return _process_class(cls, init, repr, eq, order, unsafe_hash, ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/lib/python3.11/dataclasses.py", line 958, in _process_class cls_fields.append(_get_field(cls, name, type, kw_only)) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/lib/python3.11/dataclasses.py", line 815, in _get_field raise ValueError(f'mutable default {type(f.default)} for field ' ValueError: mutable default <class 'hydra.conf.JobConf.JobConfig.OverrideDirname'> for field override_dirname is not allowed: use default_factory ``` If I then manually upgrade `hydra-core` to the current latest (`1.3.2`), I get similar errors from `nemo-toolkit`: ``` Traceback (most recent call last): File "/home/alex/T7/src/speaker_verification/nemo_speaker/test.py", line 5, in <module> import nemo.collections.asr as nemo_asr File "/home/alex/T7/src/speaker_verification/nemo_speaker/venv11/lib/python3.11/site-packages/nemo/collections/asr/__init__.py", line 15, in <module> from nemo.collections.asr import data, losses, models, modules File "/home/alex/T7/src/speaker_verification/nemo_speaker/venv11/lib/python3.11/site-packages/nemo/collections/asr/losses/__init__.py", line 16, in <module> from nemo.collections.asr.losses.audio_losses import SDRLoss File "/home/alex/T7/src/speaker_verification/nemo_speaker/venv11/lib/python3.11/site-packages/nemo/collections/asr/losses/audio_losses.py", line 21, in <module> from nemo.collections.asr.parts.preprocessing.features import make_seq_mask_like File "/home/alex/T7/src/speaker_verification/nemo_speaker/venv11/lib/python3.11/site-packages/nemo/collections/asr/parts/preprocessing/__init__.py", line 16, in <module> from nemo.collections.asr.parts.preprocessing.features import FeaturizerFactory, FilterbankFeatures, WaveformFeaturizer File "/home/alex/T7/src/speaker_verification/nemo_speaker/venv11/lib/python3.11/site-packages/nemo/collections/asr/parts/preprocessing/features.py", line 44, in <module> from nemo.collections.asr.parts.preprocessing.perturb import AudioAugmentor File "/home/alex/T7/src/speaker_verification/nemo_speaker/venv11/lib/python3.11/site-packages/nemo/collections/asr/parts/preprocessing/perturb.py", line 50, in <module> from nemo.collections.common.parts.preprocessing import collections, parsers File "/home/alex/T7/src/speaker_verification/nemo_speaker/venv11/lib/python3.11/site-packages/nemo/collections/common/__init__.py", line 16, in <module> from nemo.collections.common import data, losses, parts, tokenizers File "/home/alex/T7/src/speaker_verification/nemo_speaker/venv11/lib/python3.11/site-packages/nemo/collections/common/parts/__init__.py", line 15, in <module> from nemo.collections.common.parts.adapter_modules import LinearAdapter, LinearAdapterConfig File "/home/alex/T7/src/speaker_verification/nemo_speaker/venv11/lib/python3.11/site-packages/nemo/collections/common/parts/adapter_modules.py", line 147, in <module> @dataclass ^^^^^^^^^ File "/usr/lib/python3.11/dataclasses.py", line 1230, in dataclass return wrap(cls) ^^^^^^^^^ File "/usr/lib/python3.11/dataclasses.py", line 1220, in wrap return _process_class(cls, init, repr, eq, order, unsafe_hash, ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/lib/python3.11/dataclasses.py", line 958, in _process_class cls_fields.append(_get_field(cls, name, type, kw_only)) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/lib/python3.11/dataclasses.py", line 815, in _get_field raise ValueError(f'mutable default {type(f.default)} for field ' ValueError: mutable default <class 'nemo.core.classes.mixins.adapter_mixin_strategies.ResidualAddAdapterStrategyConfig'> for field adapter_strategy is not allowed: use default_factory ``` It's easy to fix with a patch like this: ``` --- nemo/collections/common/parts/adapter_modules.py.orig 2023-08-04 13:55:53.464534800 +0200 +++ nemo/collections/common/parts/adapter_modules.py 2023-08-04 14:05:45.579537700 +0200 @@ -12,7 +12,7 @@ # See the License for the specific language governing permissions and # limitations under the License. -from dataclasses import dataclass, is_dataclass +from dataclasses import dataclass, is_dataclass, field from typing import Any, Optional from hydra.utils import instantiate @@ -151,5 +151,5 @@ class LinearAdapterConfig: activation: str = 'swish' norm_position: str = 'pre' dropout: float = 0.0 - adapter_strategy: Optional[Any] = adapter_mixin_strategies.ResidualAddAdapterStrategyConfig() + adapter_strategy: Optional[Any] = field(default_factory=adapter_mixin_strategies.ResidualAddAdapterStrategyConfig) _target_: str = "{0}.{1}".format(LinearAdapter.__module__, LinearAdapter.__name__) ``` However then another error of the kind comes up: ``` Traceback (most recent call last): File "/home/alex/T7/src/speaker_verification/nemo_speaker/test.py", line 5, in <module> import nemo.collections.asr as nemo_asr File "/home/alex/T7/src/speaker_verification/nemo_speaker/venv11/lib/python3.11/site-packages/nemo/collections/asr/__init__.py", line 15, in <module> from nemo.collections.asr import data, losses, models, modules File "/home/alex/T7/src/speaker_verification/nemo_speaker/venv11/lib/python3.11/site-packages/nemo/collections/asr/models/__init__.py", line 18, in <module> from nemo.collections.asr.models.clustering_diarizer import ClusteringDiarizer File "/home/alex/T7/src/speaker_verification/nemo_speaker/venv11/lib/python3.11/site-packages/nemo/collections/asr/models/clustering_diarizer.py", line 29, in <module> from nemo.collections.asr.metrics.der import score_labels File "/home/alex/T7/src/speaker_verification/nemo_speaker/venv11/lib/python3.11/site-packages/nemo/collections/asr/metrics/der.py", line 24, in <module> from nemo.collections.asr.metrics.wer import word_error_rate File "/home/alex/T7/src/speaker_verification/nemo_speaker/venv11/lib/python3.11/site-packages/nemo/collections/asr/metrics/wer.py", line 27, in <module> from nemo.collections.asr.parts.submodules import ctc_beam_decoding, ctc_greedy_decoding File "/home/alex/T7/src/speaker_verification/nemo_speaker/venv11/lib/python3.11/site-packages/nemo/collections/asr/parts/submodules/ctc_beam_decoding.py", line 593, in <module> @dataclass ^^^^^^^^^ File "/usr/lib/python3.11/dataclasses.py", line 1230, in dataclass return wrap(cls) ^^^^^^^^^ File "/usr/lib/python3.11/dataclasses.py", line 1220, in wrap return _process_class(cls, init, repr, eq, order, unsafe_hash, ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/lib/python3.11/dataclasses.py", line 958, in _process_class cls_fields.append(_get_field(cls, name, type, kw_only)) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/lib/python3.11/dataclasses.py", line 815, in _get_field raise ValueError(f'mutable default {type(f.default)} for field ' ValueError: mutable default <class 'nemo.collections.asr.parts.submodules.ctc_beam_decoding.FlashlightConfig'> for field flashlight_cfg is not allowed: use default_factory ``` This can also be fixed accordingly, but all in all, it appears these issues are pretty common in the code base. Looks like NeMo isn't Python 3.11 ready, at least the `asr` collection. Steps/Code to reproduce bug With Python 3.11 1. Install `nemo-toolkit[asr]` either 1.19.1, or current HEAD from git (b498d438fc4c35ebf364a9a1c5cd3e29a2c0fe50) 2. Run `import nemo.collections.asr` Expected behavior Import `nemo.collections.asr` without errors Environment overview (please complete the following information) - Environment location: Bare-metal - Method of NeMo install: `pip install nemo-toolkit==1.19.1` or from source b498d438fc4c35ebf364a9a1c5cd3e29a2c0fe50 Environment details - Ubuntu 22.04 - PyTorch version 2.0.1 - Python version 3.11.4	{ "login": "titu1994", "id": 3048602, "node_id": "MDQ6VXNlcjMwNDg2MDI=", "avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4", "gravatar_id": "", "url": "https://api.github.com/users/titu1994", "html_url": "https://github.com/titu1994", "followers_url": "https://api.github.com/users/titu1994/followers", "following_url": "https://api.github.com/users/titu1994/following{/other_user}", "gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}", "starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/titu1994/subscriptions", "organizations_url": "https://api.github.com/users/titu1994/orgs", "repos_url": "https://api.github.com/users/titu1994/repos", "events_url": "https://api.github.com/users/titu1994/events{/privacy}", "received_events_url": "https://api.github.com/users/titu1994/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7166/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7166/timeline	null	completed	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7165	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7165/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7165/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7165/events	https://github.com/NVIDIA/NeMo/issues/7165	1,836,459,005	I_kwDOC_bI7s5tdiP9	7,165	Cannot fine-tune a Hifigan model	{ "login": "Samebodyonce", "id": 38726684, "node_id": "MDQ6VXNlcjM4NzI2Njg0", "avatar_url": "https://avatars.githubusercontent.com/u/38726684?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Samebodyonce", "html_url": "https://github.com/Samebodyonce", "followers_url": "https://api.github.com/users/Samebodyonce/followers", "following_url": "https://api.github.com/users/Samebodyonce/following{/other_user}", "gists_url": "https://api.github.com/users/Samebodyonce/gists{/gist_id}", "starred_url": "https://api.github.com/users/Samebodyonce/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Samebodyonce/subscriptions", "organizations_url": "https://api.github.com/users/Samebodyonce/orgs", "repos_url": "https://api.github.com/users/Samebodyonce/repos", "events_url": "https://api.github.com/users/Samebodyonce/events{/privacy}", "received_events_url": "https://api.github.com/users/Samebodyonce/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 1485815557, "node_id": "MDU6TGFiZWwxNDg1ODE1NTU3", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/bug", "name": "bug", "color": "d73a4a", "default": true, "description": "Something isn't working" } ]	closed	false	null	[]	null	null	2023-08-04T10:10:32	2023-08-07T10:57:44	2023-08-07T10:57:44	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	My code looks like: python /workspaces/TTS/nemo/examples/tts/hifigan_finetune.py --config-path /workspaces/TTS/nemo/examples/tts/conf/hifigan --config-name hifigan.yaml model.max_steps=200000 model.optim.lr=0.00001 ~model.optim.sched train_dataset=/workspaces/TTS/data/train_manifest_mel_test.json validation_datasets=/workspaces/TTS/data/val_manifest_mel_test.json exp_manager.exp_dir=experimentsHifigan +init_from_pretrained_model=tts_en_hifigan +trainer.val_check_interval=100 trainer.check_val_every_n_epoch=null model/train_ds=train_ds_finetune model/validation_ds=val_ds_finetune exp_manager.resume_if_exists=false And I get: [NeMo W 2023-08-04 10:01:13 optimizers:54] Apex was not found. Using the lamb or fused_adam optimizer will error out. [NeMo W 2023-08-04 10:01:13 experimental:27] Module <class 'nemo.collections.asr.modules.audio_modules.SpectrogramToMultichannelFeatures'> is experimental, not ready for production and is not fully supported. Use at your own risk. [NeMo W 2023-08-04 10:01:13 experimental:27] Module <class 'nemo.collections.tts.models.fastpitch_ssl.FastPitchModel_SSL'> is experimental, not ready for production and is not fully supported. Use at your own risk. [NeMo W 2023-08-04 10:01:13 experimental:27] Module <class 'nemo.collections.common.tokenizers.text_to_speech.tts_tokenizers.IPATokenizer'> is experimental, not ready for production and is not fully supported. Use at your own risk. [NeMo W 2023-08-04 10:01:13 experimental:27] Module <class 'nemo.collections.tts.models.radtts.RadTTSModel'> is experimental, not ready for production and is not fully supported. Use at your own risk. [NeMo W 2023-08-04 10:01:14 experimental:27] Module <class 'nemo.collections.tts.models.ssl_tts.SSLDisentangler'> is experimental, not ready for production and is not fully supported. Use at your own risk. [NeMo W 2023-08-04 10:01:14 experimental:27] Module <class 'nemo.collections.tts.models.vits.VitsModel'> is experimental, not ready for production and is not fully supported. Use at your own risk. [NeMo W 2023-08-04 10:01:14 nemo_logging:349] /opt/conda/envs/nemo/lib/python3.8/site-packages/hydra/_internal/defaults_list.py:251: UserWarning: In 'hifigan.yaml': Defaults list is missing `_self_`. See https://hydra.cc/docs/upgrades/1.0_to_1.1/default_composition_order for more information warnings.warn(msg, UserWarning) [NeMo W 2023-08-04 10:01:14 nemo_logging:349] /opt/conda/envs/nemo/lib/python3.8/site-packages/hydra/_internal/hydra.py:119: UserWarning: Future Hydra versions will no longer change working directory at job runtime by default. See https://hydra.cc/docs/next/upgrades/1.1_to_1.2/changes_to_job_working_dir/ for more information. ret = run_job( GPU available: True (cuda), used: True TPU available: False, using: 0 TPU cores IPU available: False, using: 0 IPUs HPU available: False, using: 0 HPUs [NeMo I 2023-08-04 10:01:14 exp_manager:374] Experiments will be logged at experimentsHifigan/HifiGan/2023-08-04_10-01-14 [NeMo I 2023-08-04 10:01:14 exp_manager:797] TensorboardLogger has been set up [NeMo W 2023-08-04 10:01:14 exp_manager:893] The checkpoint callback was told to monitor a validation value and trainer's max_steps was set to 200000. Please ensure that max_steps will run for at least 1 epochs to ensure that checkpointing will not error out. [NeMo I 2023-08-04 10:01:14 dataset:1068] Loading dataset from /root/../workspaces/TTS/data/train_manifest_mel_test.json. 20869it [00:00, 213363.10it/s] [NeMo I 2023-08-04 10:01:14 dataset:1092] Loaded dataset with 20869 files. [NeMo I 2023-08-04 10:01:14 dataset:1094] Dataset contains 29.45 hours. [NeMo I 2023-08-04 10:01:14 dataset:376] Pruned 0 files. Final dataset contains 20869 files [NeMo I 2023-08-04 10:01:14 dataset:378] Pruned 0.00 hours. Final dataset contains 29.45 hours. [NeMo I 2023-08-04 10:01:14 dataset:1068] Loading dataset from /root/../workspaces/TTS/data/val_manifest_mel_test.json. 460it [00:00, 199419.10it/s] [NeMo I 2023-08-04 10:01:14 dataset:1092] Loaded dataset with 460 files. [NeMo I 2023-08-04 10:01:14 dataset:1094] Dataset contains 0.76 hours. [NeMo I 2023-08-04 10:01:14 dataset:376] Pruned 13 files. Final dataset contains 447 files [NeMo I 2023-08-04 10:01:14 dataset:378] Pruned 0.01 hours. Final dataset contains 0.76 hours. [NeMo I 2023-08-04 10:01:14 features:291] PADDING: 0 [NeMo I 2023-08-04 10:01:14 features:299] STFT using exact pad [NeMo I 2023-08-04 10:01:14 features:291] PADDING: 0 [NeMo I 2023-08-04 10:01:14 features:299] STFT using exact pad [NeMo I 2023-08-04 10:01:15 cloud:58] Found existing object /root/.cache/torch/NeMo/NeMo_1.19.0/tts_hifigan/e6da322f0f7e7dcf3f1900a9229a7e69/tts_hifigan.nemo. [NeMo I 2023-08-04 10:01:15 cloud:64] Re-using file from: /root/.cache/torch/NeMo/NeMo_1.19.0/tts_hifigan/e6da322f0f7e7dcf3f1900a9229a7e69/tts_hifigan.nemo [NeMo I 2023-08-04 10:01:15 common:913] Instantiating model from pre-trained checkpoint [NeMo W 2023-08-04 10:01:17 modelPT:161] If you intend to do training or fine-tuning, please call the ModelPT.setup_training_data() method and provide a valid configuration file to setup the train data loader. Train config : dataset: _target_: nemo.collections.tts.data.datalayers.MelAudioDataset manifest_filepath: /home/fkreuk/data/train_finetune.txt min_duration: 0.75 n_segments: 8192 dataloader_params: drop_last: false shuffle: true batch_size: 64 num_workers: 4 [NeMo W 2023-08-04 10:01:17 modelPT:168] If you intend to do validation, please call the ModelPT.setup_validation_data() or ModelPT.setup_multiple_validation_data() method and provide a valid configuration file to setup the validation data loader(s). Validation config : dataset: _target_: nemo.collections.tts.data.datalayers.MelAudioDataset manifest_filepath: /home/fkreuk/data/val_finetune.txt min_duration: 3 n_segments: 66150 dataloader_params: drop_last: false shuffle: false batch_size: 5 num_workers: 4 [NeMo W 2023-08-04 10:01:17 features:268] Using torch_stft is deprecated and has been removed. The values have been forcibly set to False for FilterbankFeatures and AudioToMelSpectrogramPreprocessor. Please set exact_pad to True as needed. [NeMo I 2023-08-04 10:01:17 features:291] PADDING: 0 [NeMo W 2023-08-04 10:01:17 features:268] Using torch_stft is deprecated and has been removed. The values have been forcibly set to False for FilterbankFeatures and AudioToMelSpectrogramPreprocessor. Please set exact_pad to True as needed. [NeMo I 2023-08-04 10:01:17 features:291] PADDING: 0 [NeMo I 2023-08-04 10:01:18 save_restore_connector:249] Model HifiGanModel was successfully restored from /root/.cache/torch/NeMo/NeMo_1.19.0/tts_hifigan/e6da322f0f7e7dcf3f1900a9229a7e69/tts_hifigan.nemo. [NeMo I 2023-08-04 10:01:18 modelPT:1268] Model checkpoint restored from pretrained checkpoint with name : `tts_en_hifigan` Initializing distributed: GLOBAL_RANK: 0, MEMBER: 1/1 Added key: store_based_barrier_key:1 to store for rank: 0 Rank 0: Completed store-based barrier for key:store_based_barrier_key:1 with 1 nodes. ---------------------------------------------------------------------------------------------------- distributed_backend=nccl All distributed processes registered. Starting with 1 processes ---------------------------------------------------------------------------------------------------- LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [1] \| Name \| Type \| Params ------------------------------------------------------------------------ 0 \| audio_to_melspec_precessor \| FilterbankFeatures \| 0 1 \| trg_melspec_fn \| FilterbankFeatures \| 0 2 \| generator \| Generator \| 13.9 M 3 \| mpd \| MultiPeriodDiscriminator \| 41.1 M 4 \| msd \| MultiScaleDiscriminator \| 29.6 M 5 \| feature_loss \| FeatureMatchingLoss \| 0 6 \| discriminator_loss \| DiscriminatorLoss \| 0 7 \| generator_loss \| GeneratorLoss \| 0 ------------------------------------------------------------------------ 84.7 M Trainable params 0 Non-trainable params 84.7 M Total params 338.643 Total estimated model params size (MB) Sanity Checking: 0it [00:00, ?it/s]Error executing job with overrides: ['model.max_steps=200000', 'model.optim.lr=0.00001', '~model.optim.sched', 'train_dataset=/root/../workspaces/TTS/data/train_manifest_mel_test.json', 'validation_datasets=/root/../workspaces/TTS/data/val_manifest_mel_test.json', 'exp_manager.exp_dir=experimentsHifigan', '+init_from_pretrained_model=tts_en_hifigan', '+trainer.val_check_interval=100', 'trainer.check_val_every_n_epoch=1', 'model/train_ds=train_ds_finetune', 'model/validation_ds=val_ds_finetune', 'model.train_ds.dataloader_params.batch_size=16', 'exp_manager.resume_if_exists=false'] Traceback (most recent call last): File "/root/../workspaces/TTS/nemo/examples/tts/hifigan_finetune.py", line 32, in <module> main() # noqa pylint: disable=no-value-for-parameter File "/opt/conda/envs/nemo/lib/python3.8/site-packages/nemo/core/config/hydra_runner.py", line 105, in wrapper _run_hydra( File "/opt/conda/envs/nemo/lib/python3.8/site-packages/hydra/_internal/utils.py", line 389, in _run_hydra _run_app( File "/opt/conda/envs/nemo/lib/python3.8/site-packages/hydra/_internal/utils.py", line 452, in _run_app run_and_report( File "/opt/conda/envs/nemo/lib/python3.8/site-packages/hydra/_internal/utils.py", line 216, in run_and_report raise ex File "/opt/conda/envs/nemo/lib/python3.8/site-packages/hydra/_internal/utils.py", line 213, in run_and_report return func() File "/opt/conda/envs/nemo/lib/python3.8/site-packages/hydra/_internal/utils.py", line 453, in <lambda> lambda: hydra.run( File "/opt/conda/envs/nemo/lib/python3.8/site-packages/hydra/_internal/hydra.py", line 132, in run _ = ret.return_value File "/opt/conda/envs/nemo/lib/python3.8/site-packages/hydra/core/utils.py", line 260, in return_value raise self._return_value File "/opt/conda/envs/nemo/lib/python3.8/site-packages/hydra/core/utils.py", line 186, in run_job ret.return_value = task_function(task_cfg) File "/root/../workspaces/TTS/nemo/examples/tts/hifigan_finetune.py", line 28, in main trainer.fit(model) File "/opt/conda/envs/nemo/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 608, in fit call._call_and_handle_interrupt( File "/opt/conda/envs/nemo/lib/python3.8/site-packages/pytorch_lightning/trainer/call.py", line 36, in _call_and_handle_interrupt return trainer.strategy.launcher.launch(trainer_fn, args, trainer=trainer, kwargs) File "/opt/conda/envs/nemo/lib/python3.8/site-packages/pytorch_lightning/strategies/launchers/subprocess_script.py", line 88, in launch return function(args, *kwargs) File "/opt/conda/envs/nemo/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 650, in _fit_impl self._run(model, ckpt_path=self.ckpt_path) File "/opt/conda/envs/nemo/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 1112, in _run results = self._run_stage() File "/opt/conda/envs/nemo/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 1191, in _run_stage self._run_train() File "/opt/conda/envs/nemo/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 1204, in _run_train self._run_sanity_check() File "/opt/conda/envs/nemo/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 1276, in _run_sanity_check val_loop.run() File "/opt/conda/envs/nemo/lib/python3.8/site-packages/pytorch_lightning/loops/loop.py", line 199, in run self.advance(args, *kwargs) File "/opt/conda/envs/nemo/lib/python3.8/site-packages/pytorch_lightning/loops/dataloader/evaluation_loop.py", line 152, in advance dl_outputs = self.epoch_loop.run(self._data_fetcher, dl_max_batches, kwargs) File "/opt/conda/envs/nemo/lib/python3.8/site-packages/pytorch_lightning/loops/loop.py", line 199, in run self.advance(args, **kwargs) File "/opt/conda/envs/nemo/lib/python3.8/site-packages/pytorch_lightning/loops/epoch/evaluation_epoch_loop.py", line 121, in advance batch = next(data_fetcher) File "/opt/conda/envs/nemo/lib/python3.8/site-packages/pytorch_lightning/utilities/fetching.py", line 184, in __next__ return self.fetching_function() File "/opt/conda/envs/nemo/lib/python3.8/site-packages/pytorch_lightning/utilities/fetching.py", line 265, in fetching_function self._fetch_next_batch(self.dataloader_iter) File "/opt/conda/envs/nemo/lib/python3.8/site-packages/pytorch_lightning/utilities/fetching.py", line 280, in _fetch_next_batch batch = next(iterator) File "/opt/conda/envs/nemo/lib/python3.8/site-packages/torch/utils/data/dataloader.py", line 633, in __next__ data = self._next_data() File "/opt/conda/envs/nemo/lib/python3.8/site-packages/torch/utils/data/dataloader.py", line 1345, in _next_data return self._process_data(data) File "/opt/conda/envs/nemo/lib/python3.8/site-packages/torch/utils/data/dataloader.py", line 1371, in _process_data data.reraise() File "/opt/conda/envs/nemo/lib/python3.8/site-packages/torch/_utils.py", line 644, in reraise raise exception ValueError: Caught ValueError in DataLoader worker process 0. Original Traceback (most recent call last): File "/opt/conda/envs/nemo/lib/python3.8/site-packages/torch/utils/data/_utils/worker.py", line 308, in _worker_loop data = fetcher.fetch(index) File "/opt/conda/envs/nemo/lib/python3.8/site-packages/torch/utils/data/_utils/fetch.py", line 51, in fetch data = [self.dataset[idx] for idx in possibly_batched_index] File "/opt/conda/envs/nemo/lib/python3.8/site-packages/torch/utils/data/_utils/fetch.py", line 51, in <listcomp> data = [self.dataset[idx] for idx in possibly_batched_index] File "/opt/conda/envs/nemo/lib/python3.8/site-packages/nemo/collections/tts/data/dataset.py", line 1144, in __getitem__ start = random.randint(0, mel.shape[1] - frames - 1) File "/opt/conda/envs/nemo/lib/python3.8/random.py", line 248, in randint return self.randrange(a, b+1) File "/opt/conda/envs/nemo/lib/python3.8/random.py", line 226, in randrange raise ValueError("empty range for randrange() (%d, %d, %d)" % (istart, istop, width)) ValueError: empty range for randrange() (0, -16, -16) I am using main branch of nemo.	{ "login": "Samebodyonce", "id": 38726684, "node_id": "MDQ6VXNlcjM4NzI2Njg0", "avatar_url": "https://avatars.githubusercontent.com/u/38726684?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Samebodyonce", "html_url": "https://github.com/Samebodyonce", "followers_url": "https://api.github.com/users/Samebodyonce/followers", "following_url": "https://api.github.com/users/Samebodyonce/following{/other_user}", "gists_url": "https://api.github.com/users/Samebodyonce/gists{/gist_id}", "starred_url": "https://api.github.com/users/Samebodyonce/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Samebodyonce/subscriptions", "organizations_url": "https://api.github.com/users/Samebodyonce/orgs", "repos_url": "https://api.github.com/users/Samebodyonce/repos", "events_url": "https://api.github.com/users/Samebodyonce/events{/privacy}", "received_events_url": "https://api.github.com/users/Samebodyonce/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7165/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7165/timeline	null	completed	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7162	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7162/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7162/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7162/events	https://github.com/NVIDIA/NeMo/issues/7162	1,835,969,566	I_kwDOC_bI7s5tbqwe	7,162	swiglu activation doesn't support sequence_parallel	{ "login": "flymark2010", "id": 8042816, "node_id": "MDQ6VXNlcjgwNDI4MTY=", "avatar_url": "https://avatars.githubusercontent.com/u/8042816?v=4", "gravatar_id": "", "url": "https://api.github.com/users/flymark2010", "html_url": "https://github.com/flymark2010", "followers_url": "https://api.github.com/users/flymark2010/followers", "following_url": "https://api.github.com/users/flymark2010/following{/other_user}", "gists_url": "https://api.github.com/users/flymark2010/gists{/gist_id}", "starred_url": "https://api.github.com/users/flymark2010/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/flymark2010/subscriptions", "organizations_url": "https://api.github.com/users/flymark2010/orgs", "repos_url": "https://api.github.com/users/flymark2010/repos", "events_url": "https://api.github.com/users/flymark2010/events{/privacy}", "received_events_url": "https://api.github.com/users/flymark2010/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 1485815557, "node_id": "MDU6TGFiZWwxNDg1ODE1NTU3", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/bug", "name": "bug", "color": "d73a4a", "default": true, "description": "Something isn't working" }, { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null } ]	closed	false	null	[]	null	null	2023-08-04T02:30:54	2023-09-11T01:44:26	2023-09-11T01:44:25	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	Describe the bug When changing activation to `swiglu` and enable `sequence_parallel`, the training crashes. Steps/Code to reproduce bug 1. Checkout out repo tag: `git checkout v1.19.1` 2. Config modifications: ```diff --- a/examples/nlp/language_modeling/conf/megatron_gpt_config.yaml +++ b/examples/nlp/language_modeling/conf/megatron_gpt_config.yaml @@ -51,7 +51,7 @@ model: virtual_pipeline_model_parallel_size: null # interleaved pipeline # model architecture - encoder_seq_length: 512 + encoder_seq_length: 8192 max_position_embeddings: ${.encoder_seq_length} num_layers: 12 hidden_size: 768 @@ -71,8 +71,8 @@ model: pre_process: True # add embedding post_process: True # add pooler persist_layer_norm: True # Use of persistent fused layer norm kernel. - bias: True # Whether to use bias terms in all weight matrices. - activation: 'gelu' # Options ['gelu', 'geglu', 'swiglu', 'reglu', 'squared-relu', 'fast-geglu', 'fast-swiglu', 'fast-reglu'] + bias: False # Whether to use bias terms in all weight matrices. + activation: 'swiglu' # Options ['gelu', 'geglu', 'swiglu', 'reglu', 'squared-relu', 'fast-geglu', 'fast-swiglu', 'fast-reglu'] headscale: False # Whether to learn extra parameters that scale the output of the each self-attention head. transformer_block_type: 'pre_ln' # Options ['pre_ln', 'post_ln', 'normformer'] openai_gelu: False # Use OpenAI's GELU instead of the default GeLU @@ -88,8 +88,8 @@ model: library: 'megatron' type: 'GPT2BPETokenizer' model: null - vocab_file: null - merge_file: null + vocab_file: /data/meta/gpt2/vocab.json + merge_file: /data/meta/gpt2/merges.txt delimiter: null # only used for tabular tokenizer sentencepiece_legacy: False # Legacy=True allows you to add special tokens to sentencepiece tokenizers. @@ -107,8 +107,8 @@ model: # Fusion grad_div_ar_fusion: True # Fuse grad division into torch.distributed.all_reduce. Only used with O2 and no pipeline parallelism.. gradient_accumulation_fusion: False # Fuse weight gradient accumulation to GEMMs. Only used with pipeline parallelism and O2. - bias_activation_fusion: True # Use a kernel that fuses the bias addition from weight matrices with the subsequent activation function. - bias_dropout_add_fusion: True # Use a kernel that fuses the bias addition, dropout and residual connection addition. + bias_activation_fusion: False # Use a kernel that fuses the bias addition from weight matrices with the subsequent activation function. + bias_dropout_add_fusion: False # Use a kernel that fuses the bias addition, dropout and residual connection addition. masked_softmax_fusion: True # Use a kernel that fuses the attention softmax with it's mask. get_attention_mask_from_fusion: True # When using fused softmax it will create the attention mask so we won't copy it to the pipeline stages. @@ -187,7 +187,9 @@ model: # Dictionary: can override from CLI "model.data.data_prefix"={"train":[1.0, /path/to/data], "validation":/path/to/data, "test":/path/to/test} # Or see example below: # "model.data.data_prefix: {train:[1.0,/path/to/data], validation:[/path/to/data], test:[/path/to/test]}" - data_prefix: ??? + data_prefix: + - 1.0 + - /data/pretrain/owt2_clean_v2/hfbpe_gpt_training_data_text_document index_mapping_dir: null # path to save index mapping .npy files, by default will save in the same location as data_prefix data_impl: mmap splits_string: 900,50,50 ``` 3. Run command: ``` python examples/nlp/language_modeling/megatron_gpt_pretraining.py --config-path=/workspace/project/NeMo-upstream/examples/nlp/language_modeling/conf --config-name=megatron_gpt_config trainer.devices=4 trainer.num_nodes=1 model.micro_batch_size=1 model.global_batch_size=8 model.tensor_model_parallel_size=4 model.pipeline_model_parallel_size=1 model.sequence_parallel=True ``` 4. The error trackback: ``` Traceback (most recent call last): File "/workspace/project/NeMo-upstream/examples/nlp/language_modeling/megatron_gpt_pretraining.py", line 86, in main trainer.fit(model) File "/usr/local/lib/python3.8/dist-packages/pytorch_lightning/trainer/trainer.py", line 608, in fit call._call_and_handle_interrupt( File "/usr/local/lib/python3.8/dist-packages/pytorch_lightning/trainer/call.py", line 38, in _call_and_handle_interrupt return trainer_fn(args, kwargs) File "/usr/local/lib/python3.8/dist-packages/pytorch_lightning/trainer/trainer.py", line 650, in _fit_impl self._run(model, ckpt_path=self.ckpt_path) File "/usr/local/lib/python3.8/dist-packages/pytorch_lightning/trainer/trainer.py", line 1112, in _run results = self._run_stage() File "/usr/local/lib/python3.8/dist-packages/pytorch_lightning/trainer/trainer.py", line 1191, in _run_stage self._run_train() File "/usr/local/lib/python3.8/dist-packages/pytorch_lightning/trainer/trainer.py", line 1214, in _run_train self.fit_loop.run() File "/usr/local/lib/python3.8/dist-packages/pytorch_lightning/loops/loop.py", line 199, in run self.advance(args, *kwargs) File "/usr/local/lib/python3.8/dist-packages/pytorch_lightning/loops/fit_loop.py", line 267, in advance self._outputs = self.epoch_loop.run(self._data_fetcher) File "/usr/local/lib/python3.8/dist-packages/pytorch_lightning/loops/loop.py", line 199, in run self.advance(args, *kwargs) File "/usr/local/lib/python3.8/dist-packages/pytorch_lightning/loops/epoch/training_epoch_loop.py", line 213, in advance batch_output = self.batch_loop.run(kwargs) File "/usr/local/lib/python3.8/dist-packages/pytorch_lightning/loops/loop.py", line 199, in run self.advance(args, *kwargs) File "/usr/local/lib/python3.8/dist-packages/pytorch_lightning/loops/batch/training_batch_loop.py", line 88, in advance outputs = self.optimizer_loop.run(optimizers, kwargs) File "/usr/local/lib/python3.8/dist-packages/pytorch_lightning/loops/loop.py", line 199, in run self.advance(args, *kwargs) File "/usr/local/lib/python3.8/dist-packages/pytorch_lightning/loops/optimization/optimizer_loop.py", line 202, in advance result = self._run_optimization(kwargs, self._optimizers[self.optim_progress.optimizer_position]) File "/usr/local/lib/python3.8/dist-packages/pytorch_lightning/loops/optimization/optimizer_loop.py", line 249, in _run_optimization self._optimizer_step(optimizer, opt_idx, kwargs.get("batch_idx", 0), closure) File "/usr/local/lib/python3.8/dist-packages/pytorch_lightning/loops/optimization/optimizer_loop.py", line 370, in _optimizer_step self.trainer._call_lightning_module_hook( File "/usr/local/lib/python3.8/dist-packages/pytorch_lightning/trainer/trainer.py", line 1356, in _call_lightning_module_hook output = fn(args, kwargs) File "/usr/local/lib/python3.8/dist-packages/pytorch_lightning/core/module.py", line 1742, in optimizer_step optimizer.step(closure=optimizer_closure) File "/usr/local/lib/python3.8/dist-packages/pytorch_lightning/core/optimizer.py", line 169, in step step_output = self._strategy.optimizer_step(self._optimizer, self._optimizer_idx, closure, kwargs) File "/usr/local/lib/python3.8/dist-packages/pytorch_lightning/strategies/ddp.py", line 280, in optimizer_step optimizer_output = super().optimizer_step(optimizer, opt_idx, closure, model, *kwargs) File "/usr/local/lib/python3.8/dist-packages/pytorch_lightning/strategies/strategy.py", line 234, in optimizer_step return self.precision_plugin.optimizer_step( File "/usr/local/lib/python3.8/dist-packages/pytorch_lightning/plugins/precision/native_amp.py", line 75, in optimizer_step closure_result = closure() File "/usr/local/lib/python3.8/dist-packages/pytorch_lightning/loops/optimization/optimizer_loop.py", line 149, in __call__ self._result = self.closure(args, *kwargs) File "/usr/local/lib/python3.8/dist-packages/pytorch_lightning/loops/optimization/optimizer_loop.py", line 135, in closure step_output = self._step_fn() File "/usr/local/lib/python3.8/dist-packages/pytorch_lightning/loops/optimization/optimizer_loop.py", line 419, in _training_step training_step_output = self.trainer._call_strategy_hook("training_step", kwargs.values()) File "/usr/local/lib/python3.8/dist-packages/pytorch_lightning/trainer/trainer.py", line 1494, in _call_strategy_hook output = fn(args, kwargs) File "/usr/local/lib/python3.8/dist-packages/pytorch_lightning/strategies/ddp.py", line 351, in training_step return self.model(args, *kwargs) File "/usr/local/lib/python3.8/dist-packages/torch/nn/modules/module.py", line 1533, in _call_impl return forward_call(args, *kwargs) File "/usr/local/lib/python3.8/dist-packages/torch/nn/parallel/distributed.py", line 1137, in forward output = self._run_ddp_forward(inputs, *kwargs) File "/usr/local/lib/python3.8/dist-packages/torch/nn/parallel/distributed.py", line 1094, in _run_ddp_forward return self.module(inputs, *kwargs) File "/usr/local/lib/python3.8/dist-packages/torch/nn/modules/module.py", line 1533, in _call_impl return forward_call(args, *kwargs) File "/usr/local/lib/python3.8/dist-packages/pytorch_lightning/overrides/base.py", line 98, in forward output = self._forward_module.training_step(inputs, *kwargs) File "/usr/local/lib/python3.8/dist-packages/nemo/utils/model_utils.py", line 380, in wrap_training_step output_dict = wrapped(args, kwargs) File "/usr/local/lib/python3.8/dist-packages/nemo/collections/nlp/models/language_modeling/megatron_gpt_model.py", line 469, in training_step self.allreduce_sequence_parallel_gradients() File "/usr/local/lib/python3.8/dist-packages/nemo/collections/nlp/models/language_modeling/megatron_gpt_model.py", line 571, in allreduce_sequence_parallel_gradients coalesced = torch._utils._flatten_dense_tensors(grads) File "/usr/local/lib/python3.8/dist-packages/torch/_utils.py", line 459, in _flatten_dense_tensors return torch._C._nn.flatten_dense_tensors(tensors) RuntimeError: torch.cat(): expected a non-empty list of Tensors ``` Expected behavior The training can proceed normally Environment overview (please complete the following information) - Docker images: `nvcr.io/nvidia/nemo:23.04` - Start docker container: `sudo nvidia-docker run -it --shm-size 64g --network host --hostname nemo-2304 --name nemo-2304 -v /path/on/host:/path/on/container/ nvcr.io/nvidia/nemo:23.04 bash` Environment details If NVIDIA docker image is used you don't need to specify these. Otherwise, please provide: - OS version - PyTorch version - Python version Additional context** GPU: 4 V100	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7162/reactions", "total_count": 1, "+1": 1, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7162/timeline	null	not_planned	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7161	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7161/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7161/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7161/events	https://github.com/NVIDIA/NeMo/issues/7161	1,835,761,813	I_kwDOC_bI7s5ta4CV	7,161	Installation fails on Raspberry Pi 4 8GB with aarch64 Bullseye	{ "login": "lizaibeim", "id": 38242437, "node_id": "MDQ6VXNlcjM4MjQyNDM3", "avatar_url": "https://avatars.githubusercontent.com/u/38242437?v=4", "gravatar_id": "", "url": "https://api.github.com/users/lizaibeim", "html_url": "https://github.com/lizaibeim", "followers_url": "https://api.github.com/users/lizaibeim/followers", "following_url": "https://api.github.com/users/lizaibeim/following{/other_user}", "gists_url": "https://api.github.com/users/lizaibeim/gists{/gist_id}", "starred_url": "https://api.github.com/users/lizaibeim/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/lizaibeim/subscriptions", "organizations_url": "https://api.github.com/users/lizaibeim/orgs", "repos_url": "https://api.github.com/users/lizaibeim/repos", "events_url": "https://api.github.com/users/lizaibeim/events{/privacy}", "received_events_url": "https://api.github.com/users/lizaibeim/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null } ]	closed	false	null	[]	null	null	2023-08-03T21:31:45	2023-09-17T01:44:58	2023-09-17T01:44:58	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	Hi I am trying to install NeMo on aarch64 RasberryPi 4 64bit Bulleye. I wonder if anyone have a feasible way to install NeMo and make it run on Raspberry Pi 4? I try the method Pip from source, and I run the `./reinstall.sh`, I found it fails while installing the pynini. Then I follow methods on https://pypi.org/project/pynini/ to install the pynini on my Pi step by step, I have installed the required openfst and then when I run the `pip install pynini`, I got such error ``` error: invalid cast from type ‘std::string’ {aka ‘std::__cxx11::basic_string<char>’} to type ‘PyObject’ {aka ‘_object’} ``` The full error log during installation is shown as below ``` extensions/_pywrapfst.cpp:11529:40: error: invalid cast from type ‘std::string’ {aka ‘std::__cxx11::basic_string<char>’} to type ‘PyObject’ {aka ‘_object’} 11529 \| __pyx_t_2 = __Pyx_PyObject_AsDouble(((PyObject )__pyx_t_1)); if (unlikely(__pyx_t_2 == ((double)((double)-1)) && PyErr_Occurred())) __PYX_ERR(0, 497, __pyx_L1_error) \| ^~~~~~~~~~~~~~~~~~~~~ extensions/_pywrapfst.cpp:1400:43: note: in definition of macro ‘likely’ 1400 \| #define likely(x) __builtin_expect(!!