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/4584 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4584/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4584/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4584/events | https://github.com/NVIDIA/NeMo/issues/4584 | 1,315,135,878 | I_kwDOC_bI7s5OY2GG | 4,584 | Pytorch lightning required but not installed for text_normalization | {
"login": "f4hy",
"id": 36440,
"node_id": "MDQ6VXNlcjM2NDQw",
"avatar_url": "https://avatars.githubusercontent.com/u/36440?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/f4hy",
"html_url": "https://github.com/f4hy",
"followers_url": "https://api.github.com/users/f4hy/followers",
"following_url": "https://api.github.com/users/f4hy/following{/other_user}",
"gists_url": "https://api.github.com/users/f4hy/gists{/gist_id}",
"starred_url": "https://api.github.com/users/f4hy/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/f4hy/subscriptions",
"organizations_url": "https://api.github.com/users/f4hy/orgs",
"repos_url": "https://api.github.com/users/f4hy/repos",
"events_url": "https://api.github.com/users/f4hy/events{/privacy}",
"received_events_url": "https://api.github.com/users/f4hy/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": "yzhang123",
"id": 4204271,
"node_id": "MDQ6VXNlcjQyMDQyNzE=",
"avatar_url": "https://avatars.githubusercontent.com/u/4204271?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/yzhang123",
"html_url": "https://github.com/yzhang123",
"followers_url": "https://api.github.com/users/yzhang123/followers",
"following_url": "https://api.github.com/users/yzhang123/following{/other_user}",
"gists_url": "https://api.github.com/users/yzhang123/gists{/gist_id}",
"starred_url": "https://api.github.com/users/yzhang123/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/yzhang123/subscriptions",
"organizations_url": "https://api.github.com/users/yzhang123/orgs",
"repos_url": "https://api.github.com/users/yzhang123/repos",
"events_url": "https://api.github.com/users/yzhang123/events{/privacy}",
"received_events_url": "https://api.github.com/users/yzhang123/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"login": "yzhang123",
"id": 4204271,
"node_id": "MDQ6VXNlcjQyMDQyNzE=",
"avatar_url": "https://avatars.githubusercontent.com/u/4204271?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/yzhang123",
"html_url": "https://github.com/yzhang123",
"followers_url": "https://api.github.com/users/yzhang123/followers",
"following_url": "https://api.github.com/users/yzhang123/following{/other_user}",
"gists_url": "https://api.github.com/users/yzhang123/gists{/gist_id}",
"starred_url": "https://api.github.com/users/yzhang123/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/yzhang123/subscriptions",
"organizations_url": "https://api.github.com/users/yzhang123/orgs",
"repos_url": "https://api.github.com/users/yzhang123/repos",
"events_url": "https://api.github.com/users/yzhang123/events{/privacy}",
"received_events_url": "https://api.github.com/users/yzhang123/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
}
] | null | null | 2022-07-22T15:18:30 | 2022-08-25T14:57:32 | 2022-08-25T14:57:31 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | **Describe the bug**
Installing `nemo_toolkit[nemo_text_processing]` does not bring in `pytorch_lightning` but then
```python
from nemo_text_processing.text_normalization import normalize
```
errors as pytorch_lightning is required to import
**Steps/Code to reproduce bug**
1. on a fresh venv, install `nemo_toolkit[nemo_text_processing]`
2. run `from nemo_text_processing.text_normalization import normalize`
3. See error
**Expected behavior**
If text processing requires pytorch_lightning it should be installed as a requirement
| {
"login": "yzhang123",
"id": 4204271,
"node_id": "MDQ6VXNlcjQyMDQyNzE=",
"avatar_url": "https://avatars.githubusercontent.com/u/4204271?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/yzhang123",
"html_url": "https://github.com/yzhang123",
"followers_url": "https://api.github.com/users/yzhang123/followers",
"following_url": "https://api.github.com/users/yzhang123/following{/other_user}",
"gists_url": "https://api.github.com/users/yzhang123/gists{/gist_id}",
"starred_url": "https://api.github.com/users/yzhang123/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/yzhang123/subscriptions",
"organizations_url": "https://api.github.com/users/yzhang123/orgs",
"repos_url": "https://api.github.com/users/yzhang123/repos",
"events_url": "https://api.github.com/users/yzhang123/events{/privacy}",
"received_events_url": "https://api.github.com/users/yzhang123/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4584/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/4584/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4583 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4583/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4583/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4583/events | https://github.com/NVIDIA/NeMo/issues/4583 | 1,315,051,363 | I_kwDOC_bI7s5OYhdj | 4,583 | Loss is getting zero when training Titanet with Adam | {
"login": "GrigorKirakosyan",
"id": 98385298,
"node_id": "U_kgDOBd09kg",
"avatar_url": "https://avatars.githubusercontent.com/u/98385298?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/GrigorKirakosyan",
"html_url": "https://github.com/GrigorKirakosyan",
"followers_url": "https://api.github.com/users/GrigorKirakosyan/followers",
"following_url": "https://api.github.com/users/GrigorKirakosyan/following{/other_user}",
"gists_url": "https://api.github.com/users/GrigorKirakosyan/gists{/gist_id}",
"starred_url": "https://api.github.com/users/GrigorKirakosyan/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/GrigorKirakosyan/subscriptions",
"organizations_url": "https://api.github.com/users/GrigorKirakosyan/orgs",
"repos_url": "https://api.github.com/users/GrigorKirakosyan/repos",
"events_url": "https://api.github.com/users/GrigorKirakosyan/events{/privacy}",
"received_events_url": "https://api.github.com/users/GrigorKirakosyan/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | 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
} | [
{
"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 | 2022-07-22T14:03:21 | 2022-09-21T04:35:59 | 2022-09-21T04:35:59 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | Hi
I am trying to train TitaNet with angular loss. When the optimizer is SGD everything is fine (loss decreases and TopK accuracy increases). Switching optimizer from SGD to Adam or AdamW results zero loss and zero accuracy after some steps. Do you have any clue about zero-loss?
Btw, number of speakers in my case is equal to ~25000. | {
"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/4583/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/4583/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4582 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4582/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4582/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4582/events | https://github.com/NVIDIA/NeMo/pull/4582 | 1,314,554,036 | PR_kwDOC_bI7s477Ott | 4,582 | [TTS][ASR] customize arguments for trimming the leading/trailing silence | {
"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 | 2022-07-22T07:44:51 | 2022-07-26T17:20:02 | 2022-07-26T17:19:58 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4582",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4582",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4582.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4582.patch",
"merged_at": "2022-07-26T17:19:58"
} | # What does this PR do ?
It supports overriding arguments of `librosa.effects.trim` when trimming silence. By default, we assign `trim: false` in YAML config. But when enabling it by `trim: true`, you have options to override original arguments of `librosa.effects.trim`, for example,
trim: true
trim_top_db: 50,
trim_frame_length: ${model.n_window_size},
trim_hop_length: ${model.n_window_stride},
**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": "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/4582/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/4582/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4581 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4581/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4581/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4581/events | https://github.com/NVIDIA/NeMo/issues/4581 | 1,314,515,133 | I_kwDOC_bI7s5OWei9 | 4,581 | [Question] Exporting HifiGanModel and Tacotron2Model for TTS on Android | {
"login": "inventionsbyhamid",
"id": 4223656,
"node_id": "MDQ6VXNlcjQyMjM2NTY=",
"avatar_url": "https://avatars.githubusercontent.com/u/4223656?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/inventionsbyhamid",
"html_url": "https://github.com/inventionsbyhamid",
"followers_url": "https://api.github.com/users/inventionsbyhamid/followers",
"following_url": "https://api.github.com/users/inventionsbyhamid/following{/other_user}",
"gists_url": "https://api.github.com/users/inventionsbyhamid/gists{/gist_id}",
"starred_url": "https://api.github.com/users/inventionsbyhamid/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/inventionsbyhamid/subscriptions",
"organizations_url": "https://api.github.com/users/inventionsbyhamid/orgs",
"repos_url": "https://api.github.com/users/inventionsbyhamid/repos",
"events_url": "https://api.github.com/users/inventionsbyhamid/events{/privacy}",
"received_events_url": "https://api.github.com/users/inventionsbyhamid/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": "borisfom",
"id": 14189615,
"node_id": "MDQ6VXNlcjE0MTg5NjE1",
"avatar_url": "https://avatars.githubusercontent.com/u/14189615?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/borisfom",
"html_url": "https://github.com/borisfom",
"followers_url": "https://api.github.com/users/borisfom/followers",
"following_url": "https://api.github.com/users/borisfom/following{/other_user}",
"gists_url": "https://api.github.com/users/borisfom/gists{/gist_id}",
"starred_url": "https://api.github.com/users/borisfom/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/borisfom/subscriptions",
"organizations_url": "https://api.github.com/users/borisfom/orgs",
"repos_url": "https://api.github.com/users/borisfom/repos",
"events_url": "https://api.github.com/users/borisfom/events{/privacy}",
"received_events_url": "https://api.github.com/users/borisfom/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"login": "borisfom",
"id": 14189615,
"node_id": "MDQ6VXNlcjE0MTg5NjE1",
"avatar_url": "https://avatars.githubusercontent.com/u/14189615?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/borisfom",
"html_url": "https://github.com/borisfom",
"followers_url": "https://api.github.com/users/borisfom/followers",
"following_url": "https://api.github.com/users/borisfom/following{/other_user}",
"gists_url": "https://api.github.com/users/borisfom/gists{/gist_id}",
"starred_url": "https://api.github.com/users/borisfom/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/borisfom/subscriptions",
"organizations_url": "https://api.github.com/users/borisfom/orgs",
"repos_url": "https://api.github.com/users/borisfom/repos",
"events_url": "https://api.github.com/users/borisfom/events{/privacy}",
"received_events_url": "https://api.github.com/users/borisfom/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
}
] | null | null | 2022-07-22T07:25:12 | 2022-10-14T02:13:49 | 2022-10-14T02:13:49 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | I am using TTS using the below code. I want to run TTS offline on an Android device as a next step using the same pretrained models.
```python
import soundfile as sf
from nemo.collections.tts.models import HifiGanModel
from nemo.collections.tts.models import Tacotron2Model
spec_generator = Tacotron2Model.restore_from("tts_hi_female_tacotron2.nemo")
spec_generator.eval()
vocoder = HifiGanModel.from_pretrained(model_name="tts_hifigan")
parsed = spec_generator.parse("न्यू यॉर्क में सड़क धंसने से एक विशाल सिंकहोल बनने का वीडियो वायरल हो गया है। इसमें भारी बारिश के बाद इलाके में एक वैन को सिंकहोल में गिरते हुए देखा जा सकता है।")
spectrogram = spec_generator.generate_spectrogram(tokens=parsed)
audio = vocoder.convert_spectrogram_to_audio(spec=spectrogram)
sf.write("speech.wav", audio.to('cpu').detach().numpy()[0], 22050)
```
I went through the issue [How to use/port quartzNet model on mobile?](https://github.com/NVIDIA/NeMo/issues/1394) which suggests it is possible to export the model.
Pytorch documentation for exporting model to run on android shows below code
```python
example = torch.rand(1, 3, 224, 224)
traced_script_module = torch.jit.trace(model, example)
traced_script_module_optimized = optimize_for_mobile(traced_script_module)
traced_script_module_optimized._save_for_lite_interpreter("app/src/main/assets/model.ptl")
```
Where I am stuck now ->
What to pass in example as second paramter to torch.jit.trace for tacotron2 and hifigan?
I would need to export both tacotron2 and hifigan separately?
How to pass output from one model to input to another on android? (This only if I am able to achieve first two steps)
| {
"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/4581/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/4581/timeline | null | not_planned | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4580 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4580/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4580/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4580/events | https://github.com/NVIDIA/NeMo/pull/4580 | 1,313,932,510 | PR_kwDOC_bI7s475BWF | 4,580 | Wav2vec2.0 Script add | {
"login": "tbartley94",
"id": 90423858,
"node_id": "MDQ6VXNlcjkwNDIzODU4",
"avatar_url": "https://avatars.githubusercontent.com/u/90423858?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/tbartley94",
"html_url": "https://github.com/tbartley94",
"followers_url": "https://api.github.com/users/tbartley94/followers",
"following_url": "https://api.github.com/users/tbartley94/following{/other_user}",
"gists_url": "https://api.github.com/users/tbartley94/gists{/gist_id}",
"starred_url": "https://api.github.com/users/tbartley94/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/tbartley94/subscriptions",
"organizations_url": "https://api.github.com/users/tbartley94/orgs",
"repos_url": "https://api.github.com/users/tbartley94/repos",
"events_url": "https://api.github.com/users/tbartley94/events{/privacy}",
"received_events_url": "https://api.github.com/users/tbartley94/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 | 2022-07-21T23:27:26 | 2022-10-14T02:13:51 | 2022-10-14T02:13:51 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | true | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4580",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4580",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4580.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4580.patch",
"merged_at": null
} | # What does this PR do ?
Adds a script to load Wav2Vec2.0 weights from Fair to NeMo implementation. Also adjusts NeMo implementation to be similar to Fair's.
**Collection**: [ASR]
# Changelog
- Added script to download and load state_dict from Fairseq Wav2Vec models to NeMo equivalents
- Adjusted order of normalization operations for wav2vec preprocessing
- Changed Wav2Vec module config so that projection operation of features is done in transformer encoder
- Altered contrastive loss so config can provide option to not combine groups. (Just added an optional boolean)
# Before your PR is "Ready for review"
**Pre checks**:
- [y ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md)
- [ n] Did you write any new necessary tests?
- [ n] Did you add or update any necessary documentation?
- [ n] 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**:
- [y ] New Feature
- [ y] 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/4580/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/4580/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4577 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4577/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4577/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4577/events | https://github.com/NVIDIA/NeMo/issues/4577 | 1,313,508,295 | I_kwDOC_bI7s5OSovH | 4,577 | NormalizerWithAudio error for written digits | {
"login": "ghost",
"id": 10137,
"node_id": "MDQ6VXNlcjEwMTM3",
"avatar_url": "https://avatars.githubusercontent.com/u/10137?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ghost",
"html_url": "https://github.com/ghost",
"followers_url": "https://api.github.com/users/ghost/followers",
"following_url": "https://api.github.com/users/ghost/following{/other_user}",
"gists_url": "https://api.github.com/users/ghost/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ghost/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ghost/subscriptions",
"organizations_url": "https://api.github.com/users/ghost/orgs",
"repos_url": "https://api.github.com/users/ghost/repos",
"events_url": "https://api.github.com/users/ghost/events{/privacy}",
"received_events_url": "https://api.github.com/users/ghost/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": "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
} | [
{
"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
}
] | null | null | 2022-07-21T16:22:07 | 2022-07-29T19:56:01 | 2022-07-29T19:56:01 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | **Describe the bug**
See below, the normalization is incorrect for specific regex cases.
**Steps/Code to reproduce bug**
```python
import nemo; nemo.__version__ # '1.9.0rc0'
from nemo_text_processing.text_normalization.normalize_with_audio import (
NormalizerWithAudio,
)
normalizer = NormalizerWithAudio(input_case="cased")
# SUCCESS
written_text = "01"
spoken_texts = normalizer.normalize(written_text, n_tagged=-1)
print(spoken_texts)
"""
{'oh one', 'zero one'}
"""
# SUCCESS
written_text = "010"
spoken_texts = normalizer.normalize(written_text, n_tagged=-1)
print(spoken_texts)
"""
{'oh one oh', 'oh ten', 'zero one zero', 'zero ten'}
"""
# FAIL
written_text = "0101"
spoken_texts = normalizer.normalize(written_text, n_tagged=-1)
print(spoken_texts)
"""
{'oh one hundred and one',
'oh one hundred one',
'oh one oh one',
'oh one one', ########### WRONG ###########
'zero one hundred and one',
'zero one hundred one',
'zero one one', ########### WRONG ###########
'zero one zero one'}
"""
# FAIL
written_text = "101"
spoken_texts = normalizer.normalize(written_text, n_tagged=-1)
print(spoken_texts)
"""
{'one hundred and one',
'one hundred one',
'one oh one',
'one one', ########### WRONG ###########
'one zero one'}
"""
# FAIL
written_text = "101010"
spoken_texts = normalizer.normalize(written_text, n_tagged=-1)
print(spoken_texts)
"""
{'one hundred one thousand and ten',
'one hundred one thousand ten',
'one oh one oh one oh',
'one one thousand and ten', ########### WRONG ###########
'one one thousand ten', ########### WRONG ###########
'one zero one zero one zero'}
"""
# FAIL
written_text = "507"
spoken_texts = normalizer.normalize(written_text, n_tagged=-1)
print(spoken_texts)
"""
{'five hundred and seven',
'five hundred seven',
'five oh seven',
'five seven', ########### WRONG ###########
'five zero seven'}
"""
# etc.
```
**Expected behavior**
A clear and concise description of what you expected to happen.
**Environment overview (please complete the following information)**
- Environment location: [Bare-metal, Docker, Cloud(specify cloud provider - AWS, Azure, GCP, Collab)]
- Method of NeMo install: [pip install or from source]. Please specify exact commands you used to install.
- If method of install is [Docker], provide `docker pull` & `docker run` commands used
**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**
Add any other context about the problem here.
Example: GPU model
| {
"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/4577/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/4577/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4576 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4576/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4576/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4576/events | https://github.com/NVIDIA/NeMo/pull/4576 | 1,313,433,324 | PR_kwDOC_bI7s473Tkb | 4,576 | [Add] Catalan ASR NGC Resource | {
"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 | 2022-07-21T15:23:02 | 2022-07-21T18:25:42 | 2022-07-21T18:25:42 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4576",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4576",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4576.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4576.patch",
"merged_at": "2022-07-21T18:25:42"
} | Signed-off-by: stevehuang52 <[email protected]>
The PR adds Catalan ASR models on NGC to the available pretrained models. | {
"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/4576/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/4576/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4575 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4575/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4575/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4575/events | https://github.com/NVIDIA/NeMo/issues/4575 | 1,313,265,331 | I_kwDOC_bI7s5ORtaz | 4,575 | [NLP] [question/bug?] `train_loss` falling, whereas `val_loss`, `punct_f1`, `capit_f1` all increasing | {
"login": "itzsimpl",
"id": 4784070,
"node_id": "MDQ6VXNlcjQ3ODQwNzA=",
"avatar_url": "https://avatars.githubusercontent.com/u/4784070?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/itzsimpl",
"html_url": "https://github.com/itzsimpl",
"followers_url": "https://api.github.com/users/itzsimpl/followers",
"following_url": "https://api.github.com/users/itzsimpl/following{/other_user}",
"gists_url": "https://api.github.com/users/itzsimpl/gists{/gist_id}",
"starred_url": "https://api.github.com/users/itzsimpl/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/itzsimpl/subscriptions",
"organizations_url": "https://api.github.com/users/itzsimpl/orgs",
"repos_url": "https://api.github.com/users/itzsimpl/repos",
"events_url": "https://api.github.com/users/itzsimpl/events{/privacy}",
"received_events_url": "https://api.github.com/users/itzsimpl/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 | {
"login": "PeganovAnton",
"id": 10737305,
"node_id": "MDQ6VXNlcjEwNzM3MzA1",
"avatar_url": "https://avatars.githubusercontent.com/u/10737305?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/PeganovAnton",
"html_url": "https://github.com/PeganovAnton",
"followers_url": "https://api.github.com/users/PeganovAnton/followers",
"following_url": "https://api.github.com/users/PeganovAnton/following{/other_user}",
"gists_url": "https://api.github.com/users/PeganovAnton/gists{/gist_id}",
"starred_url": "https://api.github.com/users/PeganovAnton/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/PeganovAnton/subscriptions",
"organizations_url": "https://api.github.com/users/PeganovAnton/orgs",
"repos_url": "https://api.github.com/users/PeganovAnton/repos",
"events_url": "https://api.github.com/users/PeganovAnton/events{/privacy}",
"received_events_url": "https://api.github.com/users/PeganovAnton/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"login": "PeganovAnton",
"id": 10737305,
"node_id": "MDQ6VXNlcjEwNzM3MzA1",
"avatar_url": "https://avatars.githubusercontent.com/u/10737305?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/PeganovAnton",
"html_url": "https://github.com/PeganovAnton",
"followers_url": "https://api.github.com/users/PeganovAnton/followers",
"following_url": "https://api.github.com/users/PeganovAnton/following{/other_user}",
"gists_url": "https://api.github.com/users/PeganovAnton/gists{/gist_id}",
"starred_url": "https://api.github.com/users/PeganovAnton/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/PeganovAnton/subscriptions",
"organizations_url": "https://api.github.com/users/PeganovAnton/orgs",
"repos_url": "https://api.github.com/users/PeganovAnton/repos",
"events_url": "https://api.github.com/users/PeganovAnton/events{/privacy}",
"received_events_url": "https://api.github.com/users/PeganovAnton/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
}
] | null | null | 2022-07-21T13:30:00 | 2022-10-14T02:13:53 | 2022-10-14T02:13:53 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | @PeganovAnton Training on multiple GPUs I'm noticing that `train_loss` is decreasing, and `f1` scores are increasing, but so is `val_loss`. Is `val_loss` the right metric to be monitored? Would mean `f1` score be better suited?
What does `logits_ndim` in https://github.com/NVIDIA/NeMo/blob/fea3775c00adfacfe0a414dea15544abc96db8dc/nemo/collections/nlp/models/token_classification/punctuation_capitalization_model.py#L126
stand for? Is this initialisation still correct if one changes the number of classes, i.e. includes support for additional punctuation marks? What if the numbers of classes for punctuation and capitalisation differ?
<img width="348" alt="image" src="https://user-images.githubusercontent.com/4784070/180222056-3f2d6dbc-6123-43a1-a4d1-c49b1651f979.png">
<img width="348" alt="image" src="https://user-images.githubusercontent.com/4784070/180222242-6fde5196-fc70-454e-86c7-fbe3c800f5a0.png">
<img width="348" alt="image" src="https://user-images.githubusercontent.com/4784070/180222345-ee0f65a2-7153-4d8e-b652-2b549e937be0.png">
<img width="348" alt="image" src="https://user-images.githubusercontent.com/4784070/180222409-27f760c0-4d4e-41ef-a58f-0a905d73862f.png">
**Environment details**
`pytorch:22.06-py3` + `nemo:1.10.0` + PR #4553 patch
| {
"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/4575/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/4575/timeline | null | not_planned | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4574 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4574/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4574/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4574/events | https://github.com/NVIDIA/NeMo/issues/4574 | 1,312,872,944 | I_kwDOC_bI7s5OQNnw | 4,574 | parser not found during inference of Tacotron2 TTS on non-english language | {
"login": "prashantskit",
"id": 101550102,
"node_id": "U_kgDOBg2IFg",
"avatar_url": "https://avatars.githubusercontent.com/u/101550102?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/prashantskit",
"html_url": "https://github.com/prashantskit",
"followers_url": "https://api.github.com/users/prashantskit/followers",
"following_url": "https://api.github.com/users/prashantskit/following{/other_user}",
"gists_url": "https://api.github.com/users/prashantskit/gists{/gist_id}",
"starred_url": "https://api.github.com/users/prashantskit/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/prashantskit/subscriptions",
"organizations_url": "https://api.github.com/users/prashantskit/orgs",
"repos_url": "https://api.github.com/users/prashantskit/repos",
"events_url": "https://api.github.com/users/prashantskit/events{/privacy}",
"received_events_url": "https://api.github.com/users/prashantskit/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 | 2022-07-21T08:24:30 | 2022-07-26T17:53:41 | 2022-07-25T21:33:32 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | **Describe the bug**
I am facing potentially the following two issue:
First
I am training a non-english language TTS model using Tacotron2. During inference I try to pre-process the input text using the parse() function but I get the error `AttributeError: 'Tacotron2Model' object has no attribute 'parser'`.
Second
On looking deeper into parser and it's initialisation. The property parser function return the parser created in dataset loader class _("AudioToCharDataset")_ but the parser object created inside the class is not instance variable. It is local to the class and not available with dataset object.
Moreover, in the `nemo.collections.tts.models.Tacotron2Model` there is not placeholder for defining the name of the parser. Ideally it should come from config file.
**Steps/Code to reproduce bug**
Train the TTS on any non-english language dataset.
During inference use _parse_ function to pre-process the text to get tokens.
**Expected behavior**
An error saying `AttributeError: 'Tacotron2Model' object has no attribute 'parser'`
**Environment overview (please complete the following information)**
- Environment location: [Bare-metal, Docker, Cloud(specify cloud provider - AWS, Azure, GCP, Collab)]
- Method of NeMo install: pip install nemo-toolkit\['all'\]
- commands used `docker pull nvcr.io/nvidia/pytorch:22.05-py3` & `docker run --gpus all -it --rm -v /home/prashant/code/NeMo:/NeMo --shm-size=8g -p 8888:8888 -p 6006:6006 --ulimit memlock=-1 --ulimit stack=67108864 --device=/dev/snd nvcr.io/nvidia/pytorch:22.05-py3`
**Environment details**
Nvidia docker image used mentioned in the README
| {
"login": "redoctopus",
"id": 4268876,
"node_id": "MDQ6VXNlcjQyNjg4NzY=",
"avatar_url": "https://avatars.githubusercontent.com/u/4268876?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/redoctopus",
"html_url": "https://github.com/redoctopus",
"followers_url": "https://api.github.com/users/redoctopus/followers",
"following_url": "https://api.github.com/users/redoctopus/following{/other_user}",
"gists_url": "https://api.github.com/users/redoctopus/gists{/gist_id}",
"starred_url": "https://api.github.com/users/redoctopus/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/redoctopus/subscriptions",
"organizations_url": "https://api.github.com/users/redoctopus/orgs",
"repos_url": "https://api.github.com/users/redoctopus/repos",
"events_url": "https://api.github.com/users/redoctopus/events{/privacy}",
"received_events_url": "https://api.github.com/users/redoctopus/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4574/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/4574/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4573 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4573/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4573/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4573/events | https://github.com/NVIDIA/NeMo/issues/4573 | 1,312,501,446 | I_kwDOC_bI7s5OOy7G | 4,573 | Input features for MarbleNet-VAD | {
"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 | {
"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
} | [
{
"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
}
] | null | null | 2022-07-21T03:54:22 | 2022-09-21T04:38:13 | 2022-09-21T04:38:13 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | I am performing VAD using examples/asr/vad_infer.py . I see that, from the code, MarbleNet directly operates on raw audio data using convolution filters. There is no separate feature extraction from the audio.
But , when I see the published MarbleNet paper from Nvidia, looks like first MFCC is extracted from audio and then model operates on those MFCCs.
Am I correct with this observation ? | {
"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/4573/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/4573/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4572 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4572/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4572/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4572/events | https://github.com/NVIDIA/NeMo/issues/4572 | 1,312,312,499 | I_kwDOC_bI7s5OOEyz | 4,572 | Speaker Identification in Speaker dilation | {
"login": "sujeetboran",
"id": 44828536,
"node_id": "MDQ6VXNlcjQ0ODI4NTM2",
"avatar_url": "https://avatars.githubusercontent.com/u/44828536?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sujeetboran",
"html_url": "https://github.com/sujeetboran",
"followers_url": "https://api.github.com/users/sujeetboran/followers",
"following_url": "https://api.github.com/users/sujeetboran/following{/other_user}",
"gists_url": "https://api.github.com/users/sujeetboran/gists{/gist_id}",
"starred_url": "https://api.github.com/users/sujeetboran/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/sujeetboran/subscriptions",
"organizations_url": "https://api.github.com/users/sujeetboran/orgs",
"repos_url": "https://api.github.com/users/sujeetboran/repos",
"events_url": "https://api.github.com/users/sujeetboran/events{/privacy}",
"received_events_url": "https://api.github.com/users/sujeetboran/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": "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
}
] | null | null | 2022-07-21T01:51:56 | 2022-08-12T17:14:14 | 2022-08-12T17:14:14 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | Hello,
I am using speaker dilation for finding out number of speaker and time when which speaker has spoken. when i am setting up meta data like following where i am providing num_spearker = num_of_speaker then it is working perfectly and showing correct result.
T1 T2 Speaker_1
T3 T4 Speaker_0 #Correct result
meta = {
'audio_filepath': an4_audio,
'offset': 0,
'duration':None,
'label': 'infer',
'text': '-',
'num_speakers': 2,
'rttm_filepath': None,
'uem_filepath' : None
}
But, when i am initializing num_spearker = None, Its just showing number of speaker=1 and it's initializing all the spoken time speaker to speaker_0.
T1 T2 Speaker_0
T3 T4 Speaker_0 #Wrong result
But, I have worked with speaker dilation few days back and it was working fine for both num_spearker = num_of_speaker, num_spearker = None. But, now it's not showing correct result for num_spearker = None.
Did this is happening because of new changes in repository? Or do i need to made some changes in tutorial to make it work??
Tutorial link:- https://colab.research.google.com/github/NVIDIA/NeMo/blob/stable/tutorials/speaker_tasks/Speaker_Diarization_Inference.ipynb#scrollTo=CbKTDFr8SZ9F
| {
"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/4572/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/4572/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4571 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4571/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4571/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4571/events | https://github.com/NVIDIA/NeMo/pull/4571 | 1,312,202,878 | PR_kwDOC_bI7s47y_9W | 4,571 | Fp16 support for Conformer | {
"login": "bmwshop",
"id": 482416,
"node_id": "MDQ6VXNlcjQ4MjQxNg==",
"avatar_url": "https://avatars.githubusercontent.com/u/482416?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/bmwshop",
"html_url": "https://github.com/bmwshop",
"followers_url": "https://api.github.com/users/bmwshop/followers",
"following_url": "https://api.github.com/users/bmwshop/following{/other_user}",
"gists_url": "https://api.github.com/users/bmwshop/gists{/gist_id}",
"starred_url": "https://api.github.com/users/bmwshop/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/bmwshop/subscriptions",
"organizations_url": "https://api.github.com/users/bmwshop/orgs",
"repos_url": "https://api.github.com/users/bmwshop/repos",
"events_url": "https://api.github.com/users/bmwshop/events{/privacy}",
"received_events_url": "https://api.github.com/users/bmwshop/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | null | 2022-07-21T00:35:14 | 2022-07-30T02:18:52 | 2022-07-30T02:18:45 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4571",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4571",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4571.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4571.patch",
"merged_at": "2022-07-30T02:18:45"
} | # What does this PR do ?
enables support for fp16 training
Add a one line overview of what this PR aims to accomplish.
enables support for fp16 training by forcing MHSA to float32 or bfloat16 if available.
**Collection**: [Note which collection this PR will affect]
ASR
# Changelog
- Add specific line by line info of high level changes in this PR.
# Usage
* You can potentially add a usage example below
trainer.precision=16
```python
# Add a code snippet demonstrating how to use this
```
n/a
# 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": "bmwshop",
"id": 482416,
"node_id": "MDQ6VXNlcjQ4MjQxNg==",
"avatar_url": "https://avatars.githubusercontent.com/u/482416?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/bmwshop",
"html_url": "https://github.com/bmwshop",
"followers_url": "https://api.github.com/users/bmwshop/followers",
"following_url": "https://api.github.com/users/bmwshop/following{/other_user}",
"gists_url": "https://api.github.com/users/bmwshop/gists{/gist_id}",
"starred_url": "https://api.github.com/users/bmwshop/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/bmwshop/subscriptions",
"organizations_url": "https://api.github.com/users/bmwshop/orgs",
"repos_url": "https://api.github.com/users/bmwshop/repos",
"events_url": "https://api.github.com/users/bmwshop/events{/privacy}",
"received_events_url": "https://api.github.com/users/bmwshop/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4571/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/4571/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4570 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4570/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4570/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4570/events | https://github.com/NVIDIA/NeMo/pull/4570 | 1,312,060,907 | PR_kwDOC_bI7s47yfUi | 4,570 | [Fix] Relative audio path in speech data explorer | {
"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
} | [] | closed | false | null | [] | null | null | 2022-07-20T22:42:31 | 2022-07-21T16:44:21 | 2022-07-21T16:44:20 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4570",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4570",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4570.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4570.patch",
"merged_at": "2022-07-21T16:44:20"
} | Signed-off-by: Ante Jukić <[email protected]>
# What does this PR do ?
Enables audio file loading for manifests which use audio_filepath relative to the location of the manifest.
**Collection**: Tools
# Changelog
- Added `absolute_audio_filepath`
- Use `absolute_audio_filepath` instead of `item['audio_filepath']`
Note:
Another option may be to include the above path conversion when data is loaded (cf. `data.append(...)`).
However, that would mean each audio file is checked even if it's not loaded.
# 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": "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/4570/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/4570/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4569 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4569/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4569/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4569/events | https://github.com/NVIDIA/NeMo/issues/4569 | 1,310,771,241 | I_kwDOC_bI7s5OIMgp | 4,569 | Quantize conformer model that is trained before | {
"login": "alexandercesarr",
"id": 105767588,
"node_id": "U_kgDOBk3ipA",
"avatar_url": "https://avatars.githubusercontent.com/u/105767588?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/alexandercesarr",
"html_url": "https://github.com/alexandercesarr",
"followers_url": "https://api.github.com/users/alexandercesarr/followers",
"following_url": "https://api.github.com/users/alexandercesarr/following{/other_user}",
"gists_url": "https://api.github.com/users/alexandercesarr/gists{/gist_id}",
"starred_url": "https://api.github.com/users/alexandercesarr/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/alexandercesarr/subscriptions",
"organizations_url": "https://api.github.com/users/alexandercesarr/orgs",
"repos_url": "https://api.github.com/users/alexandercesarr/repos",
"events_url": "https://api.github.com/users/alexandercesarr/events{/privacy}",
"received_events_url": "https://api.github.com/users/alexandercesarr/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 | 2022-07-20T09:53:29 | 2022-07-30T09:38:34 | 2022-07-30T09:38:34 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | Hi, I trained a conformer model before. Now I wanna quantize that model and convert it to TRT. But when I run the **speech_to_text_quant_infer_trt.py** an error occur. The error is:
`TypeError: Error instantiating 'nemo.collections.asr.modules.conformer_encoder.ConformerEncoder' : __init__() got an unexpected keyword argument 'quantize'
`
Could you please help me to solve it? | {
"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/4569/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/4569/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4568 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4568/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4568/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4568/events | https://github.com/NVIDIA/NeMo/pull/4568 | 1,310,458,163 | PR_kwDOC_bI7s47s3z7 | 4,568 | Asymmetric Encoder and Decoder Configuration for Megatron Models | {
"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
} | [] | closed | false | null | [] | null | null | 2022-07-20T05:27:05 | 2022-08-02T02:24:33 | 2022-08-02T02:24:30 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4568",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4568",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4568.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4568.patch",
"merged_at": "2022-08-02T02:24:30"
} | Signed-off-by: MaximumEntropy <[email protected]>
# What does this PR do ?
Adds the ability to configure encoder and decoder asymmetrically for encoder and decoder.
**Collection**: NLP
# Changelog
- Make YAML 1-step hierarchical
- Function in lm_encoder_decoder to maintain backward compatibility.
# Usage
* You can potentially add a usage example below
```python
model.encoder.num_layers=12
model.decoder.num_layers=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": "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/4568/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/4568/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4567 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4567/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4567/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4567/events | https://github.com/NVIDIA/NeMo/pull/4567 | 1,310,387,070 | PR_kwDOC_bI7s47soLe | 4,567 | Update diarization data loader to train meeting data | {
"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
} | [] | closed | false | null | [] | null | null | 2022-07-20T03:52:04 | 2022-08-02T01:38:03 | 2022-07-28T07:33:26 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4567",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4567",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4567.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4567.patch",
"merged_at": "2022-07-28T07:33:26"
} | Signed-off-by: Taejin Park <[email protected]>
# What does this PR do ?
This PR adds a new feature to `audio_to_diar_label.py`.
So far, we have only trained with 2 speaker audio where one of the speakers always exists.
This PR makes dataloader to take 2 speaker subset from more-than-2-speaker audio clips with corresponding groundtruth.
Thus, there are cases where grround-truth-cluster label is "-1" where speakers are not included in either of two targeted speakers.
ex) Two speaker ground-truth label ([0,0,1,1,-1,-1,-1,1,1,1,0]
Also removed unnecessary variables.
**Collection**: [Note which collection this PR will affect]
ASR
# Changelog
- `audio_to_diar_label.py`
-- Removed unused variables in the functions.
-- base_clus_label now can have -1 label.
-- Few cosmetic changes
- `collections.py`
A few changes include:
-- Added get_rttm_speaker_index() function since sess_spk_dict could have more than 2 speakers.
- `speaker_utils.py`
-- Added get_rttm_speaker_index that takes speakers from RTTM files.
# Usage
The usage will be provided when model and module codes are published.
# 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": "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/4567/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/4567/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4566 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4566/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4566/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4566/events | https://github.com/NVIDIA/NeMo/pull/4566 | 1,310,138,132 | PR_kwDOC_bI7s47rwqu | 4,566 | Prompt Learning Inference Improvements | {
"login": "vadam5",
"id": 78445382,
"node_id": "MDQ6VXNlcjc4NDQ1Mzgy",
"avatar_url": "https://avatars.githubusercontent.com/u/78445382?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/vadam5",
"html_url": "https://github.com/vadam5",
"followers_url": "https://api.github.com/users/vadam5/followers",
"following_url": "https://api.github.com/users/vadam5/following{/other_user}",
"gists_url": "https://api.github.com/users/vadam5/gists{/gist_id}",
"starred_url": "https://api.github.com/users/vadam5/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/vadam5/subscriptions",
"organizations_url": "https://api.github.com/users/vadam5/orgs",
"repos_url": "https://api.github.com/users/vadam5/repos",
"events_url": "https://api.github.com/users/vadam5/events{/privacy}",
"received_events_url": "https://api.github.com/users/vadam5/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | null | 2022-07-19T22:45:28 | 2022-09-20T23:55:51 | 2022-08-04T05:34:47 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4566",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4566",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4566.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4566.patch",
"merged_at": "2022-08-04T05:34:47"
} | Signed-off-by: Virginia Adams <[email protected]>
Separates the gpt prompt learning eval script from the gpt model eval script to simplify inference script files and match t5 prompt learning inference workflow.
**Collection**: NLP
# Changelog
- Adds megatron_gpt_prompt_learning_eval.py script
- Adds prompt learning inference config
- Updated megatron_gpt_eval.py script
- Updates gpt inference config
- Updated gpt prompt learning dataset class
- Updated gpt prompt learning model class generate and predict step methods
# Usage
```
python megatron_gpt_prompt_learning_eval.py \
virtual_prompt_model_file=models/squad_p_tuning_126m_tokens100_epochs100_layers4_1e-4.nemo \
gpt_model_file=models/gpt_126m.nemo \
inference.greedy=True \
inference.add_BOS=False \
trainer.devices=1 \
tensor_model_parallel_size=1 \
data_paths=["data/squad_short_test.jsonl"]
python megatron_gpt_prompt_learning_eval.py \
virtual_prompt_model_file=models/squad_prompt_tuning_5b_tokens100_steps35000_2e-4.nemo \
gpt_model_file=models/gpt_5b.nemo \
inference.greedy=True \
inference.add_BOS=False \
trainer.devices=2 \
tensor_model_parallel_size=2 \
data_paths=["data/squad_short_test.jsonl"]
```
# 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
- [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": "vadam5",
"id": 78445382,
"node_id": "MDQ6VXNlcjc4NDQ1Mzgy",
"avatar_url": "https://avatars.githubusercontent.com/u/78445382?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/vadam5",
"html_url": "https://github.com/vadam5",
"followers_url": "https://api.github.com/users/vadam5/followers",
"following_url": "https://api.github.com/users/vadam5/following{/other_user}",
"gists_url": "https://api.github.com/users/vadam5/gists{/gist_id}",
"starred_url": "https://api.github.com/users/vadam5/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/vadam5/subscriptions",
"organizations_url": "https://api.github.com/users/vadam5/orgs",
"repos_url": "https://api.github.com/users/vadam5/repos",
"events_url": "https://api.github.com/users/vadam5/events{/privacy}",
"received_events_url": "https://api.github.com/users/vadam5/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4566/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/4566/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4565 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4565/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4565/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4565/events | https://github.com/NVIDIA/NeMo/pull/4565 | 1,310,073,917 | PR_kwDOC_bI7s47riLo | 4,565 | Make RETRO SP compatible | {
"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
} | [] | closed | false | null | [] | null | null | 2022-07-19T21:41:06 | 2022-07-19T21:52:36 | 2022-07-19T21:52:32 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4565",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4565",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4565.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4565.patch",
"merged_at": "2022-07-19T21:52:32"
} | # What does this PR do ?
Fix all the unit tests and CI tests to make RETRO model works with the current transformer model that supports the SP. | {
"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/4565/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/4565/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4564 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4564/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4564/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4564/events | https://github.com/NVIDIA/NeMo/pull/4564 | 1,310,047,170 | PR_kwDOC_bI7s47rcaj | 4,564 | [Fix] schedulers with no max_steps param | {
"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": 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 | 2022-07-19T21:07:34 | 2022-10-07T15:12:50 | 2022-10-07T15:12:49 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4564",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4564",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4564.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4564.patch",
"merged_at": "2022-10-07T15:12:49"
} | All LR schedulers in PyTorch do not have the `max_steps` parameter, so we should not add `max_steps` to their `scheduler_args`. Previous code tackle the problem in case-by-case manner, while here we try to solve the problem in a more general way.
To do so, we define a set of schedulers that don't have the `max_steps` parameter, and set the `add_max_args_flag` accordingly.
Similarly, we define a set of epoch-based schedulers to set the `interval`, while previous code only checks for `ReduceLROnPlateau` scheduler.
Besides, the previous code also has bugs, since it doesn't set the correct `add_max_args_flag` and `interval` when `if 'args' in scheduler_config` is `True`. | {
"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/4564/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/4564/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4563 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4563/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4563/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4563/events | https://github.com/NVIDIA/NeMo/pull/4563 | 1,308,849,461 | PR_kwDOC_bI7s47nb8k | 4,563 | Zero Shot Slot Filling Model | {
"login": "finalily",
"id": 14979030,
"node_id": "MDQ6VXNlcjE0OTc5MDMw",
"avatar_url": "https://avatars.githubusercontent.com/u/14979030?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/finalily",
"html_url": "https://github.com/finalily",
"followers_url": "https://api.github.com/users/finalily/followers",
"following_url": "https://api.github.com/users/finalily/following{/other_user}",
"gists_url": "https://api.github.com/users/finalily/gists{/gist_id}",
"starred_url": "https://api.github.com/users/finalily/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/finalily/subscriptions",
"organizations_url": "https://api.github.com/users/finalily/orgs",
"repos_url": "https://api.github.com/users/finalily/repos",
"events_url": "https://api.github.com/users/finalily/events{/privacy}",
"received_events_url": "https://api.github.com/users/finalily/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": 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 | 2022-07-19T01:05:46 | 2023-03-05T02:08:26 | 2023-03-05T02:08:26 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4563",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4563",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4563.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4563.patch",
"merged_at": null
} | Signed-off-by: Lily Lee <[email protected]>
# What does this PR do ?
Zero Shot Slot Filling Model
**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**:
- [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": "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/4563/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/4563/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4562 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4562/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4562/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4562/events | https://github.com/NVIDIA/NeMo/issues/4562 | 1,308,032,033 | I_kwDOC_bI7s5N9vwh | 4,562 | Getting Key Error 21 while finetuning Citrinet-1024-8x-Stride | {
"login": "iddqd2d",
"id": 40464324,
"node_id": "MDQ6VXNlcjQwNDY0MzI0",
"avatar_url": "https://avatars.githubusercontent.com/u/40464324?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/iddqd2d",
"html_url": "https://github.com/iddqd2d",
"followers_url": "https://api.github.com/users/iddqd2d/followers",
"following_url": "https://api.github.com/users/iddqd2d/following{/other_user}",
"gists_url": "https://api.github.com/users/iddqd2d/gists{/gist_id}",
"starred_url": "https://api.github.com/users/iddqd2d/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/iddqd2d/subscriptions",
"organizations_url": "https://api.github.com/users/iddqd2d/orgs",
"repos_url": "https://api.github.com/users/iddqd2d/repos",
"events_url": "https://api.github.com/users/iddqd2d/events{/privacy}",
"received_events_url": "https://api.github.com/users/iddqd2d/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 | 2022-07-18T14:17:22 | 2023-09-14T01:55:11 | 2022-07-25T19:22:05 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | Hi, I try this code:
Help, Please...
```
import torch
import pytorch_lightning as pl
import nemo.collections.asr as nemo_asr
from ruamel.yaml import YAML
from omegaconf import DictConfig, OmegaConf
from nemo.utils.exp_manager import exp_manager
model_conf_path = '/home/denis/tim/yaml/citrinet_1024.yaml'
params = OmegaConf.load(model_conf_path)
trainer = pl.Trainer(**params.trainer)
model = nemo_asr.models.EncDecCTCModel(cfg=params.model, trainer=trainer)
exp_manager(trainer=trainer, cfg=params.exp_manager)
trainer.fit(model)
```
My YAML config:
[https://github.com/iddqd2d/files/blob/main/citrinet_1024.yaml](url)
```
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
`Trainer(val_check_interval=1.0)` was configured so validation will run at the end of the training epoch..
[NeMo I 2022-07-18 14:14:16 audio_to_text_dataset:41] Model level config does not contain `labels`, please explicitly provide `labels` to the dataloaders.
[NeMo W 2022-07-18 14:14:16 audio_to_text_dataset:84] dataset does not have explicitly defined labels
[NeMo I 2022-07-18 14:14:16 collections:193] Dataset loaded with 2 files totalling 0.01 hours
[NeMo I 2022-07-18 14:14:16 collections:194] 0 files were filtered totalling 0.00 hours
[NeMo I 2022-07-18 14:14:16 audio_to_text_dataset:41] Model level config does not contain `labels`, please explicitly provide `labels` to the dataloaders.
[NeMo W 2022-07-18 14:14:16 audio_to_text_dataset:84] dataset does not have explicitly defined labels
[NeMo I 2022-07-18 14:14:16 collections:193] Dataset loaded with 2 files totalling 0.01 hours
[NeMo I 2022-07-18 14:14:16 collections:194] 0 files were filtered totalling 0.00 hours
[NeMo I 2022-07-18 14:14:16 audio_to_text_dataset:41] Model level config does not contain `labels`, please explicitly provide `labels` to the dataloaders.
[NeMo W 2022-07-18 14:14:16 audio_to_text_dataset:84] dataset does not have explicitly defined labels
[NeMo I 2022-07-18 14:14:16 collections:193] Dataset loaded with 2 files totalling 0.01 hours
[NeMo I 2022-07-18 14:14:16 collections:194] 0 files were filtered totalling 0.00 hours
[NeMo I 2022-07-18 14:14:16 features:200] PADDING: 16
[NeMo I 2022-07-18 14:14:18 ctc_models:64]
Replacing placeholder number of classes (-1) with actual number of classes - 0
[NeMo I 2022-07-18 14:14:18 conv_asr:427] num_classes of ConvASRDecoder is set to the size of the vocabulary: 0.
[NeMo I 2022-07-18 14:14:18 exp_manager:286] Experiments will be logged at /home/denis/tim/save_tr_nemo/Citrinet-1024-8x-Stride/2022-07-18_14-14-18
[NeMo I 2022-07-18 14:14:18 exp_manager:660] TensorboardLogger has been set up
[NeMo W 2022-07-18 14:14:18 nemo_logging:349] /home/denis/anaconda3/lib/python3.9/site-packages/pytorch_lightning/trainer/trainer.py:2267: LightningDeprecationWarning: `Trainer.weights_save_path` has been deprecated in v1.6 and will be removed in v1.8.
rank_zero_deprecation("`Trainer.weights_save_path` has been deprecated in v1.6 and will be removed in v1.8.")
[NeMo W 2022-07-18 14:14:18 exp_manager:899] The checkpoint callback was told to monitor a validation value and trainer's max_steps was set to -1. Please ensure that max_steps will run for at least 1 epochs to ensure that checkpointing will not error out.
Initializing distributed: GLOBAL_RANK: 0, MEMBER: 1/1
----------------------------------------------------------------------------------------------------
distributed_backend=nccl
All distributed processes registered. Starting with 1 processes
----------------------------------------------------------------------------------------------------
Numba: Attempted to fork from a non-main thread, the TBB library may be in an invalid state in the child process.
Numba: Attempted to fork from a non-main thread, the TBB library may be in an invalid state in the child process.
Numba: Attempted to fork from a non-main thread, the TBB library may be in an invalid state in the child process.
Numba: Attempted to fork from a non-main thread, the TBB library may be in an invalid state in the child process.
Numba: Attempted to fork from a non-main thread, the TBB library may be in an invalid state in the child process.
Numba: Attempted to fork from a non-main thread, the TBB library may be in an invalid state in the child process.
LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]
[NeMo I 2022-07-18 14:14:21 modelPT:587] Optimizer config = Novograd (
Parameter Group 0
amsgrad: False
betas: [0.8, 0.25]
eps: 1e-08
grad_averaging: False
lr: 0.05
weight_decay: 0.001
)
[NeMo I 2022-07-18 14:14:21 lr_scheduler:837] Scheduler "<nemo.core.optim.lr_scheduler.CosineAnnealing object at 0x7fd6d02abd90>"
will be used during training (effective maximum steps = 100) -
Parameters :
(warmup_steps: 5000
warmup_ratio: null
min_lr: 1.0e-05
last_epoch: -1
max_steps: 100
)
| Name | Type | Params
------------------------------------------------------------------------
0 | preprocessor | AudioToMelSpectrogramPreprocessor | 0
1 | encoder | ConvASREncoder | 140 M
2 | decoder | ConvASRDecoder | 1.0 K
3 | loss | CTCLoss | 0
4 | spec_augmentation | SpectrogramAugmentation | 0
5 | _wer | WER | 0
------------------------------------------------------------------------
140 M Trainable params
0 Non-trainable params
140 M Total params
560.699 Total estimated model params size (MB)
Sanity Checking DataLoader 0: 0%| | 0/1 [00:00<?, ?it/s]Traceback (most recent call last):
File "/home/denis/tim/new.py", line 20, in <module>
trainer.fit(model)
File "/home/denis/anaconda3/lib/python3.9/site-packages/pytorch_lightning/trainer/trainer.py", line 699, in fit
self._call_and_handle_interrupt(
File "/home/denis/anaconda3/lib/python3.9/site-packages/pytorch_lightning/trainer/trainer.py", line 651, in _call_and_handle_interrupt
return self.strategy.launcher.launch(trainer_fn, *args, trainer=self, **kwargs)
File "/home/denis/anaconda3/lib/python3.9/site-packages/pytorch_lightning/strategies/launchers/subprocess_script.py", line 93, in launch
return function(*args, **kwargs)
File "/home/denis/anaconda3/lib/python3.9/site-packages/pytorch_lightning/trainer/trainer.py", line 740, in _fit_impl
results = self._run(model, ckpt_path=self.ckpt_path)
File "/home/denis/anaconda3/lib/python3.9/site-packages/pytorch_lightning/trainer/trainer.py", line 1165, in _run
results = self._run_stage()
File "/home/denis/anaconda3/lib/python3.9/site-packages/pytorch_lightning/trainer/trainer.py", line 1251, in _run_stage
return self._run_train()
File "/home/denis/anaconda3/lib/python3.9/site-packages/pytorch_lightning/trainer/trainer.py", line 1273, in _run_train
self._run_sanity_check()
File "/home/denis/anaconda3/lib/python3.9/site-packages/pytorch_lightning/trainer/trainer.py", line 1341, in _run_sanity_check
val_loop.run()
File "/home/denis/anaconda3/lib/python3.9/site-packages/pytorch_lightning/loops/loop.py", line 204, in run
self.advance(*args, **kwargs)
File "/home/denis/anaconda3/lib/python3.9/site-packages/pytorch_lightning/loops/dataloader/evaluation_loop.py", line 155, in advance
dl_outputs = self.epoch_loop.run(self._data_fetcher, dl_max_batches, kwargs)
File "/home/denis/anaconda3/lib/python3.9/site-packages/pytorch_lightning/loops/loop.py", line 204, in run
self.advance(*args, **kwargs)
File "/home/denis/anaconda3/lib/python3.9/site-packages/pytorch_lightning/loops/epoch/evaluation_epoch_loop.py", line 143, in advance
output = self._evaluation_step(**kwargs)
File "/home/denis/anaconda3/lib/python3.9/site-packages/pytorch_lightning/loops/epoch/evaluation_epoch_loop.py", line 239, in _evaluation_step
output = self.trainer._call_strategy_hook(hook_name, *kwargs.values())
File "/home/denis/anaconda3/lib/python3.9/site-packages/pytorch_lightning/trainer/trainer.py", line 1702, in _call_strategy_hook
output = fn(*args, **kwargs)
File "/home/denis/anaconda3/lib/python3.9/site-packages/pytorch_lightning/strategies/ddp.py", line 355, in validation_step
return self.model(*args, **kwargs)
File "/home/denis/anaconda3/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl
return forward_call(*input, **kwargs)
File "/home/denis/anaconda3/lib/python3.9/site-packages/torch/nn/parallel/distributed.py", line 963, in forward
output = self.module(*inputs[0], **kwargs[0])
File "/home/denis/anaconda3/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl
return forward_call(*input, **kwargs)
File "/home/denis/anaconda3/lib/python3.9/site-packages/pytorch_lightning/overrides/base.py", line 90, in forward
return self.module.validation_step(*inputs, **kwargs)
File "/home/denis/anaconda3/lib/python3.9/site-packages/nemo/collections/asr/models/ctc_models.py", line 626, in validation_step
self._wer.update(
File "/home/denis/anaconda3/lib/python3.9/site-packages/torchmetrics/metric.py", line 383, in wrapped_func
update(*args, **kwargs)
File "/home/denis/anaconda3/lib/python3.9/site-packages/nemo/collections/asr/metrics/wer.py", line 753, in update
reference = self.decoding.decode_tokens_to_str(target)
File "/home/denis/anaconda3/lib/python3.9/site-packages/nemo/collections/asr/metrics/wer.py", line 652, in decode_tokens_to_str
hypothesis = ''.join(self.decode_ids_to_tokens(tokens))
File "/home/denis/anaconda3/lib/python3.9/site-packages/nemo/collections/asr/metrics/wer.py", line 666, in decode_ids_to_tokens
token_list = [self.labels_map[c] for c in tokens if c != self.blank_id]
File "/home/denis/anaconda3/lib/python3.9/site-packages/nemo/collections/asr/metrics/wer.py", line 666, in <listcomp>
token_list = [self.labels_map[c] for c in tokens if c != self.blank_id]
KeyError: 21
```
| {
"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/4562/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/4562/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4561 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4561/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4561/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4561/events | https://github.com/NVIDIA/NeMo/issues/4561 | 1,307,574,660 | I_kwDOC_bI7s5N8AGE | 4,561 | Getting Key Error while finetuning Speaker Recognition model | {
"login": "alamnasim",
"id": 26841776,
"node_id": "MDQ6VXNlcjI2ODQxNzc2",
"avatar_url": "https://avatars.githubusercontent.com/u/26841776?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/alamnasim",
"html_url": "https://github.com/alamnasim",
"followers_url": "https://api.github.com/users/alamnasim/followers",
"following_url": "https://api.github.com/users/alamnasim/following{/other_user}",
"gists_url": "https://api.github.com/users/alamnasim/gists{/gist_id}",
"starred_url": "https://api.github.com/users/alamnasim/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/alamnasim/subscriptions",
"organizations_url": "https://api.github.com/users/alamnasim/orgs",
"repos_url": "https://api.github.com/users/alamnasim/repos",
"events_url": "https://api.github.com/users/alamnasim/events{/privacy}",
"received_events_url": "https://api.github.com/users/alamnasim/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 | 2022-07-18T08:20:10 | 2022-10-06T02:13:26 | 2022-10-06T02:13:26 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | Hi @nithinraok ,
Please have a look into it.
I am trying to finetune speaker recognition model on open source dataset.
When I set train_manifest_path and dev_manifest_path both as same either train.json or dev.json. Its working fine for me.
When I set train_manifest and dev_manifest as train.jsom and dev.json (two different file) then I get key error in dev_manifest.