(x), 1) \| ^ /usr/include/python3.9/object.h:130:42: note: in expansion of macro ‘_PyObject_CAST_CONST’ 130 \| #define Py_IS_TYPE(ob, type) _Py_IS_TYPE(_PyObject_CAST_CONST(ob), type) \| ^~~~~~~~~~~~~~~~~~~~ /usr/include/python3.9/floatobject.h:24:32: note: in expansion of macro ‘Py_IS_TYPE’ 24 \| #define PyFloat_CheckExact(op) Py_IS_TYPE(op, &PyFloat_Type) \| ^~~~~~~~~~ extensions/_pywrapfst.cpp:3196:10: note: in expansion of macro ‘PyFloat_CheckExact’ 3196 \| ((likely(PyFloat_CheckExact(obj))) ? PyFloat_AS_DOUBLE(obj) :\ \| ^~~~~~~~~~~~~~~~~~ extensions/_pywrapfst.cpp:11529:15: note: in expansion of macro ‘__Pyx_PyObject_AsDouble’ 11529 \| __pyx_t_2 = __Pyx_PyObject_AsDouble(((PyObject )__pyx_t_1)); if (unlikely(__pyx_t_2 == ((double)((double)-1)) && PyErr_Occurred())) __PYX_ERR(0, 497, __pyx_L1_error) \| ^~~~~~~~~~~~~~~~~~~~~~~ In file included from /usr/include/python3.9/Python.h:101, from extensions/_pywrapfst.cpp:46: extensions/_pywrapfst.cpp:11529:40: error: invalid cast from type ‘std::string’ {aka ‘std::__cxx11::basic_string<char>’} to type ‘PyObject’ {aka ‘_object’} 11529 \| __pyx_t_2 = __Pyx_PyObject_AsDouble(((PyObject )__pyx_t_1)); if (unlikely(__pyx_t_2 == ((double)((double)-1)) && PyErr_Occurred())) __PYX_ERR(0, 497, __pyx_L1_error) \| ^~~~~~~~~~~~~~~~~~~~~ /usr/include/python3.9/floatobject.h:51:51: note: in definition of macro ‘PyFloat_AS_DOUBLE’ 51 \| #define PyFloat_AS_DOUBLE(op) (((PyFloatObject )(op))->ob_fval) \| ^~ extensions/_pywrapfst.cpp:11529:15: note: in expansion of macro ‘__Pyx_PyObject_AsDouble’ 11529 \| __pyx_t_2 = __Pyx_PyObject_AsDouble(((PyObject )__pyx_t_1)); if (unlikely(__pyx_t_2 == ((double)((double)-1)) && PyErr_Occurred())) __PYX_ERR(0, 497, __pyx_L1_error) \| ^~~~~~~~~~~~~~~~~~~~~~~ extensions/_pywrapfst.cpp:11529:40: error: invalid cast from type ‘std::string’ {aka ‘std::__cxx11::basic_string<char>’} to type ‘PyObject’ {aka ‘_object’} 11529 \| __pyx_t_2 = __Pyx_PyObject_AsDouble(((PyObject )__pyx_t_1)); if (unlikely(__pyx_t_2 == ((double)((double)-1)) && PyErr_Occurred())) __PYX_ERR(0, 497, __pyx_L1_error) \| ^~~~~~~~~~~~~~~~~~~~~ extensions/_pywrapfst.cpp:1400:43: note: in definition of macro ‘likely’ 1400 \| #define likely(x) __builtin_expect(!!(x), 1) \| ^ /usr/include/python3.9/object.h:130:42: note: in expansion of macro ‘_PyObject_CAST_CONST’ 130 \| #define Py_IS_TYPE(ob, type) _Py_IS_TYPE(_PyObject_CAST_CONST(ob), type) \| ^~~~~~~~~~~~~~~~~~~~ /usr/include/python3.9/longobject.h:16:31: note: in expansion of macro ‘Py_IS_TYPE’ 16 \| #define PyLong_CheckExact(op) Py_IS_TYPE(op, &PyLong_Type) \| ^~~~~~~~~~ extensions/_pywrapfst.cpp:3197:9: note: in expansion of macro ‘PyLong_CheckExact’ 3197 \| likely(PyLong_CheckExact(obj)) ?\ \| ^~~~~~~~~~~~~~~~~ extensions/_pywrapfst.cpp:11529:15: note: in expansion of macro ‘__Pyx_PyObject_AsDouble’ 11529 \| __pyx_t_2 = __Pyx_PyObject_AsDouble(((PyObject )__pyx_t_1)); if (unlikely(__pyx_t_2 == ((double)((double)-1)) && PyErr_Occurred())) __PYX_ERR(0, 497, __pyx_L1_error) \| ^~~~~~~~~~~~~~~~~~~~~~~ extensions/_pywrapfst.cpp:11529:40: error: invalid cast from type ‘std::string’ {aka ‘std::__cxx11::basic_string<char>’} to type ‘PyObject’ {aka ‘_object’} 11529 \| __pyx_t_2 = __Pyx_PyObject_AsDouble(((PyObject )__pyx_t_1)); if (unlikely(__pyx_t_2 == ((double)((double)-1)) && PyErr_Occurred())) __PYX_ERR(0, 497, __pyx_L1_error) \| ^~~~~~~~~~~~~~~~~~~~~ extensions/_pywrapfst.cpp:3198:18: note: in definition of macro ‘__Pyx_PyObject_AsDouble’ 3198 \| PyLong_AsDouble(obj) : __Pyx__PyObject_AsDouble(obj)) \| ^~~ extensions/_pywrapfst.cpp:11529:40: error: invalid cast from type ‘std::string’ {aka ‘std::__cxx11::basic_string<char>’} to type ‘PyObject’ {aka ‘_object’} 11529 \| __pyx_t_2 = __Pyx_PyObject_AsDouble(((PyObject *)__pyx_t_1)); if (unlikely(__pyx_t_2 == ((double)((double)-1)) && PyErr_Occurred())) __PYX_ERR(0, 497, __pyx_L1_error) \| ^~~~~~~~~~~~~~~~~~~~~ extensions/_pywrapfst.cpp:3198:50: note: in definition of macro ‘__Pyx_PyObject_AsDouble’ 3198 \| PyLong_AsDouble(obj) : __Pyx__PyObject_AsDouble(obj)) ``` My GCC version is `gcc (Debian 10.2.1-6) 10.2.1 20210110`, and my `python version is 3.9.2`, I am trying to install `openfst 1.8.2` and install `pynini 2.1.5`	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7161/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7161/timeline	null	not_planned	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7160	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7160/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7160/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7160/events	https://github.com/NVIDIA/NeMo/pull/7160	1,835,756,596	PR_kwDOC_bI7s5XJWoP	7,160	NFA subtitle file config - specify colors and vertical alignment	{ "login": "erastorgueva-nv", "id": 80532067, "node_id": "MDQ6VXNlcjgwNTMyMDY3", "avatar_url": "https://avatars.githubusercontent.com/u/80532067?v=4", "gravatar_id": "", "url": "https://api.github.com/users/erastorgueva-nv", "html_url": "https://github.com/erastorgueva-nv", "followers_url": "https://api.github.com/users/erastorgueva-nv/followers", "following_url": "https://api.github.com/users/erastorgueva-nv/following{/other_user}", "gists_url": "https://api.github.com/users/erastorgueva-nv/gists{/gist_id}", "starred_url": "https://api.github.com/users/erastorgueva-nv/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/erastorgueva-nv/subscriptions", "organizations_url": "https://api.github.com/users/erastorgueva-nv/orgs", "repos_url": "https://api.github.com/users/erastorgueva-nv/repos", "events_url": "https://api.github.com/users/erastorgueva-nv/events{/privacy}", "received_events_url": "https://api.github.com/users/erastorgueva-nv/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[]	closed	false	null	[]	null	null	2023-08-03T21:26:47	2023-08-03T21:41:11	2023-08-03T21:41:10	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7160", "html_url": "https://github.com/NVIDIA/NeMo/pull/7160", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7160.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7160.patch", "merged_at": "2023-08-03T21:41:10" }	# What does this PR do ? Modify the `ASSFileConfig` so we can specify the colors of the text and its vertical alignment. Collection: `tools/nemo_forced_aligner` # Changelog For specifying vertical alignment, removed `ass_file_config.marginv` parameter, which was being used as a rough way to modify the vertical placement of the text. Now, there is a parameter `ass_file_config.vertical_alignment`, which can take the values `"center"` (default), `"top"`, or `"bottom"` For specifying the colors of the text, added 3 new parameters: - `ass_file_config.text_already_spoken_rgb` - `ass_file_config.text_being_spoken_rgb` - `ass_file_config.text_not_yet_spoken_rgb` # Usage Can specify new parameters in `ASSFileConfig` e.g.: ```bash python <path to NeMo>/tools/nemo_forced_aligner/align.py \ pretrained_name="stt_en_fastconformer_ctc_large" \ manifest_filepath="manifest.json" \ output_dir=nfa_output/ \ additional_segment_grouping_separator="\|" \ ass_file_config.text_already_spoken_rgb=[252,186,3] \ # bright yellow/orange ass_file_config.text_being_spoken_rgb=[240,26,240] \ # bright pink ass_file_config.text_not_yet_spoken_rgb=[62,26,240] \ # bright blue ass_file_config.vertical_alignment="top" ``` # Before your PR is "Ready for review" Pre checks: - [x] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ N/A ] Did you write any new necessary tests? - [x] Did you add or update any necessary documentation? - [ N/A ] 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: - [x] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "titu1994", "id": 3048602, "node_id": "MDQ6VXNlcjMwNDg2MDI=", "avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4", "gravatar_id": "", "url": "https://api.github.com/users/titu1994", "html_url": "https://github.com/titu1994", "followers_url": "https://api.github.com/users/titu1994/followers", "following_url": "https://api.github.com/users/titu1994/following{/other_user}", "gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}", "starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/titu1994/subscriptions", "organizations_url": "https://api.github.com/users/titu1994/orgs", "repos_url": "https://api.github.com/users/titu1994/repos", "events_url": "https://api.github.com/users/titu1994/events{/privacy}", "received_events_url": "https://api.github.com/users/titu1994/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7160/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7160/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7159	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7159/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7159/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7159/events	https://github.com/NVIDIA/NeMo/pull/7159	1,835,671,566	PR_kwDOC_bI7s5XJFAl	7,159	Refactor LLM pretraining examples	{ "login": "maanug-nv", "id": 109391026, "node_id": "U_kgDOBoUssg", "avatar_url": "https://avatars.githubusercontent.com/u/109391026?v=4", "gravatar_id": "", "url": "https://api.github.com/users/maanug-nv", "html_url": "https://github.com/maanug-nv", "followers_url": "https://api.github.com/users/maanug-nv/followers", "following_url": "https://api.github.com/users/maanug-nv/following{/other_user}", "gists_url": "https://api.github.com/users/maanug-nv/gists{/gist_id}", "starred_url": "https://api.github.com/users/maanug-nv/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/maanug-nv/subscriptions", "organizations_url": "https://api.github.com/users/maanug-nv/orgs", "repos_url": "https://api.github.com/users/maanug-nv/repos", "events_url": "https://api.github.com/users/maanug-nv/events{/privacy}", "received_events_url": "https://api.github.com/users/maanug-nv/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-08-03T20:12:19	2023-08-16T18:08:38	2023-08-16T18:08:38	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7159", "html_url": "https://github.com/NVIDIA/NeMo/pull/7159", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7159.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7159.patch", "merged_at": "2023-08-16T18:08:38" }	# What does this PR do ? Simplify LLM pretraining example scripts by moving common logic into a new `TrainerBuilder` and `exp_manager`. Collection: [Note which collection this PR will affect] # Changelog - add a `TrainerBuilder` type that hides some common logic for setting up a `Trainer` - ~~move logic to handle `resume_from_checkpoint` arg to `exp_manager`~~ Current disabling `resume_from_checkpoint` logic and leaving a TODO comment since its current behavior is not a desirable user experience. Will make a separate PR to improve this. - utilize above refactors to reduce length of Megatron LLM pretraining example scripts # Usage * You can potentially add a usage example below ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [x] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] Did you write any new necessary tests? - [x] 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 - [x] Refactor 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "ericharper", "id": 11999610, "node_id": "MDQ6VXNlcjExOTk5NjEw", "avatar_url": "https://avatars.githubusercontent.com/u/11999610?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ericharper", "html_url": "https://github.com/ericharper", "followers_url": "https://api.github.com/users/ericharper/followers", "following_url": "https://api.github.com/users/ericharper/following{/other_user}", "gists_url": "https://api.github.com/users/ericharper/gists{/gist_id}", "starred_url": "https://api.github.com/users/ericharper/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ericharper/subscriptions", "organizations_url": "https://api.github.com/users/ericharper/orgs", "repos_url": "https://api.github.com/users/ericharper/repos", "events_url": "https://api.github.com/users/ericharper/events{/privacy}", "received_events_url": "https://api.github.com/users/ericharper/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7159/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7159/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7158	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7158/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7158/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7158/events	https://github.com/NVIDIA/NeMo/issues/7158	1,835,279,279	I_kwDOC_bI7s5tZCOv	7,158	Incorrect padding for get_seq_len of features.py	{ "login": "EndOfTheGlory", "id": 52138220, "node_id": "MDQ6VXNlcjUyMTM4MjIw", "avatar_url": "https://avatars.githubusercontent.com/u/52138220?v=4", "gravatar_id": "", "url": "https://api.github.com/users/EndOfTheGlory", "html_url": "https://github.com/EndOfTheGlory", "followers_url": "https://api.github.com/users/EndOfTheGlory/followers", "following_url": "https://api.github.com/users/EndOfTheGlory/following{/other_user}", "gists_url": "https://api.github.com/users/EndOfTheGlory/gists{/gist_id}", "starred_url": "https://api.github.com/users/EndOfTheGlory/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/EndOfTheGlory/subscriptions", "organizations_url": "https://api.github.com/users/EndOfTheGlory/orgs", "repos_url": "https://api.github.com/users/EndOfTheGlory/repos", "events_url": "https://api.github.com/users/EndOfTheGlory/events{/privacy}", "received_events_url": "https://api.github.com/users/EndOfTheGlory/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null } ]	closed	false	null	[]	null	null	2023-08-03T15:24:30	2023-09-10T01:44:54	2023-09-10T01:44:54	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	https://github.com/NVIDIA/NeMo/blob/4d78737f612964ff9cce2120bb28ec94f874ab07/nemo/collections/asr/parts/preprocessing/features.py#L392C1-L392C32 I don't know what was intended here, but to me it's clear there is redundency or error, because `self.n_fft // 2 * 2` is the same as just `self.n_fft`. Perhaps it was intended to be the power of two (aka `2 ** 2`)? ``` # Assuming that center is True is stft_pad_amount = 0 pad_amount = self.stft_pad_amount * 2 if self.stft_pad_amount is not None else self.n_fft // 2 * 2 ``` What was the intended purpose of the line? Perhaps we want n_fft to be even?	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7158/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7158/timeline	null	not_planned	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7157	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7157/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7157/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7157/events	https://github.com/NVIDIA/NeMo/pull/7157	1,834,953,330	PR_kwDOC_bI7s5XGsDZ	7,157	Changed vocoder dataset	{ "login": "styagi130", "id": 26252097, "node_id": "MDQ6VXNlcjI2MjUyMDk3", "avatar_url": "https://avatars.githubusercontent.com/u/26252097?v=4", "gravatar_id": "", "url": "https://api.github.com/users/styagi130", "html_url": "https://github.com/styagi130", "followers_url": "https://api.github.com/users/styagi130/followers", "following_url": "https://api.github.com/users/styagi130/following{/other_user}", "gists_url": "https://api.github.com/users/styagi130/gists{/gist_id}", "starred_url": "https://api.github.com/users/styagi130/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/styagi130/subscriptions", "organizations_url": "https://api.github.com/users/styagi130/orgs", "repos_url": "https://api.github.com/users/styagi130/repos", "events_url": "https://api.github.com/users/styagi130/events{/privacy}", "received_events_url": "https://api.github.com/users/styagi130/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4418253970, "node_id": "LA_kwDOC_bI7s8AAAABB1k0kg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/TTS", "name": "TTS", "color": "49AF28", "default": false, "description": "" }, { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null } ]	closed	false	null	[]	null	null	2023-08-03T12:20:03	2023-08-25T01:44:07	2023-08-25T01:44:07	CONTRIBUTOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	true	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7157", "html_url": "https://github.com/NVIDIA/NeMo/pull/7157", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7157.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7157.patch", "merged_at": null }	# 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 ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7157/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7157/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7156	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7156/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7156/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7156/events	https://github.com/NVIDIA/NeMo/pull/7156	1,833,991,097	PR_kwDOC_bI7s5XDaXb	7,156	[TTS] Support audio offsets in TTS data loaders	{ "login": "rlangman", "id": 9942053, "node_id": "MDQ6VXNlcjk5NDIwNTM=", "avatar_url": "https://avatars.githubusercontent.com/u/9942053?v=4", "gravatar_id": "", "url": "https://api.github.com/users/rlangman", "html_url": "https://github.com/rlangman", "followers_url": "https://api.github.com/users/rlangman/followers", "following_url": "https://api.github.com/users/rlangman/following{/other_user}", "gists_url": "https://api.github.com/users/rlangman/gists{/gist_id}", "starred_url": "https://api.github.com/users/rlangman/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/rlangman/subscriptions", "organizations_url": "https://api.github.com/users/rlangman/orgs", "repos_url": "https://api.github.com/users/rlangman/repos", "events_url": "https://api.github.com/users/rlangman/events{/privacy}", "received_events_url": "https://api.github.com/users/rlangman/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4418253970, "node_id": "LA_kwDOC_bI7s8AAAABB1k0kg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/TTS", "name": "TTS", "color": "49AF28", "default": false, "description": "" } ]	closed	false	{ "login": "XuesongYang", "id": 1646669, "node_id": "MDQ6VXNlcjE2NDY2Njk=", "avatar_url": "https://avatars.githubusercontent.com/u/1646669?v=4", "gravatar_id": "", "url": "https://api.github.com/users/XuesongYang", "html_url": "https://github.com/XuesongYang", "followers_url": "https://api.github.com/users/XuesongYang/followers", "following_url": "https://api.github.com/users/XuesongYang/following{/other_user}", "gists_url": "https://api.github.com/users/XuesongYang/gists{/gist_id}", "starred_url": "https://api.github.com/users/XuesongYang/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/XuesongYang/subscriptions", "organizations_url": "https://api.github.com/users/XuesongYang/orgs", "repos_url": "https://api.github.com/users/XuesongYang/repos", "events_url": "https://api.github.com/users/XuesongYang/events{/privacy}", "received_events_url": "https://api.github.com/users/XuesongYang/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "login": "XuesongYang", "id": 1646669, "node_id": "MDQ6VXNlcjE2NDY2Njk=", "avatar_url": "https://avatars.githubusercontent.com/u/1646669?v=4", "gravatar_id": "", "url": "https://api.github.com/users/XuesongYang", "html_url": "https://github.com/XuesongYang", "followers_url": "https://api.github.com/users/XuesongYang/followers", "following_url": "https://api.github.com/users/XuesongYang/following{/other_user}", "gists_url": "https://api.github.com/users/XuesongYang/gists{/gist_id}", "starred_url": "https://api.github.com/users/XuesongYang/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/XuesongYang/subscriptions", "organizations_url": "https://api.github.com/users/XuesongYang/orgs", "repos_url": "https://api.github.com/users/XuesongYang/repos", "events_url": "https://api.github.com/users/XuesongYang/events{/privacy}", "received_events_url": "https://api.github.com/users/XuesongYang/received_events", "type": "User", "user_view_type": "public", "site_admin": false }, { "login": "racoiaws", "id": 8864149, "node_id": "MDQ6VXNlcjg4NjQxNDk=", "avatar_url": "https://avatars.githubusercontent.com/u/8864149?v=4", "gravatar_id": "", "url": "https://api.github.com/users/racoiaws", "html_url": "https://github.com/racoiaws", "followers_url": "https://api.github.com/users/racoiaws/followers", "following_url": "https://api.github.com/users/racoiaws/following{/other_user}", "gists_url": "https://api.github.com/users/racoiaws/gists{/gist_id}", "starred_url": "https://api.github.com/users/racoiaws/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/racoiaws/subscriptions", "organizations_url": "https://api.github.com/users/racoiaws/orgs", "repos_url": "https://api.github.com/users/racoiaws/repos", "events_url": "https://api.github.com/users/racoiaws/events{/privacy}", "received_events_url": "https://api.github.com/users/racoiaws/received_events", "type": "User", "user_view_type": "public", "site_admin": false } ]	null	null	2023-08-02T22:41:34	2023-10-26T17:56:58	2023-10-26T17:56:57	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7156", "html_url": "https://github.com/NVIDIA/NeMo/pull/7156", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7156.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7156.patch", "merged_at": "2023-10-26T17:56:57" }	# What does this PR do ? This PR contains all of the changes needed to support training TTS models with large, multi-utterance files. Related features are all experimental, and the change is not backwards compatible as I had to change feature files from _.pt_ to _.npy_ so I could memory map them to avoid reading the entire feature file into memory. Collection: [TTS] # Changelog - Add support for "offset" and "duration" in manifest (similar to ASR) to allow reading a subset of audio in the data loader. - Modify featurizer classes so that they save features as .npy files, and reads subsets of the files using numpy memory mapping when "offset" is present in the manifest. - Add logic for computing pitch in small chunks for large audio files (to avoid out of memory issues). - Add on-the-fly volume normalization to featurizers and data loader, as we can not do it as a preprocessing step for multi-utterance files. Volume effects feature computation, as energy is proportional to volume, and volume can change the voiced/unvoiced classification during pitch computation. - Unit tests to validate that the saving/loading logic works as expected. # Before your PR is "Ready for review" Pre checks: - [x] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [x] Did you write any new necessary tests? - [x] 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: - [x] New Feature - [ ] Bugfix - [ ] Documentation	{ "login": "rlangman", "id": 9942053, "node_id": "MDQ6VXNlcjk5NDIwNTM=", "avatar_url": "https://avatars.githubusercontent.com/u/9942053?v=4", "gravatar_id": "", "url": "https://api.github.com/users/rlangman", "html_url": "https://github.com/rlangman", "followers_url": "https://api.github.com/users/rlangman/followers", "following_url": "https://api.github.com/users/rlangman/following{/other_user}", "gists_url": "https://api.github.com/users/rlangman/gists{/gist_id}", "starred_url": "https://api.github.com/users/rlangman/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/rlangman/subscriptions", "organizations_url": "https://api.github.com/users/rlangman/orgs", "repos_url": "https://api.github.com/users/rlangman/repos", "events_url": "https://api.github.com/users/rlangman/events{/privacy}", "received_events_url": "https://api.github.com/users/rlangman/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7156/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7156/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7155	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7155/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7155/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7155/events	https://github.com/NVIDIA/NeMo/pull/7155	1,833,978,158	PR_kwDOC_bI7s5XDXki	7,155	NFA bugfix: remove any empty segments	{ "login": "erastorgueva-nv", "id": 80532067, "node_id": "MDQ6VXNlcjgwNTMyMDY3", "avatar_url": "https://avatars.githubusercontent.com/u/80532067?v=4", "gravatar_id": "", "url": "https://api.github.com/users/erastorgueva-nv", "html_url": "https://github.com/erastorgueva-nv", "followers_url": "https://api.github.com/users/erastorgueva-nv/followers", "following_url": "https://api.github.com/users/erastorgueva-nv/following{/other_user}", "gists_url": "https://api.github.com/users/erastorgueva-nv/gists{/gist_id}", "starred_url": "https://api.github.com/users/erastorgueva-nv/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/erastorgueva-nv/subscriptions", "organizations_url": "https://api.github.com/users/erastorgueva-nv/orgs", "repos_url": "https://api.github.com/users/erastorgueva-nv/repos", "events_url": "https://api.github.com/users/erastorgueva-nv/events{/privacy}", "received_events_url": "https://api.github.com/users/erastorgueva-nv/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[]	closed	false	null	[]	null	null	2023-08-02T22:23:36	2023-08-02T22:26:23	2023-08-02T22:26:22	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7155", "html_url": "https://github.com/NVIDIA/NeMo/pull/7155", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7155.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7155.patch", "merged_at": "2023-08-02T22:26:22" }	# What does this PR do ? Bugfix to prevent errors being thrown if the substring specified as an `additional_segment_grouping_separator` is placed at the beginning or end of reference text, or if 2+ of these substrings are placed next to each other in the reference text. Collection: tools/nemo_forced_aligner # Changelog - Add specific line by line info of high level changes in this PR. # Usage No usage changes # Before your PR is "Ready for review" Pre checks: - [x] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ N/A ] Did you write any new necessary tests? - [ N/A ] Did you add or update any necessary documentation? - [ N/A ] 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 - [x] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "titu1994", "id": 3048602, "node_id": "MDQ6VXNlcjMwNDg2MDI=", "avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4", "gravatar_id": "", "url": "https://api.github.com/users/titu1994", "html_url": "https://github.com/titu1994", "followers_url": "https://api.github.com/users/titu1994/followers", "following_url": "https://api.github.com/users/titu1994/following{/other_user}", "gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}", "starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/titu1994/subscriptions", "organizations_url": "https://api.github.com/users/titu1994/orgs", "repos_url": "https://api.github.com/users/titu1994/repos", "events_url": "https://api.github.com/users/titu1994/events{/privacy}", "received_events_url": "https://api.github.com/users/titu1994/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7155/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7155/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7154	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7154/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7154/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7154/events	https://github.com/NVIDIA/NeMo/pull/7154	1,833,643,077	PR_kwDOC_bI7s5XCOJd	7,154	[TTS] Remove nested TTS configs	{ "login": "rlangman", "id": 9942053, "node_id": "MDQ6VXNlcjk5NDIwNTM=", "avatar_url": "https://avatars.githubusercontent.com/u/9942053?v=4", "gravatar_id": "", "url": "https://api.github.com/users/rlangman", "html_url": "https://github.com/rlangman", "followers_url": "https://api.github.com/users/rlangman/followers", "following_url": "https://api.github.com/users/rlangman/following{/other_user}", "gists_url": "https://api.github.com/users/rlangman/gists{/gist_id}", "starred_url": "https://api.github.com/users/rlangman/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/rlangman/subscriptions", "organizations_url": "https://api.github.com/users/rlangman/orgs", "repos_url": "https://api.github.com/users/rlangman/repos", "events_url": "https://api.github.com/users/rlangman/events{/privacy}", "received_events_url": "https://api.github.com/users/rlangman/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4418253970, "node_id": "LA_kwDOC_bI7s8AAAABB1k0kg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/TTS", "name": "TTS", "color": "49AF28", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-08-02T17:48:32	2023-08-05T16:20:15	2023-08-05T02:53:08	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7154", "html_url": "https://github.com/NVIDIA/NeMo/pull/7154", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7154.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7154.patch", "merged_at": "2023-08-05T02:53:08" }	# What does this PR do ? In a separate PR (https://github.com/NVIDIA/NeMo/pull/6852) it was brought up that TTS is using nested hydra config, which is against NeMo 1 config file per model policy. This PR removes nested config from our experimental FastPitch and HiFiGan in favor of a flat structure with duplicated code between 22kHz and 44kHz versions. Our older non-experimental configs are left unchanged for backwards compatibility. Collection: [TTS] # Changelog - Remove nested configuration from FastPitch and HiFiGan configs using new data loader - Add separate 22kHz and 44kHz examples for each model. - Moved hifigan_data.yaml config to separate "hifigan_dataset" folder (not sure if that is a good name) now that it does not need to reuse the generator config files. # Before your PR is "Ready for review" Pre checks: - [x] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] Did you write any new necessary tests? - [x] 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 - [x] Bugfix - [ ] Documentation	{ "login": "rlangman", "id": 9942053, "node_id": "MDQ6VXNlcjk5NDIwNTM=", "avatar_url": "https://avatars.githubusercontent.com/u/9942053?v=4", "gravatar_id": "", "url": "https://api.github.com/users/rlangman", "html_url": "https://github.com/rlangman", "followers_url": "https://api.github.com/users/rlangman/followers", "following_url": "https://api.github.com/users/rlangman/following{/other_user}", "gists_url": "https://api.github.com/users/rlangman/gists{/gist_id}", "starred_url": "https://api.github.com/users/rlangman/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/rlangman/subscriptions", "organizations_url": "https://api.github.com/users/rlangman/orgs", "repos_url": "https://api.github.com/users/rlangman/repos", "events_url": "https://api.github.com/users/rlangman/events{/privacy}", "received_events_url": "https://api.github.com/users/rlangman/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7154/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7154/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7153	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7153/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7153/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7153/events	https://github.com/NVIDIA/NeMo/pull/7153	1,833,359,665	PR_kwDOC_bI7s5XBPvZ	7,153	Fix incorrect embedding grads with distopt BF16 grad reductions	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	[ { "id": 1923495043, "node_id": "MDU6TGFiZWwxOTIzNDk1MDQz", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/core", "name": "core", "color": "f25e3a", "default": false, "description": "Changes to NeMo Core" }, { "id": 4418253970, "node_id": "LA_kwDOC_bI7s8AAAABB1k0kg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/TTS", "name": "TTS", "color": "49AF28", "default": false, "description": "" }, { "id": 4556669155, "node_id": "LA_kwDOC_bI7s8AAAABD5lA4w", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/cherry-pick", "name": "cherry-pick", "color": "ededed", "default": false, "description": null }, { "id": 4811884691, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR", "name": "ASR", "color": "F3B33E", "default": false, "description": "" }, { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" }, { "id": 4811888028, "node_id": "LA_kwDOC_bI7s8AAAABHs-VnA", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/Speaker%20Tasks", "name": "Speaker Tasks", "color": "871AD4", "default": false, "description": "" }, { "id": 4840216855, "node_id": "LA_kwDOC_bI7s8AAAABIH_ZFw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/CI", "name": "CI", "color": "1D76DB", "default": false, "description": "" }, { "id": 4847373924, "node_id": "LA_kwDOC_bI7s8AAAABIO0OZA", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/common", "name": "common", "color": "c5def5", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-08-02T14:59:51	2023-08-04T20:45:56	2023-08-04T20:45:55	CONTRIBUTOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7153", "html_url": "https://github.com/NVIDIA/NeMo/pull/7153", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7153.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7153.patch", "merged_at": "2023-08-04T20:45:55" }	# What does this PR do ? In the case where we are training GPT with distopt, BF16 grads, and pipeline parallelism, we are not properly synchronizing the embedding grads between the first and last pipeline stage. The embeddings were being accumulated into BF16 buffers and the pipeline-parallel reduction was on FP32 buffers (so just all-reducing zeros). Note that we synchronize these buffers before the data-parallel reduction, so that is still being done properly in FP32. This PR fixes the issue by eagerly accumulating the embedding grads into FP32 buffers. Collection: NLP # Changelog - Distopt accumulates grads for FP32 params into FP32 buffers. # Usage In the config file, set the optimizer with `name: distributed_fused_adam` and `grad_sync_dtype: bf16`. # Before your PR is "Ready for review" Pre checks: - [x] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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 - [x] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Depends/includes https://github.com/NVIDIA/NeMo/pull/6886	{ "login": "ericharper", "id": 11999610, "node_id": "MDQ6VXNlcjExOTk5NjEw", "avatar_url": "https://avatars.githubusercontent.com/u/11999610?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ericharper", "html_url": "https://github.com/ericharper", "followers_url": "https://api.github.com/users/ericharper/followers", "following_url": "https://api.github.com/users/ericharper/following{/other_user}", "gists_url": "https://api.github.com/users/ericharper/gists{/gist_id}", "starred_url": "https://api.github.com/users/ericharper/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ericharper/subscriptions", "organizations_url": "https://api.github.com/users/ericharper/orgs", "repos_url": "https://api.github.com/users/ericharper/repos", "events_url": "https://api.github.com/users/ericharper/events{/privacy}", "received_events_url": "https://api.github.com/users/ericharper/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7153/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7153/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7152	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7152/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7152/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7152/events	https://github.com/NVIDIA/NeMo/pull/7152	1,832,719,336	PR_kwDOC_bI7s5W_EWt	7,152	freeze base mode on init during peft	{ "login": "arendu", "id": 108822655, "node_id": "U_kgDOBnyAfw", "avatar_url": "https://avatars.githubusercontent.com/u/108822655?v=4", "gravatar_id": "", "url": "https://api.github.com/users/arendu", "html_url": "https://github.com/arendu", "followers_url": "https://api.github.com/users/arendu/followers", "following_url": "https://api.github.com/users/arendu/following{/other_user}", "gists_url": "https://api.github.com/users/arendu/gists{/gist_id}", "starred_url": "https://api.github.com/users/arendu/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/arendu/subscriptions", "organizations_url": "https://api.github.com/users/arendu/orgs", "repos_url": "https://api.github.com/users/arendu/repos", "events_url": "https://api.github.com/users/arendu/events{/privacy}", "received_events_url": "https://api.github.com/users/arendu/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-08-02T08:41:34	2023-08-02T17:08:29	2023-08-02T17:08:28	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7152", "html_url": "https://github.com/NVIDIA/NeMo/pull/7152", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7152.