I prepared data using data_preparation script from nemo and doing above experiment on opensource data only.
python speaker_reco_finetune.py
Error:
File "/home/nasim-pc/Nasim/stt_nemo_1.9.0_venv/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 768, in fit
self._call_and_handle_interrupt(
File "/home/nasim-pc/Nasim/stt_nemo_1.9.0_venv/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 719, in _call_and_handle_interrupt
return self.strategy.launcher.launch(trainer_fn, *args, trainer=self, **kwargs)
File "/home/nasim-pc/Nasim/stt_nemo_1.9.0_venv/lib/python3.8/site-packages/pytorch_lightning/strategies/launchers/subprocess_script.py", line 93, in launch
return function(*args, **kwargs)
File "/home/nasim-pc/Nasim/stt_nemo_1.9.0_venv/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 809, in _fit_impl
results = self._run(model, ckpt_path=self.ckpt_path)
File "/home/nasim-pc/Nasim/stt_nemo_1.9.0_venv/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 1234, in _run
results = self._run_stage()
File "/home/nasim-pc/Nasim/stt_nemo_1.9.0_venv/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 1321, in _run_stage
return self._run_train()
File "/home/nasim-pc/Nasim/stt_nemo_1.9.0_venv/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 1343, in _run_train
self._run_sanity_check()
File "/home/nasim-pc/Nasim/stt_nemo_1.9.0_venv/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 1411, in _run_sanity_check
val_loop.run()
File "/home/nasim-pc/Nasim/stt_nemo_1.9.0_venv/lib/python3.8/site-packages/pytorch_lightning/loops/base.py", line 204, in run
self.advance(*args, **kwargs)
File "/home/nasim-pc/Nasim/stt_nemo_1.9.0_venv/lib/python3.8/site-packages/pytorch_lightning/loops/dataloader/evaluation_loop.py", line 154, in advance
dl_outputs = self.epoch_loop.run(self._data_fetcher, dl_max_batches, kwargs)
File "/home/nasim-pc/Nasim/stt_nemo_1.9.0_venv/lib/python3.8/site-packages/pytorch_lightning/loops/base.py", line 204, in run
self.advance(*args, **kwargs)
File "/home/nasim-pc/Nasim/stt_nemo_1.9.0_venv/lib/python3.8/site-packages/pytorch_lightning/loops/epoch/evaluation_epoch_loop.py", line 111, in advance
batch = next(data_fetcher)
File "/home/nasim-pc/Nasim/stt_nemo_1.9.0_venv/lib/python3.8/site-packages/pytorch_lightning/utilities/fetching.py", line 184, in __next__
return self.fetching_function()
File "/home/nasim-pc/Nasim/stt_nemo_1.9.0_venv/lib/python3.8/site-packages/pytorch_lightning/utilities/fetching.py", line 259, in fetching_function
self._fetch_next_batch(self.dataloader_iter)
File "/home/nasim-pc/Nasim/stt_nemo_1.9.0_venv/lib/python3.8/site-packages/pytorch_lightning/utilities/fetching.py", line 273, in _fetch_next_batch
batch = next(iterator)
File "/home/nasim-pc/Nasim/stt_nemo_1.9.0_venv/lib/python3.8/site-packages/torch/utils/data/dataloader.py", line 521, in __next__
data = self._next_data()
File "/home/nasim-pc/Nasim/stt_nemo_1.9.0_venv/lib/python3.8/site-packages/torch/utils/data/dataloader.py", line 561, in _next_data
data = self._dataset_fetcher.fetch(index) # may raise StopIteration
File "/home/nasim-pc/Nasim/stt_nemo_1.9.0_venv/lib/python3.8/site-packages/torch/utils/data/_utils/fetch.py", line 49, in fetch
data = [self.dataset[idx] for idx in possibly_batched_index]
File "/home/nasim-pc/Nasim/stt_nemo_1.9.0_venv/lib/python3.8/site-packages/torch/utils/data/_utils/fetch.py", line 49, in <listcomp>
data = [self.dataset[idx] for idx in possibly_batched_index]
File "/home/nasim-pc/Nasim/stt_nemo_1.9.0_venv/lib/python3.8/site-packages/nemo/collections/asr/data/audio_to_label.py", line 314, in __getitem__
t = torch.tensor(self.label2id[sample.label]).long()
KeyError: 'SV0257'
I have attached error logs and testing_cases that I am assuming.
[https://docs.google.com/spreadsheets/d/1MqSfsFuPtg4LhJgEAl7x9tXfsOrsDIE7/edit?usp=sharing&ouid=102994177770185617606&rtpof=true&sd=true](url)
Training Log:
[https://drive.google.com/file/d/14MdNwZzQI5_RuwfFY8XRMYYtSGwewqz5/view?usp=sharing](url)
I prepared data using python scripts/dataset_processing/get_hi-mia_data.py
My Config file:
[https://drive.google.com/file/d/1ya77Dqqh-wvj0h9ufo9gl7nulHRaaY_4/view?usp=sharing](url)
Installed nemo v1.9.0, torch: 1.10.2+cu102 on my local machine through pip.
OS Version: ubuntu 1.18.04
python: 3.8.0
GPU: Quardro M2000
These two train.json and dev.json should be disjoint but if I am using same file in train and dev manifest then it work else give me key error in dev manifest as I mentioned in error log attachment.
Please help me to resolve this issue.
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/4561/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/4561/timeline | null | not_planned | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4558 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4558/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4558/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4558/events | https://github.com/NVIDIA/NeMo/pull/4558 | 1,306,886,874 | PR_kwDOC_bI7s47g7K7 | 4,558 | Fix seeding in neural machine translation examples | {
"login": "sarthmit",
"id": 19748754,
"node_id": "MDQ6VXNlcjE5NzQ4NzU0",
"avatar_url": "https://avatars.githubusercontent.com/u/19748754?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sarthmit",
"html_url": "https://github.com/sarthmit",
"followers_url": "https://api.github.com/users/sarthmit/followers",
"following_url": "https://api.github.com/users/sarthmit/following{/other_user}",
"gists_url": "https://api.github.com/users/sarthmit/gists{/gist_id}",
"starred_url": "https://api.github.com/users/sarthmit/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/sarthmit/subscriptions",
"organizations_url": "https://api.github.com/users/sarthmit/orgs",
"repos_url": "https://api.github.com/users/sarthmit/repos",
"events_url": "https://api.github.com/users/sarthmit/events{/privacy}",
"received_events_url": "https://api.github.com/users/sarthmit/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 | 2022-07-16T19:36:39 | 2022-10-11T02:13:58 | 2022-10-11T02:13:58 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4558",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4558",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4558.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4558.patch",
"merged_at": null
} | Signed-off-by: sarthmit <[email protected]>
# What does this PR do ?
Fixes the seed for everything (training/initialization/etc.) during training of nlp machine translation systems to aid reproducibility.
**Collection**: examples/nlp/machine_translation
# Changelog
- Using pytorch_lightning's seed_everything function to fix the seed within each training file for machine translation in nlp.
- Update the config files to start with default seed as 0
# Usage
* Usage is as per earlier, where the seed can be changed by just adding a seed=[seed] to override the default 0 in the configs.
**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?
- [x] 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?
@MaximumEntropy @okuchaiev | {
"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/4558/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/4558/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4557 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4557/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4557/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4557/events | https://github.com/NVIDIA/NeMo/pull/4557 | 1,306,639,116 | PR_kwDOC_bI7s47gML_ | 4,557 | Add Tokenization and Normalization pre-proecssing script for NMT | {
"login": "aklife97",
"id": 16078071,
"node_id": "MDQ6VXNlcjE2MDc4MDcx",
"avatar_url": "https://avatars.githubusercontent.com/u/16078071?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/aklife97",
"html_url": "https://github.com/aklife97",
"followers_url": "https://api.github.com/users/aklife97/followers",
"following_url": "https://api.github.com/users/aklife97/following{/other_user}",
"gists_url": "https://api.github.com/users/aklife97/gists{/gist_id}",
"starred_url": "https://api.github.com/users/aklife97/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/aklife97/subscriptions",
"organizations_url": "https://api.github.com/users/aklife97/orgs",
"repos_url": "https://api.github.com/users/aklife97/repos",
"events_url": "https://api.github.com/users/aklife97/events{/privacy}",
"received_events_url": "https://api.github.com/users/aklife97/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | null | 2022-07-16T00:08:14 | 2022-07-19T23:45:45 | 2022-07-16T03:41:36 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4557",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4557",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4557.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4557.patch",
"merged_at": "2022-07-16T03:41:36"
} | Signed-off-by: Abhinav Khattar <[email protected]>
# What does this PR do ?
Adds Tokenization and Normalization script
**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 preprocess_tokenization_normalization.py --input-src train.en \
--input-tgt train.zh \
--output-src train.tok.norm.en \
--output-tgt train.tok.norm.zh \
--source-lang en \
--target-lang zh
# 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/4557/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/4557/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4556 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4556/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4556/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4556/events | https://github.com/NVIDIA/NeMo/pull/4556 | 1,306,614,223 | PR_kwDOC_bI7s47gG44 | 4,556 | prefix tuning for Megatron gpt models | {
"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": 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 | 2022-07-15T23:04:48 | 2023-02-22T18:43:04 | 2022-10-14T02:13:55 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4556",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4556",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4556.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4556.patch",
"merged_at": null
} | # What does this PR do ?
implements prefix tuning for Megatron GPT models.
**Collection**: `examples/nlp/languge_modeling`
# Changelog
- Add specific line by line info of high level changes in this PR.
# Usage
```
/NeMo/examples/nlp/language_modeling$python megatron_gpt_prefix_tuning.py
```
The prefix tuning `.yaml` file is here: `/NeMo/examples/nlp/language_modeling/conf/megatron_gpt_prefix_tuning_config.yaml`, the settings in this config are used by default when `NeMo/examples/nlp/language_modeling/megatron_gpt_prefix_tuning.py` is launched.
Once trained, you can evaluate using:
```
python megatron_gpt_eval.py \
prefix_tuned_model_file=<PATH_.nemo_FILE_GENERATED_AT_END_OF_TRAINING> \
gpt_model_file=<PATH_TO_.nemo_FILE_WITH_PRETRAINED_GPT_MODEL> \
inference.greedy=True \
inference.add_BOS=False \
trainer.devices=1 \
trainer.num_nodes=1 \
tensor_model_parallel_size=1 \
pipeline_model_parallel_size=1 \
prompts=["<PATH_TO_.jsonl_FILE_TO_EVALUATE>"]
```
# 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/4556/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/4556/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4555 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4555/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4555/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4555/events | https://github.com/NVIDIA/NeMo/pull/4555 | 1,306,430,816 | PR_kwDOC_bI7s47feWY | 4,555 | Removed NLPDDPPlugin Import check | {
"login": "vadam5",
"id": 78445382,
"node_id": "MDQ6VXNlcjc4NDQ1Mzgy",
"avatar_url": "https://avatars.githubusercontent.com/u/78445382?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/vadam5",
"html_url": "https://github.com/vadam5",
"followers_url": "https://api.github.com/users/vadam5/followers",
"following_url": "https://api.github.com/users/vadam5/following{/other_user}",
"gists_url": "https://api.github.com/users/vadam5/gists{/gist_id}",
"starred_url": "https://api.github.com/users/vadam5/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/vadam5/subscriptions",
"organizations_url": "https://api.github.com/users/vadam5/orgs",
"repos_url": "https://api.github.com/users/vadam5/repos",
"events_url": "https://api.github.com/users/vadam5/events{/privacy}",
"received_events_url": "https://api.github.com/users/vadam5/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | null | 2022-07-15T19:55:41 | 2022-07-15T22:42:35 | 2022-07-15T22:42:31 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4555",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4555",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4555.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4555.patch",
"merged_at": "2022-07-15T22:42:31"
} | Signed-off-by: Virginia Adams <[email protected]>
Fixes `max_steps` calculation bug by setting the number of data loading workers to equal data parallel size.
**Collection**: NeMo Core
# Changelog
- Updated `ModelPT` class
# Usage
N/A same as before
# 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": "vadam5",
"id": 78445382,
"node_id": "MDQ6VXNlcjc4NDQ1Mzgy",
"avatar_url": "https://avatars.githubusercontent.com/u/78445382?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/vadam5",
"html_url": "https://github.com/vadam5",
"followers_url": "https://api.github.com/users/vadam5/followers",
"following_url": "https://api.github.com/users/vadam5/following{/other_user}",
"gists_url": "https://api.github.com/users/vadam5/gists{/gist_id}",
"starred_url": "https://api.github.com/users/vadam5/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/vadam5/subscriptions",
"organizations_url": "https://api.github.com/users/vadam5/orgs",
"repos_url": "https://api.github.com/users/vadam5/repos",
"events_url": "https://api.github.com/users/vadam5/events{/privacy}",
"received_events_url": "https://api.github.com/users/vadam5/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4555/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/4555/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4554 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4554/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4554/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4554/events | https://github.com/NVIDIA/NeMo/issues/4554 | 1,306,031,669 | I_kwDOC_bI7s5N2HY1 | 4,554 | Typo | {
"login": "grzywada",
"id": 26790024,
"node_id": "MDQ6VXNlcjI2NzkwMDI0",
"avatar_url": "https://avatars.githubusercontent.com/u/26790024?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/grzywada",
"html_url": "https://github.com/grzywada",
"followers_url": "https://api.github.com/users/grzywada/followers",
"following_url": "https://api.github.com/users/grzywada/following{/other_user}",
"gists_url": "https://api.github.com/users/grzywada/gists{/gist_id}",
"starred_url": "https://api.github.com/users/grzywada/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/grzywada/subscriptions",
"organizations_url": "https://api.github.com/users/grzywada/orgs",
"repos_url": "https://api.github.com/users/grzywada/repos",
"events_url": "https://api.github.com/users/grzywada/events{/privacy}",
"received_events_url": "https://api.github.com/users/grzywada/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | {
"login": "vadam5",
"id": 78445382,
"node_id": "MDQ6VXNlcjc4NDQ1Mzgy",
"avatar_url": "https://avatars.githubusercontent.com/u/78445382?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/vadam5",
"html_url": "https://github.com/vadam5",
"followers_url": "https://api.github.com/users/vadam5/followers",
"following_url": "https://api.github.com/users/vadam5/following{/other_user}",
"gists_url": "https://api.github.com/users/vadam5/gists{/gist_id}",
"starred_url": "https://api.github.com/users/vadam5/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/vadam5/subscriptions",
"organizations_url": "https://api.github.com/users/vadam5/orgs",
"repos_url": "https://api.github.com/users/vadam5/repos",
"events_url": "https://api.github.com/users/vadam5/events{/privacy}",
"received_events_url": "https://api.github.com/users/vadam5/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"login": "vadam5",
"id": 78445382,
"node_id": "MDQ6VXNlcjc4NDQ1Mzgy",
"avatar_url": "https://avatars.githubusercontent.com/u/78445382?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/vadam5",
"html_url": "https://github.com/vadam5",
"followers_url": "https://api.github.com/users/vadam5/followers",
"following_url": "https://api.github.com/users/vadam5/following{/other_user}",
"gists_url": "https://api.github.com/users/vadam5/gists{/gist_id}",
"starred_url": "https://api.github.com/users/vadam5/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/vadam5/subscriptions",
"organizations_url": "https://api.github.com/users/vadam5/orgs",
"repos_url": "https://api.github.com/users/vadam5/repos",
"events_url": "https://api.github.com/users/vadam5/events{/privacy}",
"received_events_url": "https://api.github.com/users/vadam5/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
}
] | null | null | 2022-07-15T13:21:23 | 2022-08-09T20:39:14 | 2022-08-09T20:39:13 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | In: https://docs.nvidia.com/deeplearning/nemo/user-guide/docs/en/main/nlp/prompt_learning.html
cartographic -> catastrophic
Instead of: "Because the original model parameters are frozen and never altered by either method, p-tuning/prompt-tuning also avoid cartographic forgetting issues often encountered when fine-tuning models."
It should be: "Because the original model parameters are frozen and never altered by either method, p-tuning/prompt-tuning also avoid catastrophic forgetting issues often encountered when fine-tuning models." | {
"login": "vadam5",
"id": 78445382,
"node_id": "MDQ6VXNlcjc4NDQ1Mzgy",
"avatar_url": "https://avatars.githubusercontent.com/u/78445382?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/vadam5",
"html_url": "https://github.com/vadam5",
"followers_url": "https://api.github.com/users/vadam5/followers",
"following_url": "https://api.github.com/users/vadam5/following{/other_user}",
"gists_url": "https://api.github.com/users/vadam5/gists{/gist_id}",
"starred_url": "https://api.github.com/users/vadam5/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/vadam5/subscriptions",
"organizations_url": "https://api.github.com/users/vadam5/orgs",
"repos_url": "https://api.github.com/users/vadam5/repos",
"events_url": "https://api.github.com/users/vadam5/events{/privacy}",
"received_events_url": "https://api.github.com/users/vadam5/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4554/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/4554/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4553 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4553/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4553/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4553/events | https://github.com/NVIDIA/NeMo/pull/4553 | 1,305,865,800 | PR_kwDOC_bI7s47dkoP | 4,553 | fix tarred dataset len when num shards is not divisible by workers | {
"login": "itzsimpl",
"id": 4784070,
"node_id": "MDQ6VXNlcjQ3ODQwNzA=",
"avatar_url": "https://avatars.githubusercontent.com/u/4784070?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/itzsimpl",
"html_url": "https://github.com/itzsimpl",
"followers_url": "https://api.github.com/users/itzsimpl/followers",
"following_url": "https://api.github.com/users/itzsimpl/following{/other_user}",
"gists_url": "https://api.github.com/users/itzsimpl/gists{/gist_id}",
"starred_url": "https://api.github.com/users/itzsimpl/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/itzsimpl/subscriptions",
"organizations_url": "https://api.github.com/users/itzsimpl/orgs",
"repos_url": "https://api.github.com/users/itzsimpl/repos",
"events_url": "https://api.github.com/users/itzsimpl/events{/privacy}",
"received_events_url": "https://api.github.com/users/itzsimpl/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": "itzsimpl",
"id": 4784070,
"node_id": "MDQ6VXNlcjQ3ODQwNzA=",
"avatar_url": "https://avatars.githubusercontent.com/u/4784070?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/itzsimpl",
"html_url": "https://github.com/itzsimpl",
"followers_url": "https://api.github.com/users/itzsimpl/followers",
"following_url": "https://api.github.com/users/itzsimpl/following{/other_user}",
"gists_url": "https://api.github.com/users/itzsimpl/gists{/gist_id}",
"starred_url": "https://api.github.com/users/itzsimpl/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/itzsimpl/subscriptions",
"organizations_url": "https://api.github.com/users/itzsimpl/orgs",
"repos_url": "https://api.github.com/users/itzsimpl/repos",
"events_url": "https://api.github.com/users/itzsimpl/events{/privacy}",
"received_events_url": "https://api.github.com/users/itzsimpl/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"login": "itzsimpl",
"id": 4784070,
"node_id": "MDQ6VXNlcjQ3ODQwNzA=",
"avatar_url": "https://avatars.githubusercontent.com/u/4784070?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/itzsimpl",
"html_url": "https://github.com/itzsimpl",
"followers_url": "https://api.github.com/users/itzsimpl/followers",
"following_url": "https://api.github.com/users/itzsimpl/following{/other_user}",
"gists_url": "https://api.github.com/users/itzsimpl/gists{/gist_id}",
"starred_url": "https://api.github.com/users/itzsimpl/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/itzsimpl/subscriptions",
"organizations_url": "https://api.github.com/users/itzsimpl/orgs",
"repos_url": "https://api.github.com/users/itzsimpl/repos",
"events_url": "https://api.github.com/users/itzsimpl/events{/privacy}",
"received_events_url": "https://api.github.com/users/itzsimpl/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
}
] | null | null | 2022-07-15T10:27:31 | 2022-07-26T13:51:17 | 2022-07-26T13:51:16 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4553",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4553",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4553.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4553.patch",
"merged_at": "2022-07-26T13:51:16"
} | Signed-off-by: Iztok Lebar Bajec <[email protected]>
# What does this PR do ?
Fixes #4522. When using tarred datasets and the number of shards is not divisible by `world_size`, nemo logs a warning, but training continues. However, in this case the computation of `DataLoader.len()` becomes incorrect. This leads to invalid `lr_scheduler.max_steps`, and other issues down the road. One example being `Validation check` not running at end of epoch.
**Collection**: [nlp]
- language_modelling
- machine_translation
- text_normalization
- token_classification
# Changelog
In case of `shard_strategy='scatter'` the `len()` is computed based on the amount of taken shards and assuming total `num_baches` are equally distributed over all shards, i.e. that each shard contains the same number of batches. In addition, following other implementations this PR adds `shard_strategy` support to NLP/PC tarred datasets.
There are two additional occasions, where tarred datasets are used
https://github.com/NVIDIA/NeMo/blob/fa2e55e607f787d129141523c724b29dab7a411c/nemo/collections/asr/data/audio_to_text.py#L186
which based on the implementation does not seem to be affected by this issue, and
https://github.com/NVIDIA/NeMo/blob/fa2e55e607f787d129141523c724b29dab7a411c/nemo/collections/nlp/data/language_modeling/l2r_lm_dataset.py#L169-L184
which might, but I chose to leave it as is, as the computation of `len()`
https://github.com/NVIDIA/NeMo/blob/fa2e55e607f787d129141523c724b29dab7a411c/nemo/collections/nlp/data/language_modeling/l2r_lm_dataset.py#L239-L240
greatly differs from others
**PR Type**:
- [ ] New Feature
- [x] Bugfix
- [x] 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.
note to @PeganovAnton for NLP/C
# Additional Information
* Related to #4522
| {
"login": "PeganovAnton",
"id": 10737305,
"node_id": "MDQ6VXNlcjEwNzM3MzA1",
"avatar_url": "https://avatars.githubusercontent.com/u/10737305?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/PeganovAnton",
"html_url": "https://github.com/PeganovAnton",
"followers_url": "https://api.github.com/users/PeganovAnton/followers",
"following_url": "https://api.github.com/users/PeganovAnton/following{/other_user}",
"gists_url": "https://api.github.com/users/PeganovAnton/gists{/gist_id}",
"starred_url": "https://api.github.com/users/PeganovAnton/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/PeganovAnton/subscriptions",
"organizations_url": "https://api.github.com/users/PeganovAnton/orgs",
"repos_url": "https://api.github.com/users/PeganovAnton/repos",
"events_url": "https://api.github.com/users/PeganovAnton/events{/privacy}",
"received_events_url": "https://api.github.com/users/PeganovAnton/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4553/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/4553/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4552 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4552/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4552/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4552/events | https://github.com/NVIDIA/NeMo/issues/4552 | 1,305,709,528 | I_kwDOC_bI7s5N04vY | 4,552 | FastSpeech2 training example | {
"login": "aidosRepoint",
"id": 64583282,
"node_id": "MDQ6VXNlcjY0NTgzMjgy",
"avatar_url": "https://avatars.githubusercontent.com/u/64583282?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/aidosRepoint",
"html_url": "https://github.com/aidosRepoint",
"followers_url": "https://api.github.com/users/aidosRepoint/followers",
"following_url": "https://api.github.com/users/aidosRepoint/following{/other_user}",
"gists_url": "https://api.github.com/users/aidosRepoint/gists{/gist_id}",
"starred_url": "https://api.github.com/users/aidosRepoint/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/aidosRepoint/subscriptions",
"organizations_url": "https://api.github.com/users/aidosRepoint/orgs",
"repos_url": "https://api.github.com/users/aidosRepoint/repos",
"events_url": "https://api.github.com/users/aidosRepoint/events{/privacy}",
"received_events_url": "https://api.github.com/users/aidosRepoint/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 | 2022-07-15T07:56:33 | 2022-07-15T20:17:15 | 2022-07-15T20:17:15 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | Hi all!
I couldn't find any examples of FastSpeech2 training. The only thing that I've found was the data preprocessing [tutorial](https://docs.nvidia.com/deeplearning/nemo/user-guide/docs/en/stable/tts/datasets.html)
Can anyone tell me if there is a py script such as those for tacotron2 and other models? And, is there an example of a config file? | {
"login": "redoctopus",
"id": 4268876,
"node_id": "MDQ6VXNlcjQyNjg4NzY=",
"avatar_url": "https://avatars.githubusercontent.com/u/4268876?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/redoctopus",
"html_url": "https://github.com/redoctopus",
"followers_url": "https://api.github.com/users/redoctopus/followers",
"following_url": "https://api.github.com/users/redoctopus/following{/other_user}",
"gists_url": "https://api.github.com/users/redoctopus/gists{/gist_id}",
"starred_url": "https://api.github.com/users/redoctopus/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/redoctopus/subscriptions",
"organizations_url": "https://api.github.com/users/redoctopus/orgs",
"repos_url": "https://api.github.com/users/redoctopus/repos",
"events_url": "https://api.github.com/users/redoctopus/events{/privacy}",
"received_events_url": "https://api.github.com/users/redoctopus/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4552/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/4552/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4551 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4551/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4551/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4551/events | https://github.com/NVIDIA/NeMo/pull/4551 | 1,305,442,456 | PR_kwDOC_bI7s47cLZz | 4,551 | Add length ratio filtering script | {
"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
} | [] | closed | false | null | [] | null | null | 2022-07-15T00:39:57 | 2022-07-15T23:39:23 | 2022-07-15T23:39:20 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4551",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4551",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4551.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4551.patch",
"merged_at": "2022-07-15T23:39:20"
} | Signed-off-by: MaximumEntropy <[email protected]>
# What does this PR do ?
Adds a script to filter a parallel corpus based on min, max and length ratios.
**Collection**: NLP
# Changelog
- Adds a script to filter a parallel corpus based on min, max and length ratios.
# Usage
* You can potentially add a usage example below
```python length_ratio_filter.py --input-src train.en \
--input-tgt train.de \
--output-src train_lang_filtered.en \
--output-tgt train_lang_filtered.de \
--removed-src train_removed.en \
--removed-tgt train_removed.de \
--min_length 1 \
--max_length 512 \
--ratio 1.3
```
# 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": "aklife97",
"id": 16078071,
"node_id": "MDQ6VXNlcjE2MDc4MDcx",
"avatar_url": "https://avatars.githubusercontent.com/u/16078071?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/aklife97",
"html_url": "https://github.com/aklife97",
"followers_url": "https://api.github.com/users/aklife97/followers",
"following_url": "https://api.github.com/users/aklife97/following{/other_user}",
"gists_url": "https://api.github.com/users/aklife97/gists{/gist_id}",
"starred_url": "https://api.github.com/users/aklife97/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/aklife97/subscriptions",
"organizations_url": "https://api.github.com/users/aklife97/orgs",
"repos_url": "https://api.github.com/users/aklife97/repos",
"events_url": "https://api.github.com/users/aklife97/events{/privacy}",
"received_events_url": "https://api.github.com/users/aklife97/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4551/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/4551/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4550 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4550/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4550/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4550/events | https://github.com/NVIDIA/NeMo/pull/4550 | 1,305,362,452 | PR_kwDOC_bI7s47b9AM | 4,550 | Adding multispeaker fastpitch and hifigan en model links to available… | {
"login": "subhankar-ghosh",
"id": 23703547,
"node_id": "MDQ6VXNlcjIzNzAzNTQ3",
"avatar_url": "https://avatars.githubusercontent.com/u/23703547?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/subhankar-ghosh",
"html_url": "https://github.com/subhankar-ghosh",
"followers_url": "https://api.github.com/users/subhankar-ghosh/followers",
"following_url": "https://api.github.com/users/subhankar-ghosh/following{/other_user}",
"gists_url": "https://api.github.com/users/subhankar-ghosh/gists{/gist_id}",
"starred_url": "https://api.github.com/users/subhankar-ghosh/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/subhankar-ghosh/subscriptions",
"organizations_url": "https://api.github.com/users/subhankar-ghosh/orgs",
"repos_url": "https://api.github.com/users/subhankar-ghosh/repos",
"events_url": "https://api.github.com/users/subhankar-ghosh/events{/privacy}",
"received_events_url": "https://api.github.com/users/subhankar-ghosh/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | {
"login": "redoctopus",
"id": 4268876,
"node_id": "MDQ6VXNlcjQyNjg4NzY=",
"avatar_url": "https://avatars.githubusercontent.com/u/4268876?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/redoctopus",
"html_url": "https://github.com/redoctopus",
"followers_url": "https://api.github.com/users/redoctopus/followers",
"following_url": "https://api.github.com/users/redoctopus/following{/other_user}",
"gists_url": "https://api.github.com/users/redoctopus/gists{/gist_id}",
"starred_url": "https://api.github.com/users/redoctopus/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/redoctopus/subscriptions",
"organizations_url": "https://api.github.com/users/redoctopus/orgs",
"repos_url": "https://api.github.com/users/redoctopus/repos",
"events_url": "https://api.github.com/users/redoctopus/events{/privacy}",
"received_events_url": "https://api.github.com/users/redoctopus/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"login": "redoctopus",
"id": 4268876,
"node_id": "MDQ6VXNlcjQyNjg4NzY=",
"avatar_url": "https://avatars.githubusercontent.com/u/4268876?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/redoctopus",
"html_url": "https://github.com/redoctopus",
"followers_url": "https://api.github.com/users/redoctopus/followers",
"following_url": "https://api.github.com/users/redoctopus/following{/other_user}",
"gists_url": "https://api.github.com/users/redoctopus/gists{/gist_id}",
"starred_url": "https://api.github.com/users/redoctopus/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/redoctopus/subscriptions",
"organizations_url": "https://api.github.com/users/redoctopus/orgs",
"repos_url": "https://api.github.com/users/redoctopus/repos",
"events_url": "https://api.github.com/users/redoctopus/events{/privacy}",
"received_events_url": "https://api.github.com/users/redoctopus/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
}
] | null | null | 2022-07-14T22:49:58 | 2023-02-22T18:41:25 | 2022-07-15T02:42:37 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4550",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4550",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4550.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4550.patch",
"merged_at": "2022-07-15T02:42:37"
} | … models
Signed-off-by: subhankar-ghosh <[email protected]>
# What does this PR do ?
This PR adds links to fastpitch and hifigan multispeaker en models trained on HiFiTTS to list of available models in Fastpitch and HifiGAN classes respectively.