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7152.patch", "merged_at": "2023-08-02T17:08:28" }	# What does this PR do ? adds a self.freeze() during peft model init to prevent memory spike during startup. Collection: [NLP] # Changelog - Add specific line by line info of high level changes in this PR. # Usage * You can potentially add a usage example below ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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 - [x] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "arendu", "id": 108822655, "node_id": "U_kgDOBnyAfw", "avatar_url": "https://avatars.githubusercontent.com/u/108822655?v=4", "gravatar_id": "", "url": "https://api.github.com/users/arendu", "html_url": "https://github.com/arendu", "followers_url": "https://api.github.com/users/arendu/followers", "following_url": "https://api.github.com/users/arendu/following{/other_user}", "gists_url": "https://api.github.com/users/arendu/gists{/gist_id}", "starred_url": "https://api.github.com/users/arendu/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/arendu/subscriptions", "organizations_url": "https://api.github.com/users/arendu/orgs", "repos_url": "https://api.github.com/users/arendu/repos", "events_url": "https://api.github.com/users/arendu/events{/privacy}", "received_events_url": "https://api.github.com/users/arendu/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7152/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7152/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7151	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7151/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7151/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7151/events	https://github.com/NVIDIA/NeMo/pull/7151	1,832,206,162	PR_kwDOC_bI7s5W9WTf	7,151	Update evaluator.py	{ "login": "stevehuang52", "id": 105218074, "node_id": "U_kgDOBkWAGg", "avatar_url": "https://avatars.githubusercontent.com/u/105218074?v=4", "gravatar_id": "", "url": "https://api.github.com/users/stevehuang52", "html_url": "https://github.com/stevehuang52", "followers_url": "https://api.github.com/users/stevehuang52/followers", "following_url": "https://api.github.com/users/stevehuang52/following{/other_user}", "gists_url": "https://api.github.com/users/stevehuang52/gists{/gist_id}", "starred_url": "https://api.github.com/users/stevehuang52/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/stevehuang52/subscriptions", "organizations_url": "https://api.github.com/users/stevehuang52/orgs", "repos_url": "https://api.github.com/users/stevehuang52/repos", "events_url": "https://api.github.com/users/stevehuang52/events{/privacy}", "received_events_url": "https://api.github.com/users/stevehuang52/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[]	closed	false	null	[]	null	null	2023-08-02T00:18:42	2023-08-02T01:38:01	2023-08-02T01:38:00	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7151", "html_url": "https://github.com/NVIDIA/NeMo/pull/7151", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7151.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7151.patch", "merged_at": "2023-08-02T01:38:00" }	reflecting changes in https://github.com/NVIDIA/NeMo/pull/7150	{ "login": "titu1994", "id": 3048602, "node_id": "MDQ6VXNlcjMwNDg2MDI=", "avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4", "gravatar_id": "", "url": "https://api.github.com/users/titu1994", "html_url": "https://github.com/titu1994", "followers_url": "https://api.github.com/users/titu1994/followers", "following_url": "https://api.github.com/users/titu1994/following{/other_user}", "gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}", "starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/titu1994/subscriptions", "organizations_url": "https://api.github.com/users/titu1994/orgs", "repos_url": "https://api.github.com/users/titu1994/repos", "events_url": "https://api.github.com/users/titu1994/events{/privacy}", "received_events_url": "https://api.github.com/users/titu1994/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7151/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7151/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7150	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7150/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7150/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7150/events	https://github.com/NVIDIA/NeMo/pull/7150	1,832,195,501	PR_kwDOC_bI7s5W9T_6	7,150	fix evaluator.py for various exceptions by ast	{ "login": "stevehuang52", "id": 105218074, "node_id": "U_kgDOBkWAGg", "avatar_url": "https://avatars.githubusercontent.com/u/105218074?v=4", "gravatar_id": "", "url": "https://api.github.com/users/stevehuang52", "html_url": "https://github.com/stevehuang52", "followers_url": "https://api.github.com/users/stevehuang52/followers", "following_url": "https://api.github.com/users/stevehuang52/following{/other_user}", "gists_url": "https://api.github.com/users/stevehuang52/gists{/gist_id}", "starred_url": "https://api.github.com/users/stevehuang52/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/stevehuang52/subscriptions", "organizations_url": "https://api.github.com/users/stevehuang52/orgs", "repos_url": "https://api.github.com/users/stevehuang52/repos", "events_url": "https://api.github.com/users/stevehuang52/events{/privacy}", "received_events_url": "https://api.github.com/users/stevehuang52/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[]	closed	false	null	[]	null	null	2023-08-02T00:03:49	2023-08-02T00:14:44	2023-08-02T00:14:43	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7150", "html_url": "https://github.com/NVIDIA/NeMo/pull/7150", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7150.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7150.patch", "merged_at": "2023-08-02T00:14:43" }	the python `ast` package throws both `SyntaxError` and `ValueError` during evaluating an invalid dict object, so exception handling needs to catch all possible kinds of exceptions.	{ "login": "titu1994", "id": 3048602, "node_id": "MDQ6VXNlcjMwNDg2MDI=", "avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4", "gravatar_id": "", "url": "https://api.github.com/users/titu1994", "html_url": "https://github.com/titu1994", "followers_url": "https://api.github.com/users/titu1994/followers", "following_url": "https://api.github.com/users/titu1994/following{/other_user}", "gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}", "starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/titu1994/subscriptions", "organizations_url": "https://api.github.com/users/titu1994/orgs", "repos_url": "https://api.github.com/users/titu1994/repos", "events_url": "https://api.github.com/users/titu1994/events{/privacy}", "received_events_url": "https://api.github.com/users/titu1994/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7150/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7150/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7149	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7149/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7149/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7149/events	https://github.com/NVIDIA/NeMo/issues/7149	1,831,883,165	I_kwDOC_bI7s5tMFGd	7,149	Default learning rate is too large	{ "login": "khanh14ph", "id": 72318494, "node_id": "MDQ6VXNlcjcyMzE4NDk0", "avatar_url": "https://avatars.githubusercontent.com/u/72318494?v=4", "gravatar_id": "", "url": "https://api.github.com/users/khanh14ph", "html_url": "https://github.com/khanh14ph", "followers_url": "https://api.github.com/users/khanh14ph/followers", "following_url": "https://api.github.com/users/khanh14ph/following{/other_user}", "gists_url": "https://api.github.com/users/khanh14ph/gists{/gist_id}", "starred_url": "https://api.github.com/users/khanh14ph/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/khanh14ph/subscriptions", "organizations_url": "https://api.github.com/users/khanh14ph/orgs", "repos_url": "https://api.github.com/users/khanh14ph/repos", "events_url": "https://api.github.com/users/khanh14ph/events{/privacy}", "received_events_url": "https://api.github.com/users/khanh14ph/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 1485815557, "node_id": "MDU6TGFiZWwxNDg1ODE1NTU3", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/bug", "name": "bug", "color": "d73a4a", "default": true, "description": "Something isn't working" } ]	closed	false	null	[]	null	null	2023-08-01T19:20:57	2023-08-01T19:34:50	2023-08-01T19:34:42	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	In _conformer_transducer_bpe.yaml_ and _conformer_transducer_char.yaml_, the default learning rate is too large (5.0), which may lead to bad training for beginners if they do not see it. I hope you can change it to a more suitable value.	{ "login": "khanh14ph", "id": 72318494, "node_id": "MDQ6VXNlcjcyMzE4NDk0", "avatar_url": "https://avatars.githubusercontent.com/u/72318494?v=4", "gravatar_id": "", "url": "https://api.github.com/users/khanh14ph", "html_url": "https://github.com/khanh14ph", "followers_url": "https://api.github.com/users/khanh14ph/followers", "following_url": "https://api.github.com/users/khanh14ph/following{/other_user}", "gists_url": "https://api.github.com/users/khanh14ph/gists{/gist_id}", "starred_url": "https://api.github.com/users/khanh14ph/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/khanh14ph/subscriptions", "organizations_url": "https://api.github.com/users/khanh14ph/orgs", "repos_url": "https://api.github.com/users/khanh14ph/repos", "events_url": "https://api.github.com/users/khanh14ph/events{/privacy}", "received_events_url": "https://api.github.com/users/khanh14ph/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7149/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7149/timeline	null	completed	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7148	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7148/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7148/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7148/events	https://github.com/NVIDIA/NeMo/pull/7148	1,831,877,914	PR_kwDOC_bI7s5W8Pet	7,148	FSDP + (TP)	{ "login": "erhoo82", "id": 15861259, "node_id": "MDQ6VXNlcjE1ODYxMjU5", "avatar_url": "https://avatars.githubusercontent.com/u/15861259?v=4", "gravatar_id": "", "url": "https://api.github.com/users/erhoo82", "html_url": "https://github.com/erhoo82", "followers_url": "https://api.github.com/users/erhoo82/followers", "following_url": "https://api.github.com/users/erhoo82/following{/other_user}", "gists_url": "https://api.github.com/users/erhoo82/gists{/gist_id}", "starred_url": "https://api.github.com/users/erhoo82/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/erhoo82/subscriptions", "organizations_url": "https://api.github.com/users/erhoo82/orgs", "repos_url": "https://api.github.com/users/erhoo82/repos", "events_url": "https://api.github.com/users/erhoo82/events{/privacy}", "received_events_url": "https://api.github.com/users/erhoo82/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null }, { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-08-01T19:16:45	2023-10-20T22:22:01	2023-08-29T01:44:08	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7148", "html_url": "https://github.com/NVIDIA/NeMo/pull/7148", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7148.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7148.patch", "merged_at": null }	# What does this PR do ? Support FSDP + (Tensor parallelism) for Transformer models. What's missing in the PR 1. Sharded checkpoint store and load (not critical at the moment) 2. Decouple the module granularity of full-activation recomputation and FSDP sharding to remove AG in recomputation 3. CPU initialization with TE enabled 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 ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7148/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7148/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7147	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7147/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7147/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7147/events	https://github.com/NVIDIA/NeMo/pull/7147	1,831,843,114	PR_kwDOC_bI7s5W8IFu	7,147	multiple fields can form a context	{ "login": "arendu", "id": 108822655, "node_id": "U_kgDOBnyAfw", "avatar_url": "https://avatars.githubusercontent.com/u/108822655?v=4", "gravatar_id": "", "url": "https://api.github.com/users/arendu", "html_url": "https://github.com/arendu", "followers_url": "https://api.github.com/users/arendu/followers", "following_url": "https://api.github.com/users/arendu/following{/other_user}", "gists_url": "https://api.github.com/users/arendu/gists{/gist_id}", "starred_url": "https://api.github.com/users/arendu/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/arendu/subscriptions", "organizations_url": "https://api.github.com/users/arendu/orgs", "repos_url": "https://api.github.com/users/arendu/repos", "events_url": "https://api.github.com/users/arendu/events{/privacy}", "received_events_url": "https://api.github.com/users/arendu/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" }, { "id": 4847373924, "node_id": "LA_kwDOC_bI7s8AAAABIO0OZA", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/common", "name": "common", "color": "c5def5", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-08-01T18:52:32	2023-09-01T20:31:09	2023-09-01T20:31:08	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7147", "html_url": "https://github.com/NVIDIA/NeMo/pull/7147", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7147.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7147.patch", "merged_at": "2023-09-01T20:31:08" }	# What does this PR do ? * Add support for multiple contexts and truncation_field * Add support for truncation in the middle of prompt_template * Arguments * Remove context_key * Remove separate_prompt_and_response_with_newline * Add truncate_method * Change the truncation logic to the following 1. Split prompt_template by placeholder 2. Tokenize each part to ids 3. Truncate each part 4. Concatenate all part 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 ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [x] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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: - [x] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "arendu", "id": 108822655, "node_id": "U_kgDOBnyAfw", "avatar_url": "https://avatars.githubusercontent.com/u/108822655?v=4", "gravatar_id": "", "url": "https://api.github.com/users/arendu", "html_url": "https://github.com/arendu", "followers_url": "https://api.github.com/users/arendu/followers", "following_url": "https://api.github.com/users/arendu/following{/other_user}", "gists_url": "https://api.github.com/users/arendu/gists{/gist_id}", "starred_url": "https://api.github.com/users/arendu/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/arendu/subscriptions", "organizations_url": "https://api.github.com/users/arendu/orgs", "repos_url": "https://api.github.com/users/arendu/repos", "events_url": "https://api.github.com/users/arendu/events{/privacy}", "received_events_url": "https://api.github.com/users/arendu/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7147/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7147/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7146	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7146/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7146/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7146/events	https://github.com/NVIDIA/NeMo/issues/7146	1,831,110,458	I_kwDOC_bI7s5tJIc6	7,146	Problems with MTEncDecModel model not releasing GPU memory	{ "login": "evros-chris", "id": 37453466, "node_id": "MDQ6VXNlcjM3NDUzNDY2", "avatar_url": "https://avatars.githubusercontent.com/u/37453466?v=4", "gravatar_id": "", "url": "https://api.github.com/users/evros-chris", "html_url": "https://github.com/evros-chris", "followers_url": "https://api.github.com/users/evros-chris/followers", "following_url": "https://api.github.com/users/evros-chris/following{/other_user}", "gists_url": "https://api.github.com/users/evros-chris/gists{/gist_id}", "starred_url": "https://api.github.com/users/evros-chris/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/evros-chris/subscriptions", "organizations_url": "https://api.github.com/users/evros-chris/orgs", "repos_url": "https://api.github.com/users/evros-chris/repos", "events_url": "https://api.github.com/users/evros-chris/events{/privacy}", "received_events_url": "https://api.github.com/users/evros-chris/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 1485815557, "node_id": "MDU6TGFiZWwxNDg1ODE1NTU3", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/bug", "name": "bug", "color": "d73a4a", "default": true, "description": "Something isn't working" }, { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null } ]	closed	false	null	[]	null	null	2023-08-01T11:58:53	2023-09-09T01:43:35	2023-09-09T01:43:34	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	Describe the bug I am creating an MTEncDecModel model and the gpu memory allocated never gets freed. Specifically, I want to create an MTEncDecModel model for english to german, perform some translations, delete then model, then create an MTEncDecModel model for english to french, perform some translations, delete the model, and continue this process for many language pairs. However, the gpu memory allocated to each model instantiated is not freed, so I get a `RuntimeError: CUDA Out of memory` error. I want a way to clear the gpu memory allocated without terminating the python program or restarting the jupyter kernel. Steps/Code to reproduce bug ``` import torch from nemo.collections.nlp.models import MTEncDecModel from transformers import MarianMTModel def memory_stats(): print('Memory allocated:', torch.cuda.memory_allocated()/10242) print('Memory cached:', torch.cuda.memory_cached()/10242, '\n') print('Initia memory stats') memory_stats() model = MTEncDecModel.from_pretrained('nmt_en_es_transformer24x6').cuda() print('Model created') memory_stats() del model print('Model deleted') memory_stats() torch.cuda.empty_cache() print('Cache emptied') memory_stats() ``` The output from this code is ``` Initial memory stats Memory allocated: 0.0 Memory cached: 0.0 Model created Memory allocated: 1793.8173828125 Memory cached: 3714.0 Model deleted Memory allocated: 1793.8173828125 Memory cached: 3714.0 Cache emptied Memory allocated: 1793.8173828125 Memory cached: 1802.0 ``` Expected behavior Deleting all objects and references pointing to objects allocating GPU memory should free the memory. Link: https://discuss.pytorch.org/t/free-all-gpu-memory-used-in-between-runs/168202/2 For example, If I create a PyTorch tensor instead of the MTEncDecModel, then the output is as expected: ``` print('Initial memory stats') memory_stats() x = torch.randn(10241024, device='cuda') print('Tensor created') memory_stats() del x print('Tensor deleted') memory_stats() torch.cuda.empty_cache() print('Cache emptied') memory_stats() ``` Output: ``` Initial memory stats Memory allocated: 0.0 Memory cached: 0.0 Tensor created Memory allocated: 4.0 Memory cached: 20.0 Tensor deleted Memory allocated: 0.0 Memory cached: 20.0 Cache emptied Memory allocated: 0.0 Memory cached: 0.0 ``` Also, if I create a Marian machine translation model from HuggingFace instead of the MTEncDecModel model then again the output is as expected: ``` print('Initial memory stats') memory_stats() model = MarianMTModel.from_pretrained('Helsinki-NLP/opus-mt-tc-big-en-it').cuda() print('Model created') memory_stats() del model print('Model deleted') memory_stats() torch.cuda.empty_cache() print('Cache emptied') memory_stats() ``` Output ``` Initial memory stats Memory allocated: 0.0 Memory cached: 0.0 Model created Memory allocated: 893.5439453125 Memory cached: 896.0 Model deleted Memory allocated: 0.0 Memory cached: 896.0 Cache emptied Memory allocated: 0.0 Memory cached: 0.0 ``` In fact, even instantiating an MTEncDecModel without sending to model to `cuda()` and without storing the model to a variable, causes the memory allocated to increate every time: ``` print('Initial memory stats') memory_stats() MTEncDecModel.from_pretrained('nmt_en_es_transformer24x6') torch.cuda.empty_cache() print('After one instantiation') memory_stats() MTEncDecModel.from_pretrained('nmt_en_es_transformer24x6') torch.cuda.empty_cache() print('After two instantiations') memory_stats() MTEncDecModel.from_pretrained('nmt_en_es_transformer24x6') torch.cuda.empty_cache() print('After three instantiations') memory_stats() ``` Output: ``` Initial memory stats Memory allocated: 0.0 Memory cached: 0.0 After one instantiation Memory allocated: 1793.8173828125 Memory cached: 1802.0 After two instantiations Memory allocated: 3587.634765625 Memory cached: 3588.0 After three instantiations Memory allocated: 5381.4521484375 Memory cached: 5390.0 ``` Environment overview (please complete the following information)* - Environment location: Docker on Kubernetes - Method of NeMo install: using helm chart using nvcr.io/nvidia/nemo: 1.5.1	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7146/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7146/timeline	null	not_planned	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7145	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7145/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7145/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7145/events	https://github.com/NVIDIA/NeMo/issues/7145	1,830,919,584	I_kwDOC_bI7s5tIZ2g	7,145	ASR Validation step predicts "???" when adding noise augmentation	{ "login": "gabitza-tech", "id": 53811655, "node_id": "MDQ6VXNlcjUzODExNjU1", "avatar_url": "https://avatars.githubusercontent.com/u/53811655?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gabitza-tech", "html_url": "https://github.com/gabitza-tech", "followers_url": "https://api.github.com/users/gabitza-tech/followers", "following_url": "https://api.github.com/users/gabitza-tech/following{/other_user}", "gists_url": "https://api.github.com/users/gabitza-tech/gists{/gist_id}", "starred_url": "https://api.github.com/users/gabitza-tech/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gabitza-tech/subscriptions", "organizations_url": "https://api.github.com/users/gabitza-tech/orgs", "repos_url": "https://api.github.com/users/gabitza-tech/repos", "events_url": "https://api.github.com/users/gabitza-tech/events{/privacy}", "received_events_url": "https://api.github.com/users/gabitza-tech/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 1485815557, "node_id": "MDU6TGFiZWwxNDg1ODE1NTU3", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/bug", "name": "bug", "color": "d73a4a", "default": true, "description": "Something isn't working" } ]	closed	false	null	[]	null	null	2023-08-01T10:08:03	2023-10-17T01:17:33	2023-08-29T14:00:22	CONTRIBUTOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	Describe the bug I am using nemo docker v23.04 , but I also observe the following behavior in a conda environment too. I am training several conformer models ( same behavior using fastconformer transducer or conformer transducer) with several perturbations ( speed and noise from musan) and at the end of the epoch, when validation is performed, i observe the following behavior: ``` [NeMo I 2023-08-01 09:14:24 rnnt_wer_bpe:395] reference :el își joacă postul în meciul cu sportul studențesc [NeMo I 2023-08-01 09:14:24 rnnt_wer_bpe:396] predicted : ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ Epoch 0: 100%\|██████████████████████████████\| 16235/16235 [2:21:02<00:00, 1.92it/s, loss=5.14, v_num=2]Epoch 0, global step 503: 'val_wer' reached 57.20575 (best 57.20575), saving model to '/mnt/zfs-mirror-02/homes/gpirlogeanu/git_repos/nemoasr/examples/asr/train_dir/nemo_experiments/fastconformer_BAS/checkpoints/fastconformer_BAS--val_wer=57.2057-epoch=0.ckpt' as top 5 ``` If I don't use augmentations, it shows the predictions and works ok. The val_loss is NaN and the val_wer displays a random value. I think it might be a problem with my noise manifest, but I am unsure why, when I was augmenting VAD audios, I remember making the segments equal to 0.63s, but I thought in case of ASR, that perturb automatically cuts the noise audios. It looks like this: ``` {"audio_filepath": "/home/gpirlogeanu/musan/noise/all_noise/noise-free-sound-0000.wav", "duration": 17.65, "label": "noise", "text": "_", "offset": 0.0} {"audio_filepath": "/home/gpirlogeanu/musan/noise/all_noise/noise-free-sound-0001.wav", "duration": 40.53, "label": "noise", "text": "_", "offset": 0.0} {"audio_filepath": "/home/gpirlogeanu/musan/noise/all_noise/noise-free-sound-0002.wav", "duration": 71.21, "label": "noise", "text": "_", "offset": 0.0} {"audio_filepath": "/home/gpirlogeanu/musan/noise/all_noise/noise-free-sound-0003.wav", "duration": 14.23, "label": "noise", "text": "_", "offset": 0.0} {"audio_filepath": "/home/gpirlogeanu/musan/noise/all_noise/noise-free-sound-0004.wav", "duration": 2.98, "label": "noise", "text": "_", "offset": 0.0} {"audio_filepath": "/home/gpirlogeanu/musan/noise/all_noise/noise-free-sound-0005.wav", "duration": 21.18, "label": "noise", "text": "_", "offset": 0.0} {"audio_filepath": "/home/gpirlogeanu/musan/noise/all_noise/noise-free-sound-0006.wav", "duration": 7.77, "label": "noise", "text": "_", "offset": 0.0} {"audio_filepath": "/home/gpirlogeanu/musan/noise/all_noise/noise-free-sound-0007.wav", "duration": 16.61, "label": "noise", "text": "_", "offset": 0.0} ``` This is my config file: ``` # It contains the default values for training a Fast Conformer-Transducer ASR model, large size (~120M) with Transducer loss and sub-word encoding. # You may find more info about FastConformer here: https://docs.nvidia.com/deeplearning/nemo/user-guide/docs/en/stable/asr/models.html#fast-conformer # We suggest to use trainer.precision=bf16 for GPUs which support it otherwise trainer.precision=16 is recommended. # Using bf16 or 16 would make it possible to double the batch size and speedup training/inference. If fp16 is not stable and model diverges after some epochs, you may use fp32. # Here are the suggested batch size per GPU for each precision and memory sizes: # +-----------+------------+------------+ # \| Precision \| GPU Memory \| Batch Size \| # +===========+============+============+ # \| 32 \| 16GB \| 16 \| # \| \| 32GB \| 32 \| # \| \| 80GB \| 64 \| # +-----------+------------+------------+ # \| fp16 or \| 16GB \| 32 \| # \| bf16 \| 32GB \| 64 \| # \| \| 80GB \| 128 \| # +-----------+------------+------------+ # Here are the recommended configs for different variants of FastConformer-Transducer-BPE, other parameters are the same as in this config file. # # +--------------+---------+---------+----------+----------------+--------------+--------------------------+-----------------+------------+ # \| Model \| d_model \| n_heads \| n_layers \|conv_kernel_size\| weight_decay \| pred_hidden/joint_hidden \| pred_rnn_layers \| xscaling \| # +==============+=========+========+===========+================+==============+==========================+=================+============+ # \| Small (14M) \| 176 \| 4 \| 16 \| 9 \| 0.0 \| 320 \| 1 \| True \| # +--------------+---------+--------+-----------+----------------+--------------+--------------------------+-----------------+------------+ # \| Medium (32M) \| 256 \| 4 \| 16 \| 9 \| 1e-3 \| 640 \| 1 \| True \| # +--------------+---------+--------+-----------+----------------+--------------+--------------------------+-----------------+------------+ # \| Large (120M) \| 512 \| 8 \| 17 \| 9 \| 1e-3 \| 640 \| 1 \| True \| # +--------------+---------+--------+-----------+----------------+--------------+--------------------------+-----------------+------------+ # \| XLarge (616M)\| 1024 \| 8 \| 24 \| 9 \| 1e-3 \| 640 \| 2 \| True \| # +--------------+---------+--------+-----------+----------------+--------------+--------------------------+-----------------+------------+ # \| XXLarge(1.2B)\| 1024 \| 8 \| 42 \| 5 \| 1e-3 \| 640 \| 2 \| False \| # +--------------------------------------------------------------+--------------+--------------------------+-----------------+------------+ # Note: They are based on the assumption of max_duration of 20. If you have longer or shorter max_duration, then batch sizes may need to get updated accordingly. # Default learning parameters in this config are set for global batch size of 2K while you may use lower values. # To increase the global batch size with limited number of GPUs, you may use higher accumulate_grad_batches. # However accumulate_grad_batches is better to be avoided as long as the global batch size is large enough and training is stable. name: "FastConformer-Transducer-BPE" init_from_nemo_model: "models/stt_en_fastconformer_transducer_large.nemo" init_strict: false model: sample_rate: 16000 compute_eval_loss: true # eval samples can be very long and exhaust memory. Disable computation of transducer loss during validation/testing with this flag. log_prediction: true # enables logging sample predictions in the output during training rnnt_reduction: 'mean_volume' skip_nan_grad: false model_defaults: enc_hidden: ${model.encoder.d_model} pred_hidden: 640 joint_hidden: 640 train_ds: manifest_filepath: "manifests/RSC_train.json" sample_rate: ${model.sample_rate} batch_size: 32 # you may increase batch_size if your memory allows shuffle: true num_workers: 8 pin_memory: true max_duration: 20 # it is set for LibriSpeech, you may need to update it for your dataset min_duration: 0.1 # tarred datasets is_tarred: false tarred_audio_filepaths: null shuffle_n: 2048 # bucketing params bucketing_strategy: "fully_randomized" bucketing_batch_size: null augmentor: speed: prob: 0.4 sr: 16000 resample_type: "kaiser_best" min_speed_rate: 0.9 max_speed_rate: 1.1 noise: prob: 0.4 manifest_path: "manifests/noise_musan.json" min_snr_db: 10 max_snr_db: 30 rng: null orig_sr: 16000 validation_ds: manifest_filepath: "manifests/RSC_eval.json" sample_rate: ${model.sample_rate} batch_size: 16 shuffle: false use_start_end_token: false num_workers: 8 pin_memory: true test_ds: manifest_filepath: null sample_rate: ${model.sample_rate} batch_size: 16 shuffle: false use_start_end_token: false num_workers: 8 pin_memory: true # You may find more detail on how to train a tokenizer at: /scripts/tokenizers/process_asr_text_tokenizer.py tokenizer: dir: "tokenizers/5_normal_BAS+news_unigram/tokenizer_spe_unigram_v1024_max_5" # path to directory which contains either tokenizer.model (bpe) or vocab.txt (for wpe) type: bpe # Can be either bpe (SentencePiece tokenizer) or wpe (WordPiece tokenizer) preprocessor: _target_: nemo.collections.asr.modules.AudioToMelSpectrogramPreprocessor sample_rate: ${model.sample_rate} normalize: "per_feature" window_size: 0.025 window_stride: 0.01 window: "hann" features: 80 n_fft: 512 frame_splicing: 1 dither: 0.00001 pad_to: 0 spec_augment: _target_: nemo.collections.asr.modules.SpectrogramAugmentation freq_masks: 2 # set to zero to disable it time_masks: 10 # set to zero to disable it freq_width: 27 time_width: 0.05 encoder: _target_: nemo.collections.asr.modules.ConformerEncoder feat_in: ${model.preprocessor.features} feat_out: -1 # you may set it if you need different output size other than the default d_model n_layers: 17 d_model: 512 # Sub-sampling parameters subsampling: dw_striding # vggnet, striding, stacking or stacking_norm, dw_striding subsampling_factor: 8 # must be power of 2 for striding and vggnet subsampling_conv_channels: 256 # set to -1 to make it equal to the d_model causal_downsampling: false # Reduction parameters: Can be used to add another subsampling layer at a given position. # Having a 2x reduction will speedup the training and inference speech while keeping similar WER. # Adding it at the end will give the best WER while adding it at the beginning will give the best speedup. reduction: null # pooling, striding, or null reduction_position: null # Encoder block index or -1 for subsampling at the end of encoder reduction_factor: 1 # Feed forward module's params ff_expansion_factor: 4 # Multi-headed Attention Module's params self_attention_model: rel_pos # rel_pos or abs_pos n_heads: 8 # may need to be lower for smaller d_models # [left, right] specifies the number of steps to be seen from left and right of each step in self-attention att_context_size: [-1, -1] # -1 means unlimited context att_context_style: regular # regular or chunked_limited xscaling: true # scales up the input embeddings by sqrt(d_model) untie_biases: true # unties the biases of the TransformerXL layers pos_emb_max_len: 5000 # Convolution module's params conv_kernel_size: 9 conv_norm_type: 'batch_norm' # batch_norm or layer_norm or groupnormN (N specifies the number of groups) # conv_context_size can be"causal" or a list of two integers while conv_context_size[0]+conv_context_size[1]+1==conv_kernel_size # null means [(kernel_size-1)//2, (kernel_size-1)//2], and 'causal' means [(kernel_size-1), 0] conv_context_size: null ### regularization dropout: 0.1 # The dropout used in most of the Conformer Modules dropout_pre_encoder: 0.1 # The dropout used before the encoder dropout_emb: 0.0 # The dropout used for embeddings dropout_att: 0.1 # The dropout for multi-headed attention modules # set to non-zero to enable stochastic depth stochastic_depth_drop_prob: 0.0 stochastic_depth_mode: linear # linear or uniform stochastic_depth_start_layer: 1 decoder: _target_: nemo.collections.asr.modules.RNNTDecoder normalization_mode: null # Currently only null is supported for export. random_state_sampling: false # Random state sampling: https://arxiv.org/pdf/1910.11455.pdf blank_as_pad: true # This flag must be set in order to support exporting of RNNT models + efficient inference. prednet: pred_hidden: ${model.model_defaults.pred_hidden} pred_rnn_layers: 1 t_max: null dropout: 0.2 # if a large vocabulary size is desired, you may wish to use SampleRNNTJoint module # _target_: nemo.collections.asr.modules.SampledRNNTJoint # n_samples: 500 # Specifies the minimum number of tokens to sample from the vocabulary space, excluding # the RNNT blank token. If a given value is larger than the entire vocabulary size, then the full # vocabulary will be used joint: _target_: nemo.collections.asr.modules.RNNTJoint log_softmax: null # 'null' would set it automatically according to CPU/GPU device preserve_memory: false # dramatically slows down training, but might preserve some memory # Fuses the computation of prediction net + joint net + loss + WER calculation # to be run on sub-batches of size `fused_batch_size`. # When this flag is set to true, consider the `batch_size` of _ds to be just `encoder` batch size. # `fused_batch_size` is the actual batch size of the prediction net, joint net and transducer loss. # Using small values here will preserve a lot of memory during training, but will make training slower as well. # An optimal ratio of fused_batch_size : _ds.batch_size is 1:1. # However, to preserve memory, this ratio can be 1:8 or even 1:16. # Extreme case of 1:B (i.e. fused_batch_size=1) should be avoided as training speed would be very slow. fuse_loss_wer: true fused_batch_size: 4 jointnet: joint_hidden: ${model.model_defaults.joint_hidden} activation: "relu" dropout: 0.2 decoding: strategy: "greedy_batch" # can be greedy, greedy_batch, beam, tsd, alsd. # greedy strategy config greedy: max_symbols: 10 # beam strategy config beam: beam_size: 2 return_best_hypothesis: False score_norm: true tsd_max_sym_exp: 50 # for Time Synchronous Decoding alsd_max_target_len: 2.0 # for Alignment-Length Synchronous Decoding loss: loss_name: "default" warprnnt_numba_kwargs: # FastEmit regularization: https://arxiv.org/abs/2010.11148 # You may enable FastEmit to reduce the latency of the model for streaming fastemit_lambda: 0.0 # Recommended values to be in range [1e-4, 1e-2], 0.001 is a good start. clamp: -1.0 # if > 0, applies gradient clamping in range [-clamp, clamp] for the joint tensor only. optim: name: adamw lr: 5e-3 # optimizer arguments betas: [0.9, 0.98] weight_decay: 1e-3 # scheduler setup sched: name: CosineAnnealing # scheduler config override warmup_steps: 15000 warmup_ratio: null min_lr: 5e-4 trainer: devices: -1 # number of GPUs, -1 would use all available GPUs num_nodes: 1 max_epochs: 500 max_steps: -1 # computed at runtime if not set val_check_interval: 1.0 # Set to 0.25 to check 4 times per epoch, or an int for number of iterations accelerator: auto strategy: ddp accumulate_grad_batches: 16 gradient_clip_val: 0.0 precision: bf16 # 16, 32, or bf16 log_every_n_steps: 100 # Interval of logging. enable_progress_bar: True resume_from_checkpoint: null # The path to a checkpoint file to continue the training, restores the whole state including the epoch, step, LR schedulers, apex, etc. num_sanity_val_steps: 0 # number of steps to perform validation steps for sanity check the validation process before starting the training, setting to 0 disables it check_val_every_n_epoch: 1 # number of evaluations on validation every n epochs sync_batchnorm: true enable_checkpointing: False # Provided by exp_manager logger: false # Provided by exp_manager benchmark: false # needs to be false for models with variable-length speech input as it slows down training exp_manager: exp_dir: fastconformer name: ${name} create_tensorboard_logger: true create_checkpoint_callback: true checkpoint_callback_params: # in case of multiple validation sets, first one is used monitor: "val_wer" mode: "min" save_top_k: 5 always_save_nemo: True # saves the checkpoints as nemo files instead of PTL checkpoints resume_if_exists: True resume_ignore_no_checkpoint: True create_wandb_logger: false wandb_logger_kwargs: name: null project: null ``` Thanks in advance!	