**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": "subhankar-ghosh",
"id": 23703547,
"node_id": "MDQ6VXNlcjIzNzAzNTQ3",
"avatar_url": "https://avatars.githubusercontent.com/u/23703547?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/subhankar-ghosh",
"html_url": "https://github.com/subhankar-ghosh",
"followers_url": "https://api.github.com/users/subhankar-ghosh/followers",
"following_url": "https://api.github.com/users/subhankar-ghosh/following{/other_user}",
"gists_url": "https://api.github.com/users/subhankar-ghosh/gists{/gist_id}",
"starred_url": "https://api.github.com/users/subhankar-ghosh/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/subhankar-ghosh/subscriptions",
"organizations_url": "https://api.github.com/users/subhankar-ghosh/orgs",
"repos_url": "https://api.github.com/users/subhankar-ghosh/repos",
"events_url": "https://api.github.com/users/subhankar-ghosh/events{/privacy}",
"received_events_url": "https://api.github.com/users/subhankar-ghosh/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4550/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/4550/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4549 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4549/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4549/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4549/events | https://github.com/NVIDIA/NeMo/issues/4549 | 1,305,103,178 | I_kwDOC_bI7s5NyktK | 4,549 | loss equals to nan when I use rir and noise augmentor | {
"login": "ZurabDz",
"id": 34181252,
"node_id": "MDQ6VXNlcjM0MTgxMjUy",
"avatar_url": "https://avatars.githubusercontent.com/u/34181252?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ZurabDz",
"html_url": "https://github.com/ZurabDz",
"followers_url": "https://api.github.com/users/ZurabDz/followers",
"following_url": "https://api.github.com/users/ZurabDz/following{/other_user}",
"gists_url": "https://api.github.com/users/ZurabDz/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ZurabDz/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ZurabDz/subscriptions",
"organizations_url": "https://api.github.com/users/ZurabDz/orgs",
"repos_url": "https://api.github.com/users/ZurabDz/repos",
"events_url": "https://api.github.com/users/ZurabDz/events{/privacy}",
"received_events_url": "https://api.github.com/users/ZurabDz/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 | 2022-07-14T17:40:26 | 2022-07-14T18:25:47 | 2022-07-14T18:25:47 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | this one was closed but bug is not fixed and I have no idea where should I add a check for nan.
https://github.com/NVIDIA/NeMo/issues/2246
Basically this issue, I am trying to train conformer-char with rir | {
"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/4549/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/4549/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4548 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4548/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4548/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4548/events | https://github.com/NVIDIA/NeMo/issues/4548 | 1,304,476,835 | I_kwDOC_bI7s5NwLyj | 4,548 | Facing Error while Speaker Diarization Inference | {
"login": "sujeetboran",
"id": 44828536,
"node_id": "MDQ6VXNlcjQ0ODI4NTM2",
"avatar_url": "https://avatars.githubusercontent.com/u/44828536?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sujeetboran",
"html_url": "https://github.com/sujeetboran",
"followers_url": "https://api.github.com/users/sujeetboran/followers",
"following_url": "https://api.github.com/users/sujeetboran/following{/other_user}",
"gists_url": "https://api.github.com/users/sujeetboran/gists{/gist_id}",
"starred_url": "https://api.github.com/users/sujeetboran/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/sujeetboran/subscriptions",
"organizations_url": "https://api.github.com/users/sujeetboran/orgs",
"repos_url": "https://api.github.com/users/sujeetboran/repos",
"events_url": "https://api.github.com/users/sujeetboran/events{/privacy}",
"received_events_url": "https://api.github.com/users/sujeetboran/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": "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 | 2022-07-14T08:56:30 | 2022-07-21T01:40:35 | 2022-07-21T01:40:35 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | Facing error while running tutorial code for speaker Diarization from NVIDIA tutorial site
after setting up config for ORACLE-VAD DIARIZATION in code when I am trying to compute.
oracle_model.diarize()
it's showing following error:-
RuntimeError: The following operation failed in the TorchScript interpreter.
Traceback of TorchScript (most recent call last):
File "/usr/local/lib/python3.7/dist-packages/nemo/collections/asr/parts/utils/nmesc_clustering.py", line 450, in getCosAffinityMatrix
"""
emb = _emb.half()
sim_d = cos_similarity(emb, emb)
~~~~~~~~~~~~~~ <--- HERE
sim_d = ScalerMinMax(sim_d)
return sim_d
File "/usr/local/lib/python3.7/dist-packages/nemo/collections/asr/parts/utils/nmesc_clustering.py", line 55, in cos_similarity
a_norm = a / (torch.norm(a, dim=1).unsqueeze(1) + eps)
b_norm = b / (torch.norm(a, dim=1).unsqueeze(1) + eps)
res = torch.mm(a_norm, b_norm.transpose(0, 1))
~~~~~~~~ <--- HERE
res.fill_diagonal_(1)
return res
RuntimeError: "addmm_impl_cpu_" not implemented for 'Half'
Few days back when i tried to run this same tutorial it was running successfully and it was giving correct out put after doing diarize(). But from 2-3 dyas i am facing this issue with doing diarize() with model.
Kindly help me with this. | {
"login": "sujeetboran",
"id": 44828536,
"node_id": "MDQ6VXNlcjQ0ODI4NTM2",
"avatar_url": "https://avatars.githubusercontent.com/u/44828536?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sujeetboran",
"html_url": "https://github.com/sujeetboran",
"followers_url": "https://api.github.com/users/sujeetboran/followers",
"following_url": "https://api.github.com/users/sujeetboran/following{/other_user}",
"gists_url": "https://api.github.com/users/sujeetboran/gists{/gist_id}",
"starred_url": "https://api.github.com/users/sujeetboran/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/sujeetboran/subscriptions",
"organizations_url": "https://api.github.com/users/sujeetboran/orgs",
"repos_url": "https://api.github.com/users/sujeetboran/repos",
"events_url": "https://api.github.com/users/sujeetboran/events{/privacy}",
"received_events_url": "https://api.github.com/users/sujeetboran/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4548/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/4548/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4547 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4547/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4547/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4547/events | https://github.com/NVIDIA/NeMo/pull/4547 | 1,304,311,116 | PR_kwDOC_bI7s47Ybp- | 4,547 | remove the variable that is not used in the context. | {
"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 | 2022-07-14T06:19:08 | 2022-07-14T16:40:12 | 2022-07-14T16:40:08 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4547",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4547",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4547.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4547.patch",
"merged_at": "2022-07-14T16:40:08"
} | Signed-off-by: Xuesong Yang <[email protected]>
# 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": "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/4547/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/4547/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4546 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4546/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4546/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4546/events | https://github.com/NVIDIA/NeMo/issues/4546 | 1,304,015,619 | I_kwDOC_bI7s5NubMD | 4,546 | ImportError: Could not compile megatron dataset C++ helper functions in Synthetic tabular data turorial | {
"login": "leoqzm",
"id": 26887149,
"node_id": "MDQ6VXNlcjI2ODg3MTQ5",
"avatar_url": "https://avatars.githubusercontent.com/u/26887149?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/leoqzm",
"html_url": "https://github.com/leoqzm",
"followers_url": "https://api.github.com/users/leoqzm/followers",
"following_url": "https://api.github.com/users/leoqzm/following{/other_user}",
"gists_url": "https://api.github.com/users/leoqzm/gists{/gist_id}",
"starred_url": "https://api.github.com/users/leoqzm/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/leoqzm/subscriptions",
"organizations_url": "https://api.github.com/users/leoqzm/orgs",
"repos_url": "https://api.github.com/users/leoqzm/repos",
"events_url": "https://api.github.com/users/leoqzm/events{/privacy}",
"received_events_url": "https://api.github.com/users/leoqzm/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 | {
"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
} | [
{
"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
}
] | null | null | 2022-07-13T22:38:49 | 2024-12-12T02:01:27 | 2022-10-17T02:13:40 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | **Describe the bug**
In NeMo/tutorials/nlp/Megatron_Synthetic_Tabular_Data_Generation.ipynb notebook, when I wanted to run it in Google Colab, the ImportError occurs. And I am not sure what package should I install to resolve this error.
**Steps/Code to reproduce bug**
It occurs in this step: ( I got the main branch of the NeMo, and follow the steps in notebook)
```
!python megatron_gpt_pretraining.py \
trainer.devices={NUM_GPUS} \
trainer.accelerator=gpu \
trainer.log_every_n_steps=100 \
trainer.val_check_interval=500 \
trainer.accumulate_grad_batches=1 \
trainer.max_steps=10000 \
trainer.precision=16 \
trainer.gradient_clip_val=1.0 \
exp_manager.exp_dir=gpt_creditcard_results \
model.tensor_model_parallel_size={TENSOR_MP_SIZE} \
model.pipeline_model_parallel_size={PIPELINE_MP_SIZE} \
model.optim.name=fused_adam \
model.optim.lr=2e-4 \
model.optim.sched.warmup_steps=2 \
model.optim.sched.constant_steps=2 \
model.optim.sched.min_lr=8e-5 \
model.max_position_embeddings={SEQ_LENGTH} \
model.encoder_seq_length={SEQ_LENGTH} \
model.data.seq_length={SEQ_LENGTH} \
model.tokenizer.type=Tabular \
model.tokenizer.library=tabular \
model.tokenizer.vocab_file={CC_OUTPUT_P} \
model.tokenizer.delimiter=\',\' \
model.data.eod_mask_loss=True \
model.data.splits_string=\'3800,198,2\' \
model.num_layers={NUM_LAYERS} \
model.hidden_size={HIDDEN_SIZE} \
model.num_attention_heads={NUM_ATTENTION_HEADS} \
model.activations_checkpoint_method=\'block\' \
model.activations_checkpoint_num_layers=1 \
model.data.data_prefix=[tabular_data_text_document]
``` | {
"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/4546/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/4546/timeline | null | not_planned | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4545 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4545/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4545/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4545/events | https://github.com/NVIDIA/NeMo/pull/4545 | 1,303,990,591 | PR_kwDOC_bI7s47XaGs | 4,545 | Update megatron prompt learning interface to dialogue | {
"login": "Zhilin123",
"id": 29811458,
"node_id": "MDQ6VXNlcjI5ODExNDU4",
"avatar_url": "https://avatars.githubusercontent.com/u/29811458?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Zhilin123",
"html_url": "https://github.com/Zhilin123",
"followers_url": "https://api.github.com/users/Zhilin123/followers",
"following_url": "https://api.github.com/users/Zhilin123/following{/other_user}",
"gists_url": "https://api.github.com/users/Zhilin123/gists{/gist_id}",
"starred_url": "https://api.github.com/users/Zhilin123/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/Zhilin123/subscriptions",
"organizations_url": "https://api.github.com/users/Zhilin123/orgs",
"repos_url": "https://api.github.com/users/Zhilin123/repos",
"events_url": "https://api.github.com/users/Zhilin123/events{/privacy}",
"received_events_url": "https://api.github.com/users/Zhilin123/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | null | 2022-07-13T22:03:16 | 2023-02-23T01:59:25 | 2022-07-14T01:05:54 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4545",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4545",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4545.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4545.patch",
"merged_at": "2022-07-14T01:05:54"
} | # What does this PR do ?
Update megatron prompt learning interface to dialogue
**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**:
- [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": "Zhilin123",
"id": 29811458,
"node_id": "MDQ6VXNlcjI5ODExNDU4",
"avatar_url": "https://avatars.githubusercontent.com/u/29811458?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Zhilin123",
"html_url": "https://github.com/Zhilin123",
"followers_url": "https://api.github.com/users/Zhilin123/followers",
"following_url": "https://api.github.com/users/Zhilin123/following{/other_user}",
"gists_url": "https://api.github.com/users/Zhilin123/gists{/gist_id}",
"starred_url": "https://api.github.com/users/Zhilin123/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/Zhilin123/subscriptions",
"organizations_url": "https://api.github.com/users/Zhilin123/orgs",
"repos_url": "https://api.github.com/users/Zhilin123/repos",
"events_url": "https://api.github.com/users/Zhilin123/events{/privacy}",
"received_events_url": "https://api.github.com/users/Zhilin123/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4545/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/4545/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4544 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4544/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4544/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4544/events | https://github.com/NVIDIA/NeMo/pull/4544 | 1,303,974,820 | PR_kwDOC_bI7s47XWoP | 4,544 | wait for global_rank_zero() to move old files | {
"login": "itzsimpl",
"id": 4784070,
"node_id": "MDQ6VXNlcjQ3ODQwNzA=",
"avatar_url": "https://avatars.githubusercontent.com/u/4784070?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/itzsimpl",
"html_url": "https://github.com/itzsimpl",
"followers_url": "https://api.github.com/users/itzsimpl/followers",
"following_url": "https://api.github.com/users/itzsimpl/following{/other_user}",
"gists_url": "https://api.github.com/users/itzsimpl/gists{/gist_id}",
"starred_url": "https://api.github.com/users/itzsimpl/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/itzsimpl/subscriptions",
"organizations_url": "https://api.github.com/users/itzsimpl/orgs",
"repos_url": "https://api.github.com/users/itzsimpl/repos",
"events_url": "https://api.github.com/users/itzsimpl/events{/privacy}",
"received_events_url": "https://api.github.com/users/itzsimpl/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | null | 2022-07-13T21:42:46 | 2022-07-24T21:23:57 | 2022-07-24T21:23:57 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4544",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4544",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4544.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4544.patch",
"merged_at": null
} | Signed-off-by: Iztok Lebar Bajec <[email protected]>
# What does this PR do ?
In multi-gpu settings there is no guarantee that `global_rank_zero()` will be the first process to start. Assuming a shared storage this means that when resuming training with `log_global_rank_0_only=False`, this may lead to previous run `nemo_log_globalrank-*_localrank-*.log` files being overwritten and current run files being moved to the `run_*` subfolder.
**Collection**: [nemo/utils]
# Changelog
Ensure all files from previous run are moved before continuing execution of non `global_rank_zero()` processes.
**PR Type**:
- [ ] New Feature
- [x] 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
* Related to # (issue)
| {
"login": "itzsimpl",
"id": 4784070,
"node_id": "MDQ6VXNlcjQ3ODQwNzA=",
"avatar_url": "https://avatars.githubusercontent.com/u/4784070?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/itzsimpl",
"html_url": "https://github.com/itzsimpl",
"followers_url": "https://api.github.com/users/itzsimpl/followers",
"following_url": "https://api.github.com/users/itzsimpl/following{/other_user}",
"gists_url": "https://api.github.com/users/itzsimpl/gists{/gist_id}",
"starred_url": "https://api.github.com/users/itzsimpl/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/itzsimpl/subscriptions",
"organizations_url": "https://api.github.com/users/itzsimpl/orgs",
"repos_url": "https://api.github.com/users/itzsimpl/repos",
"events_url": "https://api.github.com/users/itzsimpl/events{/privacy}",
"received_events_url": "https://api.github.com/users/itzsimpl/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4544/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/4544/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4543 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4543/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4543/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4543/events | https://github.com/NVIDIA/NeMo/issues/4543 | 1,303,241,621 | I_kwDOC_bI7s5NreOV | 4,543 | [text processing] Chinese text normalization PR | {
"login": "dophist",
"id": 4462723,
"node_id": "MDQ6VXNlcjQ0NjI3MjM=",
"avatar_url": "https://avatars.githubusercontent.com/u/4462723?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/dophist",
"html_url": "https://github.com/dophist",
"followers_url": "https://api.github.com/users/dophist/followers",
"following_url": "https://api.github.com/users/dophist/following{/other_user}",
"gists_url": "https://api.github.com/users/dophist/gists{/gist_id}",
"starred_url": "https://api.github.com/users/dophist/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/dophist/subscriptions",
"organizations_url": "https://api.github.com/users/dophist/orgs",
"repos_url": "https://api.github.com/users/dophist/repos",
"events_url": "https://api.github.com/users/dophist/events{/privacy}",
"received_events_url": "https://api.github.com/users/dophist/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1719393562,
"node_id": "MDU6TGFiZWwxNzE5MzkzNTYy",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/feature",
"name": "feature",
"color": "68F3F0",
"default": false,
"description": "request/PR for a new feature"
},
{
"id": 4910680898,
"node_id": "LA_kwDOC_bI7s8AAAABJLMLQg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/TN/ITN",
"name": "TN/ITN",
"color": "0052cc",
"default": false,
"description": ""
}
] | 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 | 2022-07-13T10:33:52 | 2022-12-19T21:46:33 | 2022-08-10T13:48:36 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | This is a preliminary inquiry if NeMo team is open to accept PR for Chinese text normalization, or you guys have internal development plans for it ?
[SpeechColab](https://github.com/SpeechColab) is assigning 1 intern student to work on Chinese text normalization(based on Pynini), we'd like to contribute this work to NeMo, hopefully in the next couple of weeks. Basically what we will have is a set of WFST rewrite rules just like existing [NeMo en TN](https://github.com/NVIDIA/NeMo/tree/main/nemo_text_processing/text_normalization/en), reproducing TN behaviors in [this project](https://github.com/speechio/chinese_text_normalization), which handles many typical TN scenarios for Chinese.
**Known Issue**
The intern student will no longer be maintaining the grammars after the internship, but the resulting PR is guaranteed to establish a foundation, enabling Chinese TN in NeMo.
Would like to hear your thoughts on this. @yzhang123 @ekmb
| {
"login": "yzhang123",
"id": 4204271,
"node_id": "MDQ6VXNlcjQyMDQyNzE=",
"avatar_url": "https://avatars.githubusercontent.com/u/4204271?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/yzhang123",
"html_url": "https://github.com/yzhang123",
"followers_url": "https://api.github.com/users/yzhang123/followers",
"following_url": "https://api.github.com/users/yzhang123/following{/other_user}",
"gists_url": "https://api.github.com/users/yzhang123/gists{/gist_id}",
"starred_url": "https://api.github.com/users/yzhang123/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/yzhang123/subscriptions",
"organizations_url": "https://api.github.com/users/yzhang123/orgs",
"repos_url": "https://api.github.com/users/yzhang123/repos",
"events_url": "https://api.github.com/users/yzhang123/events{/privacy}",
"received_events_url": "https://api.github.com/users/yzhang123/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4543/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/4543/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4542 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4542/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4542/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4542/events | https://github.com/NVIDIA/NeMo/issues/4542 | 1,302,702,393 | I_kwDOC_bI7s5Npak5 | 4,542 | Fine-tuning Voice Activity Detection (VAD) models | {
"login": "gradient-ascent-ai-lab",
"id": 60101623,
"node_id": "MDQ6VXNlcjYwMTAxNjIz",
"avatar_url": "https://avatars.githubusercontent.com/u/60101623?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/gradient-ascent-ai-lab",
"html_url": "https://github.com/gradient-ascent-ai-lab",
"followers_url": "https://api.github.com/users/gradient-ascent-ai-lab/followers",
"following_url": "https://api.github.com/users/gradient-ascent-ai-lab/following{/other_user}",
"gists_url": "https://api.github.com/users/gradient-ascent-ai-lab/gists{/gist_id}",
"starred_url": "https://api.github.com/users/gradient-ascent-ai-lab/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/gradient-ascent-ai-lab/subscriptions",
"organizations_url": "https://api.github.com/users/gradient-ascent-ai-lab/orgs",
"repos_url": "https://api.github.com/users/gradient-ascent-ai-lab/repos",
"events_url": "https://api.github.com/users/gradient-ascent-ai-lab/events{/privacy}",
"received_events_url": "https://api.github.com/users/gradient-ascent-ai-lab/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1719393562,
"node_id": "MDU6TGFiZWwxNzE5MzkzNTYy",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/feature",
"name": "feature",
"color": "68F3F0",
"default": false,
"description": "request/PR for a new feature"
}
] | closed | false | {
"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
} | [
{
"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
}
] | null | null | 2022-07-12T22:56:37 | 2022-07-15T16:53:14 | 2022-07-15T16:53:14 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | **Is your feature request related to a problem? Please describe.**
After running diarization baselines, I wanted to fine-tune the component models (VAD + embedding) on my own labeled diarization dataset. For the embedding model this has gone well, and for VAD I have been able to set up training using Google Speech Recognition V2 dataset, as in the examples.
Now that I try to create/adapt the Nemo manifests for VAD training on my dataset, I realise that the manifest format does not suffice for this:
`{"audio_filepath": "<audio_path>", "duration": <full_file_duration>, "label": "speech", "text": "_", "offset": <start_of_speech_segment}`
If I understand correctly, this only allows for a single speech segment per audio file. This works for the Google dataset where each file is short and has a single utterance, but my (and I believe most) diarization dataset with RTTM labels defines many segments of speech per file.
A workaround is technically possible where each file is cut into smaller segments, each with a single utterance, but this seems excessive. The requirement is also surprising, as other tools (e.g. PyAnnote) allow for training VAD models using the RTTM/diarization annotation format.
I'd love to know if I understand the situation correctly, or whether perhaps I've missed something.
Otherwise, thanks for a great framework! Especially the neural types is a feature that would enhance many other DL repositories.
**Describe the solution you'd like**
The ability to fine-tune VAD models using files with more than one utterance.
**Describe alternatives you've considered**
See workaround above
**Additional context**
- | {
"login": "gradient-ascent-ai-lab",
"id": 60101623,
"node_id": "MDQ6VXNlcjYwMTAxNjIz",
"avatar_url": "https://avatars.githubusercontent.com/u/60101623?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/gradient-ascent-ai-lab",
"html_url": "https://github.com/gradient-ascent-ai-lab",
"followers_url": "https://api.github.com/users/gradient-ascent-ai-lab/followers",
"following_url": "https://api.github.com/users/gradient-ascent-ai-lab/following{/other_user}",
"gists_url": "https://api.github.com/users/gradient-ascent-ai-lab/gists{/gist_id}",
"starred_url": "https://api.github.com/users/gradient-ascent-ai-lab/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/gradient-ascent-ai-lab/subscriptions",
"organizations_url": "https://api.github.com/users/gradient-ascent-ai-lab/orgs",
"repos_url": "https://api.github.com/users/gradient-ascent-ai-lab/repos",
"events_url": "https://api.github.com/users/gradient-ascent-ai-lab/events{/privacy}",
"received_events_url": "https://api.github.com/users/gradient-ascent-ai-lab/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4542/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/4542/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4541 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4541/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4541/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4541/events | https://github.com/NVIDIA/NeMo/issues/4541 | 1,302,274,315 | I_kwDOC_bI7s5NnyEL | 4,541 | Tensorboard step on resume training restarts from 0 | {
"login": "itzsimpl",
"id": 4784070,
"node_id": "MDQ6VXNlcjQ3ODQwNzA=",
"avatar_url": "https://avatars.githubusercontent.com/u/4784070?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/itzsimpl",
"html_url": "https://github.com/itzsimpl",
"followers_url": "https://api.github.com/users/itzsimpl/followers",
"following_url": "https://api.github.com/users/itzsimpl/following{/other_user}",
"gists_url": "https://api.github.com/users/itzsimpl/gists{/gist_id}",
"starred_url": "https://api.github.com/users/itzsimpl/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/itzsimpl/subscriptions",
"organizations_url": "https://api.github.com/users/itzsimpl/orgs",
"repos_url": "https://api.github.com/users/itzsimpl/repos",
"events_url": "https://api.github.com/users/itzsimpl/events{/privacy}",
"received_events_url": "https://api.github.com/users/itzsimpl/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 | 2022-07-12T16:05:59 | 2022-07-26T19:49:25 | 2022-07-26T19:49:25 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | **Describe the bug**
I have noticed that in previous nemo versions, e.g. 83444044327d8f0b697bbf0729bd96b7e16e3a56, Tensorboard showed consecutive steps when training was continued from the last checkpoint of an interrupted run; recent versions e.g. 77be76349901d0dd99ad6778509ef5c0663a65c3, as well as 1.10.0 do not; they always start from 0. Is this by intention or bug?
with 83444044327d8f0b697bbf0729bd96b7e16e3a56
<img width="365" alt="image" src="https://user-images.githubusercontent.com/4784070/178539639-a0e1cd2a-daf0-4460-96d1-20524bf4c969.png">
with 77be76349901d0dd99ad6778509ef5c0663a65c3
<img width="365" alt="image" src="https://user-images.githubusercontent.com/4784070/178539957-30a5d0cf-064d-4cbb-b850-e0b9d91aa4a9.png">
| {
"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/4541/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/4541/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4540 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4540/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4540/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4540/events | https://github.com/NVIDIA/NeMo/pull/4540 | 1,301,445,085 | PR_kwDOC_bI7s47O3MQ | 4,540 | Add MTEncDec Finetune support | {
"login": "aklife97",
"id": 16078071,
"node_id": "MDQ6VXNlcjE2MDc4MDcx",
"avatar_url": "https://avatars.githubusercontent.com/u/16078071?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/aklife97",
"html_url": "https://github.com/aklife97",
"followers_url": "https://api.github.com/users/aklife97/followers",
"following_url": "https://api.github.com/users/aklife97/following{/other_user}",
"gists_url": "https://api.github.com/users/aklife97/gists{/gist_id}",
"starred_url": "https://api.github.com/users/aklife97/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/aklife97/subscriptions",
"organizations_url": "https://api.github.com/users/aklife97/orgs",
"repos_url": "https://api.github.com/users/aklife97/repos",
"events_url": "https://api.github.com/users/aklife97/events{/privacy}",
"received_events_url": "https://api.github.com/users/aklife97/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | null | 2022-07-12T02:16:34 | 2022-07-13T19:31:08 | 2022-07-13T19:31:04 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4540",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4540",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4540.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4540.patch",
"merged_at": "2022-07-13T19:31:04"
} | Signed-off-by: Abhinav Khattar <[email protected]>
# What does this PR do ?
Add finetune support for NMT
**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**:
- [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": "aklife97",
"id": 16078071,
"node_id": "MDQ6VXNlcjE2MDc4MDcx",
"avatar_url": "https://avatars.githubusercontent.com/u/16078071?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/aklife97",
"html_url": "https://github.com/aklife97",
"followers_url": "https://api.github.com/users/aklife97/followers",
"following_url": "https://api.github.com/users/aklife97/following{/other_user}",
"gists_url": "https://api.github.com/users/aklife97/gists{/gist_id}",
"starred_url": "https://api.github.com/users/aklife97/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/aklife97/subscriptions",
"organizations_url": "https://api.github.com/users/aklife97/orgs",
"repos_url": "https://api.github.com/users/aklife97/repos",
"events_url": "https://api.github.com/users/aklife97/events{/privacy}",
"received_events_url": "https://api.github.com/users/aklife97/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4540/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/4540/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4539 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4539/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4539/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4539/events | https://github.com/NVIDIA/NeMo/pull/4539 | 1,301,308,196 | PR_kwDOC_bI7s47OaHr | 4,539 | Add nsys profiling | {
"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
} | [] | closed | false | null | [] | null | null | 2022-07-11T22:31:23 | 2022-07-13T20:48:46 | 2022-07-13T20:48:42 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4539",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4539",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4539.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4539.patch",
"merged_at": "2022-07-13T20:48:42"
} | # What does this PR do ?