{ "login": "gabitza-tech", "id": 53811655, "node_id": "MDQ6VXNlcjUzODExNjU1", "avatar_url": "https://avatars.githubusercontent.com/u/53811655?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gabitza-tech", "html_url": "https://github.com/gabitza-tech", "followers_url": "https://api.github.com/users/gabitza-tech/followers", "following_url": "https://api.github.com/users/gabitza-tech/following{/other_user}", "gists_url": "https://api.github.com/users/gabitza-tech/gists{/gist_id}", "starred_url": "https://api.github.com/users/gabitza-tech/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gabitza-tech/subscriptions", "organizations_url": "https://api.github.com/users/gabitza-tech/orgs", "repos_url": "https://api.github.com/users/gabitza-tech/repos", "events_url": "https://api.github.com/users/gabitza-tech/events{/privacy}", "received_events_url": "https://api.github.com/users/gabitza-tech/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7145/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7145/timeline	null	not_planned	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7143	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7143/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7143/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7143/events	https://github.com/NVIDIA/NeMo/pull/7143	1,830,724,452	PR_kwDOC_bI7s5W4YCX	7,143	Fix default context size	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	[ { "id": 1923495043, "node_id": "MDU6TGFiZWwxOTIzNDk1MDQz", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/core", "name": "core", "color": "f25e3a", "default": false, "description": "Changes to NeMo Core" }, { "id": 4418253970, "node_id": "LA_kwDOC_bI7s8AAAABB1k0kg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/TTS", "name": "TTS", "color": "49AF28", "default": false, "description": "" }, { "id": 4556669155, "node_id": "LA_kwDOC_bI7s8AAAABD5lA4w", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/cherry-pick", "name": "cherry-pick", "color": "ededed", "default": false, "description": null }, { "id": 4811884691, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR", "name": "ASR", "color": "F3B33E", "default": false, "description": "" }, { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" }, { "id": 4840216855, "node_id": "LA_kwDOC_bI7s8AAAABIH_ZFw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/CI", "name": "CI", "color": "1D76DB", "default": false, "description": "" }, { "id": 4847373924, "node_id": "LA_kwDOC_bI7s8AAAABIO0OZA", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/common", "name": "common", "color": "c5def5", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-08-01T08:22:00	2023-08-01T15:18:58	2023-08-01T15:18:57	CONTRIBUTOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7143", "html_url": "https://github.com/NVIDIA/NeMo/pull/7143", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7143.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7143.patch", "merged_at": "2023-08-01T15:18:57" }	# What does this PR do ? There is a bug when performing RNNT/CTC decoding with hybrid models. By default `att_context_size` should be [-1,-1] which is what set as default, but if None is passed through cfg, it replaces the current value, which shouldn't be the case. Collection: ASR # Changelog - change value only if `cfg.att_context_size` is not None # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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 - [x] Bugfix - [ ] Documentation	{ "login": "nithinraok", "id": 19668129, "node_id": "MDQ6VXNlcjE5NjY4MTI5", "avatar_url": "https://avatars.githubusercontent.com/u/19668129?v=4", "gravatar_id": "", "url": "https://api.github.com/users/nithinraok", "html_url": "https://github.com/nithinraok", "followers_url": "https://api.github.com/users/nithinraok/followers", "following_url": "https://api.github.com/users/nithinraok/following{/other_user}", "gists_url": "https://api.github.com/users/nithinraok/gists{/gist_id}", "starred_url": "https://api.github.com/users/nithinraok/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/nithinraok/subscriptions", "organizations_url": "https://api.github.com/users/nithinraok/orgs", "repos_url": "https://api.github.com/users/nithinraok/repos", "events_url": "https://api.github.com/users/nithinraok/events{/privacy}", "received_events_url": "https://api.github.com/users/nithinraok/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7143/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7143/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7142	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7142/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7142/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7142/events	https://github.com/NVIDIA/NeMo/issues/7142	1,830,637,905	I_kwDOC_bI7s5tHVFR	7,142	Follow the README guide. The installation of the apex failed	{ "login": "shjwudp", "id": 11439912, "node_id": "MDQ6VXNlcjExNDM5OTEy", "avatar_url": "https://avatars.githubusercontent.com/u/11439912?v=4", "gravatar_id": "", "url": "https://api.github.com/users/shjwudp", "html_url": "https://github.com/shjwudp", "followers_url": "https://api.github.com/users/shjwudp/followers", "following_url": "https://api.github.com/users/shjwudp/following{/other_user}", "gists_url": "https://api.github.com/users/shjwudp/gists{/gist_id}", "starred_url": "https://api.github.com/users/shjwudp/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/shjwudp/subscriptions", "organizations_url": "https://api.github.com/users/shjwudp/orgs", "repos_url": "https://api.github.com/users/shjwudp/repos", "events_url": "https://api.github.com/users/shjwudp/events{/privacy}", "received_events_url": "https://api.github.com/users/shjwudp/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 1485815557, "node_id": "MDU6TGFiZWwxNDg1ODE1NTU3", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/bug", "name": "bug", "color": "d73a4a", "default": true, "description": "Something isn't working" }, { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null } ]	closed	false	null	[]	null	null	2023-08-01T07:28:20	2023-09-01T01:51:09	2023-09-01T01:51:09	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	Hi, team, I found that the apex version specified in README is a problematic version, and fast_layer_norm cannot be built normally, related issue https://github.com/NVIDIA/apex/issues/1636	{ "login": "shjwudp", "id": 11439912, "node_id": "MDQ6VXNlcjExNDM5OTEy", "avatar_url": "https://avatars.githubusercontent.com/u/11439912?v=4", "gravatar_id": "", "url": "https://api.github.com/users/shjwudp", "html_url": "https://github.com/shjwudp", "followers_url": "https://api.github.com/users/shjwudp/followers", "following_url": "https://api.github.com/users/shjwudp/following{/other_user}", "gists_url": "https://api.github.com/users/shjwudp/gists{/gist_id}", "starred_url": "https://api.github.com/users/shjwudp/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/shjwudp/subscriptions", "organizations_url": "https://api.github.com/users/shjwudp/orgs", "repos_url": "https://api.github.com/users/shjwudp/repos", "events_url": "https://api.github.com/users/shjwudp/events{/privacy}", "received_events_url": "https://api.github.com/users/shjwudp/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7142/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7142/timeline	null	completed	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7141	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7141/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7141/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7141/events	https://github.com/NVIDIA/NeMo/pull/7141	1,830,522,454	PR_kwDOC_bI7s5W3ssd	7,141	Fix default context size	{ "login": "nithinraok", "id": 19668129, "node_id": "MDQ6VXNlcjE5NjY4MTI5", "avatar_url": "https://avatars.githubusercontent.com/u/19668129?v=4", "gravatar_id": "", "url": "https://api.github.com/users/nithinraok", "html_url": "https://github.com/nithinraok", "followers_url": "https://api.github.com/users/nithinraok/followers", "following_url": "https://api.github.com/users/nithinraok/following{/other_user}", "gists_url": "https://api.github.com/users/nithinraok/gists{/gist_id}", "starred_url": "https://api.github.com/users/nithinraok/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/nithinraok/subscriptions", "organizations_url": "https://api.github.com/users/nithinraok/orgs", "repos_url": "https://api.github.com/users/nithinraok/repos", "events_url": "https://api.github.com/users/nithinraok/events{/privacy}", "received_events_url": "https://api.github.com/users/nithinraok/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 1923495043, "node_id": "MDU6TGFiZWwxOTIzNDk1MDQz", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/core", "name": "core", "color": "f25e3a", "default": false, "description": "Changes to NeMo Core" }, { "id": 4418253970, "node_id": "LA_kwDOC_bI7s8AAAABB1k0kg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/TTS", "name": "TTS", "color": "49AF28", "default": false, "description": "" }, { "id": 4811884691, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR", "name": "ASR", "color": "F3B33E", "default": false, "description": "" }, { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" }, { "id": 4840216855, "node_id": "LA_kwDOC_bI7s8AAAABIH_ZFw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/CI", "name": "CI", "color": "1D76DB", "default": false, "description": "" }, { "id": 4847373924, "node_id": "LA_kwDOC_bI7s8AAAABIO0OZA", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/common", "name": "common", "color": "c5def5", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-08-01T06:00:01	2023-08-01T08:21:31	2023-08-01T08:21:30	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7141", "html_url": "https://github.com/NVIDIA/NeMo/pull/7141", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7141.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7141.patch", "merged_at": "2023-08-01T08:21:30" }	# What does this PR do ? There is a bug when performing RNNT/CTC decoding with hybrid models. By default `att_context_size` should be [-1,-1] which is what set as default, but if None is passed through cfg, it replaces the current value, which shouldn't be the case. Collection: ASR # Changelog - change value only if `cfg.att_context_size` is not None # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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 - [x] Bugfix - [ ] Documentation	{ "login": "titu1994", "id": 3048602, "node_id": "MDQ6VXNlcjMwNDg2MDI=", "avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4", "gravatar_id": "", "url": "https://api.github.com/users/titu1994", "html_url": "https://github.com/titu1994", "followers_url": "https://api.github.com/users/titu1994/followers", "following_url": "https://api.github.com/users/titu1994/following{/other_user}", "gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}", "starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/titu1994/subscriptions", "organizations_url": "https://api.github.com/users/titu1994/orgs", "repos_url": "https://api.github.com/users/titu1994/repos", "events_url": "https://api.github.com/users/titu1994/events{/privacy}", "received_events_url": "https://api.github.com/users/titu1994/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7141/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7141/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7140	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7140/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7140/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7140/events	https://github.com/NVIDIA/NeMo/pull/7140	1,830,083,334	PR_kwDOC_bI7s5W2PBx	7,140	[DRAFT] DistributedOptimizer integration for distr ckpt	{ "login": "mikolajblaz", "id": 9979479, "node_id": "MDQ6VXNlcjk5Nzk0Nzk=", "avatar_url": "https://avatars.githubusercontent.com/u/9979479?v=4", "gravatar_id": "", "url": "https://api.github.com/users/mikolajblaz", "html_url": "https://github.com/mikolajblaz", "followers_url": "https://api.github.com/users/mikolajblaz/followers", "following_url": "https://api.github.com/users/mikolajblaz/following{/other_user}", "gists_url": "https://api.github.com/users/mikolajblaz/gists{/gist_id}", "starred_url": "https://api.github.com/users/mikolajblaz/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/mikolajblaz/subscriptions", "organizations_url": "https://api.github.com/users/mikolajblaz/orgs", "repos_url": "https://api.github.com/users/mikolajblaz/repos", "events_url": "https://api.github.com/users/mikolajblaz/events{/privacy}", "received_events_url": "https://api.github.com/users/mikolajblaz/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 1923495043, "node_id": "MDU6TGFiZWwxOTIzNDk1MDQz", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/core", "name": "core", "color": "f25e3a", "default": false, "description": "Changes to NeMo Core" } ]	closed	false	null	[]	null	null	2023-07-31T21:41:59	2023-08-14T21:37:48	2023-08-14T21:37:48	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7140", "html_url": "https://github.com/NVIDIA/NeMo/pull/7140", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7140.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7140.patch", "merged_at": null }	# 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 ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "ericharper", "id": 11999610, "node_id": "MDQ6VXNlcjExOTk5NjEw", "avatar_url": "https://avatars.githubusercontent.com/u/11999610?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ericharper", "html_url": "https://github.com/ericharper", "followers_url": "https://api.github.com/users/ericharper/followers", "following_url": "https://api.github.com/users/ericharper/following{/other_user}", "gists_url": "https://api.github.com/users/ericharper/gists{/gist_id}", "starred_url": "https://api.github.com/users/ericharper/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ericharper/subscriptions", "organizations_url": "https://api.github.com/users/ericharper/orgs", "repos_url": "https://api.github.com/users/ericharper/repos", "events_url": "https://api.github.com/users/ericharper/events{/privacy}", "received_events_url": "https://api.github.com/users/ericharper/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7140/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7140/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7139	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7139/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7139/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7139/events	https://github.com/NVIDIA/NeMo/issues/7139	1,829,887,732	I_kwDOC_bI7s5tEd70	7,139	How can I run NeMo for BERT-LARGE finetuning on SQuAD dataset for multinode multigpu configuration using Megatron-LM checkpoint?	{ "login": "Druva24", "id": 41902883, "node_id": "MDQ6VXNlcjQxOTAyODgz", "avatar_url": "https://avatars.githubusercontent.com/u/41902883?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Druva24", "html_url": "https://github.com/Druva24", "followers_url": "https://api.github.com/users/Druva24/followers", "following_url": "https://api.github.com/users/Druva24/following{/other_user}", "gists_url": "https://api.github.com/users/Druva24/gists{/gist_id}", "starred_url": "https://api.github.com/users/Druva24/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Druva24/subscriptions", "organizations_url": "https://api.github.com/users/Druva24/orgs", "repos_url": "https://api.github.com/users/Druva24/repos", "events_url": "https://api.github.com/users/Druva24/events{/privacy}", "received_events_url": "https://api.github.com/users/Druva24/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null } ]	closed	false	null	[]	null	null	2023-07-31T19:36:53	2023-09-07T01:44:08	2023-09-07T01:44:08	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	I have searched for documentation on how can I use NeMo for downstream tasks, I found this : https://docs.nvidia.com/deeplearning/nemo/user-guide/docs/en/v1.0.0/nlp/megatron_finetuning.html However, I haven't found any steps to get started with that do I need to make any code changes so that I can run the code on multinode-multiple config? Thanks	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7139/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7139/timeline	null	not_planned	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7138	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7138/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7138/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7138/events	https://github.com/NVIDIA/NeMo/pull/7138	1,829,618,434	PR_kwDOC_bI7s5W0pqF	7,138	Rename `megatron_amp_o2` to `megatron_amp_O2` in source code to be consistent	{ "login": "yaoyu-33", "id": 54727607, "node_id": "MDQ6VXNlcjU0NzI3NjA3", "avatar_url": "https://avatars.githubusercontent.com/u/54727607?v=4", "gravatar_id": "", "url": "https://api.github.com/users/yaoyu-33", "html_url": "https://github.com/yaoyu-33", "followers_url": "https://api.github.com/users/yaoyu-33/followers", "following_url": "https://api.github.com/users/yaoyu-33/following{/other_user}", "gists_url": "https://api.github.com/users/yaoyu-33/gists{/gist_id}", "starred_url": "https://api.github.com/users/yaoyu-33/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/yaoyu-33/subscriptions", "organizations_url": "https://api.github.com/users/yaoyu-33/orgs", "repos_url": "https://api.github.com/users/yaoyu-33/repos", "events_url": "https://api.github.com/users/yaoyu-33/events{/privacy}", "received_events_url": "https://api.github.com/users/yaoyu-33/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null }, { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-07-31T16:27:39	2023-09-06T01:44:16	2023-09-06T01:44:16	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7138", "html_url": "https://github.com/NVIDIA/NeMo/pull/7138", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7138.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7138.patch", "merged_at": null }	# What does this PR do ? Rename `megatron_amp_o2` to `megatron_amp_O2` in source code to be consistent 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 ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7138/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7138/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7137	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7137/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7137/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7137/events	https://github.com/NVIDIA/NeMo/pull/7137	1,829,572,045	PR_kwDOC_bI7s5W0fT3	7,137	Fixed main and merging this to r1.20	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	[ { "id": 4556669155, "node_id": "LA_kwDOC_bI7s8AAAABD5lA4w", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/cherry-pick", "name": "cherry-pick", "color": "ededed", "default": false, "description": null }, { "id": 4811884691, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR", "name": "ASR", "color": "F3B33E", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-07-31T16:04:00	2023-08-04T20:46:14	2023-08-04T20:46:14	CONTRIBUTOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7137", "html_url": "https://github.com/NVIDIA/NeMo/pull/7137", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7137.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7137.patch", "merged_at": null }	# What does this PR do ? Fix vad_utils.py for r1.12.0 Redoing https://github.com/NVIDIA/NeMo/pull/7113 for r1.20.0 Collection: ASR # Changelog The lines that are treating "None" in vad_utils.py # Usage Run Speaker_Diarization_Inference.ipynb # Before your PR is "Ready for review" Pre checks: - [x] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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 - [x] Bugfix - [ ] Documentation If you haven't finished some of the above items you can still open "Draft" PR. ## Who can review? Anyone in NeMo ASR # Additional Information * Related to # (issue)	{ "login": "ericharper", "id": 11999610, "node_id": "MDQ6VXNlcjExOTk5NjEw", "avatar_url": "https://avatars.githubusercontent.com/u/11999610?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ericharper", "html_url": "https://github.com/ericharper", "followers_url": "https://api.github.com/users/ericharper/followers", "following_url": "https://api.github.com/users/ericharper/following{/other_user}", "gists_url": "https://api.github.com/users/ericharper/gists{/gist_id}", "starred_url": "https://api.github.com/users/ericharper/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ericharper/subscriptions", "organizations_url": "https://api.github.com/users/ericharper/orgs", "repos_url": "https://api.github.com/users/ericharper/repos", "events_url": "https://api.github.com/users/ericharper/events{/privacy}", "received_events_url": "https://api.github.com/users/ericharper/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7137/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7137/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7136	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7136/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7136/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7136/events	https://github.com/NVIDIA/NeMo/pull/7136	1,829,559,735	PR_kwDOC_bI7s5W0cfb	7,136	Update notebook branch	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	[ { "id": 4556669155, "node_id": "LA_kwDOC_bI7s8AAAABD5lA4w", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/cherry-pick", "name": "cherry-pick", "color": "ededed", "default": false, "description": null }, { "id": 4811884691, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR", "name": "ASR", "color": "F3B33E", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-07-31T15:59:36	2023-08-04T20:46:40	2023-08-04T20:46:35	CONTRIBUTOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7136", "html_url": "https://github.com/NVIDIA/NeMo/pull/7136", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7136.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7136.patch", "merged_at": null }	# 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 ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "ericharper", "id": 11999610, "node_id": "MDQ6VXNlcjExOTk5NjEw", "avatar_url": "https://avatars.githubusercontent.com/u/11999610?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ericharper", "html_url": "https://github.com/ericharper", "followers_url": "https://api.github.com/users/ericharper/followers", "following_url": "https://api.github.com/users/ericharper/following{/other_user}", "gists_url": "https://api.github.com/users/ericharper/gists{/gist_id}", "starred_url": "https://api.github.com/users/ericharper/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ericharper/subscriptions", "organizations_url": "https://api.github.com/users/ericharper/orgs", "repos_url": "https://api.github.com/users/ericharper/repos", "events_url": "https://api.github.com/users/ericharper/events{/privacy}", "received_events_url": "https://api.github.com/users/ericharper/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7136/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7136/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7135	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7135/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7135/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7135/events	https://github.com/NVIDIA/NeMo/pull/7135	1,829,479,428	PR_kwDOC_bI7s5W0KMu	7,135	Update notebook branch	{ "login": "ericharper", "id": 11999610, "node_id": "MDQ6VXNlcjExOTk5NjEw", "avatar_url": "https://avatars.githubusercontent.com/u/11999610?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ericharper", "html_url": "https://github.com/ericharper", "followers_url": "https://api.github.com/users/ericharper/followers", "following_url": "https://api.github.com/users/ericharper/following{/other_user}", "gists_url": "https://api.github.com/users/ericharper/gists{/gist_id}", "starred_url": "https://api.github.com/users/ericharper/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ericharper/subscriptions", "organizations_url": "https://api.github.com/users/ericharper/orgs", "repos_url": "https://api.github.com/users/ericharper/repos", "events_url": "https://api.github.com/users/ericharper/events{/privacy}", "received_events_url": "https://api.github.com/users/ericharper/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4811884691, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR", "name": "ASR", "color": "F3B33E", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-07-31T15:30:36	2023-07-31T15:59:08	2023-07-31T15:59:06	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7135", "html_url": "https://github.com/NVIDIA/NeMo/pull/7135", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7135.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7135.patch", "merged_at": "2023-07-31T15:59:06" }	# 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 ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "ericharper", "id": 11999610, "node_id": "MDQ6VXNlcjExOTk5NjEw", "avatar_url": "https://avatars.githubusercontent.com/u/11999610?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ericharper", "html_url": "https://github.com/ericharper", "followers_url": "https://api.github.com/users/ericharper/followers", "following_url": "https://api.github.com/users/ericharper/following{/other_user}", "gists_url": "https://api.github.com/users/ericharper/gists{/gist_id}", "starred_url": "https://api.github.com/users/ericharper/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ericharper/subscriptions", "organizations_url": "https://api.github.com/users/ericharper/orgs", "repos_url": "https://api.github.com/users/ericharper/repos", "events_url": "https://api.github.com/users/ericharper/events{/privacy}", "received_events_url": "https://api.github.com/users/ericharper/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7135/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7135/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7134	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7134/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7134/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7134/events	https://github.com/NVIDIA/NeMo/issues/7134	1,828,847,935	I_kwDOC_bI7s5tAgE_	7,134	SpellMapper / Missing generate_configs.sh	{ "login": "thomaschhh", "id": 64537745, "node_id": "MDQ6VXNlcjY0NTM3NzQ1", "avatar_url": "https://avatars.githubusercontent.com/u/64537745?v=4", "gravatar_id": "", "url": "https://api.github.com/users/thomaschhh", "html_url": "https://github.com/thomaschhh", "followers_url": "https://api.github.com/users/thomaschhh/followers", "following_url": "https://api.github.com/users/thomaschhh/following{/other_user}", "gists_url": "https://api.github.com/users/thomaschhh/gists{/gist_id}", "starred_url": "https://api.github.com/users/thomaschhh/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/thomaschhh/subscriptions", "organizations_url": "https://api.github.com/users/thomaschhh/orgs", "repos_url": "https://api.github.com/users/thomaschhh/repos", "events_url": "https://api.github.com/users/thomaschhh/events{/privacy}", "received_events_url": "https://api.github.com/users/thomaschhh/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 1485815557, "node_id": "MDU6TGFiZWwxNDg1ODE1NTU3", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/bug", "name": "bug", "color": "d73a4a", "default": true, "description": "Something isn't working" } ]	closed	false	null	[]	null	null	2023-07-31T10:05:43	2023-08-25T06:25:39	2023-08-25T06:25:38	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	I am currently looking into your code base and am wondering where to find the above-mentioned script that you are referring to [here](https://github.com/NVIDIA/NeMo/blob/2fa7abfb7311f1a4115dfa47afb80339697ec88d/examples/nlp/spellchecking_asr_customization/run_training.sh#L34C29-L34C29). I have already searched in your repo but can't seem to find it. Any help would be greatly appreciated.	{ "login": "thomaschhh", "id": 64537745, "node_id": "MDQ6VXNlcjY0NTM3NzQ1", "avatar_url": "https://avatars.githubusercontent.com/u/64537745?v=4", "gravatar_id": "", "url": "https://api.github.com/users/thomaschhh", "html_url": "https://github.com/thomaschhh", "followers_url": "https://api.github.com/users/thomaschhh/followers", "following_url": "https://api.github.com/users/thomaschhh/following{/other_user}", "gists_url": "https://api.github.com/users/thomaschhh/gists{/gist_id}", "starred_url": "https://api.github.com/users/thomaschhh/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/thomaschhh/subscriptions", "organizations_url": "https://api.github.com/users/thomaschhh/orgs", "repos_url": "https://api.github.com/users/thomaschhh/repos", "events_url": "https://api.github.com/users/thomaschhh/events{/privacy}", "received_events_url": "https://api.github.com/users/thomaschhh/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7134/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7134/timeline	null	completed	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7133	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7133/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7133/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7133/events	https://github.com/NVIDIA/NeMo/issues/7133	1,828,497,055	I_kwDOC_bI7s5s_Kaf	7,133	Regarding Synthetic tabular data generation	{ "login": "Vathsa28", "id": 44828631, "node_id": "MDQ6VXNlcjQ0ODI4NjMx", "avatar_url": "https://avatars.githubusercontent.com/u/44828631?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Vathsa28", "html_url": "https://github.com/Vathsa28", "followers_url": "https://api.github.com/users/Vathsa28/followers", "following_url": "https://api.github.com/users/Vathsa28/following{/other_user}", "gists_url": "https://api.github.com/users/Vathsa28/gists{/gist_id}", "starred_url": "https://api.github.com/users/Vathsa28/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Vathsa28/subscriptions", "organizations_url": "https://api.github.com/users/Vathsa28/orgs", "repos_url": "https://api.github.com/users/Vathsa28/repos", "events_url": "https://api.github.com/users/Vathsa28/events{/privacy}", "received_events_url": "https://api.github.com/users/Vathsa28/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null } ]	closed	false	{ "login": "okuchaiev", "id": 22035961, "node_id": "MDQ6VXNlcjIyMDM1OTYx", "avatar_url": "https://avatars.githubusercontent.com/u/22035961?v=4", "gravatar_id": "", "url": "https://api.github.com/users/okuchaiev", "html_url": "https://github.com/okuchaiev", "followers_url": "https://api.github.com/users/okuchaiev/followers", "following_url": "https://api.github.com/users/okuchaiev/following{/other_user}", "gists_url": "https://api.github.com/users/okuchaiev/gists{/gist_id}", "starred_url": "https://api.github.com/users/okuchaiev/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/okuchaiev/subscriptions", "organizations_url": "https://api.github.com/users/okuchaiev/orgs", "repos_url": "https://api.github.com/users/okuchaiev/repos", "events_url": "https://api.github.com/users/okuchaiev/events{/privacy}", "received_events_url": "https://api.github.com/users/okuchaiev/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "login": "okuchaiev", "id": 22035961, "node_id": "MDQ6VXNlcjIyMDM1OTYx", "avatar_url": "https://avatars.githubusercontent.com/u/22035961?v=4", "gravatar_id": "", "url": "https://api.github.com/users/okuchaiev", "html_url": "https://github.com/okuchaiev", "followers_url": "https://api.github.com/users/okuchaiev/followers", "following_url": "https://api.github.com/users/okuchaiev/following{/other_user}", "gists_url": "https://api.github.com/users/okuchaiev/gists{/gist_id}", "starred_url": "https://api.github.com/users/okuchaiev/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/okuchaiev/subscriptions", "organizations_url": "https://api.github.com/users/okuchaiev/orgs", "repos_url": "https://api.github.com/users/okuchaiev/repos", "events_url": "https://api.github.com/users/okuchaiev/events{/privacy}", "received_events_url": "https://api.github.com/users/okuchaiev/received_events", "type": "User", "user_view_type": "public", "site_admin": false } ]	null	null	2023-07-31T06:29:41	2023-09-07T01:44:10	2023-09-07T01:44:10	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	Hello, I was hoping the NVIDIA team would release a script to evaluate the generated synthetic data appropriately, Fine grained , course grained and medium grained evaluations shown in the video was done very nicely. Please release the script for that. Best Regards	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7133/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7133/timeline	null	not_planned	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7132	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7132/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7132/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7132/events	https://github.com/NVIDIA/NeMo/pull/7132	1,828,005,334	PR_kwDOC_bI7s5WvI8n	7,132	Grouped query attention support	{ "login": "LydiaXiaohongLi", "id": 35672632, "node_id": "MDQ6VXNlcjM1NjcyNjMy", "avatar_url": "https://avatars.githubusercontent.com/u/35672632?v=4", "gravatar_id": "", "url": "https://api.github.com/users/LydiaXiaohongLi", "html_url": "https://github.com/LydiaXiaohongLi", "followers_url": "https://api.github.com/users/LydiaXiaohongLi/followers", "following_url": "https://api.github.com/users/LydiaXiaohongLi/following{/other_user}", "gists_url": "https://api.github.com/users/LydiaXiaohongLi/gists{/gist_id}", "starred_url": "https://api.github.com/users/LydiaXiaohongLi/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/LydiaXiaohongLi/subscriptions", "organizations_url": "https://api.github.com/users/LydiaXiaohongLi/orgs", "repos_url": "https://api.github.com/users/LydiaXiaohongLi/repos", "events_url": "https://api.github.com/users/LydiaXiaohongLi/events{/privacy}", "received_events_url": "https://api.github.com/users/LydiaXiaohongLi/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null }, { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-07-30T17:11:13	2023-08-22T01:43:50	2023-08-22T01:43:50	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7132", "html_url": "https://github.com/NVIDIA/NeMo/pull/7132", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7132.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7132.patch", "merged_at": null }	# What does this PR do ? Support grouped query attention for language modelling, with torch attention and flash attention 2. Collection: NLP # Changelog Support grouped query attention for language modelling, with torch attention and flash attention 2. - Added parameter parse to enable GQA for gpt like models (change in nemo/collections/nlp/models/language_modeling/megatron_gpt_model.py, nemo/collections/nlp/models/language_modeling/megatron/gpt_model.py) - Added support for grouped query attention for language modeling (change in nemo/collections/nlp/modules/common/megatron/language_model.py, nemo/collections/nlp/modules/common/megatron/transformer.py, nemo/collections/nlp/modules/common/megatron/attention.py) - Not implemented for flash attention triton yet, as it is marked as experimental - Not implemented for xpos yet, as it is marked as experimental # Usage Configure num_kv_attention_heads in model config yaml file, e.g. add in examples/nlp/language_modeling/conf/megatron_gpt_config.yaml, in model configuration. num_kv_attention_heads is optional configuraiton, if not configured or configured same as num_attention_heads, the existing multi head attention will be used instead. Requires num_attention_heads as same or multiple of num_kv_attention_heads (same -> multi head attention), num_kv_attention_heads as same or multiple of tensor_model_parallel_size. Same convention as flash attention, e.g. if Q has 6 heads and K, V have 2 heads, head 0, 1, 2 of Q will attention to head 0 of K, V, and head 3, 4, 5 of Q will attention to head 1 of K, V. # Before your PR is "Ready for review" Pre checks: - [x] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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: - [x] New Feature - [ ] Bugfix - [ ] Documentation ## Who can review? Anyone in the community is free to review the PR once the checks have passed. [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7132/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7132/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7131	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7131/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7131/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7131/events	https://github.com/NVIDIA/NeMo/issues/7131	1,827,967,478	I_kwDOC_bI7s5s9JH2	7,131	Convert ckpt file to nemo	{ "login": "SagyHarpazGong", "id": 99274232, "node_id": "U_kgDOBerN-A", "avatar_url": "https://avatars.githubusercontent.com/u/99274232?