Enable nsys profiling for any NeMo model from CLI or YAML.
Note this PR was originally: #4531 , authored by @erhoo82
**Collection**: [Core]
# Changelog
- Add setup nsys method to ModelPt
- Add on_train_batch_start/end hooks to enable nsys profiling
# Usage
* Add the following to your NeMo Model config:
```
# Nsys profiling options
nsys_profile:
enabled: False
start_step: 10 # Global batch to start profiling
end_step: 10 # Global batch to end profiling
ranks: [0] # Global rank IDs to profile
gen_shape: False # Generate model and kernel details including input shapes
```
Then wrap your training command with:
```
nsys profile -s none -o <path_to_output> -t cuda,nvtx --force-overwrite true --capture-range=cudaProfilerApi --capture-range-end=stop python ...
```
See https://docs.nvidia.com/nsight-systems/UserGuide/index.html#cli-profiling for more details.
# 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
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/4539/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/4539/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4538 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4538/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4538/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4538/events | https://github.com/NVIDIA/NeMo/pull/4538 | 1,300,640,640 | PR_kwDOC_bI7s47MKG7 | 4,538 | RADTTS ADLR-NEMO porting | {
"login": "MikyasDesta",
"id": 4801137,
"node_id": "MDQ6VXNlcjQ4MDExMzc=",
"avatar_url": "https://avatars.githubusercontent.com/u/4801137?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/MikyasDesta",
"html_url": "https://github.com/MikyasDesta",
"followers_url": "https://api.github.com/users/MikyasDesta/followers",
"following_url": "https://api.github.com/users/MikyasDesta/following{/other_user}",
"gists_url": "https://api.github.com/users/MikyasDesta/gists{/gist_id}",
"starred_url": "https://api.github.com/users/MikyasDesta/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/MikyasDesta/subscriptions",
"organizations_url": "https://api.github.com/users/MikyasDesta/orgs",
"repos_url": "https://api.github.com/users/MikyasDesta/repos",
"events_url": "https://api.github.com/users/MikyasDesta/events{/privacy}",
"received_events_url": "https://api.github.com/users/MikyasDesta/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | null | 2022-07-11T12:27:47 | 2022-08-04T14:16:51 | 2022-08-04T14:16:51 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4538",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4538",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4538.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4538.patch",
"merged_at": "2022-08-04T14:16:51"
} | # What does this PR do ?
This PR adds RADTTS code to NeMo
**Collection**: [Note which collection this PR will affect]
# Changelog
- Add specific line by line info of high level changes in this PR.
# Usage
```python
# Add a code snippet demonstrating how to use this
python examples/tts/radtts_train.py trainer.max_epochs=1000 export_dir="directory to export checkpoints and logs" model.optim.lr=0.0001 model.train_ds.dataloader_params.batch_size=32 model.validation_ds.dataloader_params.batch_size=32 model.modelConfig.decoder_use_partial_padding=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)
- [X] Did you write any new necessary tests?
- [X] Did you add or update any necessary documentation?
- [X] 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**:
- [X] New Feature
- [ ] Bugfix
- [ ] Documentation
This code base uses pytorch implemenation of RAdam which hasn't been tested fully when this is released we are waiting for SWIPAT approval to use https://github.com/LiyuanLucasLiu/RAdam/blob/master/LICENSE. Until that you can integrate this and train. we will switch code to this implementation once we get SWIPAT approval which could take a couple of days. | {
"login": "borisfom",
"id": 14189615,
"node_id": "MDQ6VXNlcjE0MTg5NjE1",
"avatar_url": "https://avatars.githubusercontent.com/u/14189615?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/borisfom",
"html_url": "https://github.com/borisfom",
"followers_url": "https://api.github.com/users/borisfom/followers",
"following_url": "https://api.github.com/users/borisfom/following{/other_user}",
"gists_url": "https://api.github.com/users/borisfom/gists{/gist_id}",
"starred_url": "https://api.github.com/users/borisfom/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/borisfom/subscriptions",
"organizations_url": "https://api.github.com/users/borisfom/orgs",
"repos_url": "https://api.github.com/users/borisfom/repos",
"events_url": "https://api.github.com/users/borisfom/events{/privacy}",
"received_events_url": "https://api.github.com/users/borisfom/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4538/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/4538/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4537 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4537/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4537/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4537/events | https://github.com/NVIDIA/NeMo/issues/4537 | 1,300,133,430 | I_kwDOC_bI7s5NfnY2 | 4,537 | the NMT infer OOM | {
"login": "XiaoqingNLP",
"id": 12473617,
"node_id": "MDQ6VXNlcjEyNDczNjE3",
"avatar_url": "https://avatars.githubusercontent.com/u/12473617?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/XiaoqingNLP",
"html_url": "https://github.com/XiaoqingNLP",
"followers_url": "https://api.github.com/users/XiaoqingNLP/followers",
"following_url": "https://api.github.com/users/XiaoqingNLP/following{/other_user}",
"gists_url": "https://api.github.com/users/XiaoqingNLP/gists{/gist_id}",
"starred_url": "https://api.github.com/users/XiaoqingNLP/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/XiaoqingNLP/subscriptions",
"organizations_url": "https://api.github.com/users/XiaoqingNLP/orgs",
"repos_url": "https://api.github.com/users/XiaoqingNLP/repos",
"events_url": "https://api.github.com/users/XiaoqingNLP/events{/privacy}",
"received_events_url": "https://api.github.com/users/XiaoqingNLP/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1719393562,
"node_id": "MDU6TGFiZWwxNzE5MzkzNTYy",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/feature",
"name": "feature",
"color": "68F3F0",
"default": false,
"description": "request/PR for a new feature"
}
] | 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 | 2022-07-11T02:48:33 | 2022-09-21T11:38:35 | 2022-09-21T11:38:35 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | **Is your feature request related to a problem? Please describe.**
1. Inference does not support automatic batch processing according to length, which leads to the OOM of excessively long sentences and makes it difficult to enlarge the beam.
2. And I found that as the inference goes on, the occupation of CUDA memory is gradually increasing.
A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]
**Describe the solution you'd like**
1. support automatic batch processing according to length
2. I have try to solve it by `torch.cuda.empty_cache()`
| {
"login": "XiaoqingNLP",
"id": 12473617,
"node_id": "MDQ6VXNlcjEyNDczNjE3",
"avatar_url": "https://avatars.githubusercontent.com/u/12473617?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/XiaoqingNLP",
"html_url": "https://github.com/XiaoqingNLP",
"followers_url": "https://api.github.com/users/XiaoqingNLP/followers",
"following_url": "https://api.github.com/users/XiaoqingNLP/following{/other_user}",
"gists_url": "https://api.github.com/users/XiaoqingNLP/gists{/gist_id}",
"starred_url": "https://api.github.com/users/XiaoqingNLP/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/XiaoqingNLP/subscriptions",
"organizations_url": "https://api.github.com/users/XiaoqingNLP/orgs",
"repos_url": "https://api.github.com/users/XiaoqingNLP/repos",
"events_url": "https://api.github.com/users/XiaoqingNLP/events{/privacy}",
"received_events_url": "https://api.github.com/users/XiaoqingNLP/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4537/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/4537/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4536 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4536/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4536/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4536/events | https://github.com/NVIDIA/NeMo/pull/4536 | 1,300,056,293 | PR_kwDOC_bI7s47KM9B | 4,536 | Update megatron interface to dialogue | {
"login": "Zhilin123",
"id": 29811458,
"node_id": "MDQ6VXNlcjI5ODExNDU4",
"avatar_url": "https://avatars.githubusercontent.com/u/29811458?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Zhilin123",
"html_url": "https://github.com/Zhilin123",
"followers_url": "https://api.github.com/users/Zhilin123/followers",
"following_url": "https://api.github.com/users/Zhilin123/following{/other_user}",
"gists_url": "https://api.github.com/users/Zhilin123/gists{/gist_id}",
"starred_url": "https://api.github.com/users/Zhilin123/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/Zhilin123/subscriptions",
"organizations_url": "https://api.github.com/users/Zhilin123/orgs",
"repos_url": "https://api.github.com/users/Zhilin123/repos",
"events_url": "https://api.github.com/users/Zhilin123/events{/privacy}",
"received_events_url": "https://api.github.com/users/Zhilin123/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | null | 2022-07-11T00:19:30 | 2023-02-23T01:59:25 | 2022-07-13T22:16:01 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4536",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4536",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4536.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4536.patch",
"merged_at": null
} | # What does this PR do ?
Update megatron interface to dialogue
**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**:
- [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": "Zhilin123",
"id": 29811458,
"node_id": "MDQ6VXNlcjI5ODExNDU4",
"avatar_url": "https://avatars.githubusercontent.com/u/29811458?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Zhilin123",
"html_url": "https://github.com/Zhilin123",
"followers_url": "https://api.github.com/users/Zhilin123/followers",
"following_url": "https://api.github.com/users/Zhilin123/following{/other_user}",
"gists_url": "https://api.github.com/users/Zhilin123/gists{/gist_id}",
"starred_url": "https://api.github.com/users/Zhilin123/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/Zhilin123/subscriptions",
"organizations_url": "https://api.github.com/users/Zhilin123/orgs",
"repos_url": "https://api.github.com/users/Zhilin123/repos",
"events_url": "https://api.github.com/users/Zhilin123/events{/privacy}",
"received_events_url": "https://api.github.com/users/Zhilin123/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4536/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/4536/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4535 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4535/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4535/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4535/events | https://github.com/NVIDIA/NeMo/issues/4535 | 1,299,870,936 | I_kwDOC_bI7s5NenTY | 4,535 | Speaker Diarization with Marblenet and ClusterDiarizer issue | {
"login": "ddebnath228",
"id": 66294050,
"node_id": "MDQ6VXNlcjY2Mjk0MDUw",
"avatar_url": "https://avatars.githubusercontent.com/u/66294050?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ddebnath228",
"html_url": "https://github.com/ddebnath228",
"followers_url": "https://api.github.com/users/ddebnath228/followers",
"following_url": "https://api.github.com/users/ddebnath228/following{/other_user}",
"gists_url": "https://api.github.com/users/ddebnath228/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ddebnath228/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ddebnath228/subscriptions",
"organizations_url": "https://api.github.com/users/ddebnath228/orgs",
"repos_url": "https://api.github.com/users/ddebnath228/repos",
"events_url": "https://api.github.com/users/ddebnath228/events{/privacy}",
"received_events_url": "https://api.github.com/users/ddebnath228/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | 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
} | [
{
"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 | 2022-07-10T09:16:10 | 2022-09-21T04:38:34 | 2022-09-21T04:38:34 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | **Describe the bug**
I am not getting correct time stamps for speech segment and many speech chunks are removed. I am using pretrained Marblenet and speakerdiarization_speakernet models. It removes lots of speech data in between the chunks as below:
['0.0 0.23 speaker_1', '1.55 1.99 speaker_2', '3.18 4.25 speaker_2', '5.77 11.35 speaker_1',
**Steps/Code to reproduce bug**
from nemo.collections.asr.models import ClusteringDiarizer
sd_model = ClusteringDiarizer(cfg=cfg)
sd_model.diarize()
It generates RTTM file with speaker and timestamp which itself is wrong. Tried with multiple CFG parameter changes but no change in the output.
**Expected behavior**
The chunks should be continous one after another so that no loss of actual speech data.
**Environment overview (please complete the following information)**
- Environment location: Collab
- Method of NeMo install: [pip install or from source]. PIP install from git as per the tutorials and git
- If method of install is [Docker], NO
**Environment details**
If NVIDIA docker image is used you don't need to specify these.
Otherwise, please provide:
- OS version Windows 10
- PyTorch version 1.11
- Python version 3.7
**Additional context**
Add any other context about the problem here.
Example: GPU model
| {
"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/4535/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/4535/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4533 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4533/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4533/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4533/events | https://github.com/NVIDIA/NeMo/issues/4533 | 1,299,765,514 | I_kwDOC_bI7s5NeNkK | 4,533 | Train Speaker Diarization model | {
"login": "AMITKESARI2000",
"id": 66141447,
"node_id": "MDQ6VXNlcjY2MTQxNDQ3",
"avatar_url": "https://avatars.githubusercontent.com/u/66141447?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/AMITKESARI2000",
"html_url": "https://github.com/AMITKESARI2000",
"followers_url": "https://api.github.com/users/AMITKESARI2000/followers",
"following_url": "https://api.github.com/users/AMITKESARI2000/following{/other_user}",
"gists_url": "https://api.github.com/users/AMITKESARI2000/gists{/gist_id}",
"starred_url": "https://api.github.com/users/AMITKESARI2000/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/AMITKESARI2000/subscriptions",
"organizations_url": "https://api.github.com/users/AMITKESARI2000/orgs",
"repos_url": "https://api.github.com/users/AMITKESARI2000/repos",
"events_url": "https://api.github.com/users/AMITKESARI2000/events{/privacy}",
"received_events_url": "https://api.github.com/users/AMITKESARI2000/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | null | 2022-07-09T20:56:13 | 2022-09-21T04:42:27 | 2022-09-21T04:42:27 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | Hello,
Can you guide me how to train and fine tune the Speaker Diarization model taught in https://github.com/NVIDIA/NeMo/blob/main/tutorials/speaker_tasks/Speaker_Diarization_Inference.ipynb. I was unable to find any documentation on this
Thanks | {
"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/4533/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/4533/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4532 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4532/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4532/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4532/events | https://github.com/NVIDIA/NeMo/pull/4532 | 1,299,496,768 | PR_kwDOC_bI7s47IgYa | 4,532 | Integrating support for GPT/T5/BART for Question Answering | {
"login": "ameyasm1154",
"id": 34514696,
"node_id": "MDQ6VXNlcjM0NTE0Njk2",
"avatar_url": "https://avatars.githubusercontent.com/u/34514696?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ameyasm1154",
"html_url": "https://github.com/ameyasm1154",
"followers_url": "https://api.github.com/users/ameyasm1154/followers",
"following_url": "https://api.github.com/users/ameyasm1154/following{/other_user}",
"gists_url": "https://api.github.com/users/ameyasm1154/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ameyasm1154/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ameyasm1154/subscriptions",
"organizations_url": "https://api.github.com/users/ameyasm1154/orgs",
"repos_url": "https://api.github.com/users/ameyasm1154/repos",
"events_url": "https://api.github.com/users/ameyasm1154/events{/privacy}",
"received_events_url": "https://api.github.com/users/ameyasm1154/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | null | 2022-07-08T22:48:11 | 2022-07-25T16:33:53 | 2022-07-25T16:33:31 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4532",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4532",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4532.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4532.patch",
"merged_at": "2022-07-25T16:33:31"
} | Signed-off-by: Ameya Mahabaleshwarkar <[email protected]>
# What does this PR do ?
Integrating support for GPT/T5/BART for Question Answering
**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**:
- [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
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": "ameyasm1154",
"id": 34514696,
"node_id": "MDQ6VXNlcjM0NTE0Njk2",
"avatar_url": "https://avatars.githubusercontent.com/u/34514696?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ameyasm1154",
"html_url": "https://github.com/ameyasm1154",
"followers_url": "https://api.github.com/users/ameyasm1154/followers",
"following_url": "https://api.github.com/users/ameyasm1154/following{/other_user}",
"gists_url": "https://api.github.com/users/ameyasm1154/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ameyasm1154/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ameyasm1154/subscriptions",
"organizations_url": "https://api.github.com/users/ameyasm1154/orgs",
"repos_url": "https://api.github.com/users/ameyasm1154/repos",
"events_url": "https://api.github.com/users/ameyasm1154/events{/privacy}",
"received_events_url": "https://api.github.com/users/ameyasm1154/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4532/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/4532/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4531 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4531/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4531/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4531/events | https://github.com/NVIDIA/NeMo/pull/4531 | 1,299,488,693 | PR_kwDOC_bI7s47Iern | 4,531 | Add NSys Profiling Interface to NeMo Core | {
"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
} | [] | closed | false | null | [] | null | null | 2022-07-08T22:27:53 | 2022-09-20T23:56:00 | 2022-07-11T22:33:04 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4531",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4531",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4531.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4531.patch",
"merged_at": null
} | # What does this PR do ?
The interface to generate nsys profile results.
**Collection**: [Core]
# Changelog
- Enable profile
- Specify range to profile by batch step
- Specify rank to profile
- Specify profiling details
# Usage
Enable knobs to profile
```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/4531/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/4531/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4530 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4530/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4530/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4530/events | https://github.com/NVIDIA/NeMo/pull/4530 | 1,299,424,748 | PR_kwDOC_bI7s47IRFR | 4,530 | Weighted bucketing | {
"login": "tbartley94",
"id": 90423858,
"node_id": "MDQ6VXNlcjkwNDIzODU4",
"avatar_url": "https://avatars.githubusercontent.com/u/90423858?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/tbartley94",
"html_url": "https://github.com/tbartley94",
"followers_url": "https://api.github.com/users/tbartley94/followers",
"following_url": "https://api.github.com/users/tbartley94/following{/other_user}",
"gists_url": "https://api.github.com/users/tbartley94/gists{/gist_id}",
"starred_url": "https://api.github.com/users/tbartley94/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/tbartley94/subscriptions",
"organizations_url": "https://api.github.com/users/tbartley94/orgs",
"repos_url": "https://api.github.com/users/tbartley94/repos",
"events_url": "https://api.github.com/users/tbartley94/events{/privacy}",
"received_events_url": "https://api.github.com/users/tbartley94/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | null | 2022-07-08T20:36:37 | 2022-07-11T17:16:08 | 2022-07-11T17:16:08 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4530",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4530",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4530.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4530.patch",
"merged_at": "2022-07-11T17:16:08"
} | # What does this PR do ?
- Adds weighted bucketing to speaker classification.
**Collection**: [Note which collection this PR will affect]
- ASR
- Copied bucketing logic as used in https://github.com/NVIDIA/NeMo/pull/4474 to `audio_to_label_dataset` for Speaker classification
# Before your PR is "Ready for review"
**Pre checks**:
- [Y ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md)
- [ N] Did you write any new necessary tests?
- [ N] Did you add or update any necessary documentation?
- [ N] 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**:
- [Y ] 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": "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/4530/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/4530/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4529 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4529/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4529/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4529/events | https://github.com/NVIDIA/NeMo/pull/4529 | 1,299,388,087 | PR_kwDOC_bI7s47IJOX | 4,529 | Sequence parallel rebase with bugfixes | {
"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
} | [] | closed | false | null | [] | null | null | 2022-07-08T19:44:02 | 2022-07-08T20:07:03 | 2022-07-08T20:07:00 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4529",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4529",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4529.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4529.patch",
"merged_at": "2022-07-08T20:07:00"
} | # 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/4529/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/4529/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4528 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4528/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4528/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4528/events | https://github.com/NVIDIA/NeMo/issues/4528 | 1,298,867,698 | I_kwDOC_bI7s5NayXy | 4,528 | Megatron: NameError: name 'ensure_divisibility' is not defined | {
"login": "MatejUlcar",
"id": 26550612,
"node_id": "MDQ6VXNlcjI2NTUwNjEy",
"avatar_url": "https://avatars.githubusercontent.com/u/26550612?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/MatejUlcar",
"html_url": "https://github.com/MatejUlcar",
"followers_url": "https://api.github.com/users/MatejUlcar/followers",
"following_url": "https://api.github.com/users/MatejUlcar/following{/other_user}",
"gists_url": "https://api.github.com/users/MatejUlcar/gists{/gist_id}",
"starred_url": "https://api.github.com/users/MatejUlcar/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/MatejUlcar/subscriptions",
"organizations_url": "https://api.github.com/users/MatejUlcar/orgs",
"repos_url": "https://api.github.com/users/MatejUlcar/repos",
"events_url": "https://api.github.com/users/MatejUlcar/events{/privacy}",
"received_events_url": "https://api.github.com/users/MatejUlcar/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 | 2022-07-08T11:04:38 | 2022-10-07T02:13:38 | 2022-10-07T02:13:38 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | **Describe the bug**
Trying to run `megatron_gpt_pretraining.py`. During the init, the NameError occurs, Traceback below:
```
Traceback (most recent call last):
File "/NeMo/examples/nlp/language_modeling/megatron_gpt_pretraining.py", line 92, in <module>
main()
File "/NeMo/nemo/core/config/hydra_runner.py", line 104, in wrapper
_run_hydra(
File "/opt/conda/lib/python3.8/site-packages/hydra/_internal/utils.py", line 377, in _run_hydra
run_and_report(
File "/opt/conda/lib/python3.8/site-packages/hydra/_internal/utils.py", line 214, in run_and_report
raise ex
File "/opt/conda/lib/python3.8/site-packages/hydra/_internal/utils.py", line 211, in run_and_report
return func()
File "/opt/conda/lib/python3.8/site-packages/hydra/_internal/utils.py", line 378, in <lambda>
lambda: hydra.run(
File "/opt/conda/lib/python3.8/site-packages/hydra/_internal/hydra.py", line 111, in run
_ = ret.return_value
File "/opt/conda/lib/python3.8/site-packages/hydra/core/utils.py", line 233, in return_value
raise self._return_value
File "/opt/conda/lib/python3.8/site-packages/hydra/core/utils.py", line 160, in run_job
ret.return_value = task_function(task_cfg)
File "/NeMo/examples/nlp/language_modeling/megatron_gpt_pretraining.py", line 86, in main
model = MegatronGPTModel(cfg.model, trainer)
File "/NeMo/nemo/collections/nlp/models/language_modeling/megatron_gpt_model.py", line 79, in __init__
super().__init__(cfg, trainer=trainer, no_lm_init=True)
File "/NeMo/nemo/collections/nlp/models/language_modeling/megatron_base_model.py", line 73, in __init__
initialize_model_parallel_for_nemo(
File "/NeMo/nemo/collections/nlp/modules/common/megatron/megatron_init.py", line 68, in initialize_model_parallel_for_nemo
) = fake_initialize_model_parallel(
File "/NeMo/nemo/collections/nlp/modules/common/megatron/megatron_init.py", line 181, in fake_initialize_model_parallel
ensure_divisibility(world_size, tensor_model_parallel_size * pipeline_model_parallel_size)
NameError: name 'ensure_divisibility' is not defined
```
**Expected behavior**
Manually importing `ensure_divisibility` from apex.transformer.utils works fine. Running the code should properly import it as well.
**Environment overview (please complete the following information)**
singularity image, based on nvcr.io/nvidia/pytorch:22.06-py3 docker image, NeMo installed with
```
git clone https://github.com/NVIDIA/NeMo
cd NeMo
python -m pip install -r requirements/requirements.txt \
-r requirements/requirements_common.txt \
-r requirements/requirements_lightning.txt \
-r requirements/requirements_nlp.txt \
-r requirements/requirements_test.txt
./reinstall.sh
```
**Update**
Update: no issue when using 22.05 image AND NeMo release v1.10.0. I did not yet check which of the two changes is responsible for the 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/4528/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/4528/timeline | null | not_planned | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4526 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4526/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4526/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4526/events | https://github.com/NVIDIA/NeMo/issues/4526 | 1,298,670,162 | I_kwDOC_bI7s5NaCJS | 4,526 | Entity Linking gives error while training | {
"login": "FatimaArshad-DS",
"id": 78070220,
"node_id": "MDQ6VXNlcjc4MDcwMjIw",
"avatar_url": "https://avatars.githubusercontent.com/u/78070220?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/FatimaArshad-DS",
"html_url": "https://github.com/FatimaArshad-DS",
"followers_url": "https://api.github.com/users/FatimaArshad-DS/followers",
"following_url": "https://api.github.com/users/FatimaArshad-DS/following{/other_user}",
"gists_url": "https://api.github.com/users/FatimaArshad-DS/gists{/gist_id}",
"starred_url": "https://api.github.com/users/FatimaArshad-DS/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/FatimaArshad-DS/subscriptions",
"organizations_url": "https://api.github.com/users/FatimaArshad-DS/orgs",
"repos_url": "https://api.github.com/users/FatimaArshad-DS/repos",
"events_url": "https://api.github.com/users/FatimaArshad-DS/events{/privacy}",
"received_events_url": "https://api.github.com/users/FatimaArshad-DS/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 | {
"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
} | [
{
"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
},
{
"login": "vadam5",
"id": 78445382,
"node_id": "MDQ6VXNlcjc4NDQ1Mzgy",
"avatar_url": "https://avatars.githubusercontent.com/u/78445382?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/vadam5",
"html_url": "https://github.com/vadam5",
"followers_url": "https://api.github.com/users/vadam5/followers",
"following_url": "https://api.github.com/users/vadam5/following{/other_user}",
"gists_url": "https://api.github.com/users/vadam5/gists{/gist_id}",
"starred_url": "https://api.github.com/users/vadam5/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/vadam5/subscriptions",
"organizations_url": "https://api.github.com/users/vadam5/orgs",
"repos_url": "https://api.github.com/users/vadam5/repos",
"events_url": "https://api.github.com/users/vadam5/events{/privacy}",
"received_events_url": "https://api.github.com/users/vadam5/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
}
] | null | null | 2022-07-08T08:25:01 | 2022-10-09T02:13:22 | 2022-10-09T02:13:22 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | Hi,
I followed this tutorial completely:
https://colab.research.google.com/github/NVIDIA/NeMo/blob/stable/tutorials/nlp/Entity_Linking_Medical.ipynb#scrollTo=gO3t67PnmtL7
Now, I'm following these steps to train model on entire UMLs data:
!conda install python=3.7.13 --y
!pip install Cython
!python -m pip install 'git+https://github.com/NVIDIA/[email protected]#egg=nemo_toolkit[all]'
!pip install -q faiss-gpu
!pip install -q folium==0.2.1
!pip install -q imgaug==0.2.5
!git clone https://github.com/NVIDIA/apex
!cd apex
!pip install -v --disable-pip-version-check --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" ./apex
!python "/home/ml-002/Fatima/NeMo/examples/nlp/entity_linking/data/umls_dataset_processing.py" --project_dir "./NeMo/examples/nlp/entity_linking"
!python self_alignment_pretraining.py project_dir=.
However, as soon as epoch 0 reaches 1% I get this error:
Epoch 0: 1%| | 16870/2147821 [5:23:29<681:01:39, 1.15s/it, loss=0.093, v_num=
Epoch 0, global step 7000: 'val_loss' reached 0.31622 (best 0.31622), saving model to '/home/ml-002/Fatima/NeMo/examples/nlp/entity_linking/medical_entity_linking_experiments/sap_bert_umls/2022-07-07_20-20-04/checkpoints/sap_bert_umls--val_loss=0.3162-epoch=0.ckpt' as top 3
Epoch 0: 1%| | 17283/2147821 [5:25:50<669:28:28, 1.13s/it, loss=0.0938, v_num[NeMo I 2022-07-08 01:46:18 multi_similarity_loss:91] Encountered zero loss in multisimloss, loss = 0.0. No hard examples found in the batch
Error executing job with overrides: ['project_dir=.']