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SagyHarpazGong", "html_url": "https://github.com/SagyHarpazGong", "followers_url": "https://api.github.com/users/SagyHarpazGong/followers", "following_url": "https://api.github.com/users/SagyHarpazGong/following{/other_user}", "gists_url": "https://api.github.com/users/SagyHarpazGong/gists{/gist_id}", "starred_url": "https://api.github.com/users/SagyHarpazGong/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SagyHarpazGong/subscriptions", "organizations_url": "https://api.github.com/users/SagyHarpazGong/orgs", "repos_url": "https://api.github.com/users/SagyHarpazGong/repos", "events_url": "https://api.github.com/users/SagyHarpazGong/events{/privacy}", "received_events_url": "https://api.github.com/users/SagyHarpazGong/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null } ]	closed	false	{ "login": "okuchaiev", "id": 22035961, "node_id": "MDQ6VXNlcjIyMDM1OTYx", "avatar_url": "https://avatars.githubusercontent.com/u/22035961?v=4", "gravatar_id": "", "url": "https://api.github.com/users/okuchaiev", "html_url": "https://github.com/okuchaiev", "followers_url": "https://api.github.com/users/okuchaiev/followers", "following_url": "https://api.github.com/users/okuchaiev/following{/other_user}", "gists_url": "https://api.github.com/users/okuchaiev/gists{/gist_id}", "starred_url": "https://api.github.com/users/okuchaiev/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/okuchaiev/subscriptions", "organizations_url": "https://api.github.com/users/okuchaiev/orgs", "repos_url": "https://api.github.com/users/okuchaiev/repos", "events_url": "https://api.github.com/users/okuchaiev/events{/privacy}", "received_events_url": "https://api.github.com/users/okuchaiev/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "login": "okuchaiev", "id": 22035961, "node_id": "MDQ6VXNlcjIyMDM1OTYx", "avatar_url": "https://avatars.githubusercontent.com/u/22035961?v=4", "gravatar_id": "", "url": "https://api.github.com/users/okuchaiev", "html_url": "https://github.com/okuchaiev", "followers_url": "https://api.github.com/users/okuchaiev/followers", "following_url": "https://api.github.com/users/okuchaiev/following{/other_user}", "gists_url": "https://api.github.com/users/okuchaiev/gists{/gist_id}", "starred_url": "https://api.github.com/users/okuchaiev/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/okuchaiev/subscriptions", "organizations_url": "https://api.github.com/users/okuchaiev/orgs", "repos_url": "https://api.github.com/users/okuchaiev/repos", "events_url": "https://api.github.com/users/okuchaiev/events{/privacy}", "received_events_url": "https://api.github.com/users/okuchaiev/received_events", "type": "User", "user_view_type": "public", "site_admin": false } ]	null	null	2023-07-30T15:05:51	2023-09-09T01:43:37	2023-09-09T01:43:37	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	Hi. I trained msdd model but I stopped the training (not waited until the max_epoch value), is there any script to convert the ckpt file into nemo file, the ckpt file contain all the optimizer params and gradients. please advise.	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7131/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7131/timeline	null	not_planned	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7130	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7130/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7130/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7130/events	https://github.com/NVIDIA/NeMo/pull/7130	1,827,240,746	PR_kwDOC_bI7s5WstDj	7,130	Add updated fc ctc and rnnt xxl models	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	[ { "id": 1923495043, "node_id": "MDU6TGFiZWwxOTIzNDk1MDQz", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/core", "name": "core", "color": "f25e3a", "default": false, "description": "Changes to NeMo Core" }, { "id": 4418253970, "node_id": "LA_kwDOC_bI7s8AAAABB1k0kg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/TTS", "name": "TTS", "color": "49AF28", "default": false, "description": "" }, { "id": 4556669155, "node_id": "LA_kwDOC_bI7s8AAAABD5lA4w", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/cherry-pick", "name": "cherry-pick", "color": "ededed", "default": false, "description": null }, { "id": 4811884691, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR", "name": "ASR", "color": "F3B33E", "default": false, "description": "" }, { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" }, { "id": 4811888028, "node_id": "LA_kwDOC_bI7s8AAAABHs-VnA", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/Speaker%20Tasks", "name": "Speaker Tasks", "color": "871AD4", "default": false, "description": "" }, { "id": 4840216855, "node_id": "LA_kwDOC_bI7s8AAAABIH_ZFw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/CI", "name": "CI", "color": "1D76DB", "default": false, "description": "" }, { "id": 4847373924, "node_id": "LA_kwDOC_bI7s8AAAABIO0OZA", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/common", "name": "common", "color": "c5def5", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-07-29T00:37:46	2023-07-29T03:24:07	2023-07-29T03:24:07	CONTRIBUTOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7130", "html_url": "https://github.com/NVIDIA/NeMo/pull/7130", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7130.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7130.patch", "merged_at": "2023-07-29T03:24:06" }	# What does this PR do ? Add updated FC-XXL RNNT and CTC models Collection: ASR # Changelog - Add pre trained models to code - Add to docs # Usage ```python # FC Transducer BPE models from nemo.collections.asr.models import EncDecRNNTBPEModel fc_xxl_rnnt = EncDecRNNTBPEModel.from_pretrained('stt_en_fastconformer_transducer_xxlarge') # FC CTC BPE models from nemo.collections.asr.models import EncDecCTCModelBPE fc_xxl_ctc = EncDecCTCModelBPE.from_pretrained('stt_en_fastconformer_ctc_xxlarge') ``` # Before your PR is "Ready for review" Pre checks: - [x] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] Did you write any new necessary tests? - [x] 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: - [x] New Feature - [ ] Bugfix - [x] Documentation	{ "login": "nithinraok", "id": 19668129, "node_id": "MDQ6VXNlcjE5NjY4MTI5", "avatar_url": "https://avatars.githubusercontent.com/u/19668129?v=4", "gravatar_id": "", "url": "https://api.github.com/users/nithinraok", "html_url": "https://github.com/nithinraok", "followers_url": "https://api.github.com/users/nithinraok/followers", "following_url": "https://api.github.com/users/nithinraok/following{/other_user}", "gists_url": "https://api.github.com/users/nithinraok/gists{/gist_id}", "starred_url": "https://api.github.com/users/nithinraok/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/nithinraok/subscriptions", "organizations_url": "https://api.github.com/users/nithinraok/orgs", "repos_url": "https://api.github.com/users/nithinraok/repos", "events_url": "https://api.github.com/users/nithinraok/events{/privacy}", "received_events_url": "https://api.github.com/users/nithinraok/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7130/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7130/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7129	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7129/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7129/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7129/events	https://github.com/NVIDIA/NeMo/issues/7129	1,827,131,611	I_kwDOC_bI7s5s59Db	7,129	ONNX model takes too long to run	{ "login": "divyansh2681", "id": 46630567, "node_id": "MDQ6VXNlcjQ2NjMwNTY3", "avatar_url": "https://avatars.githubusercontent.com/u/46630567?v=4", "gravatar_id": "", "url": "https://api.github.com/users/divyansh2681", "html_url": "https://github.com/divyansh2681", "followers_url": "https://api.github.com/users/divyansh2681/followers", "following_url": "https://api.github.com/users/divyansh2681/following{/other_user}", "gists_url": "https://api.github.com/users/divyansh2681/gists{/gist_id}", "starred_url": "https://api.github.com/users/divyansh2681/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/divyansh2681/subscriptions", "organizations_url": "https://api.github.com/users/divyansh2681/orgs", "repos_url": "https://api.github.com/users/divyansh2681/repos", "events_url": "https://api.github.com/users/divyansh2681/events{/privacy}", "received_events_url": "https://api.github.com/users/divyansh2681/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[]	closed	false	null	[]	null	null	2023-07-28T22:02:32	2023-10-31T16:10:53	2023-09-01T02:24:51	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	I have exported a NeMo ASR model to ONNX. When I run the ONNX model for inference, it takes much much more time as compared to the original .nemo model. I am running both of them on the same machine in the same conda environment.	{ "login": "titu1994", "id": 3048602, "node_id": "MDQ6VXNlcjMwNDg2MDI=", "avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4", "gravatar_id": "", "url": "https://api.github.com/users/titu1994", "html_url": "https://github.com/titu1994", "followers_url": "https://api.github.com/users/titu1994/followers", "following_url": "https://api.github.com/users/titu1994/following{/other_user}", "gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}", "starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/titu1994/subscriptions", "organizations_url": "https://api.github.com/users/titu1994/orgs", "repos_url": "https://api.github.com/users/titu1994/repos", "events_url": "https://api.github.com/users/titu1994/events{/privacy}", "received_events_url": "https://api.github.com/users/titu1994/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7129/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7129/timeline	null	completed	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7128	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7128/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7128/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7128/events	https://github.com/NVIDIA/NeMo/pull/7128	1,827,098,092	PR_kwDOC_bI7s5WsOJD	7,128	Add updated fc ctc and rnnt xxl models	{ "login": "nithinraok", "id": 19668129, "node_id": "MDQ6VXNlcjE5NjY4MTI5", "avatar_url": "https://avatars.githubusercontent.com/u/19668129?v=4", "gravatar_id": "", "url": "https://api.github.com/users/nithinraok", "html_url": "https://github.com/nithinraok", "followers_url": "https://api.github.com/users/nithinraok/followers", "following_url": "https://api.github.com/users/nithinraok/following{/other_user}", "gists_url": "https://api.github.com/users/nithinraok/gists{/gist_id}", "starred_url": "https://api.github.com/users/nithinraok/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/nithinraok/subscriptions", "organizations_url": "https://api.github.com/users/nithinraok/orgs", "repos_url": "https://api.github.com/users/nithinraok/repos", "events_url": "https://api.github.com/users/nithinraok/events{/privacy}", "received_events_url": "https://api.github.com/users/nithinraok/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 1923495043, "node_id": "MDU6TGFiZWwxOTIzNDk1MDQz", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/core", "name": "core", "color": "f25e3a", "default": false, "description": "Changes to NeMo Core" }, { "id": 4418253970, "node_id": "LA_kwDOC_bI7s8AAAABB1k0kg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/TTS", "name": "TTS", "color": "49AF28", "default": false, "description": "" }, { "id": 4811884691, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR", "name": "ASR", "color": "F3B33E", "default": false, "description": "" }, { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" }, { "id": 4811888028, "node_id": "LA_kwDOC_bI7s8AAAABHs-VnA", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/Speaker%20Tasks", "name": "Speaker Tasks", "color": "871AD4", "default": false, "description": "" }, { "id": 4840216855, "node_id": "LA_kwDOC_bI7s8AAAABIH_ZFw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/CI", "name": "CI", "color": "1D76DB", "default": false, "description": "" }, { "id": 4847373924, "node_id": "LA_kwDOC_bI7s8AAAABIO0OZA", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/common", "name": "common", "color": "c5def5", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-07-28T21:16:39	2023-07-29T00:37:20	2023-07-29T00:37:19	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7128", "html_url": "https://github.com/NVIDIA/NeMo/pull/7128", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7128.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7128.patch", "merged_at": "2023-07-29T00:37:19" }	# What does this PR do ? Add updated FC-XXL RNNT and CTC models Collection: ASR # Changelog - Add pre trained models to code - Add to docs # Usage ```python # FC Transducer BPE models from nemo.collections.asr.models import EncDecRNNTBPEModel fc_xxl_rnnt = EncDecRNNTBPEModel.from_pretrained('stt_en_fastconformer_transducer_xxlarge') # FC CTC BPE models from nemo.collections.asr.models import EncDecCTCModelBPE fc_xxl_ctc = EncDecCTCModelBPE.from_pretrained('stt_en_fastconformer_ctc_xxlarge') ``` # Before your PR is "Ready for review" Pre checks: - [x] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] Did you write any new necessary tests? - [x] 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: - [x] New Feature - [ ] Bugfix - [x] Documentation	{ "login": "titu1994", "id": 3048602, "node_id": "MDQ6VXNlcjMwNDg2MDI=", "avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4", "gravatar_id": "", "url": "https://api.github.com/users/titu1994", "html_url": "https://github.com/titu1994", "followers_url": "https://api.github.com/users/titu1994/followers", "following_url": "https://api.github.com/users/titu1994/following{/other_user}", "gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}", "starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/titu1994/subscriptions", "organizations_url": "https://api.github.com/users/titu1994/orgs", "repos_url": "https://api.github.com/users/titu1994/repos", "events_url": "https://api.github.com/users/titu1994/events{/privacy}", "received_events_url": "https://api.github.com/users/titu1994/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7128/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7128/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7127	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7127/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7127/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7127/events	https://github.com/NVIDIA/NeMo/pull/7127	1,827,090,301	PR_kwDOC_bI7s5WsMdx	7,127	Fixed main and merging this to r1.20	{ "login": "tango4j", "id": 13211483, "node_id": "MDQ6VXNlcjEzMjExNDgz", "avatar_url": "https://avatars.githubusercontent.com/u/13211483?v=4", "gravatar_id": "", "url": "https://api.github.com/users/tango4j", "html_url": "https://github.com/tango4j", "followers_url": "https://api.github.com/users/tango4j/followers", "following_url": "https://api.github.com/users/tango4j/following{/other_user}", "gists_url": "https://api.github.com/users/tango4j/gists{/gist_id}", "starred_url": "https://api.github.com/users/tango4j/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/tango4j/subscriptions", "organizations_url": "https://api.github.com/users/tango4j/orgs", "repos_url": "https://api.github.com/users/tango4j/repos", "events_url": "https://api.github.com/users/tango4j/events{/privacy}", "received_events_url": "https://api.github.com/users/tango4j/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4811884691, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR", "name": "ASR", "color": "F3B33E", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-07-28T21:06:51	2023-07-31T16:03:29	2023-07-31T16:03:28	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7127", "html_url": "https://github.com/NVIDIA/NeMo/pull/7127", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7127.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7127.patch", "merged_at": "2023-07-31T16:03:28" }	# What does this PR do ? Fix vad_utils.py for r1.12.0 Redoing https://github.com/NVIDIA/NeMo/pull/7113 for r1.20.0 Collection: ASR # Changelog The lines that are treating "None" in vad_utils.py # Usage Run Speaker_Diarization_Inference.ipynb # Before your PR is "Ready for review" Pre checks: - [x] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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 - [x] Bugfix - [ ] Documentation If you haven't finished some of the above items you can still open "Draft" PR. ## Who can review? Anyone in NeMo ASR # Additional Information * Related to # (issue)	{ "login": "tango4j", "id": 13211483, "node_id": "MDQ6VXNlcjEzMjExNDgz", "avatar_url": "https://avatars.githubusercontent.com/u/13211483?v=4", "gravatar_id": "", "url": "https://api.github.com/users/tango4j", "html_url": "https://github.com/tango4j", "followers_url": "https://api.github.com/users/tango4j/followers", "following_url": "https://api.github.com/users/tango4j/following{/other_user}", "gists_url": "https://api.github.com/users/tango4j/gists{/gist_id}", "starred_url": "https://api.github.com/users/tango4j/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/tango4j/subscriptions", "organizations_url": "https://api.github.com/users/tango4j/orgs", "repos_url": "https://api.github.com/users/tango4j/repos", "events_url": "https://api.github.com/users/tango4j/events{/privacy}", "received_events_url": "https://api.github.com/users/tango4j/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7127/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7127/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7126	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7126/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7126/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7126/events	https://github.com/NVIDIA/NeMo/pull/7126	1,827,002,497	PR_kwDOC_bI7s5Wr5Ob	7,126	Fix import guard checks	{ "login": "titu1994", "id": 3048602, "node_id": "MDQ6VXNlcjMwNDg2MDI=", "avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4", "gravatar_id": "", "url": "https://api.github.com/users/titu1994", "html_url": "https://github.com/titu1994", "followers_url": "https://api.github.com/users/titu1994/followers", "following_url": "https://api.github.com/users/titu1994/following{/other_user}", "gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}", "starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/titu1994/subscriptions", "organizations_url": "https://api.github.com/users/titu1994/orgs", "repos_url": "https://api.github.com/users/titu1994/repos", "events_url": "https://api.github.com/users/titu1994/events{/privacy}", "received_events_url": "https://api.github.com/users/titu1994/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[]	closed	false	null	[]	null	null	2023-07-28T19:41:23	2023-07-28T22:01:42	2023-07-28T22:01:38	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7126", "html_url": "https://github.com/NVIDIA/NeMo/pull/7126", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7126.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7126.patch", "merged_at": "2023-07-28T22:01:38" }	# What does this PR do ? Fix import guards properly Collection: [ASR, NLP] PR Type: - [ ] New Feature - [x] Bugfix - [ ] Documentation	{ "login": "titu1994", "id": 3048602, "node_id": "MDQ6VXNlcjMwNDg2MDI=", "avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4", "gravatar_id": "", "url": "https://api.github.com/users/titu1994", "html_url": "https://github.com/titu1994", "followers_url": "https://api.github.com/users/titu1994/followers", "following_url": "https://api.github.com/users/titu1994/following{/other_user}", "gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}", "starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/titu1994/subscriptions", "organizations_url": "https://api.github.com/users/titu1994/orgs", "repos_url": "https://api.github.com/users/titu1994/repos", "events_url": "https://api.github.com/users/titu1994/events{/privacy}", "received_events_url": "https://api.github.com/users/titu1994/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7126/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7126/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7125	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7125/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7125/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7125/events	https://github.com/NVIDIA/NeMo/pull/7125	1,827,000,599	PR_kwDOC_bI7s5Wr4z1	7,125	Revert "Fix import guard checks"	{ "login": "titu1994", "id": 3048602, "node_id": "MDQ6VXNlcjMwNDg2MDI=", "avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4", "gravatar_id": "", "url": "https://api.github.com/users/titu1994", "html_url": "https://github.com/titu1994", "followers_url": "https://api.github.com/users/titu1994/followers", "following_url": "https://api.github.com/users/titu1994/following{/other_user}", "gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}", "starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/titu1994/subscriptions", "organizations_url": "https://api.github.com/users/titu1994/orgs", "repos_url": "https://api.github.com/users/titu1994/repos", "events_url": "https://api.github.com/users/titu1994/events{/privacy}", "received_events_url": "https://api.github.com/users/titu1994/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[]	closed	false	null	[]	null	null	2023-07-28T19:39:20	2023-07-28T19:39:33	2023-07-28T19:39:32	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7125", "html_url": "https://github.com/NVIDIA/NeMo/pull/7125", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7125.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7125.patch", "merged_at": "2023-07-28T19:39:32" }	Reverts NVIDIA/NeMo#7124	{ "login": "titu1994", "id": 3048602, "node_id": "MDQ6VXNlcjMwNDg2MDI=", "avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4", "gravatar_id": "", "url": "https://api.github.com/users/titu1994", "html_url": "https://github.com/titu1994", "followers_url": "https://api.github.com/users/titu1994/followers", "following_url": "https://api.github.com/users/titu1994/following{/other_user}", "gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}", "starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/titu1994/subscriptions", "organizations_url": "https://api.github.com/users/titu1994/orgs", "repos_url": "https://api.github.com/users/titu1994/repos", "events_url": "https://api.github.com/users/titu1994/events{/privacy}", "received_events_url": "https://api.github.com/users/titu1994/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7125/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7125/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7124	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7124/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7124/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7124/events	https://github.com/NVIDIA/NeMo/pull/7124	1,826,995,448	PR_kwDOC_bI7s5Wr3sy	7,124	Fix import guard checks	{ "login": "titu1994", "id": 3048602, "node_id": "MDQ6VXNlcjMwNDg2MDI=", "avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4", "gravatar_id": "", "url": "https://api.github.com/users/titu1994", "html_url": "https://github.com/titu1994", "followers_url": "https://api.github.com/users/titu1994/followers", "following_url": "https://api.github.com/users/titu1994/following{/other_user}", "gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}", "starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/titu1994/subscriptions", "organizations_url": "https://api.github.com/users/titu1994/orgs", "repos_url": "https://api.github.com/users/titu1994/repos", "events_url": "https://api.github.com/users/titu1994/events{/privacy}", "received_events_url": "https://api.github.com/users/titu1994/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[]	closed	false	null	[]	null	null	2023-07-28T19:34:30	2023-07-28T19:36:20	2023-07-28T19:34:53	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7124", "html_url": "https://github.com/NVIDIA/NeMo/pull/7124", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7124.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7124.patch", "merged_at": "2023-07-28T19:34:53" }	# What does this PR do ? Fix import guards for flash attention and DALI Collection: [ASR, NLP] # Changelog - Add AttributeError support for import where classpath resolution fails. - Add import guards to pynvml # Before your PR is "Ready for review" Pre checks: - [x] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [x] 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 - [x] Bugfix - [ ] Documentation Closes #7122	{ "login": "titu1994", "id": 3048602, "node_id": "MDQ6VXNlcjMwNDg2MDI=", "avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4", "gravatar_id": "", "url": "https://api.github.com/users/titu1994", "html_url": "https://github.com/titu1994", "followers_url": "https://api.github.com/users/titu1994/followers", "following_url": "https://api.github.com/users/titu1994/following{/other_user}", "gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}", "starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/titu1994/subscriptions", "organizations_url": "https://api.github.com/users/titu1994/orgs", "repos_url": "https://api.github.com/users/titu1994/repos", "events_url": "https://api.github.com/users/titu1994/events{/privacy}", "received_events_url": "https://api.github.com/users/titu1994/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7124/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7124/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7123	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7123/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7123/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7123/events	https://github.com/NVIDIA/NeMo/pull/7123	1,826,799,436	PR_kwDOC_bI7s5WrL5_	7,123	Refactoring to make MegatronGPTModel dataset building more flexible	{ "login": "odelalleau", "id": 507137, "node_id": "MDQ6VXNlcjUwNzEzNw==", "avatar_url": "https://avatars.githubusercontent.com/u/507137?v=4", "gravatar_id": "", "url": "https://api.github.com/users/odelalleau", "html_url": "https://github.com/odelalleau", "followers_url": "https://api.github.com/users/odelalleau/followers", "following_url": "https://api.github.com/users/odelalleau/following{/other_user}", "gists_url": "https://api.github.com/users/odelalleau/gists{/gist_id}", "starred_url": "https://api.github.com/users/odelalleau/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/odelalleau/subscriptions", "organizations_url": "https://api.github.com/users/odelalleau/orgs", "repos_url": "https://api.github.com/users/odelalleau/repos", "events_url": "https://api.github.com/users/odelalleau/events{/privacy}", "received_events_url": "https://api.github.com/users/odelalleau/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null }, { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-07-28T17:23:45	2023-08-26T01:43:12	2023-08-26T01:43:11	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7123", "html_url": "https://github.com/NVIDIA/NeMo/pull/7123", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7123.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7123.patch", "merged_at": null }	# What does this PR do ? Refactoring to make MegatronGPTModel dataset building more flexible. In particular this allows subclasses to use a different `build_datasets_strategy` without having to copy / paste the full `build_train_valid_test_datasets()` method. Collection: `nlp` # Changelog - Add a new attribute `build_datasets_strategy` to the constructor of `MegatronGPTModel` to make it easier to override the dataset building logic # Usage ```python from my_module import my_build_datasets_strategy class MyClass(MegatronGPTModel): def __init__(self): super().__init__(build_datasets_strategy=my_build_datasets_strategy) ``` # Discussion There are at least two potential controversial design choices, happy to discuss them here (marking PR as draft in the meantime): 1. Naming: the most "natural" name would be `build_train_valid_test_datasets_strategy`, but I felt it was starting to get annoyingly long, which is why I picked a shorter variant 2. Implementation: alternatively, we could (a) avoid changing the signature of `__init__()`, letting subclasses override `self._build_datasets_strategy` as they see fit (I didn't go this route because I tend to prefer customization options to be clearly exposed through the API), or (b) make it part of the config, using Hydra's instantiate feature to obtain the strategy (although I actually like this option better, I wanted to discuss it here before making changes to the config). [edit: see [comment below](https://github.com/NVIDIA/NeMo/pull/7123#issuecomment-1665501843) for a follow-up discussion] I'll look into adding a test once the implementation is approved. # Before your PR is "Ready for review" Pre checks: - [X] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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: - [X] New Feature - [ ] Bugfix - [ ] Documentation	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7123/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7123/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7122	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7122/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7122/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7122/events	https://github.com/NVIDIA/NeMo/issues/7122	1,826,758,614	I_kwDOC_bI7s5s4h_W	7,122	Installation instructions should better indicate mandatory steps to make tests pass (or reinstall.sh needs an update)	{ "login": "odelalleau", "id": 507137, "node_id": "MDQ6VXNlcjUwNzEzNw==", "avatar_url": "https://avatars.githubusercontent.com/u/507137?v=4", "gravatar_id": "", "url": "https://api.github.com/users/odelalleau", "html_url": "https://github.com/odelalleau", "followers_url": "https://api.github.com/users/odelalleau/followers", "following_url": "https://api.github.com/users/odelalleau/following{/other_user}", "gists_url": "https://api.github.com/users/odelalleau/gists{/gist_id}", "starred_url": "https://api.github.com/users/odelalleau/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/odelalleau/subscriptions", "organizations_url": "https://api.github.com/users/odelalleau/orgs", "repos_url": "https://api.github.com/users/odelalleau/repos", "events_url": "https://api.github.com/users/odelalleau/events{/privacy}", "received_events_url": "https://api.github.com/users/odelalleau/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[]	closed	false	{ "login": "titu1994", "id": 3048602, "node_id": "MDQ6VXNlcjMwNDg2MDI=", "avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4", "gravatar_id": "", "url": "https://api.github.com/users/titu1994", "html_url": "https://github.com/titu1994", "followers_url": "https://api.github.com/users/titu1994/followers", "following_url": "https://api.github.com/users/titu1994/following{/other_user}", "gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}", "starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/titu1994/subscriptions", "organizations_url": "https://api.github.com/users/titu1994/orgs", "repos_url": "https://api.github.com/users/titu1994/repos", "events_url": "https://api.github.com/users/titu1994/events{/privacy}", "received_events_url": "https://api.github.com/users/titu1994/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "login": "titu1994", "id": 3048602, "node_id": "MDQ6VXNlcjMwNDg2MDI=", "avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4", "gravatar_id": "", "url": "https://api.github.com/users/titu1994", "html_url": "https://github.com/titu1994", "followers_url": "https://api.github.com/users/titu1994/followers", "following_url": "https://api.github.com/users/titu1994/following{/other_user}", "gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}", "starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/titu1994/subscriptions", "organizations_url": "https://api.github.com/users/titu1994/orgs", "repos_url": "https://api.github.com/users/titu1994/repos", "events_url": "https://api.github.com/users/titu1994/events{/privacy}", "received_events_url": "https://api.github.com/users/titu1994/received_events", "type": "User", "user_view_type": "public", "site_admin": false } ]	null	null	2023-07-28T16:56:00	2023-07-28T19:34:54	2023-07-28T19:34:54	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	Is your feature request related to a problem? Please describe. I wanted to setup a dev conda environment for NeMo, so I followed steps at https://github.com/NVIDIA/NeMo/tree/main#from-source Afterwards `pytest --cpu` was failing (before it could even run any test) with two errors: * `module 'nvidia' has no attribute 'dali'` * `No module named 'pynvml'` Describe the solution you'd like After manually installing both libraries with * pip install --extra-index-url https://developer.download.nvidia.com/compute/redist --upgrade nvidia-dali-cuda120 * pip install pynvml the tests were able to pass (`1420 passed, 304 skipped, 261 warnings`). Ideally these libraries would be installed automatically by the `reinstall.sh` script.	{ "login": "titu1994", "id": 3048602, "node_id": "MDQ6VXNlcjMwNDg2MDI=", "avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4", "gravatar_id": "", "url": "https://api.github.com/users/titu1994", "html_url": "https://github.com/titu1994", "followers_url": "https://api.github.com/users/titu1994/followers", "following_url": "https://api.github.com/users/titu1994/following{/other_user}", "gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}", "starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/titu1994/subscriptions", "organizations_url": "https://api.github.com/users/titu1994/orgs", "repos_url": "https://api.github.com/users/titu1994/repos", "events_url": "https://api.github.com/users/titu1994/events{/privacy}", "received_events_url": "https://api.github.com/users/titu1994/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7122/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7122/timeline	null	completed	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7121	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7121/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7121/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7121/events	https://github.com/NVIDIA/NeMo/pull/7121	1,826,613,552	PR_kwDOC_bI7s5Wqje4	7,121	RoPE length extrapolation with interpolation (#7005)	{ "login": "dimapihtar", "id": 37850217, "node_id": "MDQ6VXNlcjM3ODUwMjE3", "avatar_url": "https://avatars.githubusercontent.com/u/37850217?v=4", "gravatar_id": "", "url": "https://api.github.com/users/dimapihtar", "html_url": "https://github.com/dimapihtar", "followers_url": "https://api.github.com/users/dimapihtar/followers", "following_url": "https://api.github.com/users/dimapihtar/following{/other_user}", "gists_url": "https://api.github.com/users/dimapihtar/gists{/gist_id}", "starred_url": "https://api.github.com/users/dimapihtar/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/dimapihtar/subscriptions", "organizations_url": "https://api.github.com/users/dimapihtar/orgs", "repos_url": "https://api.github.com/users/dimapihtar/repos", "events_url": "https://api.github.com/users/dimapihtar/events{/privacy}", "received_events_url": "https://api.github.com/users/dimapihtar/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null }, { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-07-28T15:15:22	2023-08-22T01:43:52	2023-08-22T01:43:52	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7121", "html_url": "https://github.com/NVIDIA/NeMo/pull/7121", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7121.