Traceback (most recent call last):
File "self_alignment_pretraining.py", line 38, in main
trainer.fit(model)
File "/home/ml-002/anaconda3/envs/py39_ft/lib/python3.7/site-packages/pytorch_lightning/trainer/trainer.py", line 771, in fit
self._fit_impl, model, train_dataloaders, val_dataloaders, datamodule, ckpt_path
File "/home/ml-002/anaconda3/envs/py39_ft/lib/python3.7/site-packages/pytorch_lightning/trainer/trainer.py", line 721, in _call_and_handle_interrupt
return self.strategy.launcher.launch(trainer_fn, *args, trainer=self, **kwargs)
File "/home/ml-002/anaconda3/envs/py39_ft/lib/python3.7/site-packages/pytorch_lightning/strategies/launchers/subprocess_script.py", line 93, in launch
return function(*args, **kwargs)
File "/home/ml-002/anaconda3/envs/py39_ft/lib/python3.7/site-packages/pytorch_lightning/trainer/trainer.py", line 811, in _fit_impl
results = self._run(model, ckpt_path=self.ckpt_path)
File "/home/ml-002/anaconda3/envs/py39_ft/lib/python3.7/site-packages/pytorch_lightning/trainer/trainer.py", line 1236, in _run
results = self._run_stage()
File "/home/ml-002/anaconda3/envs/py39_ft/lib/python3.7/site-packages/pytorch_lightning/trainer/trainer.py", line 1323, in _run_stage
return self._run_train()
File "/home/ml-002/anaconda3/envs/py39_ft/lib/python3.7/site-packages/pytorch_lightning/trainer/trainer.py", line 1353, in _run_train
self.fit_loop.run()
File "/home/ml-002/anaconda3/envs/py39_ft/lib/python3.7/site-packages/pytorch_lightning/loops/base.py", line 204, in run
self.advance(*args, **kwargs)
File "/home/ml-002/anaconda3/envs/py39_ft/lib/python3.7/site-packages/pytorch_lightning/loops/fit_loop.py", line 269, in advance
self._outputs = self.epoch_loop.run(self._data_fetcher)
File "/home/ml-002/anaconda3/envs/py39_ft/lib/python3.7/site-packages/pytorch_lightning/loops/base.py", line 204, in run
self.advance(*args, **kwargs)
File "/home/ml-002/anaconda3/envs/py39_ft/lib/python3.7/site-packages/pytorch_lightning/loops/epoch/training_epoch_loop.py", line 208, in advance
batch_output = self.batch_loop.run(batch, batch_idx)
File "/home/ml-002/anaconda3/envs/py39_ft/lib/python3.7/site-packages/pytorch_lightning/loops/base.py", line 204, in run
self.advance(*args, **kwargs)
File "/home/ml-002/anaconda3/envs/py39_ft/lib/python3.7/site-packages/pytorch_lightning/loops/batch/training_batch_loop.py", line 88, in advance
outputs = self.optimizer_loop.run(split_batch, optimizers, batch_idx)
File "/home/ml-002/anaconda3/envs/py39_ft/lib/python3.7/site-packages/pytorch_lightning/loops/base.py", line 204, in run
self.advance(*args, **kwargs)
File "/home/ml-002/anaconda3/envs/py39_ft/lib/python3.7/site-packages/pytorch_lightning/loops/optimization/optimizer_loop.py", line 207, in advance
self.optimizer_idx,
File "/home/ml-002/anaconda3/envs/py39_ft/lib/python3.7/site-packages/pytorch_lightning/loops/optimization/optimizer_loop.py", line 256, in _run_optimization
self._optimizer_step(optimizer, opt_idx, batch_idx, closure)
File "/home/ml-002/anaconda3/envs/py39_ft/lib/python3.7/site-packages/pytorch_lightning/loops/optimization/optimizer_loop.py", line 378, in _optimizer_step
using_lbfgs=is_lbfgs,
File "/home/ml-002/anaconda3/envs/py39_ft/lib/python3.7/site-packages/pytorch_lightning/trainer/trainer.py", line 1595, in _call_lightning_module_hook
output = fn(*args, **kwargs)
File "/home/ml-002/anaconda3/envs/py39_ft/lib/python3.7/site-packages/pytorch_lightning/core/lightning.py", line 1646, in optimizer_step
optimizer.step(closure=optimizer_closure)
File "/home/ml-002/anaconda3/envs/py39_ft/lib/python3.7/site-packages/pytorch_lightning/core/optimizer.py", line 168, in step
step_output = self._strategy.optimizer_step(self._optimizer, self._optimizer_idx, closure, **kwargs)
File "/home/ml-002/anaconda3/envs/py39_ft/lib/python3.7/site-packages/pytorch_lightning/strategies/ddp.py", line 286, in optimizer_step
optimizer_output = super().optimizer_step(optimizer, opt_idx, closure, model, **kwargs)
File "/home/ml-002/anaconda3/envs/py39_ft/lib/python3.7/site-packages/pytorch_lightning/strategies/strategy.py", line 193, in optimizer_step
return self.precision_plugin.optimizer_step(model, optimizer, opt_idx, closure, **kwargs)
File "/home/ml-002/anaconda3/envs/py39_ft/lib/python3.7/site-packages/pytorch_lightning/plugins/precision/native_amp.py", line 85, in optimizer_step
closure_result = closure()
File "/home/ml-002/anaconda3/envs/py39_ft/lib/python3.7/site-packages/pytorch_lightning/loops/optimization/optimizer_loop.py", line 148, in __call__
self._result = self.closure(*args, **kwargs)
File "/home/ml-002/anaconda3/envs/py39_ft/lib/python3.7/site-packages/pytorch_lightning/loops/optimization/optimizer_loop.py", line 134, in closure
step_output = self._step_fn()
File "/home/ml-002/anaconda3/envs/py39_ft/lib/python3.7/site-packages/pytorch_lightning/loops/optimization/optimizer_loop.py", line 437, in _training_step
training_step_output, self.trainer.accumulate_grad_batches
File "/home/ml-002/anaconda3/envs/py39_ft/lib/python3.7/site-packages/pytorch_lightning/loops/optimization/optimizer_loop.py", line 75, in from_training_step_output
"In automatic_optimization, when `training_step` returns a dict, the 'loss' key needs to be present"
pytorch_lightning.utilities.exceptions.MisconfigurationException: In automatic_optimization, when `training_step` returns a dict, the 'loss' key needs to be present
Set the environment variable HYDRA_FULL_ERROR=1 for a complete stack trace.
Epoch 0: 1%| | 17283/2147821 [5:25:53<669:33:18, 1.13s/it, loss=0.0938, v_num
I checked file and training_step does have loss key. What is the reason behind this error? How do I catch this exception and continue with execution? Please help!! I have been stuck on this for 2 weeks.
Im using Ubuntu | {
"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/4526/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/4526/timeline | null | not_planned | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4525 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4525/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4525/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4525/events | https://github.com/NVIDIA/NeMo/pull/4525 | 1,298,513,660 | PR_kwDOC_bI7s47FLRi | 4,525 | Additional sentencepiece args - Byte fallback, split digits, split_on_whitespace | {
"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
} | [] | closed | false | null | [] | null | null | 2022-07-08T05:35:38 | 2022-07-11T22:45:09 | 2022-07-11T22:45:05 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4525",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4525",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4525.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4525.patch",
"merged_at": "2022-07-11T22:45:05"
} | # What does this PR do ?
Adds some sentencepiece args useful for NLP.
**Collection**: NLP
# Changelog
- Adds byte_level fallback, split_digits and split_on_whitespace args for spm.
# Usage
# 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/4525/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/4525/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4523 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4523/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4523/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4523/events | https://github.com/NVIDIA/NeMo/pull/4523 | 1,298,294,109 | PR_kwDOC_bI7s47EbkZ | 4,523 | [tutorials][asr] Fixed WER initialization in ASR_with_Nemo notebook | {
"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
} | [] | closed | false | null | [] | null | null | 2022-07-08T00:01:52 | 2022-07-08T02:47:50 | 2022-07-08T02:47:50 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4523",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4523",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4523.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4523.patch",
"merged_at": "2022-07-08T02:47:50"
} | Signed-off-by: Ante Jukić <[email protected]>
# 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": "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/4523/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/4523/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4522 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4522/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4522/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4522/events | https://github.com/NVIDIA/NeMo/issues/4522 | 1,298,143,175 | I_kwDOC_bI7s5NYBfH | 4,522 | [NLP] punctuation_capitalization resume training fails | {
"login": "itzsimpl",
"id": 4784070,
"node_id": "MDQ6VXNlcjQ3ODQwNzA=",
"avatar_url": "https://avatars.githubusercontent.com/u/4784070?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/itzsimpl",
"html_url": "https://github.com/itzsimpl",
"followers_url": "https://api.github.com/users/itzsimpl/followers",
"following_url": "https://api.github.com/users/itzsimpl/following{/other_user}",
"gists_url": "https://api.github.com/users/itzsimpl/gists{/gist_id}",
"starred_url": "https://api.github.com/users/itzsimpl/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/itzsimpl/subscriptions",
"organizations_url": "https://api.github.com/users/itzsimpl/orgs",
"repos_url": "https://api.github.com/users/itzsimpl/repos",
"events_url": "https://api.github.com/users/itzsimpl/events{/privacy}",
"received_events_url": "https://api.github.com/users/itzsimpl/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": "PeganovAnton",
"id": 10737305,
"node_id": "MDQ6VXNlcjEwNzM3MzA1",
"avatar_url": "https://avatars.githubusercontent.com/u/10737305?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/PeganovAnton",
"html_url": "https://github.com/PeganovAnton",
"followers_url": "https://api.github.com/users/PeganovAnton/followers",
"following_url": "https://api.github.com/users/PeganovAnton/following{/other_user}",
"gists_url": "https://api.github.com/users/PeganovAnton/gists{/gist_id}",
"starred_url": "https://api.github.com/users/PeganovAnton/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/PeganovAnton/subscriptions",
"organizations_url": "https://api.github.com/users/PeganovAnton/orgs",
"repos_url": "https://api.github.com/users/PeganovAnton/repos",
"events_url": "https://api.github.com/users/PeganovAnton/events{/privacy}",
"received_events_url": "https://api.github.com/users/PeganovAnton/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"login": "PeganovAnton",
"id": 10737305,
"node_id": "MDQ6VXNlcjEwNzM3MzA1",
"avatar_url": "https://avatars.githubusercontent.com/u/10737305?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/PeganovAnton",
"html_url": "https://github.com/PeganovAnton",
"followers_url": "https://api.github.com/users/PeganovAnton/followers",
"following_url": "https://api.github.com/users/PeganovAnton/following{/other_user}",
"gists_url": "https://api.github.com/users/PeganovAnton/gists{/gist_id}",
"starred_url": "https://api.github.com/users/PeganovAnton/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/PeganovAnton/subscriptions",
"organizations_url": "https://api.github.com/users/PeganovAnton/orgs",
"repos_url": "https://api.github.com/users/PeganovAnton/repos",
"events_url": "https://api.github.com/users/PeganovAnton/events{/privacy}",
"received_events_url": "https://api.github.com/users/PeganovAnton/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
}
] | null | null | 2022-07-07T21:55:09 | 2022-07-26T13:51:17 | 2022-07-26T13:51:17 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | **Describe the bug**
When training PC from scratch, using the example configs, everything works fine, however if I try to resume an interrupted training from the last checkpoint, I get the following log entry
```
[NeMo W 2022-07-07 21:18:38 nemo_logging:349] /opt/conda/lib/python3.8/site-packages/pytorch_lightning/callbacks/model_checkpoint.py:378: UserWarning: `ModelCheckpoint(monitor='val_loss')` could not find the monitored key in the returned metrics: ['epoch', 'step']. HINT: Did you call `log('val_loss', value)` in the `LightningModule`?
warning_cache.warn(m)
```
and all new checkpoints get a `val_loss=0.0000`
**Environment details**
- pytorch:22.04-py3
- nemo:1.8.2 installed with ./reinstall.sh, without numba install
| {
"login": "PeganovAnton",
"id": 10737305,
"node_id": "MDQ6VXNlcjEwNzM3MzA1",
"avatar_url": "https://avatars.githubusercontent.com/u/10737305?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/PeganovAnton",
"html_url": "https://github.com/PeganovAnton",
"followers_url": "https://api.github.com/users/PeganovAnton/followers",
"following_url": "https://api.github.com/users/PeganovAnton/following{/other_user}",
"gists_url": "https://api.github.com/users/PeganovAnton/gists{/gist_id}",
"starred_url": "https://api.github.com/users/PeganovAnton/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/PeganovAnton/subscriptions",
"organizations_url": "https://api.github.com/users/PeganovAnton/orgs",
"repos_url": "https://api.github.com/users/PeganovAnton/repos",
"events_url": "https://api.github.com/users/PeganovAnton/events{/privacy}",
"received_events_url": "https://api.github.com/users/PeganovAnton/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4522/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/4522/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4521 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4521/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4521/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4521/events | https://github.com/NVIDIA/NeMo/pull/4521 | 1,298,117,886 | PR_kwDOC_bI7s47DzVG | 4,521 | [TTS] [bugfix] German tutorial directory correction | {
"login": "aroraakshit",
"id": 30349184,
"node_id": "MDQ6VXNlcjMwMzQ5MTg0",
"avatar_url": "https://avatars.githubusercontent.com/u/30349184?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/aroraakshit",
"html_url": "https://github.com/aroraakshit",
"followers_url": "https://api.github.com/users/aroraakshit/followers",
"following_url": "https://api.github.com/users/aroraakshit/following{/other_user}",
"gists_url": "https://api.github.com/users/aroraakshit/gists{/gist_id}",
"starred_url": "https://api.github.com/users/aroraakshit/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/aroraakshit/subscriptions",
"organizations_url": "https://api.github.com/users/aroraakshit/orgs",
"repos_url": "https://api.github.com/users/aroraakshit/repos",
"events_url": "https://api.github.com/users/aroraakshit/events{/privacy}",
"received_events_url": "https://api.github.com/users/aroraakshit/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | null | 2022-07-07T21:19:46 | 2022-07-11T23:02:28 | 2022-07-07T23:55:05 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4521",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4521",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4521.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4521.patch",
"merged_at": "2022-07-07T23:55:05"
} | Signed-off-by: Akshit Arora <[email protected]>
# What does this PR do ?
Corrected the directory created to download validation_ds, required for HiFi-GAN fine-tuning.
**Collection**: TTS
# Changelog
From
```
!cd NeMoGermanTTS && mkdir -p model/train_ds && cd model/validation_ds && wget https://raw.githubusercontent.com/nvidia/NeMo/$BRANCH/examples/tts/conf/hifigan/model/validation_ds/val_ds_finetune.yaml
```
to:
```
!cd NeMoGermanTTS && mkdir -p model/validation_ds && cd model/validation_ds && wget https://raw.githubusercontent.com/nvidia/NeMo/$BRANCH/examples/tts/conf/hifigan/model/validation_ds/val_ds_finetune.yaml
```
**PR Type**:
- [ ] New Feature
- [x] Bugfix
- [ ] Documentation | {
"login": "aroraakshit",
"id": 30349184,
"node_id": "MDQ6VXNlcjMwMzQ5MTg0",
"avatar_url": "https://avatars.githubusercontent.com/u/30349184?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/aroraakshit",
"html_url": "https://github.com/aroraakshit",
"followers_url": "https://api.github.com/users/aroraakshit/followers",
"following_url": "https://api.github.com/users/aroraakshit/following{/other_user}",
"gists_url": "https://api.github.com/users/aroraakshit/gists{/gist_id}",
"starred_url": "https://api.github.com/users/aroraakshit/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/aroraakshit/subscriptions",
"organizations_url": "https://api.github.com/users/aroraakshit/orgs",
"repos_url": "https://api.github.com/users/aroraakshit/repos",
"events_url": "https://api.github.com/users/aroraakshit/events{/privacy}",
"received_events_url": "https://api.github.com/users/aroraakshit/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4521/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/4521/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4520 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4520/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4520/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4520/events | https://github.com/NVIDIA/NeMo/pull/4520 | 1,298,009,710 | PR_kwDOC_bI7s47DbL- | 4,520 | [Fix] Relative Audio Path | {
"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 | 2022-07-07T19:49:23 | 2022-07-12T16:15:03 | 2022-07-12T16:15:03 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4520",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4520",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4520.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4520.patch",
"merged_at": "2022-07-12T16:15:02"
} | Fix the too long audio filenames for tarred datasets. If the audio filename is too long, just keep it as it is. | {
"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/4520/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/4520/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4519 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4519/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4519/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4519/events | https://github.com/NVIDIA/NeMo/pull/4519 | 1,297,913,308 | PR_kwDOC_bI7s47DF3B | 4,519 | r1.10.0 untie embeddings weights | {
"login": "michalivne",
"id": 7453913,
"node_id": "MDQ6VXNlcjc0NTM5MTM=",
"avatar_url": "https://avatars.githubusercontent.com/u/7453913?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/michalivne",
"html_url": "https://github.com/michalivne",
"followers_url": "https://api.github.com/users/michalivne/followers",
"following_url": "https://api.github.com/users/michalivne/following{/other_user}",
"gists_url": "https://api.github.com/users/michalivne/gists{/gist_id}",
"starred_url": "https://api.github.com/users/michalivne/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/michalivne/subscriptions",
"organizations_url": "https://api.github.com/users/michalivne/orgs",
"repos_url": "https://api.github.com/users/michalivne/repos",
"events_url": "https://api.github.com/users/michalivne/events{/privacy}",
"received_events_url": "https://api.github.com/users/michalivne/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | null | 2022-07-07T18:41:02 | 2022-08-04T18:09:52 | 2022-07-14T18:35:08 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4519",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4519",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4519.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4519.patch",
"merged_at": "2022-07-14T18:35:08"
} | # What does this PR do ?
This PR adds the config option to prevent weights tying between encoder embeddings, decoder embeddings, and decoder tokens head.
This PR does not support PP, only TP.
## Added configs
```
model:
share_word_embeddings: True # If True share encoder/decoder embeddings
share_decoder_tokens_head_embeddings: True # If True share decoder embeddings and decoder projection to logits
```
**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)
- [x] Did you write any new necessary tests?
- [x] Did you add or update any necessary documentation?
- [x] 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": "michalivne",
"id": 7453913,
"node_id": "MDQ6VXNlcjc0NTM5MTM=",
"avatar_url": "https://avatars.githubusercontent.com/u/7453913?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/michalivne",
"html_url": "https://github.com/michalivne",
"followers_url": "https://api.github.com/users/michalivne/followers",
"following_url": "https://api.github.com/users/michalivne/following{/other_user}",
"gists_url": "https://api.github.com/users/michalivne/gists{/gist_id}",
"starred_url": "https://api.github.com/users/michalivne/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/michalivne/subscriptions",
"organizations_url": "https://api.github.com/users/michalivne/orgs",
"repos_url": "https://api.github.com/users/michalivne/repos",
"events_url": "https://api.github.com/users/michalivne/events{/privacy}",
"received_events_url": "https://api.github.com/users/michalivne/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4519/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/4519/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4518 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4518/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4518/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4518/events | https://github.com/NVIDIA/NeMo/pull/4518 | 1,297,900,500 | PR_kwDOC_bI7s47DDBe | 4,518 | [TTS] bug fix - sample rate was being ignored in vocoder dataset | {
"login": "paarthneekhara",
"id": 7913368,
"node_id": "MDQ6VXNlcjc5MTMzNjg=",
"avatar_url": "https://avatars.githubusercontent.com/u/7913368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/paarthneekhara",
"html_url": "https://github.com/paarthneekhara",
"followers_url": "https://api.github.com/users/paarthneekhara/followers",
"following_url": "https://api.github.com/users/paarthneekhara/following{/other_user}",
"gists_url": "https://api.github.com/users/paarthneekhara/gists{/gist_id}",
"starred_url": "https://api.github.com/users/paarthneekhara/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/paarthneekhara/subscriptions",
"organizations_url": "https://api.github.com/users/paarthneekhara/orgs",
"repos_url": "https://api.github.com/users/paarthneekhara/repos",
"events_url": "https://api.github.com/users/paarthneekhara/events{/privacy}",
"received_events_url": "https://api.github.com/users/paarthneekhara/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | null | 2022-07-07T18:30:38 | 2022-10-13T07:38:32 | 2022-10-13T07:38:32 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4518",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4518",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4518.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4518.patch",
"merged_at": "2022-10-13T07:38:32"
} | Signed-off-by: Paarth Neekhara <[email protected]>
# What does this PR do ?
Fixes a bug where sample rate was being ignored in VocoderDataset when loading just the audio and audio lengths and not the precomputed mel.
**Collection**: tts
# Changelog
- Add specific line by line info of high level changes in this PR.
# Usage
* You can potentially add a usage example below
```python
dataset = VocoderDataset(manifest_filepath=manifest_path, sample_rate=16000)
```
# 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": "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/4518/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/4518/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4517 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4517/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4517/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4517/events | https://github.com/NVIDIA/NeMo/pull/4517 | 1,297,663,005 | PR_kwDOC_bI7s47COmi | 4,517 | fix duplex inference with grammars | {
"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 | 2022-07-07T15:41:18 | 2022-07-12T17:15:04 | 2022-07-12T17:15:00 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4517",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4517",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4517.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4517.patch",
"merged_at": "2022-07-12T17:15:00"
} | Signed-off-by: ekmb <[email protected]>
# 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": "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/4517/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/4517/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4516 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4516/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4516/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4516/events | https://github.com/NVIDIA/NeMo/pull/4516 | 1,297,490,450 | PR_kwDOC_bI7s47Bod5 | 4,516 | Add ITN pt | {
"login": "guidefloripa",
"id": 6698381,
"node_id": "MDQ6VXNlcjY2OTgzODE=",
"avatar_url": "https://avatars.githubusercontent.com/u/6698381?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/guidefloripa",
"html_url": "https://github.com/guidefloripa",
"followers_url": "https://api.github.com/users/guidefloripa/followers",
"following_url": "https://api.github.com/users/guidefloripa/following{/other_user}",
"gists_url": "https://api.github.com/users/guidefloripa/gists{/gist_id}",
"starred_url": "https://api.github.com/users/guidefloripa/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/guidefloripa/subscriptions",
"organizations_url": "https://api.github.com/users/guidefloripa/orgs",
"repos_url": "https://api.github.com/users/guidefloripa/repos",
"events_url": "https://api.github.com/users/guidefloripa/events{/privacy}",
"received_events_url": "https://api.github.com/users/guidefloripa/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | null | 2022-07-07T13:56:09 | 2022-07-08T12:13:21 | 2022-07-07T20:40:09 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4516",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4516",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4516.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4516.patch",
"merged_at": "2022-07-07T20:40:09"
} | # What does this PR do ?
Adds ITN rules for Portuguese.
In addiction, tests for these rules were added.
**Collection**: [Text Processing]
# 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**:
- [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.
| {
"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/4516/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/4516/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4515 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4515/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4515/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4515/events | https://github.com/NVIDIA/NeMo/pull/4515 | 1,297,471,109 | PR_kwDOC_bI7s47BkJX | 4,515 | Add ITN pt | {
"login": "guidefloripa",
"id": 6698381,
"node_id": "MDQ6VXNlcjY2OTgzODE=",
"avatar_url": "https://avatars.githubusercontent.com/u/6698381?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/guidefloripa",
"html_url": "https://github.com/guidefloripa",
"followers_url": "https://api.github.com/users/guidefloripa/followers",
"following_url": "https://api.github.com/users/guidefloripa/following{/other_user}",
"gists_url": "https://api.github.com/users/guidefloripa/gists{/gist_id}",
"starred_url": "https://api.github.com/users/guidefloripa/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/guidefloripa/subscriptions",
"organizations_url": "https://api.github.com/users/guidefloripa/orgs",
"repos_url": "https://api.github.com/users/guidefloripa/repos",
"events_url": "https://api.github.com/users/guidefloripa/events{/privacy}",
"received_events_url": "https://api.github.com/users/guidefloripa/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | null | 2022-07-07T13:44:36 | 2022-07-07T13:51:47 | 2022-07-07T13:51:37 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4515",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4515",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4515.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4515.patch",
"merged_at": null
} | # What does this PR do ?
Adds ITN rules for Portuguese.
In addiction, tests for these rules were added.
**Collection**: [Text Processing]
# 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**:
- [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.
| {
"login": "guidefloripa",
"id": 6698381,
"node_id": "MDQ6VXNlcjY2OTgzODE=",
"avatar_url": "https://avatars.githubusercontent.com/u/6698381?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/guidefloripa",
"html_url": "https://github.com/guidefloripa",
"followers_url": "https://api.github.com/users/guidefloripa/followers",
"following_url": "https://api.github.com/users/guidefloripa/following{/other_user}",
"gists_url": "https://api.github.com/users/guidefloripa/gists{/gist_id}",
"starred_url": "https://api.github.com/users/guidefloripa/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/guidefloripa/subscriptions",
"organizations_url": "https://api.github.com/users/guidefloripa/orgs",
"repos_url": "https://api.github.com/users/guidefloripa/repos",
"events_url": "https://api.github.com/users/guidefloripa/events{/privacy}",
"received_events_url": "https://api.github.com/users/guidefloripa/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4515/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/4515/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4514 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4514/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4514/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4514/events | https://github.com/NVIDIA/NeMo/issues/4514 | 1,296,992,267 | I_kwDOC_bI7s5NTogL | 4,514 | problem of load ASR lstm trained model | {
"login": "sarasesar",
"id": 95332871,
"node_id": "U_kgDOBa6qBw",
"avatar_url": "https://avatars.githubusercontent.com/u/95332871?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sarasesar",
"html_url": "https://github.com/sarasesar",
"followers_url": "https://api.github.com/users/sarasesar/followers",
"following_url": "https://api.github.com/users/sarasesar/following{/other_user}",
"gists_url": "https://api.github.com/users/sarasesar/gists{/gist_id}",
"starred_url": "https://api.github.com/users/sarasesar/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/sarasesar/subscriptions",
"organizations_url": "https://api.github.com/users/sarasesar/orgs",
"repos_url": "https://api.github.com/users/sarasesar/repos",
"events_url": "https://api.github.com/users/sarasesar/events{/privacy}",
"received_events_url": "https://api.github.com/users/sarasesar/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 | 2022-07-07T08:02:41 | 2022-09-29T04:39:15 | 2022-09-29T04:39:15 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | hello. I have train a "LSTM-Transducer-BPE" model some days ago. and to save the model, I have used checkpoint. for this train, I have used the default config.
after that, when i want to load this model by `EncDecRNNTBPEModel` method,
i got an error and i could not use this model:
```
RuntimeError: Error(s) in loading state_dict for EncDecRNNTBPEModel:
Missing key(s) in state_dict: "decoder.prediction.dec_rnn.lstm.weight_hr_l0", "decoder.prediction.dec_rnn.lstm.weight_hr_l1".
size mismatch for decoder.prediction.dec_rnn.lstm.weight_ih_l0: copying a param with shape torch.Size([2560, 640]) from checkpoint, the shape in current model is torch.Size([8192, 640]).
size mismatch for decoder.prediction.dec_rnn.lstm.weight_hh_l0: copying a param with shape torch.Size([2560, 640]) from checkpoint, the shape in current model is torch.Size([8192, 640]).
size mismatch for decoder.prediction.dec_rnn.lstm.bias_ih_l0: copying a param with shape torch.Size([2560]) from checkpoint, the shape in current model is torch.Size([8192]).
size mismatch for decoder.prediction.dec_rnn.lstm.bias_hh_l0: copying a param with shape torch.Size([2560]) from checkpoint, the shape in current model is torch.Size([8192]).
size mismatch for decoder.prediction.dec_rnn.lstm.weight_ih_l1: copying a param with shape torch.Size([2560, 640]) from checkpoint, the shape in current model is torch.Size([8192, 640]).
size mismatch for decoder.prediction.dec_rnn.lstm.weight_hh_l1: copying a param with shape torch.Size([2560, 640]) from checkpoint, the shape in current model is torch.Size([8192, 640]).
size mismatch for decoder.prediction.dec_rnn.lstm.bias_ih_l1: copying a param with shape torch.Size([2560]) from checkpoint, the shape in current model is torch.Size([8192]).
size mismatch for decoder.prediction.dec_rnn.lstm.bias_hh_l1: copying a param with shape torch.Size([2560]) from checkpoint, the shape in current model is torch.Size([8192]).
```
please help me.
thank you. | {
"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/4514/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/4514/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4513 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4513/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4513/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4513/events | https://github.com/NVIDIA/NeMo/issues/4513 | 1,296,974,047 | I_kwDOC_bI7s5NTkDf | 4,513 | OSError: libtorch_hip.so: cannot open shared object file in Speaker_Diarization_Inference | {
"login": "AMITKESARI2000",
"id": 66141447,
"node_id": "MDQ6VXNlcjY2MTQxNDQ3",
"avatar_url": "https://avatars.githubusercontent.com/u/66141447?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/AMITKESARI2000",
"html_url": "https://github.com/AMITKESARI2000",
"followers_url": "https://api.github.com/users/AMITKESARI2000/followers",
"following_url": "https://api.github.com/users/AMITKESARI2000/following{/other_user}",
"gists_url": "https://api.github.com/users/AMITKESARI2000/gists{/gist_id}",
"starred_url": "https://api.github.com/users/AMITKESARI2000/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/AMITKESARI2000/subscriptions",
"organizations_url": "https://api.github.com/users/AMITKESARI2000/orgs",
"repos_url": "https://api.github.com/users/AMITKESARI2000/repos",
"events_url": "https://api.github.com/users/AMITKESARI2000/events{/privacy}",
"received_events_url": "https://api.github.com/users/AMITKESARI2000/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 | 2022-07-07T07:46:06 | 2022-08-05T19:26:39 | 2022-08-05T19:25:03 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null |
Followed the steps here : https://github.com/NVIDIA/NeMo/blob/main/tutorials/speaker_tasks/Speaker_Diarization_Inference.ipynb
to run in jupyter notebook and just ran the whole notebook as it is. But got the error `OSError: libtorch_hip.so: cannot open shared object file: No such file or directory` when importing `nemo.collections.asr.models`
**Environment details**
- OS version Ubuntu 20.04
- PyTorch version 1.11.0
- Python version 3.9.12
| {
"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/4513/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/4513/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4512 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4512/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4512/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4512/events | https://github.com/NVIDIA/NeMo/pull/4512 | 1,296,880,940 | PR_kwDOC_bI7s46_iyq | 4,512 | Add silence handling for speaker diarization pipeline | {
"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
} | [] | closed | false | null | [] | null | null | 2022-07-07T06:10:22 | 2022-07-08T19:43:59 | 2022-07-08T19:43:56 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4512",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4512",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4512.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4512.patch",
"merged_at": "2022-07-08T19:43:56"
} | Signed-off-by: nithinraok <[email protected]>
# What does this PR do ?
Provides fix for handling silence wav files for diarization
**Collection**: [ASR - Speaker Diarization]
# Changelog
- Added `validate_vad_manifest` function to compare the uniq_ids in vad manifest file to manifest file provided by user for diarization
- Fixed an issue to check for a VAD `.frame` file, even if the duration of that audio is <=0.