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7121.patch", "merged_at": null }	* Push changes * Fixes * add continue training script * [WIP] nonlinear interp * Fix * override encoder_seq_len * Remove nonlinear * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * sft with pi (#7006) * sft with pi * update values only if not None" --------- * Address comments * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Add info * Empty --------- # 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 ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7121/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7121/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7120	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7120/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7120/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7120/events	https://github.com/NVIDIA/NeMo/issues/7120	1,826,357,947	I_kwDOC_bI7s5s3AK7	7,120	In MegatronGPTModels, Creation of AutoCastTransformerLayer with TransformerEngine ignores chosen activation	{ "login": "shaltielshmid", "id": 20192446, "node_id": "MDQ6VXNlcjIwMTkyNDQ2", "avatar_url": "https://avatars.githubusercontent.com/u/20192446?v=4", "gravatar_id": "", "url": "https://api.github.com/users/shaltielshmid", "html_url": "https://github.com/shaltielshmid", "followers_url": "https://api.github.com/users/shaltielshmid/followers", "following_url": "https://api.github.com/users/shaltielshmid/following{/other_user}", "gists_url": "https://api.github.com/users/shaltielshmid/gists{/gist_id}", "starred_url": "https://api.github.com/users/shaltielshmid/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/shaltielshmid/subscriptions", "organizations_url": "https://api.github.com/users/shaltielshmid/orgs", "repos_url": "https://api.github.com/users/shaltielshmid/repos", "events_url": "https://api.github.com/users/shaltielshmid/events{/privacy}", "received_events_url": "https://api.github.com/users/shaltielshmid/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 1485815557, "node_id": "MDU6TGFiZWwxNDg1ODE1NTU3", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/bug", "name": "bug", "color": "d73a4a", "default": true, "description": "Something isn't working" }, { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null } ]	closed	false	null	[]	null	null	2023-07-28T12:41:37	2023-09-04T01:44:25	2023-09-04T01:44:25	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	Describe the bug I was training a MegatronGPTModel using `megatron_gpt_pretraining.py` and a machine with H100 chips. When I set `transformer_engine=true` I later noticed that my requested `activation=fast-swiglu` was ignored. I traced the issue back to the [transformer.py](./nemo/collections/nlp/modules/common/megatron/transformer.py) file, in the definition of the class `AutocastTransformerLayer`. In the initializer there is no `activation` parameter being accepted or passed into the layer. Steps/Code to reproduce bug python megatron_gpt_pretraining.py model.activation=swiglu Expected behavior The model to use the SwiGLU activation function, which can be seen by the `layernorm_mlp.fc1_weight` parameter (or `mlp.dense_h_to_4h` in the non transformer_engine code) being of dimension `2 * config.ffn_hidden_size` . Environment overview (please complete the following information) - Environment location: [Docker] - Method of NeMo install: [from source]. Please specify exact commands you used to install. ```bash git clone https://github.com/NVIDIA/NeMo.git DOCKER_BUILDKIT=1 docker build -f Dockerfile -t nemo:latest . ```	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7120/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7120/timeline	null	not_planned	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7119	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7119/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7119/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7119/events	https://github.com/NVIDIA/NeMo/pull/7119	1,825,859,048	PR_kwDOC_bI7s5Wn--b	7,119	Pass `rotary_pos_emb` and `relative_position_bias` to TE	{ "login": "yaox12", "id": 3831900, "node_id": "MDQ6VXNlcjM4MzE5MDA=", "avatar_url": "https://avatars.githubusercontent.com/u/3831900?v=4", "gravatar_id": "", "url": "https://api.github.com/users/yaox12", "html_url": "https://github.com/yaox12", "followers_url": "https://api.github.com/users/yaox12/followers", "following_url": "https://api.github.com/users/yaox12/following{/other_user}", "gists_url": "https://api.github.com/users/yaox12/gists{/gist_id}", "starred_url": "https://api.github.com/users/yaox12/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/yaox12/subscriptions", "organizations_url": "https://api.github.com/users/yaox12/orgs", "repos_url": "https://api.github.com/users/yaox12/repos", "events_url": "https://api.github.com/users/yaox12/events{/privacy}", "received_events_url": "https://api.github.com/users/yaox12/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null }, { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-07-28T07:21:11	2023-11-14T02:32:35	2023-10-05T01:45:20	CONTRIBUTOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7119", "html_url": "https://github.com/NVIDIA/NeMo/pull/7119", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7119.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7119.patch", "merged_at": null }	# What does this PR do ? Correctly pass `rotary_pos_emb` and `relative_position_bias` to TE. Collection: [Note which collection this PR will affect] [NLP] # Changelog - nemo/collections/nlp/modules/common/megatron/transformer.py # Usage * You can potentially add a usage example below ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7119/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7119/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7118	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7118/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7118/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7118/events	https://github.com/NVIDIA/NeMo/pull/7118	1,825,614,842	PR_kwDOC_bI7s5WnMUj	7,118	Fixed small bug with NoisePerturbationWithNormalization	{ "login": "trias702", "id": 25867060, "node_id": "MDQ6VXNlcjI1ODY3MDYw", "avatar_url": "https://avatars.githubusercontent.com/u/25867060?v=4", "gravatar_id": "", "url": "https://api.github.com/users/trias702", "html_url": "https://github.com/trias702", "followers_url": "https://api.github.com/users/trias702/followers", "following_url": "https://api.github.com/users/trias702/following{/other_user}", "gists_url": "https://api.github.com/users/trias702/gists{/gist_id}", "starred_url": "https://api.github.com/users/trias702/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/trias702/subscriptions", "organizations_url": "https://api.github.com/users/trias702/orgs", "repos_url": "https://api.github.com/users/trias702/repos", "events_url": "https://api.github.com/users/trias702/events{/privacy}", "received_events_url": "https://api.github.com/users/trias702/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4811884691, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR", "name": "ASR", "color": "F3B33E", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-07-28T03:01:04	2023-07-28T17:08:08	2023-07-28T17:08:08	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7118", "html_url": "https://github.com/NVIDIA/NeMo/pull/7118", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7118.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7118.patch", "merged_at": "2023-07-28T17:08:07" }	# What does this PR do ? Fixes a small indexing bug inside the NoisePerturbationWithNormalization class in perturb.py Collection: ASR # Changelog - nemo/collections/asr/parts/preprocessing/perturb.py: added safety check to ensure you don't index into scalar value # Before your PR is "Ready for review" Pre checks: - [X] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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 - [X] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "trias702", "id": 25867060, "node_id": "MDQ6VXNlcjI1ODY3MDYw", "avatar_url": "https://avatars.githubusercontent.com/u/25867060?v=4", "gravatar_id": "", "url": "https://api.github.com/users/trias702", "html_url": "https://github.com/trias702", "followers_url": "https://api.github.com/users/trias702/followers", "following_url": "https://api.github.com/users/trias702/following{/other_user}", "gists_url": "https://api.github.com/users/trias702/gists{/gist_id}", "starred_url": "https://api.github.com/users/trias702/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/trias702/subscriptions", "organizations_url": "https://api.github.com/users/trias702/orgs", "repos_url": "https://api.github.com/users/trias702/repos", "events_url": "https://api.github.com/users/trias702/events{/privacy}", "received_events_url": "https://api.github.com/users/trias702/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7118/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7118/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7117	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7117/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7117/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7117/events	https://github.com/NVIDIA/NeMo/pull/7117	1,825,249,223	PR_kwDOC_bI7s5Wl8FJ	7,117	Fix links in Segmentation tutorial	{ "login": "ekmb", "id": 10428420, "node_id": "MDQ6VXNlcjEwNDI4NDIw", "avatar_url": "https://avatars.githubusercontent.com/u/10428420?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ekmb", "html_url": "https://github.com/ekmb", "followers_url": "https://api.github.com/users/ekmb/followers", "following_url": "https://api.github.com/users/ekmb/following{/other_user}", "gists_url": "https://api.github.com/users/ekmb/gists{/gist_id}", "starred_url": "https://api.github.com/users/ekmb/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ekmb/subscriptions", "organizations_url": "https://api.github.com/users/ekmb/orgs", "repos_url": "https://api.github.com/users/ekmb/repos", "events_url": "https://api.github.com/users/ekmb/events{/privacy}", "received_events_url": "https://api.github.com/users/ekmb/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[]	closed	false	null	[]	null	null	2023-07-27T21:23:41	2023-07-27T21:32:04	2023-07-27T21:32:03	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7117", "html_url": "https://github.com/NVIDIA/NeMo/pull/7117", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7117.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7117.patch", "merged_at": "2023-07-27T21:32:03" }	# 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 ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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 - [ X] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "ericharper", "id": 11999610, "node_id": "MDQ6VXNlcjExOTk5NjEw", "avatar_url": "https://avatars.githubusercontent.com/u/11999610?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ericharper", "html_url": "https://github.com/ericharper", "followers_url": "https://api.github.com/users/ericharper/followers", "following_url": "https://api.github.com/users/ericharper/following{/other_user}", "gists_url": "https://api.github.com/users/ericharper/gists{/gist_id}", "starred_url": "https://api.github.com/users/ericharper/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ericharper/subscriptions", "organizations_url": "https://api.github.com/users/ericharper/orgs", "repos_url": "https://api.github.com/users/ericharper/repos", "events_url": "https://api.github.com/users/ericharper/events{/privacy}", "received_events_url": "https://api.github.com/users/ericharper/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7117/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7117/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7116	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7116/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7116/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7116/events	https://github.com/NVIDIA/NeMo/pull/7116	1,824,963,983	PR_kwDOC_bI7s5Wk8LX	7,116	Distributed checkpointing with mcore GPT	{ "login": "ericharper", "id": 11999610, "node_id": "MDQ6VXNlcjExOTk5NjEw", "avatar_url": "https://avatars.githubusercontent.com/u/11999610?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ericharper", "html_url": "https://github.com/ericharper", "followers_url": "https://api.github.com/users/ericharper/followers", "following_url": "https://api.github.com/users/ericharper/following{/other_user}", "gists_url": "https://api.github.com/users/ericharper/gists{/gist_id}", "starred_url": "https://api.github.com/users/ericharper/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ericharper/subscriptions", "organizations_url": "https://api.github.com/users/ericharper/orgs", "repos_url": "https://api.github.com/users/ericharper/repos", "events_url": "https://api.github.com/users/ericharper/events{/privacy}", "received_events_url": "https://api.github.com/users/ericharper/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 1923495043, "node_id": "MDU6TGFiZWwxOTIzNDk1MDQz", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/core", "name": "core", "color": "f25e3a", "default": false, "description": "Changes to NeMo Core" }, { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" }, { "id": 4840216855, "node_id": "LA_kwDOC_bI7s8AAAABIH_ZFw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/CI", "name": "CI", "color": "1D76DB", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-07-27T18:39:03	2023-08-28T21:52:55	2023-08-28T21:52:54	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7116", "html_url": "https://github.com/NVIDIA/NeMo/pull/7116", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7116.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7116.patch", "merged_at": "2023-08-28T21:52:54" }	This PR needs mcore dist ckpt for GPT PR to be pushed before merging. # What does this PR do ? Adds distributed checkpointing when using mcore gpt. Distributed checkpointing enables training runs to restart automatically with different model parallel configs. The checkpoint is saved to disk according to the `sharded_state_dict`: Below is a sample of what the checkpoint looks like on disk. ``` bash common.pt model.decoder.layers.self_attention.linear_qkv.weight optimizer.state.exp_avg.model.embedding.word_embeddings.weight optimizer.state.fp32_from_fp16.model.decoder.final_layernorm.bias metadata.json model.embedding.position_embeddings.weight optimizer.state.exp_avg.model.output_layer.weight optimizer.state.fp32_from_fp16.model.decoder.final_layernorm.weight model.decoder.final_layernorm.bias model.embedding.word_embeddings.weight optimizer.state.exp_avg_sq.model.decoder.final_layernorm.bias optimizer.state.fp32_from_fp16.model.decoder.layers.input_layernorm.bias model.decoder.final_layernorm.weight model.output_layer.weight optimizer.state.exp_avg_sq.model.decoder.final_layernorm.weight optimizer.state.fp32_from_fp16.model.decoder.layers.input_layernorm.weight model.decoder.layers.input_layernorm.bias optimizer.state.exp_avg.model.decoder.final_layernorm.bias optimizer.state.exp_avg_sq.model.decoder.layers.input_layernorm.bias optimizer.state.fp32_from_fp16.model.decoder.layers.mlp.linear_fc1.bias model.decoder.layers.input_layernorm.weight optimizer.state.exp_avg.model.decoder.final_layernorm.weight optimizer.state.exp_avg_sq.model.decoder.layers.input_layernorm.weight optimizer.state.fp32_from_fp16.model.decoder.layers.mlp.linear_fc1.weight model.decoder.layers.mlp.linear_fc1.bias optimizer.state.exp_avg.model.decoder.layers.input_layernorm.bias optimizer.state.exp_avg_sq.model.decoder.layers.mlp.linear_fc1.bias optimizer.state.fp32_from_fp16.model.decoder.layers.mlp.linear_fc2.bias model.decoder.layers.mlp.linear_fc1._extra_state optimizer.state.exp_avg.model.decoder.layers.input_layernorm.weight optimizer.state.exp_avg_sq.model.decoder.layers.mlp.linear_fc1.weight optimizer.state.fp32_from_fp16.model.decoder.layers.mlp.linear_fc2.weight model.decoder.layers.mlp.linear_fc1.weight optimizer.state.exp_avg.model.decoder.layers.mlp.linear_fc1.bias optimizer.state.exp_avg_sq.model.decoder.layers.mlp.linear_fc2.bias optimizer.state.fp32_from_fp16.model.decoder.layers.post_self_attn_layernorm.bias model.decoder.layers.mlp.linear_fc2.bias optimizer.state.exp_avg.model.decoder.layers.mlp.linear_fc1.weight optimizer.state.exp_avg_sq.model.decoder.layers.mlp.linear_fc2.weight optimizer.state.fp32_from_fp16.model.decoder.layers.post_self_attn_layernorm.weight model.decoder.layers.mlp.linear_fc2._extra_state optimizer.state.exp_avg.model.decoder.layers.mlp.linear_fc2.bias optimizer.state.exp_avg_sq.model.decoder.layers.post_self_attn_layernorm.bias optimizer.state.fp32_from_fp16.model.decoder.layers.self_attention.linear_proj.bias model.decoder.layers.mlp.linear_fc2.weight optimizer.state.exp_avg.model.decoder.layers.mlp.linear_fc2.weight optimizer.state.exp_avg_sq.model.decoder.layers.post_self_attn_layernorm.weight optimizer.state.fp32_from_fp16.model.decoder.layers.self_attention.linear_proj.weight model.decoder.layers.post_self_attn_layernorm.bias optimizer.state.exp_avg.model.decoder.layers.post_self_attn_layernorm.bias optimizer.state.exp_avg_sq.model.decoder.layers.self_attention.linear_proj.bias optimizer.state.fp32_from_fp16.model.decoder.layers.self_attention.linear_qkv.bias model.decoder.layers.post_self_attn_layernorm.weight optimizer.state.exp_avg.model.decoder.layers.post_self_attn_layernorm.weight optimizer.state.exp_avg_sq.model.decoder.layers.self_attention.linear_proj.weight optimizer.state.fp32_from_fp16.model.decoder.layers.self_attention.linear_qkv.weight model.decoder.layers.self_attention.linear_proj.bias optimizer.state.exp_avg.model.decoder.layers.self_attention.linear_proj.bias optimizer.state.exp_avg_sq.model.decoder.layers.self_attention.linear_qkv.bias optimizer.state.fp32_from_fp16.model.embedding.position_embeddings.weight model.decoder.layers.self_attention.linear_proj._extra_state optimizer.state.exp_avg.model.decoder.layers.self_attention.linear_proj.weight optimizer.state.exp_avg_sq.model.decoder.layers.self_attention.linear_qkv.weight optimizer.state.fp32_from_fp16.model.embedding.word_embeddings.weight model.decoder.layers.self_attention.linear_proj.weight optimizer.state.exp_avg.model.decoder.layers.self_attention.linear_qkv.bias optimizer.state.exp_avg_sq.model.embedding.position_embeddings.weight optimizer.state.fp32_from_fp16.model.output_layer.weight model.decoder.layers.self_attention.linear_qkv.bias optimizer.state.exp_avg.model.decoder.layers.self_attention.linear_qkv.weight optimizer.state.exp_avg_sq.model.embedding.word_embeddings.weight model.decoder.layers.self_attention.linear_qkv._extra_state optimizer.state.exp_avg.model.embedding.position_embeddings.weight optimizer.state.exp_avg_sq.model.output_layer.weight ``` Then inside a module directory we have the sharded tensor: ``` bash ls model.decoder.layers.mlp.linear_fc1.weight/ 0.0.0 1.0.0 10.0.0 11.0.0 12.0.0 13.0.0 14.0.0 15.0.0 2.0.0 3.0.0 4.0.0 5.0.0 6.0.0 7.0.0 8.0.0 9.0.0 ``` To implement distributed checkpointing for a model, the `sharded_state_dict` has to be defined. This is done in megatron core so that in NeMo, if the module is from mcore, we only have to call `module.sharded_state_dict()`. Collection: NLP # Usage Usage is automatic when using mcore: ``` bash model.mcore_gpt=True ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "ericharper", "id": 11999610, "node_id": "MDQ6VXNlcjExOTk5NjEw", "avatar_url": "https://avatars.githubusercontent.com/u/11999610?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ericharper", "html_url": "https://github.com/ericharper", "followers_url": "https://api.github.com/users/ericharper/followers", "following_url": "https://api.github.com/users/ericharper/following{/other_user}", "gists_url": "https://api.github.com/users/ericharper/gists{/gist_id}", "starred_url": "https://api.github.com/users/ericharper/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ericharper/subscriptions", "organizations_url": "https://api.github.com/users/ericharper/orgs", "repos_url": "https://api.github.com/users/ericharper/repos", "events_url": "https://api.github.com/users/ericharper/events{/privacy}", "received_events_url": "https://api.github.com/users/ericharper/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7116/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7116/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7115	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7115/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7115/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7115/events	https://github.com/NVIDIA/NeMo/pull/7115	1,823,714,193	PR_kwDOC_bI7s5WgtxW	7,115	Set a default value when getting `rampup_batch_size`	{ "login": "yaox12", "id": 3831900, "node_id": "MDQ6VXNlcjM4MzE5MDA=", "avatar_url": "https://avatars.githubusercontent.com/u/3831900?v=4", "gravatar_id": "", "url": "https://api.github.com/users/yaox12", "html_url": "https://github.com/yaox12", "followers_url": "https://api.github.com/users/yaox12/followers", "following_url": "https://api.github.com/users/yaox12/following{/other_user}", "gists_url": "https://api.github.com/users/yaox12/gists{/gist_id}", "starred_url": "https://api.github.com/users/yaox12/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/yaox12/subscriptions", "organizations_url": "https://api.github.com/users/yaox12/orgs", "repos_url": "https://api.github.com/users/yaox12/repos", "events_url": "https://api.github.com/users/yaox12/events{/privacy}", "received_events_url": "https://api.github.com/users/yaox12/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null }, { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-07-27T06:36:22	2023-08-29T01:50:33	2023-08-28T19:53:31	CONTRIBUTOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7115", "html_url": "https://github.com/NVIDIA/NeMo/pull/7115", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7115.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7115.patch", "merged_at": "2023-08-28T19:53:31" }	# What does this PR do ? Set a default value when getting `rampup_batch_size`. Otherwise, configs without setting the value would result in errors. Collection: [Note which collection this PR will affect] [NLP] # Changelog - Add specific line by line info of high level changes in this PR. # Usage * You can potentially add a usage example below ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "MaximumEntropy", "id": 9114321, "node_id": "MDQ6VXNlcjkxMTQzMjE=", "avatar_url": "https://avatars.githubusercontent.com/u/9114321?v=4", "gravatar_id": "", "url": "https://api.github.com/users/MaximumEntropy", "html_url": "https://github.com/MaximumEntropy", "followers_url": "https://api.github.com/users/MaximumEntropy/followers", "following_url": "https://api.github.com/users/MaximumEntropy/following{/other_user}", "gists_url": "https://api.github.com/users/MaximumEntropy/gists{/gist_id}", "starred_url": "https://api.github.com/users/MaximumEntropy/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/MaximumEntropy/subscriptions", "organizations_url": "https://api.github.com/users/MaximumEntropy/orgs", "repos_url": "https://api.github.com/users/MaximumEntropy/repos", "events_url": "https://api.github.com/users/MaximumEntropy/events{/privacy}", "received_events_url": "https://api.github.com/users/MaximumEntropy/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7115/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7115/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7114	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7114/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7114/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7114/events	https://github.com/NVIDIA/NeMo/issues/7114	1,823,560,959	I_kwDOC_bI7s5ssVT_	7,114	online diarizer	{ "login": "jeremy110", "id": 52516810, "node_id": "MDQ6VXNlcjUyNTE2ODEw", "avatar_url": "https://avatars.githubusercontent.com/u/52516810?v=4", "gravatar_id": "", "url": "https://api.github.com/users/jeremy110", "html_url": "https://github.com/jeremy110", "followers_url": "https://api.github.com/users/jeremy110/followers", "following_url": "https://api.github.com/users/jeremy110/following{/other_user}", "gists_url": "https://api.github.com/users/jeremy110/gists{/gist_id}", "starred_url": "https://api.github.com/users/jeremy110/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/jeremy110/subscriptions", "organizations_url": "https://api.github.com/users/jeremy110/orgs", "repos_url": "https://api.github.com/users/jeremy110/repos", "events_url": "https://api.github.com/users/jeremy110/events{/privacy}", "received_events_url": "https://api.github.com/users/jeremy110/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null } ]	closed	false	null	[]	null	null	2023-07-27T03:51:14	2023-09-03T01:44:45	2023-09-03T01:44:45	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	Hi! I was trying an online diarizer with reference to this PR(https://github.com/NVIDIA/NeMo/pull/5609). I am using the Nemo Frame VAD and modified it to directly output vad_timestamps (e.g., tensor([[0.0000, 1.5200, 1.5300]])). Then, I call diarize_step function every two seconds, as I'm not quite sure what one "step" (1 frame) represents. However, I only get one speaker output each time, despite calling diarize_step multiple times. Could you please help me identify what might be going wrong? I will provide some relevant code snippets and logs below. Thank you in advance. ```python MODEL_CONFIG = "/workspace/diar_infer.yaml" cfg = OmegaConf.load(MODEL_CONFIG) online_diar = OnlineClusteringDiarizer(cfg) filepath = "/workspace/test_60s.wav" waveform, _ = torchaudio.load(filepath) vad_model = fvad("/workspace/model/vad_multilingual_frame_marblenet.nemo", "/workspace/diar_infer.yaml") print(f"waveform.size() {waveform.size())}") online_diar.buffer_end = 60 scale = 2 for step in range(20): start = step * scale online_diar.frame_start = start s = start * 16000 e = s + scale * 16000 #(step + 1) * scale * 16000 cur_wav = waveform[:,: e] vad_timestamps = vad_model.infer(cur_wav[:, s: ]) + start print(f"vad_timestamps: {vad_timestamps}") diar_hyp = online_diar.diarize_step(cur_wav.squeeze(0), vad_timestamps) print(f"step: {step}, diar_hyp: {diar_hyp}") ``` [log_online1.txt](https://github.com/NVIDIA/NeMo/files/12179113/log_online1.txt)	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7114/reactions", "total_count": 1, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 1 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7114/timeline	null	not_planned	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7113	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7113/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7113/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7113/events	https://github.com/NVIDIA/NeMo/pull/7113	1,823,468,254	PR_kwDOC_bI7s5Wf5aY	7,113	Fix plot function in vad_utils.py	{ "login": "stevehuang52", "id": 105218074, "node_id": "U_kgDOBkWAGg", "avatar_url": "https://avatars.githubusercontent.com/u/105218074?v=4", "gravatar_id": "", "url": "https://api.github.com/users/stevehuang52", "html_url": "https://github.com/stevehuang52", "followers_url": "https://api.github.com/users/stevehuang52/followers", "following_url": "https://api.github.com/users/stevehuang52/following{/other_user}", "gists_url": "https://api.github.com/users/stevehuang52/gists{/gist_id}", "starred_url": "https://api.github.com/users/stevehuang52/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/stevehuang52/subscriptions", "organizations_url": "https://api.github.com/users/stevehuang52/orgs", "repos_url": "https://api.github.com/users/stevehuang52/repos", "events_url": "https://api.github.com/users/stevehuang52/events{/privacy}", "received_events_url": "https://api.github.com/users/stevehuang52/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4811884691, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR", "name": "ASR", "color": "F3B33E", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-07-27T01:47:45	2023-07-27T12:22:39	2023-07-27T12:22:38	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7113", "html_url": "https://github.com/NVIDIA/NeMo/pull/7113", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7113.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7113.patch", "merged_at": "2023-07-27T12:22:38" }	Fix plot function in vad_utils.py	{ "login": "stevehuang52", "id": 105218074, "node_id": "U_kgDOBkWAGg", "avatar_url": "https://avatars.githubusercontent.com/u/105218074?v=4", "gravatar_id": "", "url": "https://api.github.com/users/stevehuang52", "html_url": "https://github.com/stevehuang52", "followers_url": "https://api.github.com/users/stevehuang52/followers", "following_url": "https://api.github.com/users/stevehuang52/following{/other_user}", "gists_url": "https://api.github.com/users/stevehuang52/gists{/gist_id}", "starred_url": "https://api.github.com/users/stevehuang52/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/stevehuang52/subscriptions", "organizations_url": "https://api.github.com/users/stevehuang52/orgs", "repos_url": "https://api.github.com/users/stevehuang52/repos", "events_url": "https://api.github.com/users/stevehuang52/events{/privacy}", "received_events_url": "https://api.github.com/users/stevehuang52/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7113/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7113/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7112	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7112/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7112/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7112/events	https://github.com/NVIDIA/NeMo/pull/7112	1,823,221,886	PR_kwDOC_bI7s5WfEjI	7,112	Include the scripts for preprocessing OAST and unit tests for chat sft datasets	{ "login": "yidong72", "id": 43824965, "node_id": "MDQ6VXNlcjQzODI0OTY1", "avatar_url": "https://avatars.githubusercontent.com/u/43824965?v=4", "gravatar_id": "", "url": "https://api.github.com/users/yidong72", "html_url": "https://github.com/yidong72", "followers_url": "https://api.github.com/users/yidong72/followers", "following_url": "https://api.github.com/users/yidong72/following{/other_user}", "gists_url": "https://api.github.com/users/yidong72/gists{/gist_id}", "starred_url": "https://api.github.com/users/yidong72/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/yidong72/subscriptions", "organizations_url": "https://api.github.com/users/yidong72/orgs", "repos_url": "https://api.github.com/users/yidong72/repos", "events_url": "https://api.github.com/users/yidong72/events{/privacy}", "received_events_url": "https://api.github.com/users/yidong72/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-07-26T21:36:45	2023-08-07T16:34:15	2023-08-07T16:29:22	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7112", "html_url": "https://github.com/NVIDIA/NeMo/pull/7112", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7112.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7112.patch", "merged_at": "2023-08-07T16:29:22" }	In this PR. 1. include scripts for preprocess OAST datasets. 2. annotate the datasets using reward model. 3. clean up the data for training. 4. include the unit tests to make sure the Chat SFT datasets produced expected masks. 5. add support in the UI to run attribute condition generations. 6. add HF dataset as JSON dataset backend.	{ "login": "yidong72", "id": 43824965, "node_id": "MDQ6VXNlcjQzODI0OTY1", "avatar_url": "https://avatars.githubusercontent.com/u/43824965?v=4", "gravatar_id": "", "url": "https://api.github.com/users/yidong72", "html_url": "https://github.com/yidong72", "followers_url": "https://api.github.com/users/yidong72/followers", "following_url": "https://api.github.com/users/yidong72/following{/other_user}", "gists_url": "https://api.github.com/users/yidong72/gists{/gist_id}", "starred_url": "https://api.github.com/users/yidong72/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/yidong72/subscriptions", "organizations_url": "https://api.github.com/users/yidong72/orgs", "repos_url": "https://api.github.com/users/yidong72/repos", "events_url": "https://api.github.com/users/yidong72/events{/privacy}", "received_events_url": "https://api.github.com/users/yidong72/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7112/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7112/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7111	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7111/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7111/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7111/events	https://github.com/NVIDIA/NeMo/issues/7111	1,822,986,621	I_kwDOC_bI7s5sqJF9	7,111	`nemo.collections.asr.models.EncDecSpeakerLabelModel.verify_speakers()` excessive embeddings normalization	{ "login": "alllexx88", "id": 6587516, "node_id": "MDQ6VXNlcjY1ODc1MTY=", "avatar_url": "https://avatars.githubusercontent.com/u/6587516?v=4", "gravatar_id": "", "url": "https://api.github.com/users/alllexx88", "html_url": "https://github.com/alllexx88", "followers_url": "https://api.github.com/users/alllexx88/followers", "following_url": "https://api.github.com/users/alllexx88/following{/other_user}", "gists_url": "https://api.github.com/users/alllexx88/gists{/gist_id}", "starred_url": "https://api.github.com/users/alllexx88/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/alllexx88/subscriptions", "organizations_url": "https://api.github.com/users/alllexx88/orgs", "repos_url": "https://api.github.com/users/alllexx88/repos", "events_url": "https://api.github.com/users/alllexx88/events{/privacy}", "received_events_url": "https://api.github.com/users/alllexx88/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 1485815557, "node_id": "MDU6TGFiZWwxNDg1ODE1NTU3", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/bug", "name": "bug", "color": "d73a4a", "default": true, "description": "Something isn't working" }, { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null } ]	closed	false	null	[]	null	null	2023-07-26T19:00:42	2023-09-03T01:44:47	2023-09-03T01:44:47	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	Describe the bug `nemo.collections.asr.models.EncDecSpeakerLabelModel.verify_speakers` has excessive embeddings normalization. First computed ambeddings are normalized here: ``` # Length Normalize X = embs1 / torch.linalg.norm(embs1) Y = embs2 / torch.linalg.norm(embs2) ``` (https://github.com/NVIDIA/NeMo/blob/main/nemo/collections/asr/models/label_models.py#L510) Then during similarity computing the dot product is divided by `X` and `Y` norms (which are mathematically identical to `1`; computationally very close to `1`): ``` similarity_score = torch.dot(X, Y) / ((torch.dot(X, X) * torch.dot(Y, Y)) 0.5) ``` (https://github.com/NVIDIA/NeMo/blob/main/nemo/collections/asr/models/label_models.py#L513) Steps/Code to reproduce bug Just use `verify_speakers()`. The result is accurate, however doing something excessive doesn't look good. Expected behavior Don't normalize embeddings twice. Either normalize them first, or divide the dot product of (not normalized) embeddings by the norms. No sense to do both. Environment overview (please complete the following information) - Environment location: [Bare-metal] - Method of NeMo install: `pip install nemo-toolkit[asr]`. Environment details** If NVIDIA docker image is used you don't need to specify these. Otherwise, please provide: - OS: Ubuntu 22.04 - PyTorch version: 2.0.1 - Python version: 3.10.12	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7111/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7111/timeline	null	not_planned	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7110	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7110/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7110/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7110/events	https://github.com/NVIDIA/NeMo/pull/7110	1,822,960,551	PR_kwDOC_bI7s5WeLHe	7,110	NeMo ASR Demo	{ "login": "lleaver", "id": 137942999, "node_id": "U_kgDOCDjX1w", "avatar_url": "https://avatars.githubusercontent.com/u/137942999?v=4", "gravatar_id": "", "url": "https://api.github.com/users/lleaver", "html_url": "https://github.com/lleaver", "followers_url": "https://api.github.com/users/lleaver/followers", "following_url": "https://api.github.com/users/lleaver/following{/other_user}", "gists_url": "https://api.github.com/users/lleaver/gists{/gist_id}", "starred_url": "https://api.github.com/users/lleaver/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/lleaver/subscriptions", "organizations_url": "https://api.github.com/users/lleaver/orgs", "repos_url": "https://api.github.com/users/lleaver/repos", "events_url": "https://api.github.com/users/lleaver/events{/privacy}", "received_events_url": "https://api.github.com/users/lleaver/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4811884691, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR", "name": "ASR", "color": "F3B33E", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-07-26T18:39:45	2023-07-26T19:55:17	2023-07-26T19:55:17	CONTRIBUTOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7110", "html_url": "https://github.com/NVIDIA/NeMo/pull/7110", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7110.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7110.patch", "merged_at": "2023-07-26T19:55:17" }	# What does this PR do ? Hopefully fix the microphone access error for the transcription demo on the ASR docs main page. Collection: docs # Changelog - Add specific line by line info of high level changes in this PR. # Usage * You can potentially add a usage example below ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "titu1994", "id": 3048602, "node_id": "MDQ6VXNlcjMwNDg2MDI=", "avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4", "gravatar_id": "", "url": "https://api.github.com/users/titu1994", "html_url": "https://github.com/titu1994", "followers_url": "https://api.github.com/users/titu1994/followers", "following_url": "https://api.github.com/users/titu1994/following{/other_user}", "gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}", "starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/titu1994/subscriptions", "organizations_url": "https://api.github.com/users/titu1994/orgs", "repos_url": "https://api.github.com/users/titu1994/repos", "events_url": "https://api.github.com/users/titu1994/events{/privacy}", "received_events_url": "https://api.github.com/users/titu1994/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7110/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7110/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7109	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7109/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7109/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7109/events	https://github.com/NVIDIA/NeMo/issues/7109	1,822,311,801	I_kwDOC_bI7s5snkV5	7,109	Nemo toolkit installation failed in google colab	{ "login": "dawnuuuu", "id": 58391499, "node_id": "MDQ6VXNlcjU4MzkxNDk5", "avatar_url": "https://avatars.githubusercontent.com/u/58391499?v=4", "gravatar_id": "", "url": "https://api.github.com/users/dawnuuuu", "html_url": "https://github.com/dawnuuuu", "followers_url": "https://api.github.com/users/dawnuuuu/followers", "following_url": "https://api.github.com/users/dawnuuuu/following{/other_user}", "gists_url": "https://api.github.com/users/dawnuuuu/gists{/gist_id}", "starred_url": "https://api.github.com/users/dawnuuuu/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/dawnuuuu/subscriptions", "organizations_url": "https://api.github.com/users/dawnuuuu/orgs", "repos_url": "https://api.github.com/users/dawnuuuu/repos", "events_url": "https://api.github.com/users/dawnuuuu/events{/privacy}", "received_events_url": "https://api.github.com/users/dawnuuuu/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 1485815557, "node_id": "MDU6TGFiZWwxNDg1ODE1NTU3", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/bug", "name": "bug", "color": "d73a4a", "default": true, "description": "Something isn't working" }, { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null } ]	closed	false	null	[]	null	null	2023-07-26T12:28:17	2023-09-09T01:43:39	2023-09-09T01:43:39	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	I kept on trying to install nemo toolkit but I keep on getting this error: ```jupyter-notebook Collecting wget Downloading wget-3.2.zip (10 kB) Preparing metadata (setup.py) ... done Building wheels for collected packages: wget Building wheel for wget (setup.py) ... done Created wheel for wget: filename=wget-3.2-py3-none-any.whl size=9657 sha256=9d056575abefb14615c70a2e0e473b362f9821047bd7a3da45bbb2d36aab773f Stored in directory: /root/.cache/pip/wheels/8b/f1/7f/5c94f0a7a505ca1c81cd1d9208ae2064675d97582078e6c769 Successfully built wget Installing collected packages: wget Successfully installed wget-3.2 Reading package lists... Done Building dependency tree... Done Reading state information... Done libsndfile1 is already the newest version (1.0.31-2build1). ffmpeg is already the newest version (7:4.4.2-0ubuntu0.22.04.1). The following additional packages will be installed: libid3tag0 libmad0 libopencore-amrnb0 libopencore-amrwb0 libsox-fmt-alsa libsox-fmt-base libsox3 libwavpack1 Suggested packages: libsox-fmt-all The following NEW packages will be installed: libid3tag0 libmad0 libopencore-amrnb0 libopencore-amrwb0 libsox-fmt-alsa libsox-fmt-base libsox-fmt-mp3 libsox3 libwavpack1 sox 0 upgraded, 10 newly installed, 0 to remove and 8 not upgraded. Need to get 729 kB of archives. After this operation, 2,146 kB of additional disk space will be used. Get:1 http://archive.ubuntu.com/ubuntu jammy/universe amd64 libid3tag0 amd64 0.15.1b-14 [31.3 kB] Get:2 http://archive.ubuntu.com/ubuntu jammy/universe amd64 libmad0 amd64 0.15.1b-10ubuntu1 [63.1 kB] Get:3 http://archive.ubuntu.com/ubuntu jammy/universe amd64 libopencore-amrnb0 amd64 0.1.5-1 [94.8 kB] Get:4 http://archive.ubuntu.com/ubuntu jammy/universe amd64 libopencore-amrwb0 amd64 0.1.5-1 [49.1 kB] Get:5 http://archive.ubuntu.com/ubuntu jammy-updates/universe amd64 libsox3 amd64 14.4.2+git20190427-2+deb11u2build0.22.04.1 [240 kB] Get:6 http://archive.ubuntu.com/ubuntu jammy-updates/universe amd64 libsox-fmt-alsa amd64 14.4.2+git20190427-2+deb11u2build0.22.04.1 [11.2 kB] Get:7 http://archive.ubuntu.com/ubuntu jammy/main amd64 libwavpack1 amd64 5.4.0-1build2 [83.7 kB] Get:8 http://archive.ubuntu.com/ubuntu jammy-updates/universe amd64 libsox-fmt-base amd64 14.4.2+git20190427-2+deb11u2build0.22.04.1 [33.7 kB] Get:9 http://archive.ubuntu.com/ubuntu jammy-updates/universe amd64 libsox-fmt-mp3 amd64 14.4.2+git20190427-2+deb11u2build0.22.04.1 [17.4 kB] Get:10 http://archive.ubuntu.com/ubuntu jammy-updates/universe amd64 sox amd64 14.4.2+git20190427-2+deb11u2build0.22.04.1 [104 kB] Fetched 729 kB in 2s (368 kB/s) Selecting previously unselected package libid3tag0:amd64. (Reading database ... 129824 files and directories currently installed.) Preparing to unpack .../0-libid3tag0_0.15.1b-14_amd64.deb ... Unpacking libid3tag0:amd64 (0.15.1b-14) ... Selecting previously unselected package libmad0:amd64. Preparing to unpack .../1-libmad0_0.15.1b-10ubuntu1_amd64.deb ... Unpacking libmad0:amd64 (0.15.1b-10ubuntu1) ... Selecting previously unselected package libopencore-amrnb0:amd64. Preparing to unpack .../2-libopencore-amrnb0_0.1.5-1_amd64.deb ... Unpacking libopencore-amrnb0:amd64 (0.1.5-1) ... Selecting previously unselected package libopencore-amrwb0:amd64. Preparing to unpack .../3-libopencore-amrwb0_0.1.5-1_amd64.deb ... Unpacking libopencore-amrwb0:amd64 (0.1.5-1) ... Selecting previously unselected package libsox3:amd64. Preparing to unpack .../4-libsox3_14.4.2+git20190427-2+deb11u2build0.22.04.1_amd64.deb ... Unpacking libsox3:amd64 (14.4.2+git20190427-2+deb11u2build0.22.04.1) ... Selecting previously unselected package libsox-fmt-alsa:amd64. Preparing to unpack .../5-libsox-fmt-alsa_14.4.2+git20190427-2+deb11u2build0.22.04.1_amd64.deb ... Unpacking libsox-fmt-alsa:amd64 (14.4.2+git20190427-2+deb11u2build0.22.04.1) ... Selecting previously unselected package libwavpack1:amd64. Preparing to unpack .../6-libwavpack1_5.4.0-1build2_amd64.deb ... Unpacking libwavpack1:amd64 (5.4.0-1build2) ... Selecting previously unselected package libsox-fmt-base:amd64. Preparing to unpack .../7-libsox-fmt-base_14.4.2+git20190427-2+deb11u2build0.22.04.1_amd64.deb ... Unpacking libsox-fmt-base:amd64 (14.4.2+git20190427-2+deb11u2build0.22.04.1) ... Selecting previously unselected package libsox-fmt-mp3:amd64. Preparing to unpack .../8-libsox-fmt-mp3_14.4.2+git20190427-2+deb11u2build0.22.04.1_amd64.deb ... Unpacking libsox-fmt-mp3:amd64 (14.4.2+git20190427-2+deb11u2build0.22.04.1) ... Selecting previously unselected package sox. Preparing to unpack .../9-sox_14.4.2+git20190427-2+deb11u2build0.22.04.1_amd64.deb ... Unpacking sox (14.4.2+git20190427-2+deb11u2build0.22.04.1) ... Setting up libsox3:amd64 (14.4.2+git20190427-2+deb11u2build0.22.04.1) ... Setting up libid3tag0:amd64 (0.15.1b-14) ... Setting up libopencore-amrwb0:amd64 (0.1.5-1) ... Setting up libsox-fmt-alsa:amd64 (14.4.2+git20190427-2+deb11u2build0.22.04.1) ... Setting up libmad0:amd64 (0.15.1b-10ubuntu1) ... Setting up libwavpack1:amd64 (5.4.0-1build2) ... Setting up libopencore-amrnb0:amd64 (0.1.5-1) ... Setting up libsox-fmt-base:amd64 (14.4.2+git20190427-2+deb11u2build0.22.04.1) ... Setting up libsox-fmt-mp3:amd64 (14.4.2+git20190427-2+deb11u2build0.22.04.1) ... Setting up sox (14.4.2+git20190427-2+deb11u2build0.22.04.1) ... Processing triggers for man-db (2.10.2-1) ... Processing triggers for libc-bin (2.35-0ubuntu3.1) ... Requirement already satisfied: text-unidecode in /usr/local/lib/python3.10/dist-packages (1.3) DEPRECATION: git+https://github.com/NVIDIA/[email protected]#egg=nemo_toolkit[all] contains an egg fragment with a non-PEP 508 name pip 25.0 will enforce this behaviour change. A possible replacement is to use the req @ url syntax, and remove the egg fragment. Discussion can be found at https://github.com/pypa/pip/issues/11617 Collecting nemo_toolkit[all] Cloning https://github.com/NVIDIA/NeMo.git (to revision r1.19.1) to /tmp/pip-install-z_sduiuz/nemo-toolkit_20856122c60348d8bf748ed8dbe6478b Running command git clone --filter=blob:none --quiet https://github.com/NVIDIA/NeMo.git /tmp/pip-install-z_sduiuz/nemo-toolkit_20856122c60348d8bf748ed8dbe6478b Running command git checkout -b r1.19.1 --track origin/r1.19.1 Switched to a new branch 'r1.19.1' Branch 'r1.19.1' set up to track remote branch 'r1.19.1' from 'origin'. Resolved https://github.com/NVIDIA/NeMo.git to commit 55f8ce5a32ee2a2ec2c50c6df3eb8af585ea2fae Installing build dependencies ... done Getting requirements to build wheel ... done Preparing metadata (pyproject.toml) ... done Collecting huggingface-hub (from nemo_toolkit[all]) Downloading huggingface_hub-0.16.4-py3-none-any.whl (268 kB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 268.8/268.8 kB 4.7 MB/s eta 0:00:00 Requirement already satisfied: numba in /usr/local/lib/python3.10/dist-packages (from nemo_toolkit[all]) (0.56.4) Requirement already satisfied: numpy>=1.22 in /usr/local/lib/python3.10/dist-packages (from nemo_toolkit[all]) (1.22.4) Collecting onnx>=1.7.0 (from nemo_toolkit[all]) Downloading onnx-1.14.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (14.6 MB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 14.6/14.6 MB 85.3 MB/s eta 0:00:00 Requirement already satisfied: python-dateutil in /usr/local/lib/python3.10/dist-packages (from nemo_toolkit[all]) (2.8.2) Collecting ruamel.yaml (from nemo_toolkit[all]) Downloading ruamel.yaml-0.17.32-py3-none-any.whl (112 kB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 112.2/112.2 kB 18.6 MB/s eta 0:00:00 Requirement already satisfied: scikit-learn in /usr/local/lib/python3.10/dist-packages (from nemo_toolkit[all]) (1.2.2) Collecting setuptools==65.5.1 (from nemo_toolkit[all]) Downloading setuptools-65.5.1-py3-none-any.whl (1.2 MB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.2/1.2 MB 24.9 MB/s eta 0:00:00 Requirement already satisfied: tensorboard in /usr/local/lib/python3.10/dist-packages (from nemo_toolkit[all]) (2.12.3) Requirement already satisfied: text-unidecode in /usr/local/lib/python3.10/dist-packages (from nemo_toolkit[all]) (1.3) Requirement already satisfied: torch in /usr/local/lib/python3.10/dist-packages (from nemo_toolkit[all]) (2.0.1+cu118) Requirement already satisfied: tqdm>=4.41.0 in /usr/local/lib/python3.10/dist-packages (from nemo_toolkit[all]) (4.65.0) Requirement already satisfied: wget in /usr/local/lib/python3.10/dist-packages (from nemo_toolkit[all]) (3.2) Requirement already satisfied: wrapt in /usr/local/lib/python3.10/dist-packages (from nemo_toolkit[all]) (1.14.1) Collecting black==19.10b0 (from nemo_toolkit[all]) Downloading black-19.10b0-py36-none-any.whl (97 kB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 97.5/97.5 kB 14.7 MB/s eta 0:00:00 Collecting click==8.0.2 (from nemo_toolkit[all]) Downloading click-8.0.2-py3-none-any.whl (97 kB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 97.6/97.6 kB 16.4 MB/s eta 0:00:00 Collecting isort<6.0.0,>5.1.0 (from nemo_toolkit[all]) Downloading isort-5.12.0-py3-none-any.whl (91 kB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 91.2/91.2 kB 9.8 MB/s eta 0:00:00 Collecting parameterized (from nemo_toolkit[all]) Downloading parameterized-0.9.0-py2.py3-none-any.whl (20 kB) Requirement already satisfied: pytest in /usr/local/lib/python3.10/dist-packages (from nemo_toolkit[all]) (7.2.2) Collecting pytest-runner (from nemo_toolkit[all]) Downloading pytest_runner-6.0.0-py3-none-any.whl (7.2 kB) Requirement already satisfied: sphinx in /usr/local/lib/python3.10/dist-packages (from nemo_toolkit[all]) (3.5.4) Collecting sphinxcontrib-bibtex (from nemo_toolkit[all]) Downloading sphinxcontrib_bibtex-2.5.0-py3-none-any.whl (39 kB) Collecting wandb (from nemo_toolkit[all]) Downloading wandb-0.15.7-py3-none-any.whl (2.1 MB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 2.1/2.1 MB 76.5 MB/s eta 0:00:00 Collecting hydra-core<1.3,>=1.2.0 (from nemo_toolkit[all]) Downloading hydra_core-1.2.0-py3-none-any.whl (151 kB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 151.1/151.1 kB 23.8 MB/s eta 0:00:00 Collecting omegaconf<2.3,>=2.2 (from nemo_toolkit[all]) Downloading omegaconf-2.2.3-py3-none-any.whl (79 kB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 79.3/79.3 kB 7.9 MB/s eta 0:00:00 Collecting pytorch-lightning<=1.9.4,>=1.9.0 (from nemo_toolkit[all]) Downloading pytorch_lightning-1.9.4-py3-none-any.whl (827 kB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 827.8/827.8 kB 80.1 MB/s eta 0:00:00 Collecting pyyaml<6 (from nemo_toolkit[all]) Downloading PyYAML-5.4.1.tar.gz (175 kB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 175.1/175.1 kB 25.8 MB/s eta 0:00:00 Installing build dependencies ... done error: subprocess-exited-with-error × Getting requirements to build wheel did not run successfully. │ exit code: 1 ╰─> See above for output. note: This error originates from a subprocess, and is likely not a problem with pip. Getting requirements to build wheel ... error error: subprocess-exited-with-error × Getting requirements to build wheel did not run successfully. │ exit code: 1 ╰─> See above for output. ``` Environment details - Python version 3.10.6	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7109/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7109/timeline	null	not_planned	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7108	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7108/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7108/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7108/events	https://github.com/NVIDIA/NeMo/issues/7108	1,821,614,836	I_kwDOC_bI7s5sk6L0	7,108	Script is failing at "punctuator = PunctuationCapitalizationModel.from_pretrained("punctuation_en_bert")"	{ "login": "dsutradhar2", "id": 74150742, "node_id": "MDQ6VXNlcjc0MTUwNzQy", "avatar_url": "https://avatars.githubusercontent.com/u/74150742?v=4", "gravatar_id": "", "url": "https://api.github.com/users/dsutradhar2", "html_url": "https://github.com/dsutradhar2", "followers_url": "https://api.github.com/users/dsutradhar2/followers", "following_url": "https://api.github.com/users/dsutradhar2/following{/other_user}", "gists_url": "https://api.github.com/users/dsutradhar2/gists{/gist_id}", "starred_url": "https://api.github.com/users/dsutradhar2/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/dsutradhar2/subscriptions", "organizations_url": "https://api.github.com/users/dsutradhar2/orgs", "repos_url": "https://api.github.com/users/dsutradhar2/repos", "events_url": "https://api.github.com/users/dsutradhar2/events{/privacy}", "received_events_url": "https://api.github.com/users/dsutradhar2/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 1485815557, "node_id": "MDU6TGFiZWwxNDg1ODE1NTU3", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/bug", "name": "bug", "color": "d73a4a", "default": true, "description": "Something isn't working" }, { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null } ]	closed	false	null	[]	null	null	2023-07-26T05:34:10	2023-09-03T01:44:50	2023-09-03T01:44:50	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	I am getting the following error at "punctuator = PunctuationCapitalizationModel.from_pretrained("punctuation_en_bert")" I am using Ubuntu 22.04.2 LTS File "/home/deb/.local/lib/python3.10/site-packages/nemo/core/classes/common.py", line 852, in from_pretrained instance = class_.restore_from( File "/home/deb/.local/lib/python3.10/site-packages/nemo/core/classes/modelPT.py", line 435, in restore_from instance = cls._save_restore_connector.restore_from( File "/home/deb/.local/lib/python3.10/site-packages/nemo/core/connectors/save_restore_connector.py", line 248, in restore_from self.load_instance_with_state_dict(instance, state_dict, strict) File "/home/deb/.local/lib/python3.10/site-packages/nemo/core/connectors/save_restore_connector.py", line 203, in load_instance_with_state_dict instance.load_state_dict(state_dict, strict=strict) File "/home/deb/.local/lib/python3.10/site-packages/torch/nn/modules/module.py", line 2041, in load_state_dict raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format( RuntimeError: Error(s) in loading state_dict for PunctuationCapitalizationModel: Unexpected key(s) in state_dict: "bert_model.embeddings.position_ids".	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7108/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7108/timeline	null	not_planned	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7107	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7107/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7107/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7107/events	https://github.com/NVIDIA/NeMo/pull/7107	1,821,505,652	PR_kwDOC_bI7s5WZS0x	7,107	add bloom support	{ "login": "yaox12", "id": 3831900, "node_id": "MDQ6VXNlcjM4MzE5MDA=", "avatar_url": "https://avatars.githubusercontent.com/u/3831900?v=4", "gravatar_id": "", "url": "https://api.github.com/users/yaox12", "html_url": "https://github.com/yaox12", "followers_url": "https://api.github.com/users/yaox12/followers", "following_url": "https://api.github.com/users/yaox12/following{/other_user}", "gists_url": "https://api.github.com/users/yaox12/gists{/gist_id}", "starred_url": "https://api.github.com/users/yaox12/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/yaox12/subscriptions", "organizations_url": "https://api.github.com/users/yaox12/orgs", "repos_url": "https://api.github.com/users/yaox12/repos", "events_url": "https://api.github.com/users/yaox12/events{/privacy}", "received_events_url": "https://api.github.com/users/yaox12/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null }, { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-07-26T03:20:31	2023-09-01T01:45:28	2023-09-01T01:45:28	CONTRIBUTOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7107", "html_url": "https://github.com/NVIDIA/NeMo/pull/7107", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7107.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7107.patch", "merged_at": null }	# What does this PR do ? Add BLOOM model support to NeMo. Collection: [Note which collection this PR will affect] [NLP] # Changelog - Add BLOOM model architecture. - Add input embedding layernorm to the language model. - Add model configurations. - A conversion script to convert HuggingFace checkpoints to `.nemo` file. - Configurations and scripts for pertaining, evaluation, and SFT. # Usage * You can potentially add a usage example below ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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: - [x] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7107/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7107/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7106	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7106/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7106/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7106/events	https://github.com/NVIDIA/NeMo/pull/7106	1,820,506,061	PR_kwDOC_bI7s5WV7eJ	7,106	fix syntax error introduced in PR-7079	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	[ { "id": 4556669155, "node_id": "LA_kwDOC_bI7s8AAAABD5lA4w", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/cherry-pick", "name": "cherry-pick", "color": "ededed", "default": false, "description": null }, { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null }, { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-07-25T14:44:39	2023-08-17T01:43:02	2023-08-17T01:43:02	CONTRIBUTOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7106", "html_url": "https://github.com/NVIDIA/NeMo/pull/7106", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7106.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7106.patch", "merged_at": null }	# What does this PR do ? Fix quotes causing syntax error in notebook json 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 ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7106/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7106/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7105	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7105/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7105/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7105/events	https://github.com/NVIDIA/NeMo/pull/7105	1,820,334,983	PR_kwDOC_bI7s5WVWNO	7,105	bug fix - alibi causal: for ith query, bias is m · [−(i − 1), ..., −2, −1, 0] for the first i keys	{ "login": "LydiaXiaohongLi", "id": 35672632, "node_id": "MDQ6VXNlcjM1NjcyNjMy", "avatar_url": "https://avatars.githubusercontent.com/u/35672632?v=4", "gravatar_id": "", "url": "https://api.github.com/users/LydiaXiaohongLi", "html_url": "https://github.com/LydiaXiaohongLi", "followers_url": "https://api.github.com/users/LydiaXiaohongLi/followers", "following_url": "https://api.github.com/users/LydiaXiaohongLi/following{/other_user}", "gists_url": "https://api.github.com/users/LydiaXiaohongLi/gists{/gist_id}", "starred_url": "https://api.github.com/users/LydiaXiaohongLi/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/LydiaXiaohongLi/subscriptions", "organizations_url": "https://api.github.com/users/LydiaXiaohongLi/orgs", "repos_url": "https://api.github.com/users/LydiaXiaohongLi/repos", "events_url": "https://api.github.com/users/LydiaXiaohongLi/events{/privacy}", "received_events_url": "https://api.github.com/users/LydiaXiaohongLi/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[]	closed	false	null	[]	null	null	2023-07-25T13:22:26	2023-07-28T04:52:51	2023-07-28T04:52:50	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7105", "html_url": "https://github.com/NVIDIA/NeMo/pull/7105", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7105.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7105.patch", "merged_at": null }	# What does this PR do ? Fix alibi position embedding for causal attention. Collection: NLP # Changelog Existing: returns (1, num_heads, 1, key_length) Fixed: returns (1, num_heads, query_length, key_length), where for ith query, the bias is m · [−(i − 1), ..., −2, −1, 0] for the first i keys, as per the alibi paper PR Type: - [ ] Bugfix ## Who can review? Anyone in the community is free to review the PR once the checks have passed. [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "LydiaXiaohongLi", "id": 35672632, "node_id": "MDQ6VXNlcjM1NjcyNjMy", "avatar_url": "https://avatars.githubusercontent.com/u/35672632?v=4", "gravatar_id": "", "url": "https://api.github.com/users/LydiaXiaohongLi", "html_url": "https://github.com/LydiaXiaohongLi", "followers_url": "https://api.github.com/users/LydiaXiaohongLi/followers", "following_url": "https://api.github.com/users/LydiaXiaohongLi/following{/other_user}", "gists_url": "https://api.github.com/users/LydiaXiaohongLi/gists{/gist_id}", "starred_url": "https://api.github.com/users/LydiaXiaohongLi/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/LydiaXiaohongLi/subscriptions", "organizations_url": "https://api.github.com/users/LydiaXiaohongLi/orgs", "repos_url": "https://api.github.com/users/LydiaXiaohongLi/repos", "events_url": "https://api.github.com/users/LydiaXiaohongLi/events{/privacy}", "received_events_url": "https://api.github.com/users/LydiaXiaohongLi/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7105/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7105/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7104	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7104/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7104/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7104/events	https://github.com/NVIDIA/NeMo/issues/7104	1,819,829,068	I_kwDOC_bI7s5seGNM	7,104	When importing nemo.collections.asr: ImportError: cannot import name 'get_num_classes' from 'torchmetrics.utilities.data'	{ "login": "xekl", "id": 15870518, "node_id": "MDQ6VXNlcjE1ODcwNTE4", "avatar_url": "https://avatars.githubusercontent.com/u/15870518?v=4", "gravatar_id": "", "url": "https://api.github.com/users/xekl", "html_url": "https://github.com/xekl", "followers_url": "https://api.github.com/users/xekl/followers", "following_url": "https://api.github.com/users/xekl/following{/other_user}", "gists_url": "https://api.github.com/users/xekl/gists{/gist_id}", "starred_url": "https://api.github.com/users/xekl/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/xekl/subscriptions", "organizations_url": "https://api.github.com/users/xekl/orgs", "repos_url": "https://api.github.com/users/xekl/repos", "events_url": "https://api.github.com/users/xekl/events{/privacy}", "received_events_url": "https://api.github.com/users/xekl/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 1485815557, "node_id": "MDU6TGFiZWwxNDg1ODE1NTU3", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/bug", "name": "bug", "color": "d73a4a", "default": true, "description": "Something isn't working" }, { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null } ]	closed	false	null	[]	null	null	2023-07-25T08:38:32	2023-09-10T01:44:56	2023-09-10T01:44:56	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	I cannot start my app due to an import error when using NeMo ASR. The relevant stack trace: ``` (...) File "/asr/asr_solution.py", line 34, in <module> import nemo.collections.asr as nemo_asr # NeMo's ASR library File "/usr/local/lib/python3.10/site-packages/nemo/collections/asr/__init__.py", line 15, in <module> from nemo.collections.asr import data, losses, models, modules File "/usr/local/lib/python3.10/site-packages/nemo/collections/asr/losses/__init__.py", line 15, in <module> from nemo.collections.asr.losses.angularloss import AngularSoftmaxLoss File "/usr/local/lib/python3.10/site-packages/nemo/collections/asr/losses/angularloss.py", line 18, in <module> from nemo.core.classes import Loss, Typing, typecheck File "/usr/local/lib/python3.10/site-packages/nemo/core/__init__.py", line 16, in <module> from nemo.core.classes import * File "/usr/local/lib/python3.10/site-packages/nemo/core/classes/__init__.py", line 16, in <module> from nemo.core.classes.common import ( File "/usr/local/lib/python3.10/site-packages/nemo/core/classes/common.py", line 31, in <module> from nemo.utils import logging File "/usr/local/lib/python3.10/site-packages/nemo/utils/__init__.py", line 22, in <module> from nemo.utils.lightning_logger_patch import add_memory_handlers_to_pl_logger File "/usr/local/lib/python3.10/site-packages/nemo/utils/lightning_logger_patch.py", line 18, in <module> import pytorch_lightning as pl File "/usr/local/lib/python3.10/site-packages/pytorch_lightning/__init__.py", line 20, in <module> from pytorch_lightning import metrics # noqa: E402 File "/usr/local/lib/python3.10/site-packages/pytorch_lightning/metrics/__init__.py", line 15, in <module> from pytorch_lightning.metrics.classification import ( # noqa: F401 File "/usr/local/lib/python3.10/site-packages/pytorch_lightning/metrics/classification/__init__.py", line 14, in <module> from pytorch_lightning.metrics.classification.accuracy import Accuracy # noqa: F401 File "/usr/local/lib/python3.10/site-packages/pytorch_lightning/metrics/classification/accuracy.py", line 18, in <module> from pytorch_lightning.metrics.utils import deprecated_metrics, void File "/usr/local/lib/python3.10/site-packages/pytorch_lightning/metrics/utils.py", line 22, in <module> from torchmetrics.utilities.data import get_num_classes as _get_num_classes ImportError: cannot import name 'get_num_classes' from 'torchmetrics.utilities.data' (/usr/local/lib/python3.10/site-packages/torchmetrics/utilities/data.py) ``` This does not happen in my local virtual environment, but in a docker container, so I am guessing that somehow not all necessary files have been installed when building the container. However, `nemo_toolkit[asr]` is part of the libs being pipped via Dockerfile, and I thought all necessary requirements were contained in that installation. Should I add anything specific to the requirements?	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7104/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7104/timeline	null	not_planned	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7103	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7103/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7103/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7103/events	https://github.com/NVIDIA/NeMo/pull/7103	1,819,317,652	PR_kwDOC_bI7s5WR432	7,103	skip all reduce fix	{ "login": "wdykas", "id": 73254672, "node_id": "MDQ6VXNlcjczMjU0Njcy", "avatar_url": "https://avatars.githubusercontent.com/u/73254672?v=4", "gravatar_id": "", "url": "https://api.github.com/users/wdykas", "html_url": "https://github.com/wdykas", "followers_url": "https://api.github.com/users/wdykas/followers", "following_url": "https://api.github.com/users/wdykas/following{/other_user}", "gists_url": "https://api.github.com/users/wdykas/gists{/gist_id}", "starred_url": "https://api.github.com/users/wdykas/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/wdykas/subscriptions", "organizations_url": "https://api.github.com/users/wdykas/orgs", "repos_url": "https://api.github.com/users/wdykas/repos", "events_url": "https://api.github.com/users/wdykas/events{/privacy}", "received_events_url": "https://api.github.com/users/wdykas/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null }, { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-07-25T00:06:50	2023-08-17T01:43:04	2023-08-17T01:43:04	CONTRIBUTOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7103", "html_url": "https://github.com/NVIDIA/NeMo/pull/7103", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7103.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7103.patch", "merged_at": null }	# What does this PR do ? In certain scenarios(ex: single stage embeddings) there are no gradients for a rank with sequence parallel. This PR will skip the all reduce on those ranks. Collection: [NLP] # Changelog - skip seqparallel all reduce on no grads # Usage * You can potentially add a usage example below ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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 - [ x] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7103/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7103/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7102	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7102/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7102/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7102/events	https://github.com/NVIDIA/NeMo/pull/7102	1,819,182,640	PR_kwDOC_bI7s5WRbH6	7,102	fix syntax error introduced in PR-7079	{ "login": "bene-ges", "id": 61418381, "node_id": "MDQ6VXNlcjYxNDE4Mzgx", "avatar_url": "https://avatars.githubusercontent.com/u/61418381?v=4", "gravatar_id": "", "url": "https://api.github.com/users/bene-ges", "html_url": "https://github.com/bene-ges", "followers_url": "https://api.github.com/users/bene-ges/followers", "following_url": "https://api.github.com/users/bene-ges/following{/other_user}", "gists_url": "https://api.github.com/users/bene-ges/gists{/gist_id}", "starred_url": "https://api.github.com/users/bene-ges/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/bene-ges/subscriptions", "organizations_url": "https://api.github.com/users/bene-ges/orgs", "repos_url": "https://api.github.com/users/bene-ges/repos", "events_url": "https://api.github.com/users/bene-ges/events{/privacy}", "received_events_url": "https://api.github.com/users/bene-ges/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-07-24T21:35:42	2023-07-25T14:44:07	2023-07-25T14:44:07	CONTRIBUTOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7102", "html_url": "https://github.com/NVIDIA/NeMo/pull/7102", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7102.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7102.patch", "merged_at": "2023-07-25T14:44:07" }	# What does this PR do ? Fix quotes causing syntax error in notebook json 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 ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "ekmb", "id": 10428420, "node_id": "MDQ6VXNlcjEwNDI4NDIw", "avatar_url": "https://avatars.githubusercontent.com/u/10428420?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ekmb", "html_url": "https://github.com/ekmb", "followers_url": "https://api.github.com/users/ekmb/followers", "following_url": "https://api.github.com/users/ekmb/following{/other_user}", "gists_url": "https://api.github.com/users/ekmb/gists{/gist_id}", "starred_url": "https://api.github.com/users/ekmb/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ekmb/subscriptions", "organizations_url": "https://api.github.com/users/ekmb/orgs", "repos_url": "https://api.github.com/users/ekmb/repos", "events_url": "https://api.github.com/users/ekmb/events{/privacy}", "received_events_url": "https://api.github.com/users/ekmb/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7102/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7102/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7101	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7101/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7101/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7101/events	https://github.com/NVIDIA/NeMo/issues/7101	1,819,124,873	I_kwDOC_bI7s5sbaSJ	7,101	MSDD inference is too slow	{ "login": "SagyHarpazGong", "id": 99274232, "node_id": "U_kgDOBerN-A", "avatar_url": "https://avatars.githubusercontent.com/u/99274232?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SagyHarpazGong", "html_url": "https://github.com/SagyHarpazGong", "followers_url": "https://api.github.com/users/SagyHarpazGong/followers", "following_url": "https://api.github.com/users/SagyHarpazGong/following{/other_user}", "gists_url": "https://api.github.com/users/SagyHarpazGong/gists{/gist_id}", "starred_url": "https://api.github.com/users/SagyHarpazGong/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SagyHarpazGong/subscriptions", "organizations_url": "https://api.github.com/users/SagyHarpazGong/orgs", "repos_url": "https://api.github.com/users/SagyHarpazGong/repos", "events_url": "https://api.github.com/users/SagyHarpazGong/events{/privacy}", "received_events_url": "https://api.github.com/users/SagyHarpazGong/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null } ]	closed	false	{ "login": "tango4j", "id": 13211483, "node_id": "MDQ6VXNlcjEzMjExNDgz", "avatar_url": "https://avatars.githubusercontent.com/u/13211483?v=4", "gravatar_id": "", "url": "https://api.github.com/users/tango4j", "html_url": "https://github.com/tango4j", "followers_url": "https://api.github.com/users/tango4j/followers", "following_url": "https://api.github.com/users/tango4j/following{/other_user}", "gists_url": "https://api.github.com/users/tango4j/gists{/gist_id}", "starred_url": "https://api.github.com/users/tango4j/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/tango4j/subscriptions", "organizations_url": "https://api.github.com/users/tango4j/orgs", "repos_url": "https://api.github.com/users/tango4j/repos", "events_url": "https://api.github.com/users/tango4j/events{/privacy}", "received_events_url": "https://api.github.com/users/tango4j/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "login": "tango4j", "id": 13211483, "node_id": "MDQ6VXNlcjEzMjExNDgz", "avatar_url": "https://avatars.githubusercontent.com/u/13211483?v=4", "gravatar_id": "", "url": "https://api.github.com/users/tango4j", "html_url": "https://github.com/tango4j", "followers_url": "https://api.github.com/users/tango4j/followers", "following_url": "https://api.github.com/users/tango4j/following{/other_user}", "gists_url": "https://api.github.com/users/tango4j/gists{/gist_id}", "starred_url": "https://api.github.com/users/tango4j/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/tango4j/subscriptions", "organizations_url": "https://api.github.com/users/tango4j/orgs", "repos_url": "https://api.github.com/users/tango4j/repos", "events_url": "https://api.github.com/users/tango4j/events{/privacy}", "received_events_url": "https://api.github.com/users/tango4j/received_events", "type": "User", "user_view_type": "public", "site_admin": false }, { "login": "nithinraok", "id": 19668129, "node_id": "MDQ6VXNlcjE5NjY4MTI5", "avatar_url": "https://avatars.githubusercontent.com/u/19668129?v=4", "gravatar_id": "", "url": "https://api.github.com/users/nithinraok", "html_url": "https://github.com/nithinraok", "followers_url": "https://api.github.com/users/nithinraok/followers", "following_url": "https://api.github.com/users/nithinraok/following{/other_user}", "gists_url": "https://api.github.com/users/nithinraok/gists{/gist_id}", "starred_url": "https://api.github.com/users/nithinraok/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/nithinraok/subscriptions", "organizations_url": "https://api.github.com/users/nithinraok/orgs", "repos_url": "https://api.github.com/users/nithinraok/repos", "events_url": "https://api.github.com/users/nithinraok/events{/privacy}", "received_events_url": "https://api.github.com/users/nithinraok/received_events", "type": "User", "user_view_type": "public", "site_admin": false } ]	null	null	2023-07-24T20:50:28	2024-11-19T13:57:49	2024-08-01T01:55:21	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	I run the MSDD model on Nvidia A10 (24GB), but the inference is too slow, I looked on the code and there is a lot of traffic between the CPU and GPU and vice versa. most of the time GPU utilization is on 0% First the data is split into short segments according to the number of scales (I have 5 scales). After each scale splitting the embedding extraction is applied and save the embedding to pkl file. Then the the clustering is applied and finally the MSDD is applied. Is there something that can be done in order to speed up the inference? Is there any flag for parallelism the embedding extraction stage? please help.	