# Usage
```bash
python offline_diarization.py
```
# 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": "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/4512/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/4512/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4511 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4511/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4511/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4511/events | https://github.com/NVIDIA/NeMo/pull/4511 | 1,296,729,697 | PR_kwDOC_bI7s46_CR5 | 4,511 | [ASR][Breaking Change] Update signature of Hypothesis alignments | {
"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 | 2022-07-07T03:00:27 | 2022-07-09T08:53:14 | 2022-07-09T08:53:13 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4511",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4511",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4511.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4511.patch",
"merged_at": "2022-07-09T08:53:13"
} | # What does this PR do ?
Updates signature of Hypothis.alignments such that it now contains a tuple (logprob, label), such that CTC, RNNT greedy and RNNT beam search all yield the logprob tensor of shape (V+1,) where V is the vocab size (and last index is CTC/RNNT Blank). This can be used to compute beam search results.
For CTC Models, hypothesis.alignments now returns a tuple of tensors (logp, labels) rather than just logp to match the RNNT signature.
**Collection**: [ASR]
# Changelog
- [BREAKING CHANGE] Force `preserve_alignments` flag to preserve a tuple of values (logp, label) for both CTC and RNNT.
- This enables external beam search rescoring to work with CTC and RNNT models.
- Also enables the possibility of RNNT based char/word time stamp calculation
- In totality, enables CTC/RNNT models to compute word confidence.
- Add respective tests
# Usage
```python
model = EncDecRNNTModelBPE.from_pretrained(...)
model.eval()
decoding_cfg. model.cfg.decoding
decoding_cfg.preserve_alignments = True
model.change_decoding_strategy(decoding_cfg)
# For RNNT
hyps, _ = model.transcribe(..., return_hypothesis=True)
# For CTC
hyps = model.transcribe(..., return_hypothesis=True)
# select hypothesis of 1st sample
hyp = hyps[0]
# extract alignments
for t in range(len(hyp.alignments)):
for u in range(len(hyp.alignments[t])):
logprob, label = hyp[t][u]
....
```
# 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?
- [x] 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**:
- [x] New Feature
- [ ] Bugfix
- [ ] Documentation
If you haven't finished some of the above items you can still open "Draft" PR.
| {
"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/4511/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/4511/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4510 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4510/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4510/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4510/events | https://github.com/NVIDIA/NeMo/pull/4510 | 1,296,537,258 | PR_kwDOC_bI7s46-Z9M | 4,510 | [TTS] Update cmudict | {
"login": "jasro23",
"id": 108691071,
"node_id": "U_kgDOBnp-fw",
"avatar_url": "https://avatars.githubusercontent.com/u/108691071?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jasro23",
"html_url": "https://github.com/jasro23",
"followers_url": "https://api.github.com/users/jasro23/followers",
"following_url": "https://api.github.com/users/jasro23/following{/other_user}",
"gists_url": "https://api.github.com/users/jasro23/gists{/gist_id}",
"starred_url": "https://api.github.com/users/jasro23/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/jasro23/subscriptions",
"organizations_url": "https://api.github.com/users/jasro23/orgs",
"repos_url": "https://api.github.com/users/jasro23/repos",
"events_url": "https://api.github.com/users/jasro23/events{/privacy}",
"received_events_url": "https://api.github.com/users/jasro23/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | 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 | 2022-07-06T22:04:15 | 2022-07-08T06:29:15 | 2022-07-08T06:29:15 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4510",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4510",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4510.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4510.patch",
"merged_at": "2022-07-08T06:29:15"
} | phoneme IY1 -> IH1 in NVIDIA
Added phonemes for CUSTOMIZABLE
Update the cmudict version and its reference.
# What does this PR do ?
Update phoneme typo and a new phoneme word
**Collection**: [Text to speech]
# 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": "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/4510/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/4510/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4509 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4509/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4509/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4509/events | https://github.com/NVIDIA/NeMo/pull/4509 | 1,296,485,337 | PR_kwDOC_bI7s46-OYH | 4,509 | update fastpitch to add export controls | {
"login": "blisc",
"id": 4763352,
"node_id": "MDQ6VXNlcjQ3NjMzNTI=",
"avatar_url": "https://avatars.githubusercontent.com/u/4763352?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/blisc",
"html_url": "https://github.com/blisc",
"followers_url": "https://api.github.com/users/blisc/followers",
"following_url": "https://api.github.com/users/blisc/following{/other_user}",
"gists_url": "https://api.github.com/users/blisc/gists{/gist_id}",
"starred_url": "https://api.github.com/users/blisc/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/blisc/subscriptions",
"organizations_url": "https://api.github.com/users/blisc/orgs",
"repos_url": "https://api.github.com/users/blisc/repos",
"events_url": "https://api.github.com/users/blisc/events{/privacy}",
"received_events_url": "https://api.github.com/users/blisc/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | null | 2022-07-06T21:15:41 | 2022-07-14T21:30:56 | 2022-07-14T21:30:53 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4509",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4509",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4509.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4509.patch",
"merged_at": "2022-07-14T21:30:52"
} | # What does this PR do ?
Adds options for fastpitch export. One for ragged_batch and one for volume. Undos the hard assumption that `volume` must be input and output as done in #4167
**Collection**: TTS
# Changelog
- Adds ragged_batch and volume options for fastpitch export
**PR Type**:
- [x] New Feature | {
"login": "blisc",
"id": 4763352,
"node_id": "MDQ6VXNlcjQ3NjMzNTI=",
"avatar_url": "https://avatars.githubusercontent.com/u/4763352?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/blisc",
"html_url": "https://github.com/blisc",
"followers_url": "https://api.github.com/users/blisc/followers",
"following_url": "https://api.github.com/users/blisc/following{/other_user}",
"gists_url": "https://api.github.com/users/blisc/gists{/gist_id}",
"starred_url": "https://api.github.com/users/blisc/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/blisc/subscriptions",
"organizations_url": "https://api.github.com/users/blisc/orgs",
"repos_url": "https://api.github.com/users/blisc/repos",
"events_url": "https://api.github.com/users/blisc/events{/privacy}",
"received_events_url": "https://api.github.com/users/blisc/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4509/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/4509/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4508 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4508/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4508/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4508/events | https://github.com/NVIDIA/NeMo/pull/4508 | 1,296,415,992 | PR_kwDOC_bI7s469-Ye | 4,508 | [Add] Support for Different LRs with Param Groups | {
"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 | 2022-07-06T20:39:27 | 2022-07-08T13:12:58 | 2022-07-08T13:12:58 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4508",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4508",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4508.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4508.patch",
"merged_at": "2022-07-08T13:12:58"
} | This PR aims to add support for different learning rates for different modules in a model. For example, when finetuning an ASR model on small datasets, we might want the encoder to have smaller learning rate than the decoder. With this PR, we can do this by adding a `optim_param_groups` section under `model` in the config file:
```yaml
model:
optim_param_groups:
"encoder":
"lr": 1e-4
"momentum": 0.8
optim:
"name": sgd
"lr": 1e-3 # default lr for non-specified parameters
```
In this example, all parameters in the encoder will have learning rate `1e-4` while all other parameters (e.g., decoder) will have learning rate `1e-3`.
For logging learning rates, the current models only log the first group of `param_groups`, which in this case is the default `lr`. To log all learning rates, we need to modify the `training_step()` function in the model. | {
"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/4508/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/4508/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4507 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4507/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4507/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4507/events | https://github.com/NVIDIA/NeMo/issues/4507 | 1,296,202,754 | I_kwDOC_bI7s5NQnwC | 4,507 | duplex_text_normalization_infer.py FST error | {
"login": "uralik",
"id": 2152005,
"node_id": "MDQ6VXNlcjIxNTIwMDU=",
"avatar_url": "https://avatars.githubusercontent.com/u/2152005?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/uralik",
"html_url": "https://github.com/uralik",
"followers_url": "https://api.github.com/users/uralik/followers",
"following_url": "https://api.github.com/users/uralik/following{/other_user}",
"gists_url": "https://api.github.com/users/uralik/gists{/gist_id}",
"starred_url": "https://api.github.com/users/uralik/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/uralik/subscriptions",
"organizations_url": "https://api.github.com/users/uralik/orgs",
"repos_url": "https://api.github.com/users/uralik/repos",
"events_url": "https://api.github.com/users/uralik/events{/privacy}",
"received_events_url": "https://api.github.com/users/uralik/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": 4910680898,
"node_id": "LA_kwDOC_bI7s8AAAABJLMLQg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/TN/ITN",
"name": "TN/ITN",
"color": "0052cc",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | null | 2022-07-06T18:15:19 | 2022-12-19T21:46:50 | 2022-07-07T19:07:26 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | **Describe the bug**
Runtime crashes when running the inference example with T5 pretrained model.
**Steps/Code to reproduce bug**
`python examples/nlp/duplex_text_normalization/duplex_text_normalization_infer.py lang=en mode=tn tagger_pretrained_model=neural_text_normalization_t5 decoder_pretrained_model=neural_text_normalization_t5 inference.from_file=./text_en.txt`
Traceback:
```python
[NeMo I 2022-07-06 11:06:49 duplex_text_normalization_infer:83] Running inference on ./text_en.txt...
0%| | 0/1 [00:00<?, ?it/s]ERROR: StringFstToOutputLabels: Invalid start state
Error executing job with overrides: ['lang=en', 'mode=tn', 'tagger_pretrained_model=neural_text_normalization_t5', 'decoder_pretrained_model=neural_text_normalization_t5', 'inference.from_file=./text_en.txt']
An error occurred during Hydra's exception formatting:
AssertionError()
Traceback (most recent call last):
File "/private/home/kulikov/miniconda3/envs/nemo/lib/python3.8/site-packages/hydra/_internal/utils.py", line 252, in run_and_report
assert mdl is not None
AssertionError
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "examples/nlp/duplex_text_normalization/duplex_text_normalization_infer.py", line 155, in <module>
main()
File "/private/home/kulikov/code/NeMo/nemo/core/config/hydra_runner.py", line 104, in wrapper
_run_hydra(
File "/private/home/kulikov/miniconda3/envs/nemo/lib/python3.8/site-packages/hydra/_internal/utils.py", line 377, in _run_hydra
run_and_report(
File "/private/home/kulikov/miniconda3/envs/nemo/lib/python3.8/site-packages/hydra/_internal/utils.py", line 294, in run_and_report
raise ex
File "/private/home/kulikov/miniconda3/envs/nemo/lib/python3.8/site-packages/hydra/_internal/utils.py", line 211, in run_and_report
return func()
File "/private/home/kulikov/miniconda3/envs/nemo/lib/python3.8/site-packages/hydra/_internal/utils.py", line 378, in <lambda>
lambda: hydra.run(
File "/private/home/kulikov/miniconda3/envs/nemo/lib/python3.8/site-packages/hydra/_internal/hydra.py", line 111, in run
_ = ret.return_value
File "/private/home/kulikov/miniconda3/envs/nemo/lib/python3.8/site-packages/hydra/core/utils.py", line 233, in return_value
raise self._return_value
File "/private/home/kulikov/miniconda3/envs/nemo/lib/python3.8/site-packages/hydra/core/utils.py", line 160, in run_job
ret.return_value = task_function(task_cfg)
File "examples/nlp/duplex_text_normalization/duplex_text_normalization_infer.py", line 91, in main
new_lines = normalizer_electronic.normalize_list(lines)
File "/private/home/kulikov/code/NeMo/nemo_text_processing/text_normalization/normalize.py", line 150, in normalize_list
raise e
File "/private/home/kulikov/code/NeMo/nemo_text_processing/text_normalization/normalize.py", line 145, in normalize_list
normalized_texts = Parallel(n_jobs=n_jobs)(
File "/private/home/kulikov/miniconda3/envs/nemo/lib/python3.8/site-packages/joblib/parallel.py", line 1043, in __call__
if self.dispatch_one_batch(iterator):
File "/private/home/kulikov/miniconda3/envs/nemo/lib/python3.8/site-packages/joblib/parallel.py", line 861, in dispatch_one_batch
self._dispatch(tasks)
File "/private/home/kulikov/miniconda3/envs/nemo/lib/python3.8/site-packages/joblib/parallel.py", line 779, in _dispatch
job = self._backend.apply_async(batch, callback=cb)
File "/private/home/kulikov/miniconda3/envs/nemo/lib/python3.8/site-packages/joblib/_parallel_backends.py", line 208, in apply_async
result = ImmediateResult(func)
File "/private/home/kulikov/miniconda3/envs/nemo/lib/python3.8/site-packages/joblib/_parallel_backends.py", line 572, in __init__
self.results = batch()
File "/private/home/kulikov/miniconda3/envs/nemo/lib/python3.8/site-packages/joblib/parallel.py", line 262, in __call__
return [func(*args, **kwargs)
File "/private/home/kulikov/miniconda3/envs/nemo/lib/python3.8/site-packages/joblib/parallel.py", line 262, in <listcomp>
return [func(*args, **kwargs)
File "/private/home/kulikov/code/NeMo/nemo_text_processing/text_normalization/normalize.py", line 164, in __process_batch
normalized_lines = [
File "/private/home/kulikov/code/NeMo/nemo_text_processing/text_normalization/normalize.py", line 165, in <listcomp>
self.normalize(
File "/private/home/kulikov/code/NeMo/nemo_text_processing/text_normalization/normalize.py", line 268, in normalize
tagged_text = self.select_tag(tagged_lattice)
File "/private/home/kulikov/code/NeMo/nemo_text_processing/text_normalization/normalize.py", line 382, in select_tag
tagged_text = pynini.shortestpath(lattice, nshortest=1, unique=True).string()
File "extensions/_pynini.pyx", line 471, in _pynini.Fst.string
File "extensions/_pynini.pyx", line 516, in _pynini.Fst.string
_pywrapfst.FstOpError: Operation failed
```
where
```
cat text_en.txt
he was born in 1992
```
**Expected behavior**
The example is expected to work! Looks like smth is wrong with the pynini lib.
**Environment overview (please complete the following information)**
- Environment location: bare metal
- Method of NeMo install: I used conda install
- If method of install is [Docker], provide `docker pull` & `docker run` commands used
**Environment details**
If NVIDIA docker image is used you don't need to specify these.
Otherwise, please provide:
- OS version Ubuntu
- PyTorch version 1.12.0
- Python version 3.8.0
**Additional context**
pynini 2.1.4
| {
"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/4507/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/4507/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4506 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4506/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4506/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4506/events | https://github.com/NVIDIA/NeMo/issues/4506 | 1,296,088,027 | I_kwDOC_bI7s5NQLvb | 4,506 | wav2vec CTC val_wer is always 1 | {
"login": "demsarjure",
"id": 35776954,
"node_id": "MDQ6VXNlcjM1Nzc2OTU0",
"avatar_url": "https://avatars.githubusercontent.com/u/35776954?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/demsarjure",
"html_url": "https://github.com/demsarjure",
"followers_url": "https://api.github.com/users/demsarjure/followers",
"following_url": "https://api.github.com/users/demsarjure/following{/other_user}",
"gists_url": "https://api.github.com/users/demsarjure/gists{/gist_id}",
"starred_url": "https://api.github.com/users/demsarjure/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/demsarjure/subscriptions",
"organizations_url": "https://api.github.com/users/demsarjure/orgs",
"repos_url": "https://api.github.com/users/demsarjure/repos",
"events_url": "https://api.github.com/users/demsarjure/events{/privacy}",
"received_events_url": "https://api.github.com/users/demsarjure/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 | 2022-07-06T16:24:45 | 2022-07-07T13:01:13 | 2022-07-07T13:01:12 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | **Describe the bug**
Hi, I am having issues with training wav2vec models. It seems like the model starts training, there are no warnings or errors, everything runs. However, after every epoch val_wer is equal to 1.0. For [wav2vec_pretrain.yaml](https://github.com/NVIDIA/NeMo/blob/main/examples/asr/conf/ssl/wav2vec/wav2vec_pretrain.yaml) I am not sure where to start as there are no tutorials or examples. So I have been using [wav2vecCTC.yaml](https://github.com/NVIDIA/NeMo/blob/main/examples/asr/conf/wav2vec_ctc/wav2vecCTC.yaml) and the example script for training CTC models.
**Steps/Code to reproduce bug**
Below is the exact used command, used datasets work perfectly with other NeMo ASR models. It is the same command that I use for all CTC models.
```
python /workspace/nemo/examples/asr/asr_ctc/speech_to_text_ctc_bpe.py \\
--config-path=/workspace/nemo/examples/asr/conf/wav2vec_ctc \\
--config-name=wav2vecCTC \\
exp_manager.name=wav2vec} \\
+exp_manager.exp_dir=/experiments \\
exp_manager.resume_if_exists=true \\
exp_manager.resume_ignore_no_checkpoint=true \\
+exp_manager.checkpoint_callback_params.save_best_model=true \\
model.train_ds.manifest_filepath=/data/train.nemo \\
model.train_ds.batch_size=8 \\
model.train_ds.num_workers=4 \\
model.train_ds.max_duration=17 \\
+model.train_ds.min_duration=1 \\
+model.train_ds.parser=base \\
model.validation_ds.manifest_filepath=/data/validation.nemo \\
model.validation_ds.batch_size=8 \\
model.validation_ds.num_workers=4 \\
+model.validation_ds.parser=base \\
model.tokenizer.dir=/data/out/tokenizer_spe_unigram_v128 \\
model.tokenizer.type=bpe \\
trainer.precision=bf16 \\
trainer.devices=-1
```
**Expected behavior**
I expect val_wer to drop with training and not remain at its maximum value (1.0).
**Environment overview (please complete the following information)**
- Official latest NeMo container where the NeMo repository is updated to the current tip of the master branch.
**Environment details**
- Official latest NeMo container where the NeMo repository is updated to the current tip of the master branch.
**Additional context**
Running on 4 RTX A6000 GPUs.
| {
"login": "demsarjure",
"id": 35776954,
"node_id": "MDQ6VXNlcjM1Nzc2OTU0",
"avatar_url": "https://avatars.githubusercontent.com/u/35776954?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/demsarjure",
"html_url": "https://github.com/demsarjure",
"followers_url": "https://api.github.com/users/demsarjure/followers",
"following_url": "https://api.github.com/users/demsarjure/following{/other_user}",
"gists_url": "https://api.github.com/users/demsarjure/gists{/gist_id}",
"starred_url": "https://api.github.com/users/demsarjure/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/demsarjure/subscriptions",
"organizations_url": "https://api.github.com/users/demsarjure/orgs",
"repos_url": "https://api.github.com/users/demsarjure/repos",
"events_url": "https://api.github.com/users/demsarjure/events{/privacy}",
"received_events_url": "https://api.github.com/users/demsarjure/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4506/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/4506/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4504 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4504/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4504/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4504/events | https://github.com/NVIDIA/NeMo/pull/4504 | 1,295,206,361 | PR_kwDOC_bI7s4650FJ | 4,504 | Update finetune label 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
} | [] | closed | false | null | [] | null | null | 2022-07-06T05:57:27 | 2022-07-09T02:11:22 | 2022-07-09T02:11:14 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4504",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4504",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4504.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4504.patch",
"merged_at": "2022-07-09T02:11:14"
} | # What does this PR do ?
Upgraded speaker recognition fine-tune script
**Collection**: [ASR - Speaker Recognition]
# Changelog
- Removed set_up_finetune_model definition in label_models
- Updated speaker_reco_finetune.py script, to resemble PTL training method
- Now, fine-tune can be directly used from config. Added titanet-finetune.yaml config
- Updated notebook tutorials for the fine-tuning section
# Usage
* You can potentially add a usage example below
```bash
python speaker_reco_finetune.py
```
# 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/4504/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/4504/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4502 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4502/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4502/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4502/events | https://github.com/NVIDIA/NeMo/pull/4502 | 1,295,009,281 | PR_kwDOC_bI7s465I0U | 4,502 | Conditionally skip input loading at pipeline-parallel training | {
"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
} | [] | closed | false | null | [] | null | null | 2022-07-06T02:04:48 | 2022-07-06T02:28:32 | 2022-07-06T02:28:28 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4502",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4502",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4502.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4502.patch",
"merged_at": "2022-07-06T02:28:28"
} | # What does this PR do ?
At pipeline-parallel training, set the pipeline stage that the current GPU belongs to skip loading inputs
Intermediate stage: do not load any inputs
Fist pipeline stage: load only 'tokens' and 'position_ids'
Last pipeline stage: load only 'labels' and 'loss_mask'
The gain of this PR increases when used with interleaved pipelining because the baseline code loads inputs at each virtual pipe stage.
**Collection**: [Note which collection this PR will affect]
# Changelog
- Add specific line by line info of high level changes in this PR.
# Usage
This is enabled when pipeline-model size is larger than 1
```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/4502/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/4502/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4501 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4501/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4501/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4501/events | https://github.com/NVIDIA/NeMo/pull/4501 | 1,295,000,019 | PR_kwDOC_bI7s465G0D | 4,501 | Fix runtime check | {
"login": "borisfom",
"id": 14189615,
"node_id": "MDQ6VXNlcjE0MTg5NjE1",
"avatar_url": "https://avatars.githubusercontent.com/u/14189615?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/borisfom",
"html_url": "https://github.com/borisfom",
"followers_url": "https://api.github.com/users/borisfom/followers",
"following_url": "https://api.github.com/users/borisfom/following{/other_user}",
"gists_url": "https://api.github.com/users/borisfom/gists{/gist_id}",
"starred_url": "https://api.github.com/users/borisfom/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/borisfom/subscriptions",
"organizations_url": "https://api.github.com/users/borisfom/orgs",
"repos_url": "https://api.github.com/users/borisfom/repos",
"events_url": "https://api.github.com/users/borisfom/events{/privacy}",
"received_events_url": "https://api.github.com/users/borisfom/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | null | 2022-07-06T01:53:48 | 2022-07-08T22:31:05 | 2022-07-08T22:31:05 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4501",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4501",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4501.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4501.patch",
"merged_at": "2022-07-08T22:31:05"
} | # What does this PR do ?
Post-export runtime check refinements - allow checking multiple input sizes.
Restored max_dim argument functionality in export.py
Also, added a cast to ensure ASR nets are always exported with output in FP32 - even with autocast.
**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": "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/4501/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/4501/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4500 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4500/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4500/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4500/events | https://github.com/NVIDIA/NeMo/pull/4500 | 1,294,848,663 | PR_kwDOC_bI7s464mKu | 4,500 | NMESC speaker counting algorithm update | {
"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
} | [] | closed | false | null | [] | null | null | 2022-07-05T22:29:23 | 2022-07-07T05:01:24 | 2022-07-07T05:01:21 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4500",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4500",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4500.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4500.patch",
"merged_at": "2022-07-07T05:01:21"
} | # What does this PR do ?
`nmesc_clustering.py`
This PR adds an option named "maj_vote_spk_count" to NMESC clustering module.
Originally, speaker number was obtained from one p-value.
In the new implementation, maj_vote_spk_count = True, the system takes majority vote on
all p-values that are included.
This feature almost halves the speaker counting error for 1.5sec window setting.
(CH109, 1.5/0.75 win/shift, speaker counting error becomes 12.8% -> 7.3%)
Default will be False for now. Yaml file will be later updated with other updates to provide the best performance.
`diarizer.clustering.parameters.maj_vote_spk_count=False`
This PR also adds error handling for extreamly short segments (1~10 segments)
when creating p_value_list, the minimum length is now forcibly assigned so that we can avoid having an empty list.
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)
- [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
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/4500/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/4500/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4499 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4499/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4499/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4499/events | https://github.com/NVIDIA/NeMo/pull/4499 | 1,294,670,225 | PR_kwDOC_bI7s464AdI | 4,499 | Changes to make Megatron NMT exportable | {
"login": "Davood-M",
"id": 3964285,
"node_id": "MDQ6VXNlcjM5NjQyODU=",
"avatar_url": "https://avatars.githubusercontent.com/u/3964285?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Davood-M",
"html_url": "https://github.com/Davood-M",
"followers_url": "https://api.github.com/users/Davood-M/followers",
"following_url": "https://api.github.com/users/Davood-M/following{/other_user}",
"gists_url": "https://api.github.com/users/Davood-M/gists{/gist_id}",
"starred_url": "https://api.github.com/users/Davood-M/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/Davood-M/subscriptions",
"organizations_url": "https://api.github.com/users/Davood-M/orgs",
"repos_url": "https://api.github.com/users/Davood-M/repos",
"events_url": "https://api.github.com/users/Davood-M/events{/privacy}",
"received_events_url": "https://api.github.com/users/Davood-M/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | null | 2022-07-05T19:13:22 | 2022-08-09T18:55:00 | 2022-08-09T18:28:35 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4499",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4499",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4499.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4499.patch",
"merged_at": "2022-08-09T18:28:35"
} | # What does this PR do ?