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7101/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7101/timeline	null	not_planned	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7099	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7099/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7099/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7099/events	https://github.com/NVIDIA/NeMo/pull/7099	1,818,905,415	PR_kwDOC_bI7s5WQfHo	7,099	Fix absolute path in path join call	{ "login": "kingjan1999", "id": 3208269, "node_id": "MDQ6VXNlcjMyMDgyNjk=", "avatar_url": "https://avatars.githubusercontent.com/u/3208269?v=4", "gravatar_id": "", "url": "https://api.github.com/users/kingjan1999", "html_url": "https://github.com/kingjan1999", "followers_url": "https://api.github.com/users/kingjan1999/followers", "following_url": "https://api.github.com/users/kingjan1999/following{/other_user}", "gists_url": "https://api.github.com/users/kingjan1999/gists{/gist_id}", "starred_url": "https://api.github.com/users/kingjan1999/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/kingjan1999/subscriptions", "organizations_url": "https://api.github.com/users/kingjan1999/orgs", "repos_url": "https://api.github.com/users/kingjan1999/repos", "events_url": "https://api.github.com/users/kingjan1999/events{/privacy}", "received_events_url": "https://api.github.com/users/kingjan1999/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4811884691, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR", "name": "ASR", "color": "F3B33E", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-07-24T18:17:11	2023-07-24T18:28:27	2023-07-24T18:27:50	CONTRIBUTOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7099", "html_url": "https://github.com/NVIDIA/NeMo/pull/7099", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7099.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7099.patch", "merged_at": "2023-07-24T18:27:50" }	# What does this PR do ? This fixes a regression introduced in 39aff5ca3d52247e136a86429d567255c02a5d44 when joining paths was changed from string concatenation to `os.path.join`. At the moment the call disregards the first part (`frame_pred_dir`) and always returns a path starting with `/overlap_smoothing_output`. Collection: asr # Changelog - Remove leading slash in argument to `os.path.join` # Before your PR is "Ready for review" Pre checks: - [x] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [-] 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 - [x] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information	{ "login": "titu1994", "id": 3048602, "node_id": "MDQ6VXNlcjMwNDg2MDI=", "avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4", "gravatar_id": "", "url": "https://api.github.com/users/titu1994", "html_url": "https://github.com/titu1994", "followers_url": "https://api.github.com/users/titu1994/followers", "following_url": "https://api.github.com/users/titu1994/following{/other_user}", "gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}", "starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/titu1994/subscriptions", "organizations_url": "https://api.github.com/users/titu1994/orgs", "repos_url": "https://api.github.com/users/titu1994/repos", "events_url": "https://api.github.com/users/titu1994/events{/privacy}", "received_events_url": "https://api.github.com/users/titu1994/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7099/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7099/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7098	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7098/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7098/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7098/events	https://github.com/NVIDIA/NeMo/issues/7098	1,818,639,371	I_kwDOC_bI7s5sZjwL	7,098	No code for SLU metrics computation	{ "login": "maevdokimov", "id": 43167887, "node_id": "MDQ6VXNlcjQzMTY3ODg3", "avatar_url": "https://avatars.githubusercontent.com/u/43167887?v=4", "gravatar_id": "", "url": "https://api.github.com/users/maevdokimov", "html_url": "https://github.com/maevdokimov", "followers_url": "https://api.github.com/users/maevdokimov/followers", "following_url": "https://api.github.com/users/maevdokimov/following{/other_user}", "gists_url": "https://api.github.com/users/maevdokimov/gists{/gist_id}", "starred_url": "https://api.github.com/users/maevdokimov/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/maevdokimov/subscriptions", "organizations_url": "https://api.github.com/users/maevdokimov/orgs", "repos_url": "https://api.github.com/users/maevdokimov/repos", "events_url": "https://api.github.com/users/maevdokimov/events{/privacy}", "received_events_url": "https://api.github.com/users/maevdokimov/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null } ]	closed	false	{ "login": "stevehuang52", "id": 105218074, "node_id": "U_kgDOBkWAGg", "avatar_url": "https://avatars.githubusercontent.com/u/105218074?v=4", "gravatar_id": "", "url": "https://api.github.com/users/stevehuang52", "html_url": "https://github.com/stevehuang52", "followers_url": "https://api.github.com/users/stevehuang52/followers", "following_url": "https://api.github.com/users/stevehuang52/following{/other_user}", "gists_url": "https://api.github.com/users/stevehuang52/gists{/gist_id}", "starred_url": "https://api.github.com/users/stevehuang52/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/stevehuang52/subscriptions", "organizations_url": "https://api.github.com/users/stevehuang52/orgs", "repos_url": "https://api.github.com/users/stevehuang52/repos", "events_url": "https://api.github.com/users/stevehuang52/events{/privacy}", "received_events_url": "https://api.github.com/users/stevehuang52/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "login": "stevehuang52", "id": 105218074, "node_id": "U_kgDOBkWAGg", "avatar_url": "https://avatars.githubusercontent.com/u/105218074?v=4", "gravatar_id": "", "url": "https://api.github.com/users/stevehuang52", "html_url": "https://github.com/stevehuang52", "followers_url": "https://api.github.com/users/stevehuang52/followers", "following_url": "https://api.github.com/users/stevehuang52/following{/other_user}", "gists_url": "https://api.github.com/users/stevehuang52/gists{/gist_id}", "starred_url": "https://api.github.com/users/stevehuang52/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/stevehuang52/subscriptions", "organizations_url": "https://api.github.com/users/stevehuang52/orgs", "repos_url": "https://api.github.com/users/stevehuang52/repos", "events_url": "https://api.github.com/users/stevehuang52/events{/privacy}", "received_events_url": "https://api.github.com/users/stevehuang52/received_events", "type": "User", "user_view_type": "public", "site_admin": false } ]	null	null	2023-07-24T15:19:32	2023-09-03T01:44:55	2023-09-03T01:44:55	NONE	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	Hi! I was trying to find a place where SLU metrics are computed. AFAIU this is going to be [here](https://github.com/NVIDIA/NeMo/blob/main/examples/slu/speech_intent_slot/eval_utils/evaluator.py#L92). But I can't find where ```ErrorMetric ``` class is defined. Are you planning to make this code publicly available?	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7098/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7098/timeline	null	not_planned	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7097	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7097/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7097/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7097/events	https://github.com/NVIDIA/NeMo/issues/7097	1,817,142,891	I_kwDOC_bI7s5sT2Zr	7,097	[feature request] RNN-T loss to support arbitrary padding	{ "login": "vadimkantorov", "id": 1041752, "node_id": "MDQ6VXNlcjEwNDE3NTI=", "avatar_url": "https://avatars.githubusercontent.com/u/1041752?v=4", "gravatar_id": "", "url": "https://api.github.com/users/vadimkantorov", "html_url": "https://github.com/vadimkantorov", "followers_url": "https://api.github.com/users/vadimkantorov/followers", "following_url": "https://api.github.com/users/vadimkantorov/following{/other_user}", "gists_url": "https://api.github.com/users/vadimkantorov/gists{/gist_id}", "starred_url": "https://api.github.com/users/vadimkantorov/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/vadimkantorov/subscriptions", "organizations_url": "https://api.github.com/users/vadimkantorov/orgs", "repos_url": "https://api.github.com/users/vadimkantorov/repos", "events_url": "https://api.github.com/users/vadimkantorov/events{/privacy}", "received_events_url": "https://api.github.com/users/vadimkantorov/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4585431311, "node_id": "LA_kwDOC_bI7s8AAAABEVAhDw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale", "name": "stale", "color": "ededed", "default": false, "description": null } ]	closed	false	{ "login": "okuchaiev", "id": 22035961, "node_id": "MDQ6VXNlcjIyMDM1OTYx", "avatar_url": "https://avatars.githubusercontent.com/u/22035961?v=4", "gravatar_id": "", "url": "https://api.github.com/users/okuchaiev", "html_url": "https://github.com/okuchaiev", "followers_url": "https://api.github.com/users/okuchaiev/followers", "following_url": "https://api.github.com/users/okuchaiev/following{/other_user}", "gists_url": "https://api.github.com/users/okuchaiev/gists{/gist_id}", "starred_url": "https://api.github.com/users/okuchaiev/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/okuchaiev/subscriptions", "organizations_url": "https://api.github.com/users/okuchaiev/orgs", "repos_url": "https://api.github.com/users/okuchaiev/repos", "events_url": "https://api.github.com/users/okuchaiev/events{/privacy}", "received_events_url": "https://api.github.com/users/okuchaiev/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "login": "okuchaiev", "id": 22035961, "node_id": "MDQ6VXNlcjIyMDM1OTYx", "avatar_url": "https://avatars.githubusercontent.com/u/22035961?v=4", "gravatar_id": "", "url": "https://api.github.com/users/okuchaiev", "html_url": "https://github.com/okuchaiev", "followers_url": "https://api.github.com/users/okuchaiev/followers", "following_url": "https://api.github.com/users/okuchaiev/following{/other_user}", "gists_url": "https://api.github.com/users/okuchaiev/gists{/gist_id}", "starred_url": "https://api.github.com/users/okuchaiev/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/okuchaiev/subscriptions", "organizations_url": "https://api.github.com/users/okuchaiev/orgs", "repos_url": "https://api.github.com/users/okuchaiev/repos", "events_url": "https://api.github.com/users/okuchaiev/events{/privacy}", "received_events_url": "https://api.github.com/users/okuchaiev/received_events", "type": "User", "user_view_type": "public", "site_admin": false } ]	null	null	2023-07-23T14:08:41	2023-11-02T01:44:34	2023-11-02T01:44:34	CONTRIBUTOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	null	null	I think currently, the RNN-t loss supports any padding for every example in batch except the longest example. What is the nature of such limitation? (as it seems that the underlying code is already able to handle padding, so not clear why the longest example is different) Original discussion: https://github.com/NVIDIA/NeMo/issues/6797#issuecomment-1589149555	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7097/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7097/timeline	null	not_planned	false
https://api.github.com/repos/NVIDIA/NeMo/issues/7096	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7096/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7096/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7096/events	https://github.com/NVIDIA/NeMo/pull/7096	1,816,470,361	PR_kwDOC_bI7s5WIitW	7,096	Add GQA config to megatron gpt model	{ "login": "blahBlahhhJ", "id": 25292161, "node_id": "MDQ6VXNlcjI1MjkyMTYx", "avatar_url": "https://avatars.githubusercontent.com/u/25292161?v=4", "gravatar_id": "", "url": "https://api.github.com/users/blahBlahhhJ", "html_url": "https://github.com/blahBlahhhJ", "followers_url": "https://api.github.com/users/blahBlahhhJ/followers", "following_url": "https://api.github.com/users/blahBlahhhJ/following{/other_user}", "gists_url": "https://api.github.com/users/blahBlahhhJ/gists{/gist_id}", "starred_url": "https://api.github.com/users/blahBlahhhJ/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/blahBlahhhJ/subscriptions", "organizations_url": "https://api.github.com/users/blahBlahhhJ/orgs", "repos_url": "https://api.github.com/users/blahBlahhhJ/repos", "events_url": "https://api.github.com/users/blahBlahhhJ/events{/privacy}", "received_events_url": "https://api.github.com/users/blahBlahhhJ/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" } ]	closed	false	{ "login": "blahBlahhhJ", "id": 25292161, "node_id": "MDQ6VXNlcjI1MjkyMTYx", "avatar_url": "https://avatars.githubusercontent.com/u/25292161?v=4", "gravatar_id": "", "url": "https://api.github.com/users/blahBlahhhJ", "html_url": "https://github.com/blahBlahhhJ", "followers_url": "https://api.github.com/users/blahBlahhhJ/followers", "following_url": "https://api.github.com/users/blahBlahhhJ/following{/other_user}", "gists_url": "https://api.github.com/users/blahBlahhhJ/gists{/gist_id}", "starred_url": "https://api.github.com/users/blahBlahhhJ/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/blahBlahhhJ/subscriptions", "organizations_url": "https://api.github.com/users/blahBlahhhJ/orgs", "repos_url": "https://api.github.com/users/blahBlahhhJ/repos", "events_url": "https://api.github.com/users/blahBlahhhJ/events{/privacy}", "received_events_url": "https://api.github.com/users/blahBlahhhJ/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "login": "blahBlahhhJ", "id": 25292161, "node_id": "MDQ6VXNlcjI1MjkyMTYx", "avatar_url": "https://avatars.githubusercontent.com/u/25292161?v=4", "gravatar_id": "", "url": "https://api.github.com/users/blahBlahhhJ", "html_url": "https://github.com/blahBlahhhJ", "followers_url": "https://api.github.com/users/blahBlahhhJ/followers", "following_url": "https://api.github.com/users/blahBlahhhJ/following{/other_user}", "gists_url": "https://api.github.com/users/blahBlahhhJ/gists{/gist_id}", "starred_url": "https://api.github.com/users/blahBlahhhJ/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/blahBlahhhJ/subscriptions", "organizations_url": "https://api.github.com/users/blahBlahhhJ/orgs", "repos_url": "https://api.github.com/users/blahBlahhhJ/repos", "events_url": "https://api.github.com/users/blahBlahhhJ/events{/privacy}", "received_events_url": "https://api.github.com/users/blahBlahhhJ/received_events", "type": "User", "user_view_type": "public", "site_admin": false } ]	null	null	2023-07-21T23:17:15	2023-07-21T23:32:43	2023-07-21T23:32:42	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7096", "html_url": "https://github.com/NVIDIA/NeMo/pull/7096", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7096.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7096.patch", "merged_at": "2023-07-21T23:32:42" }	# What does this PR do ? Add GQA config in GPT config file. __Note__ that this PR merge into `mcore_gpt_path` not `main`. Collection: [Note which collection this PR will affect] # Changelog - Add specific line by line info of high level changes in this PR. # Usage * Set `num_query_groups` in `MegatronGPTModel` `cfg.model`. If `None`, normal (multi-head) attention is used. If `1`, multi-query attention is used. # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "ericharper", "id": 11999610, "node_id": "MDQ6VXNlcjExOTk5NjEw", "avatar_url": "https://avatars.githubusercontent.com/u/11999610?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ericharper", "html_url": "https://github.com/ericharper", "followers_url": "https://api.github.com/users/ericharper/followers", "following_url": "https://api.github.com/users/ericharper/following{/other_user}", "gists_url": "https://api.github.com/users/ericharper/gists{/gist_id}", "starred_url": "https://api.github.com/users/ericharper/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ericharper/subscriptions", "organizations_url": "https://api.github.com/users/ericharper/orgs", "repos_url": "https://api.github.com/users/ericharper/repos", "events_url": "https://api.github.com/users/ericharper/events{/privacy}", "received_events_url": "https://api.github.com/users/ericharper/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7096/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7096/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7095	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7095/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7095/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7095/events	https://github.com/NVIDIA/NeMo/pull/7095	1,816,462,677	PR_kwDOC_bI7s5WIhF8	7,095	Disable distopt contiguous param buffer by default	{ "login": "timmoon10", "id": 4406448, "node_id": "MDQ6VXNlcjQ0MDY0NDg=", "avatar_url": "https://avatars.githubusercontent.com/u/4406448?v=4", "gravatar_id": "", "url": "https://api.github.com/users/timmoon10", "html_url": "https://github.com/timmoon10", "followers_url": "https://api.github.com/users/timmoon10/followers", "following_url": "https://api.github.com/users/timmoon10/following{/other_user}", "gists_url": "https://api.github.com/users/timmoon10/gists{/gist_id}", "starred_url": "https://api.github.com/users/timmoon10/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/timmoon10/subscriptions", "organizations_url": "https://api.github.com/users/timmoon10/orgs", "repos_url": "https://api.github.com/users/timmoon10/repos", "events_url": "https://api.github.com/users/timmoon10/events{/privacy}", "received_events_url": "https://api.github.com/users/timmoon10/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 1485815557, "node_id": "MDU6TGFiZWwxNDg1ODE1NTU3", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/bug", "name": "bug", "color": "d73a4a", "default": true, "description": "Something isn't working" }, { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-07-21T23:01:08	2023-07-25T19:12:21	2023-07-25T19:12:17	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7095", "html_url": "https://github.com/NVIDIA/NeMo/pull/7095", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7095.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7095.patch", "merged_at": "2023-07-25T19:12:17" }	# What does this PR do ? This fixes a bug reported by @wdykas, where https://github.com/NVIDIA/NeMo/pull/5684 significantly increased the size of GPT checkpoints. I think this is because this PR configures the model parameters so they are all views into a single large buffer. However, PyTorch tensors save their underlying memory buffer when pickled, so checkpointing the model results in multiple redundant copies of all the parameters. This PR disables this behavior by default, at the expense of adding an extra copy after each param all-gather. Collection: NLP # Changelog - Disable distopt contiguous param buffer by default. # Usage This fix is applied by default. To restore the old behavior, configure the optimizer with `contiguous_param_buffer=True`. # Before your PR is "Ready for review" Pre checks: - [x] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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 - [x] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. Pinging @wdykas. # Additional Information * Fixes bug from https://github.com/NVIDIA/NeMo/pull/5684.	{ "login": "timmoon10", "id": 4406448, "node_id": "MDQ6VXNlcjQ0MDY0NDg=", "avatar_url": "https://avatars.githubusercontent.com/u/4406448?v=4", "gravatar_id": "", "url": "https://api.github.com/users/timmoon10", "html_url": "https://github.com/timmoon10", "followers_url": "https://api.github.com/users/timmoon10/followers", "following_url": "https://api.github.com/users/timmoon10/following{/other_user}", "gists_url": "https://api.github.com/users/timmoon10/gists{/gist_id}", "starred_url": "https://api.github.com/users/timmoon10/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/timmoon10/subscriptions", "organizations_url": "https://api.github.com/users/timmoon10/orgs", "repos_url": "https://api.github.com/users/timmoon10/repos", "events_url": "https://api.github.com/users/timmoon10/events{/privacy}", "received_events_url": "https://api.github.com/users/timmoon10/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7095/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7095/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7094	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7094/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7094/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7094/events	https://github.com/NVIDIA/NeMo/pull/7094	1,816,453,851	PR_kwDOC_bI7s5WIfNM	7,094	Adding docs and models for multiple lookahead cache-aware ASR	{ "login": "github-actions[bot]", "id": 41898282, "node_id": "MDM6Qm90NDE4OTgyODI=", "avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4", "gravatar_id": "", "url": "https://api.github.com/users/github-actions%5Bbot%5D", "html_url": "https://github.com/apps/github-actions", "followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers", "following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}", "gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}", "starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions", "organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs", "repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos", "events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}", "received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events", "type": "Bot", "user_view_type": "public", "site_admin": false }	[ { "id": 4556669155, "node_id": "LA_kwDOC_bI7s8AAAABD5lA4w", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/cherry-pick", "name": "cherry-pick", "color": "ededed", "default": false, "description": null }, { "id": 4811884691, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR", "name": "ASR", "color": "F3B33E", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-07-21T22:43:32	2023-07-22T07:12:02	2023-07-22T07:12:01	CONTRIBUTOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7094", "html_url": "https://github.com/NVIDIA/NeMo/pull/7094", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7094.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7094.patch", "merged_at": "2023-07-22T07:12:01" }	# What does this PR do ? This PR adds the following items related to multi-lookahead cache-aware streaming models: + documentation + support of multi-lookahead for transcribe_speech.py, transcribe_speech_parallel.py, and speech_to_text_eval.py, + one pre-trained model which is uploaded on NGC and supports 0s, 80ms, 480ms and 1040ms: https://ngc.nvidia.com/catalog/models/nvidia:nemo:stt_en_fastconformer_hybrid_large_streaming_multi # Changelog - Added a multi-lookahead cache-aware streaming model which supports multiple look-ahead in one single model. # Usage You may switch between different supported look-aheads by: 1- calling the set_default_att_context_size method of the model's encoder like this: ```python asr_model.encoder.set_default_att_context_size(att_context_size=[70,1]) ``` 2- set att_context_size when using transcribe_speech.py, transcribe_speech_parallel.py, or speech_to_text_eval.py. PR Type: - [X] New Feature - [ ] Bugfix - [X] Documentation	{ "login": "VahidooX", "id": 23551647, "node_id": "MDQ6VXNlcjIzNTUxNjQ3", "avatar_url": "https://avatars.githubusercontent.com/u/23551647?v=4", "gravatar_id": "", "url": "https://api.github.com/users/VahidooX", "html_url": "https://github.com/VahidooX", "followers_url": "https://api.github.com/users/VahidooX/followers", "following_url": "https://api.github.com/users/VahidooX/following{/other_user}", "gists_url": "https://api.github.com/users/VahidooX/gists{/gist_id}", "starred_url": "https://api.github.com/users/VahidooX/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/VahidooX/subscriptions", "organizations_url": "https://api.github.com/users/VahidooX/orgs", "repos_url": "https://api.github.com/users/VahidooX/repos", "events_url": "https://api.github.com/users/VahidooX/events{/privacy}", "received_events_url": "https://api.github.com/users/VahidooX/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7094/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7094/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7093	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7093/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7093/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7093/events	https://github.com/NVIDIA/NeMo/pull/7093	1,816,420,942	PR_kwDOC_bI7s5WIYF9	7,093	Use GPTModel from mcore	{ "login": "ericharper", "id": 11999610, "node_id": "MDQ6VXNlcjExOTk5NjEw", "avatar_url": "https://avatars.githubusercontent.com/u/11999610?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ericharper", "html_url": "https://github.com/ericharper", "followers_url": "https://api.github.com/users/ericharper/followers", "following_url": "https://api.github.com/users/ericharper/following{/other_user}", "gists_url": "https://api.github.com/users/ericharper/gists{/gist_id}", "starred_url": "https://api.github.com/users/ericharper/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ericharper/subscriptions", "organizations_url": "https://api.github.com/users/ericharper/orgs", "repos_url": "https://api.github.com/users/ericharper/repos", "events_url": "https://api.github.com/users/ericharper/events{/privacy}", "received_events_url": "https://api.github.com/users/ericharper/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" }, { "id": 4840216855, "node_id": "LA_kwDOC_bI7s8AAAABIH_ZFw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/CI", "name": "CI", "color": "1D76DB", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-07-21T21:51:57	2023-08-14T21:37:45	2023-08-14T21:37:44	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7093", "html_url": "https://github.com/NVIDIA/NeMo/pull/7093", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7093.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7093.patch", "merged_at": "2023-08-14T21:37:44" }	# What does this PR do ? This PR adds a path to use GPTModel from Megatron Core. Requirements: Make sure that megatron core is installed from a recent commit: ``` git clone https://github.com/NVIDIA/Megatron-LM.git && \ cd Megatron-LM && \ git checkout 3316e811cc5335ee24c2d203416d864edcf2f7a8 && \ pip install -e . ``` Collection: NLP # Changelog - Add specific line by line info of high level changes in this PR. # Usage Set `mcore_gpt=True` in the `MegatronGPTModel` config. # Before your PR is "Ready for review" Pre checks: - [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "ericharper", "id": 11999610, "node_id": "MDQ6VXNlcjExOTk5NjEw", "avatar_url": "https://avatars.githubusercontent.com/u/11999610?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ericharper", "html_url": "https://github.com/ericharper", "followers_url": "https://api.github.com/users/ericharper/followers", "following_url": "https://api.github.com/users/ericharper/following{/other_user}", "gists_url": "https://api.github.com/users/ericharper/gists{/gist_id}", "starred_url": "https://api.github.com/users/ericharper/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ericharper/subscriptions", "organizations_url": "https://api.github.com/users/ericharper/orgs", "repos_url": "https://api.github.com/users/ericharper/repos", "events_url": "https://api.github.com/users/ericharper/events{/privacy}", "received_events_url": "https://api.github.com/users/ericharper/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7093/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7093/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7092	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7092/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7092/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7092/events	https://github.com/NVIDIA/NeMo/pull/7092	1,816,328,863	PR_kwDOC_bI7s5WIEGs	7,092	Added script to extract ASR CTC and RNNT models from ASR hybrid models	{ "login": "trias702", "id": 25867060, "node_id": "MDQ6VXNlcjI1ODY3MDYw", "avatar_url": "https://avatars.githubusercontent.com/u/25867060?v=4", "gravatar_id": "", "url": "https://api.github.com/users/trias702", "html_url": "https://github.com/trias702", "followers_url": "https://api.github.com/users/trias702/followers", "following_url": "https://api.github.com/users/trias702/following{/other_user}", "gists_url": "https://api.github.com/users/trias702/gists{/gist_id}", "starred_url": "https://api.github.com/users/trias702/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/trias702/subscriptions", "organizations_url": "https://api.github.com/users/trias702/orgs", "repos_url": "https://api.github.com/users/trias702/repos", "events_url": "https://api.github.com/users/trias702/events{/privacy}", "received_events_url": "https://api.github.com/users/trias702/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4811884691, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR", "name": "ASR", "color": "F3B33E", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-07-21T20:08:06	2023-07-22T03:53:29	2023-07-22T03:53:28	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7092", "html_url": "https://github.com/NVIDIA/NeMo/pull/7092", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7092.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7092.patch", "merged_at": "2023-07-22T03:53:28" }	# What does this PR do ? Adds a script which extracts pure CTC or pure RNNT models from ASR hybrid models in .nemo format Collection: ASR # Changelog - Added new script in examples/asr/asr_hybrid_transducer_ctc/helpers/convert_nemo_asr_hybrid_to_ctc.py # Usage * You can potentially add a usage example below ```shell python convert_nemo_asr_hybrid_to_ctc.py -i /path/to/hybrid.nemo -o /path/to/saved_ctc_model.nemo -m 'ctc'\|'rnnt' ``` # Before your PR is "Ready for review" Pre checks: - [X] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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: - [X] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "trias702", "id": 25867060, "node_id": "MDQ6VXNlcjI1ODY3MDYw", "avatar_url": "https://avatars.githubusercontent.com/u/25867060?v=4", "gravatar_id": "", "url": "https://api.github.com/users/trias702", "html_url": "https://github.com/trias702", "followers_url": "https://api.github.com/users/trias702/followers", "following_url": "https://api.github.com/users/trias702/following{/other_user}", "gists_url": "https://api.github.com/users/trias702/gists{/gist_id}", "starred_url": "https://api.github.com/users/trias702/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/trias702/subscriptions", "organizations_url": "https://api.github.com/users/trias702/orgs", "repos_url": "https://api.github.com/users/trias702/repos", "events_url": "https://api.github.com/users/trias702/events{/privacy}", "received_events_url": "https://api.github.com/users/trias702/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7092/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7092/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7091	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7091/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7091/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7091/events	https://github.com/NVIDIA/NeMo/pull/7091	1,816,264,599	PR_kwDOC_bI7s5WH2Li	7,091	Fix tensor_parallel function calls to megatron-lm	{ "login": "stevehuang52", "id": 105218074, "node_id": "U_kgDOBkWAGg", "avatar_url": "https://avatars.githubusercontent.com/u/105218074?v=4", "gravatar_id": "", "url": "https://api.github.com/users/stevehuang52", "html_url": "https://github.com/stevehuang52", "followers_url": "https://api.github.com/users/stevehuang52/followers", "following_url": "https://api.github.com/users/stevehuang52/following{/other_user}", "gists_url": "https://api.github.com/users/stevehuang52/gists{/gist_id}", "starred_url": "https://api.github.com/users/stevehuang52/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/stevehuang52/subscriptions", "organizations_url": "https://api.github.com/users/stevehuang52/orgs", "repos_url": "https://api.github.com/users/stevehuang52/repos", "events_url": "https://api.github.com/users/stevehuang52/events{/privacy}", "received_events_url": "https://api.github.com/users/stevehuang52/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[ { "id": 4811886210, "node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg", "url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP", "name": "NLP", "color": "CBF09A", "default": false, "description": "" } ]	closed	false	null	[]	null	null	2023-07-21T19:01:20	2023-07-21T19:17:37	2023-07-21T19:17:02	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7091", "html_url": "https://github.com/NVIDIA/NeMo/pull/7091", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7091.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7091.patch", "merged_at": null }	# What does this PR do ? Megatron-LM has changed its function APIs in tensor_parallel, where a lot of params (e.g., `use_cpu_initialization`) are moved into a config object (e.g., `megatron.core.model_parallel_config.ModelParallelConfig`), for example [here](https://github.com/NVIDIA/Megatron-LM/blob/main/megatron/core/tensor_parallel/layers.py#L148) This PR fixes such errors when trying to use those functions with the latest Megatron-LM codebase, but there may be some other missed functions. Collection: [NLP]	{ "login": "stevehuang52", "id": 105218074, "node_id": "U_kgDOBkWAGg", "avatar_url": "https://avatars.githubusercontent.com/u/105218074?v=4", "gravatar_id": "", "url": "https://api.github.com/users/stevehuang52", "html_url": "https://github.com/stevehuang52", "followers_url": "https://api.github.com/users/stevehuang52/followers", "following_url": "https://api.github.com/users/stevehuang52/following{/other_user}", "gists_url": "https://api.github.com/users/stevehuang52/gists{/gist_id}", "starred_url": "https://api.github.com/users/stevehuang52/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/stevehuang52/subscriptions", "organizations_url": "https://api.github.com/users/stevehuang52/orgs", "repos_url": "https://api.github.com/users/stevehuang52/repos", "events_url": "https://api.github.com/users/stevehuang52/events{/privacy}", "received_events_url": "https://api.github.com/users/stevehuang52/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7091/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7091/timeline	null	null	true
https://api.github.com/repos/NVIDIA/NeMo/issues/7088	https://api.github.com/repos/NVIDIA/NeMo	https://api.github.com/repos/NVIDIA/NeMo/issues/7088/labels{/name}	https://api.github.com/repos/NVIDIA/NeMo/issues/7088/comments	https://api.github.com/repos/NVIDIA/NeMo/issues/7088/events	https://github.com/NVIDIA/NeMo/pull/7088	1,815,000,132	PR_kwDOC_bI7s5WDmOT	7,088	update TTS readme	{ "login": "XuesongYang", "id": 1646669, "node_id": "MDQ6VXNlcjE2NDY2Njk=", "avatar_url": "https://avatars.githubusercontent.com/u/1646669?v=4", "gravatar_id": "", "url": "https://api.github.com/users/XuesongYang", "html_url": "https://github.com/XuesongYang", "followers_url": "https://api.github.com/users/XuesongYang/followers", "following_url": "https://api.github.com/users/XuesongYang/following{/other_user}", "gists_url": "https://api.github.com/users/XuesongYang/gists{/gist_id}", "starred_url": "https://api.github.com/users/XuesongYang/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/XuesongYang/subscriptions", "organizations_url": "https://api.github.com/users/XuesongYang/orgs", "repos_url": "https://api.github.com/users/XuesongYang/repos", "events_url": "https://api.github.com/users/XuesongYang/events{/privacy}", "received_events_url": "https://api.github.com/users/XuesongYang/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	[]	closed	false	null	[]	null	null	2023-07-21T00:51:26	2023-07-24T06:59:02	2023-07-24T06:59:01	COLLABORATOR	{ "total": 0, "completed": 0, "percent_completed": 0 }	null	false	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/7088", "html_url": "https://github.com/NVIDIA/NeMo/pull/7088", "diff_url": "https://github.com/NVIDIA/NeMo/pull/7088.diff", "patch_url": "https://github.com/NVIDIA/NeMo/pull/7088.patch", "merged_at": "2023-07-24T06:59:01" }	# 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 ```python # Add a code snippet demonstrating how to use this ``` # Before your PR is "Ready for review" Pre checks: - [x] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) - [ ] 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 - [x] 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](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas. # Additional Information * Related to # (issue)	{ "login": "XuesongYang", "id": 1646669, "node_id": "MDQ6VXNlcjE2NDY2Njk=", "avatar_url": "https://avatars.githubusercontent.com/u/1646669?v=4", "gravatar_id": "", "url": "https://api.github.com/users/XuesongYang", "html_url": "https://github.com/XuesongYang", "followers_url": "https://api.github.com/users/XuesongYang/followers", "following_url": "https://api.github.com/users/XuesongYang/following{/other_user}", "gists_url": "https://api.github.com/users/XuesongYang/gists{/gist_id}", "starred_url": "https://api.github.com/users/XuesongYang/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/XuesongYang/subscriptions", "organizations_url": "https://api.github.com/users/XuesongYang/orgs", "repos_url": "https://api.github.com/users/XuesongYang/repos", "events_url": "https://api.github.com/users/XuesongYang/events{/privacy}", "received_events_url": "https://api.github.com/users/XuesongYang/received_events", "type": "User", "user_view_type": "public", "site_admin": false }	{ "url": "https://api.github.com/repos/NVIDIA/NeMo/issues/7088/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/NVIDIA/NeMo/issues/7088/timeline	null	null	true

Subsets and Splits

No saved queries yet

Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.