Adding requirements to make Megatron NMT exportable
**Collection**: [Note which collection this PR will affect]
# Changelog
- Added input/output types for encoder/decoder/tokens_head
- Added input_example
# 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
## 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": "Davood-M",
"id": 3964285,
"node_id": "MDQ6VXNlcjM5NjQyODU=",
"avatar_url": "https://avatars.githubusercontent.com/u/3964285?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Davood-M",
"html_url": "https://github.com/Davood-M",
"followers_url": "https://api.github.com/users/Davood-M/followers",
"following_url": "https://api.github.com/users/Davood-M/following{/other_user}",
"gists_url": "https://api.github.com/users/Davood-M/gists{/gist_id}",
"starred_url": "https://api.github.com/users/Davood-M/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/Davood-M/subscriptions",
"organizations_url": "https://api.github.com/users/Davood-M/orgs",
"repos_url": "https://api.github.com/users/Davood-M/repos",
"events_url": "https://api.github.com/users/Davood-M/events{/privacy}",
"received_events_url": "https://api.github.com/users/Davood-M/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4499/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/4499/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4498 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4498/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4498/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4498/events | https://github.com/NVIDIA/NeMo/issues/4498 | 1,293,698,925 | I_kwDOC_bI7s5NHEdt | 4,498 | Cannot run Fastpitch on CPU | {
"login": "icyda17",
"id": 73210249,
"node_id": "MDQ6VXNlcjczMjEwMjQ5",
"avatar_url": "https://avatars.githubusercontent.com/u/73210249?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/icyda17",
"html_url": "https://github.com/icyda17",
"followers_url": "https://api.github.com/users/icyda17/followers",
"following_url": "https://api.github.com/users/icyda17/following{/other_user}",
"gists_url": "https://api.github.com/users/icyda17/gists{/gist_id}",
"starred_url": "https://api.github.com/users/icyda17/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/icyda17/subscriptions",
"organizations_url": "https://api.github.com/users/icyda17/orgs",
"repos_url": "https://api.github.com/users/icyda17/repos",
"events_url": "https://api.github.com/users/icyda17/events{/privacy}",
"received_events_url": "https://api.github.com/users/icyda17/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 | 2022-07-05T02:09:08 | 2022-07-05T06:15:46 | 2022-07-05T06:15:46 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | Hi, I ran the code at tutorials/tts/FastPitch_MixerTTS_Training.ipynb and received the error as following when run python fastpitch.py
I change config file to run on CPU
- OS version: Ubuntu 20.04
- PyTorch version:
- Python version: 3.8
Is there anything need to be changed to be able to train on CPU?
| {
"login": "icyda17",
"id": 73210249,
"node_id": "MDQ6VXNlcjczMjEwMjQ5",
"avatar_url": "https://avatars.githubusercontent.com/u/73210249?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/icyda17",
"html_url": "https://github.com/icyda17",
"followers_url": "https://api.github.com/users/icyda17/followers",
"following_url": "https://api.github.com/users/icyda17/following{/other_user}",
"gists_url": "https://api.github.com/users/icyda17/gists{/gist_id}",
"starred_url": "https://api.github.com/users/icyda17/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/icyda17/subscriptions",
"organizations_url": "https://api.github.com/users/icyda17/orgs",
"repos_url": "https://api.github.com/users/icyda17/repos",
"events_url": "https://api.github.com/users/icyda17/events{/privacy}",
"received_events_url": "https://api.github.com/users/icyda17/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4498/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/4498/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4497 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4497/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4497/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4497/events | https://github.com/NVIDIA/NeMo/issues/4497 | 1,293,366,856 | I_kwDOC_bI7s5NFzZI | 4,497 | Wav2vec deviations from fairseq implementation | {
"login": "1-800-BAD-CODE",
"id": 50530592,
"node_id": "MDQ6VXNlcjUwNTMwNTky",
"avatar_url": "https://avatars.githubusercontent.com/u/50530592?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/1-800-BAD-CODE",
"html_url": "https://github.com/1-800-BAD-CODE",
"followers_url": "https://api.github.com/users/1-800-BAD-CODE/followers",
"following_url": "https://api.github.com/users/1-800-BAD-CODE/following{/other_user}",
"gists_url": "https://api.github.com/users/1-800-BAD-CODE/gists{/gist_id}",
"starred_url": "https://api.github.com/users/1-800-BAD-CODE/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/1-800-BAD-CODE/subscriptions",
"organizations_url": "https://api.github.com/users/1-800-BAD-CODE/orgs",
"repos_url": "https://api.github.com/users/1-800-BAD-CODE/repos",
"events_url": "https://api.github.com/users/1-800-BAD-CODE/events{/privacy}",
"received_events_url": "https://api.github.com/users/1-800-BAD-CODE/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 | 2022-07-04T16:26:54 | 2022-07-23T19:18:44 | 2022-07-23T19:18:44 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | **Describe the bug**
I recently loaded the weights of a pre-trained "[XLS-R](https://github.com/facebookresearch/fairseq/tree/main/examples/wav2vec/xlsr)" wav2vec SSL model into the NeMo implementation and found a couple subtle deviations from the fairseq implementation that led to a different computational graph with the same state dict.
E.g., in the fairseq feature extractor, layer norm is applied [before](https://github.com/facebookresearch/fairseq/blob/5307a0e078d7460003a86f4e2246d459d4706a1d/fairseq/models/wav2vec/wav2vec2.py#L604) projecting the embeddings to the encoder dim whereas in the NeMo implementation the layer norm is applied [after](https://github.com/NVIDIA/NeMo/blob/8a172df69250da231742893387e9a1dd01b16e3d/nemo/collections/asr/modules/wav2vec_modules.py#L196).
After a couple changes I was able to run the XLS-R fairseq model in NeMo and get the same results from the feature extractor and encoder graphs.
**Are these deviations intentional?** It seems like most of the code is just copy-pasted except for these changes.
If these are intentional deviations, I will simply keep my patch local and close the issue.
If these deviations are accidental or if it would be useful to have an option to rearrange the graph to match the fairseq implementation, I will submit my changes.
| {
"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/4497/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/4497/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4496 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4496/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4496/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4496/events | https://github.com/NVIDIA/NeMo/pull/4496 | 1,292,462,546 | PR_kwDOC_bI7s46wt8d | 4,496 | GPT Prompt Learning Improvements | {
"login": "vadam5",
"id": 78445382,
"node_id": "MDQ6VXNlcjc4NDQ1Mzgy",
"avatar_url": "https://avatars.githubusercontent.com/u/78445382?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/vadam5",
"html_url": "https://github.com/vadam5",
"followers_url": "https://api.github.com/users/vadam5/followers",
"following_url": "https://api.github.com/users/vadam5/following{/other_user}",
"gists_url": "https://api.github.com/users/vadam5/gists{/gist_id}",
"starred_url": "https://api.github.com/users/vadam5/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/vadam5/subscriptions",
"organizations_url": "https://api.github.com/users/vadam5/orgs",
"repos_url": "https://api.github.com/users/vadam5/repos",
"events_url": "https://api.github.com/users/vadam5/events{/privacy}",
"received_events_url": "https://api.github.com/users/vadam5/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | null | 2022-07-04T00:23:32 | 2022-07-06T18:31:19 | 2022-07-06T18:30:58 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4496",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4496",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4496.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4496.patch",
"merged_at": "2022-07-06T18:30:58"
} | Signed-off-by: Virginia Adams <[email protected]>
Makes several improvements to the gpt prompt learning code including:
- faster training time using less memory
- switch `global_batch_size` to refer to actual global batch size and not local mini batch size
**Collection**: BigNLP
# Changelog
- Update GPT prompt learning model class
# 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": "vadam5",
"id": 78445382,
"node_id": "MDQ6VXNlcjc4NDQ1Mzgy",
"avatar_url": "https://avatars.githubusercontent.com/u/78445382?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/vadam5",
"html_url": "https://github.com/vadam5",
"followers_url": "https://api.github.com/users/vadam5/followers",
"following_url": "https://api.github.com/users/vadam5/following{/other_user}",
"gists_url": "https://api.github.com/users/vadam5/gists{/gist_id}",
"starred_url": "https://api.github.com/users/vadam5/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/vadam5/subscriptions",
"organizations_url": "https://api.github.com/users/vadam5/orgs",
"repos_url": "https://api.github.com/users/vadam5/repos",
"events_url": "https://api.github.com/users/vadam5/events{/privacy}",
"received_events_url": "https://api.github.com/users/vadam5/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4496/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/4496/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4495 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4495/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4495/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4495/events | https://github.com/NVIDIA/NeMo/pull/4495 | 1,292,431,173 | PR_kwDOC_bI7s46wnlB | 4,495 | Megatron BART BOS / EOS bug fix | {
"login": "michalivne",
"id": 7453913,
"node_id": "MDQ6VXNlcjc0NTM5MTM=",
"avatar_url": "https://avatars.githubusercontent.com/u/7453913?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/michalivne",
"html_url": "https://github.com/michalivne",
"followers_url": "https://api.github.com/users/michalivne/followers",
"following_url": "https://api.github.com/users/michalivne/following{/other_user}",
"gists_url": "https://api.github.com/users/michalivne/gists{/gist_id}",
"starred_url": "https://api.github.com/users/michalivne/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/michalivne/subscriptions",
"organizations_url": "https://api.github.com/users/michalivne/orgs",
"repos_url": "https://api.github.com/users/michalivne/repos",
"events_url": "https://api.github.com/users/michalivne/events{/privacy}",
"received_events_url": "https://api.github.com/users/michalivne/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | null | 2022-07-03T22:04:23 | 2022-07-06T19:05:05 | 2022-07-06T17:37:47 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4495",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4495",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4495.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4495.patch",
"merged_at": "2022-07-06T17:37:47"
} | Fixes Megatron BART dataset missing <EOS> for decoder output.
`BARTDataset` output description:
1. Encoder input: tokens (no `<BOS>` or `<EOS>` added)
2. Decoder input: [`<BOS>`] + tokens
3. Decoder output: tokens + [`<EOS>`]
# 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": "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/4495/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/4495/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4494 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4494/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4494/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4494/events | https://github.com/NVIDIA/NeMo/pull/4494 | 1,292,396,523 | PR_kwDOC_bI7s46wg3o | 4,494 | Gradient Accumulation fusion to Linear layer weight gradient computation | {
"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
} | [] | closed | false | null | [] | null | null | 2022-07-03T19:04:38 | 2022-07-05T21:30:06 | 2022-07-05T21:30:01 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4494",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4494",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4494.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4494.patch",
"merged_at": "2022-07-05T21:30:01"
} | # What does this PR do ?
Fuse gradient accumulation ops to the weight gradient computes of linear layers.
This feature should be used with (1) serialized gradient allreduce at this update and (2) O2.
**Collection**: [Note which collection this PR will affect]
# Changelog
- Add specific line by line info of high level changes in this PR.
# Usage
Add below knob to the commanline.
```python
training.model.gradient_accumulation_fusion=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/4494/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/4494/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4493 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4493/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4493/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4493/events | https://github.com/NVIDIA/NeMo/issues/4493 | 1,292,298,000 | I_kwDOC_bI7s5NBucQ | 4,493 | After training the finished model stt_en_citrinet_1024 on 2 trial sentences, the finished checkpoint gives out one letter. | {
"login": "Project-fil",
"id": 82889232,
"node_id": "MDQ6VXNlcjgyODg5MjMy",
"avatar_url": "https://avatars.githubusercontent.com/u/82889232?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Project-fil",
"html_url": "https://github.com/Project-fil",
"followers_url": "https://api.github.com/users/Project-fil/followers",
"following_url": "https://api.github.com/users/Project-fil/following{/other_user}",
"gists_url": "https://api.github.com/users/Project-fil/gists{/gist_id}",
"starred_url": "https://api.github.com/users/Project-fil/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/Project-fil/subscriptions",
"organizations_url": "https://api.github.com/users/Project-fil/orgs",
"repos_url": "https://api.github.com/users/Project-fil/repos",
"events_url": "https://api.github.com/users/Project-fil/events{/privacy}",
"received_events_url": "https://api.github.com/users/Project-fil/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 | 2022-07-03T12:44:56 | 2022-07-23T19:18:30 | 2022-07-23T19:18:30 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | - yaml used originally for stt_en_citrinet_1024
- for the test and validation set, I used train data
- as a result I get the letter "u" or "k" | {
"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/4493/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/4493/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4491 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4491/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4491/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4491/events | https://github.com/NVIDIA/NeMo/issues/4491 | 1,292,164,531 | I_kwDOC_bI7s5NBN2z | 4,491 | Traing to fine tune pre-trained model but getting "...shape torch.Size([69, 512])..." | {
"login": "NK990",
"id": 7748729,
"node_id": "MDQ6VXNlcjc3NDg3Mjk=",
"avatar_url": "https://avatars.githubusercontent.com/u/7748729?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/NK990",
"html_url": "https://github.com/NK990",
"followers_url": "https://api.github.com/users/NK990/followers",
"following_url": "https://api.github.com/users/NK990/following{/other_user}",
"gists_url": "https://api.github.com/users/NK990/gists{/gist_id}",
"starred_url": "https://api.github.com/users/NK990/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/NK990/subscriptions",
"organizations_url": "https://api.github.com/users/NK990/orgs",
"repos_url": "https://api.github.com/users/NK990/repos",
"events_url": "https://api.github.com/users/NK990/events{/privacy}",
"received_events_url": "https://api.github.com/users/NK990/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": "redoctopus",
"id": 4268876,
"node_id": "MDQ6VXNlcjQyNjg4NzY=",
"avatar_url": "https://avatars.githubusercontent.com/u/4268876?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/redoctopus",
"html_url": "https://github.com/redoctopus",
"followers_url": "https://api.github.com/users/redoctopus/followers",
"following_url": "https://api.github.com/users/redoctopus/following{/other_user}",
"gists_url": "https://api.github.com/users/redoctopus/gists{/gist_id}",
"starred_url": "https://api.github.com/users/redoctopus/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/redoctopus/subscriptions",
"organizations_url": "https://api.github.com/users/redoctopus/orgs",
"repos_url": "https://api.github.com/users/redoctopus/repos",
"events_url": "https://api.github.com/users/redoctopus/events{/privacy}",
"received_events_url": "https://api.github.com/users/redoctopus/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"login": "redoctopus",
"id": 4268876,
"node_id": "MDQ6VXNlcjQyNjg4NzY=",
"avatar_url": "https://avatars.githubusercontent.com/u/4268876?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/redoctopus",
"html_url": "https://github.com/redoctopus",
"followers_url": "https://api.github.com/users/redoctopus/followers",
"following_url": "https://api.github.com/users/redoctopus/following{/other_user}",
"gists_url": "https://api.github.com/users/redoctopus/gists{/gist_id}",
"starred_url": "https://api.github.com/users/redoctopus/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/redoctopus/subscriptions",
"organizations_url": "https://api.github.com/users/redoctopus/orgs",
"repos_url": "https://api.github.com/users/redoctopus/repos",
"events_url": "https://api.github.com/users/redoctopus/events{/privacy}",
"received_events_url": "https://api.github.com/users/redoctopus/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
}
] | null | null | 2022-07-02T23:44:56 | 2022-07-05T18:08:39 | 2022-07-05T18:08:39 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | I'm traing to fine tune pre-trained model, and I'm getting this error:
`RuntimeError: Error(s) in loading state_dict for Tacotron2Model:
size mismatch for text_embedding.weight: copying a param with shape torch.Size([69, 512]) from checkpoint, the shape in current model is torch.Size([83, 512]).`
I know that is been caused by updating the charters(labels) dictionary, is there any way to compute new dict size?
Thnx in advance. | {
"login": "redoctopus",
"id": 4268876,
"node_id": "MDQ6VXNlcjQyNjg4NzY=",
"avatar_url": "https://avatars.githubusercontent.com/u/4268876?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/redoctopus",
"html_url": "https://github.com/redoctopus",
"followers_url": "https://api.github.com/users/redoctopus/followers",
"following_url": "https://api.github.com/users/redoctopus/following{/other_user}",
"gists_url": "https://api.github.com/users/redoctopus/gists{/gist_id}",
"starred_url": "https://api.github.com/users/redoctopus/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/redoctopus/subscriptions",
"organizations_url": "https://api.github.com/users/redoctopus/orgs",
"repos_url": "https://api.github.com/users/redoctopus/repos",
"events_url": "https://api.github.com/users/redoctopus/events{/privacy}",
"received_events_url": "https://api.github.com/users/redoctopus/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4491/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/4491/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4489 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4489/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4489/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4489/events | https://github.com/NVIDIA/NeMo/pull/4489 | 1,292,011,333 | PR_kwDOC_bI7s46vWgE | 4,489 | Fix for GeGLU activations with not using fusion | {
"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
} | [] | closed | false | null | [] | null | null | 2022-07-02T09:19:01 | 2023-02-22T18:47:03 | 2022-07-11T23:09:24 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4489",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4489",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4489.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4489.patch",
"merged_at": null
} | Signed-off-by: MaximumEntropy <[email protected]>
# What does this PR do ?
Fixes GeGLU when not using fusion.
**Collection**: [Note which collection this PR will affect]
# Changelog
- Updates if/else chain when deciding how to use activation functions in transformer.py
# Usage
N/A
# 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/4489/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/4489/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4487 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4487/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4487/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4487/events | https://github.com/NVIDIA/NeMo/pull/4487 | 1,291,740,275 | PR_kwDOC_bI7s46ueRg | 4,487 | Add support for Apex distributed Adam optimizer with GPT-3 | {
"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
} | [] | closed | false | null | [] | null | null | 2022-07-01T19:00:37 | 2022-08-15T23:51:27 | 2022-08-15T23:37:31 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4487",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4487",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4487.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4487.patch",
"merged_at": "2022-08-15T23:37:31"
} | # What does this PR do ?
Adds support for training GPT-3 with the [Apex implementation of the ZeRO optimizer](https://github.com/NVIDIA/apex/blob/master/apex/contrib/optimizers/distributed_fused_adam.py).
**Collection**: NLP
# Changelog
- Add option for `distributed_fused_adam` optimizer and integrate with GPT-3 training
# Usage
Set optimizer to `distributed_fused_adam` in config file:
https://github.com/NVIDIA/NeMo/blob/23f6a95fb1479f5d2ff7111adbc5720ed3c58e48/examples/nlp/language_modeling/conf/megatron_gpt_config.yaml#L122
~Note that this optimizer is incompatible with the `megatron_amp_O2` option. It internally applies similar optimizations (bf16 params, wgrad kernel fusion), so refactoring may be warranted in the future.~ This optimizer requires the `megatron_amp_O2` option.
# 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?
- [x] 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**:
- [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
* Depends on https://github.com/NVIDIA/apex/pull/1414.
* Depends on https://github.com/NVIDIA/apex/pull/1432
* This is mostly orthogonal to https://github.com/NVIDIA/NeMo/pull/4380. It just requires changing [this line](https://github.com/NVIDIA/NeMo/blob/5bb47b7638645b6b177638fff4e3b3a82b3fd76c/nemo/collections/nlp/models/language_modeling/megatron_gpt_model.py#L335) to:
```if self.megatron_amp_o2 or self.with_distributed_adam:``` | {
"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/4487/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/4487/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4486 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4486/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4486/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4486/events | https://github.com/NVIDIA/NeMo/pull/4486 | 1,291,733,073 | PR_kwDOC_bI7s46ucvp | 4,486 | Merge r1.10.0 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
} | [] | closed | false | null | [] | null | null | 2022-07-01T18:49:03 | 2022-07-01T20:16:59 | 2022-07-01T20:16:56 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4486",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4486",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4486.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4486.patch",
"merged_at": "2022-07-01T20:16:56"
} | # 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/4486/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/4486/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4485 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4485/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4485/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4485/events | https://github.com/NVIDIA/NeMo/issues/4485 | 1,291,712,280 | I_kwDOC_bI7s5M_fcY | 4,485 | How to start training again from last checkpoint | {
"login": "FatimaArshad-DS",
"id": 78070220,
"node_id": "MDQ6VXNlcjc4MDcwMjIw",
"avatar_url": "https://avatars.githubusercontent.com/u/78070220?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/FatimaArshad-DS",
"html_url": "https://github.com/FatimaArshad-DS",
"followers_url": "https://api.github.com/users/FatimaArshad-DS/followers",
"following_url": "https://api.github.com/users/FatimaArshad-DS/following{/other_user}",
"gists_url": "https://api.github.com/users/FatimaArshad-DS/gists{/gist_id}",
"starred_url": "https://api.github.com/users/FatimaArshad-DS/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/FatimaArshad-DS/subscriptions",
"organizations_url": "https://api.github.com/users/FatimaArshad-DS/orgs",
"repos_url": "https://api.github.com/users/FatimaArshad-DS/repos",
"events_url": "https://api.github.com/users/FatimaArshad-DS/events{/privacy}",
"received_events_url": "https://api.github.com/users/FatimaArshad-DS/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 | true | null | [] | null | null | 2022-07-01T18:22:00 | 2022-07-01T22:30:00 | 2022-07-01T22:30:00 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | Hi,
My training got interrupted and Im trying to restart training from last checkpoint. However, training starts from the beginning. How do I make it start from last checkpoint? | null | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4485/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/4485/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4484 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4484/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4484/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4484/events | https://github.com/NVIDIA/NeMo/pull/4484 | 1,291,703,277 | PR_kwDOC_bI7s46uWPV | 4,484 | Fix for TP=2,PP=2 decoding with megatron encoder-decoder models | {
"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
} | [] | closed | false | null | [] | null | null | 2022-07-01T18:11:15 | 2022-07-01T18:39:32 | 2022-07-01T18:39:28 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4484",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4484",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4484.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4484.patch",
"merged_at": "2022-07-01T18:39:28"
} | Signed-off-by: MaximumEntropy <[email protected]>
# What does this PR do ?
Fixes decoding for TP>1, PP>1 models
**Collection**: NLP
# Changelog
- Gather from pipeline model parallel ranks only instead of all model parallel ranks.
# Usage
* N/A
# 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/4484/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/4484/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4482 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4482/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4482/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4482/events | https://github.com/NVIDIA/NeMo/issues/4482 | 1,291,673,043 | I_kwDOC_bI7s5M_V3T | 4,482 | Problems with phone type audios | {
"login": "evilc3",
"id": 19889717,
"node_id": "MDQ6VXNlcjE5ODg5NzE3",
"avatar_url": "https://avatars.githubusercontent.com/u/19889717?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/evilc3",
"html_url": "https://github.com/evilc3",
"followers_url": "https://api.github.com/users/evilc3/followers",
"following_url": "https://api.github.com/users/evilc3/following{/other_user}",
"gists_url": "https://api.github.com/users/evilc3/gists{/gist_id}",
"starred_url": "https://api.github.com/users/evilc3/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/evilc3/subscriptions",
"organizations_url": "https://api.github.com/users/evilc3/orgs",
"repos_url": "https://api.github.com/users/evilc3/repos",
"events_url": "https://api.github.com/users/evilc3/events{/privacy}",
"received_events_url": "https://api.github.com/users/evilc3/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | true | null | [] | null | null | 2022-07-01T17:43:00 | 2022-07-01T18:10:29 | 2022-07-01T18:10:29 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | Hi, I hv trained a new conformer model.
For telephone audios this model performs worse than the pretrained model used for fine-tuning it.
For Other audios its ok
Error seen are quite odd instead of insertion or substitution all the errors deletion. It just doesn't transcribe parts of the audio that pretrain model does.
1. Changes made for training my model
- have changed tokens from 128 to 512 added. Puncts. (Same tok type used unigram).
- Trained on different dataset of size ~10k hrs. (Amond them fisher was the only telephone speech dataset)
I am unable to debug where/why this is happening
Possible ans
1. Turning data hd problems.
2. Not enough telephone type training data used. Eg ( asr is not able to transcribe becz the audio is noisy )
Can anyone please pont me in the write direction.
| null | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4482/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/4482/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4481 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4481/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4481/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4481/events | https://github.com/NVIDIA/NeMo/pull/4481 | 1,290,778,930 | PR_kwDOC_bI7s46rOtd | 4,481 | Option to disregard document boundaries for t5, bart, ul2 | {
"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
} | [] | closed | false | null | [] | null | null | 2022-07-01T01:04:17 | 2023-02-22T18:41:31 | 2022-07-22T22:16:54 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4481",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4481",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4481.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4481.patch",
"merged_at": "2022-07-22T22:16:53"
} | # What does this PR do ?
Adds the option to disregard document boundaries while constructing a training example for t5, bart, ul2.
This reduces the number of padding tokens in a batch for T5 from ~35% to ~4%.
**Collection**: NLP
# Changelog
- Changes `dataset_utils.py` for T5, BART, UL2 to use `_build_samples_mapping()` from `gpt_dataset.py`
- Switches between T5 style samples mapping construction and GPT samples mapping construction based on the respect_document_boundaries flag
# Usage
* set `model.data.respect_document_boundaries=True/False` in T5, BART, UL2
# 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/4481/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/4481/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4480 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4480/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4480/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4480/events | https://github.com/NVIDIA/NeMo/pull/4480 | 1,290,710,108 | PR_kwDOC_bI7s46q_-b | 4,480 | Add support for ASR Adapter Auxiliary Losses | {
"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 | 2022-06-30T22:48:23 | 2022-07-12T02:05:07 | 2022-07-12T02:05:05 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4480",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4480",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4480.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4480.patch",
"merged_at": "2022-07-12T02:05:05"
} | # What does this PR do ?
Add support for auxiliary regularization losses to ASR adapters.
**Collection**: [ASR]
# Changelog
- Add support for access mixin to all adapter modules
- Add utility methods for gathering the loss from auxiliary models
- Add tests for adapter modules
# Usage
```python
# Currently, setting the config value of `l2_lambda` for the Adapter strategy adds auxiliary losses.
python train_adapters.py \
... \
model.adapter.adapter_strategy.l2_lambda=0.001
```
# 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)
- [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**:
- [x] New Feature
- [ ] 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/4480/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/4480/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4479 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4479/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4479/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4479/events | https://github.com/NVIDIA/NeMo/pull/4479 | 1,290,629,587 | PR_kwDOC_bI7s46qukt | 4,479 | Add kwargs to exact string match | {
"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
} | [] | closed | false | null | [] | null | null | 2022-06-30T20:52:11 | 2022-06-30T22:06:51 | 2022-06-30T22:06:48 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4479",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4479",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4479.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4479.patch",
"merged_at": "2022-06-30T22:06:48"
} | Signed-off-by: MaximumEntropy <[email protected]>
# What does this PR do ?
Exact string match fix for finetuning.
**Collection**: NLP
# Changelog
- Exact string match fix for fine-tuning by adding kwargs.
# 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/4479/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/4479/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4478 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4478/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4478/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4478/events | https://github.com/NVIDIA/NeMo/pull/4478 | 1,290,615,365 | PR_kwDOC_bI7s46qrek | 4,478 | [TTS][DE] created the finetuning Hifigan 44100Hz recipe on HUI-Audio-Corpus-German | {
"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 | 2022-06-30T20:35:49 | 2022-06-30T22:37:27 | 2022-06-30T22:37:24 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4478",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4478",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4478.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4478.patch",
"merged_at": "2022-06-30T22:37:24"
} |
* implemented a script of generating mel-spectrograms for finetuning Hifigan using multiprocessing.
* fixed some typos.
Signed-off-by: Xuesong Yang <[email protected]>
**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": "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/4478/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/4478/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4477 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4477/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4477/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4477/events | https://github.com/NVIDIA/NeMo/pull/4477 | 1,290,604,335 | PR_kwDOC_bI7s46qpCd | 4,477 | Fix exact string match metric instantiation | {
"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
} | [] | closed | false | null | [] | null | null | 2022-06-30T20:25:19 | 2023-02-22T18:46:50 | 2022-06-30T20:25:39 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4477",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4477",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4477.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4477.patch",
"merged_at": null
} | # What does this PR do ?
Fixes an issue when using exact string match with general T5 finetuning
**Collection**: NLP
# Changelog
- Fixes exact string match metric instantiation by adding **kwargs to exact string match.
# 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/4477/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/4477/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4476 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4476/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4476/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4476/events | https://github.com/NVIDIA/NeMo/pull/4476 | 1,290,507,480 | PR_kwDOC_bI7s46qT9F | 4,476 | added MLM Scoring | {
"login": "yzhang123",
"id": 4204271,
"node_id": "MDQ6VXNlcjQyMDQyNzE=",
"avatar_url": "https://avatars.githubusercontent.com/u/4204271?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/yzhang123",
"html_url": "https://github.com/yzhang123",
"followers_url": "https://api.github.com/users/yzhang123/followers",
"following_url": "https://api.github.com/users/yzhang123/following{/other_user}",
"gists_url": "https://api.github.com/users/yzhang123/gists{/gist_id}",
"starred_url": "https://api.github.com/users/yzhang123/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/yzhang123/subscriptions",
"organizations_url": "https://api.github.com/users/yzhang123/orgs",
"repos_url": "https://api.github.com/users/yzhang123/repos",
"events_url": "https://api.github.com/users/yzhang123/events{/privacy}",
"received_events_url": "https://api.github.com/users/yzhang123/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | null | 2022-06-30T18:40:22 | 2022-07-15T19:00:56 | 2022-07-15T19:00:50 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4476",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4476",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4476.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4476.patch",
"merged_at": "2022-07-15T19:00:50"
} | WFST+LM scoring with @ekmb and @anand joseph
Signed-off-by: Yang Zhang <[email protected]>
| {
"login": "yzhang123",
"id": 4204271,
"node_id": "MDQ6VXNlcjQyMDQyNzE=",
"avatar_url": "https://avatars.githubusercontent.com/u/4204271?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/yzhang123",
"html_url": "https://github.com/yzhang123",
"followers_url": "https://api.github.com/users/yzhang123/followers",
"following_url": "https://api.github.com/users/yzhang123/following{/other_user}",
"gists_url": "https://api.github.com/users/yzhang123/gists{/gist_id}",
"starred_url": "https://api.github.com/users/yzhang123/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/yzhang123/subscriptions",
"organizations_url": "https://api.github.com/users/yzhang123/orgs",
"repos_url": "https://api.github.com/users/yzhang123/repos",
"events_url": "https://api.github.com/users/yzhang123/events{/privacy}",
"received_events_url": "https://api.github.com/users/yzhang123/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/4476/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/4476/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4475 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4475/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4475/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4475/events | https://github.com/NVIDIA/NeMo/issues/4475 | 1,290,374,421 | I_kwDOC_bI7s5M6Y0V | 4,475 | Has Biomegatron been removed from huggingface.co/models? | {
"login": "jiwonjoung",
"id": 21979456,
"node_id": "MDQ6VXNlcjIxOTc5NDU2",
"avatar_url": "https://avatars.githubusercontent.com/u/21979456?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jiwonjoung",
"html_url": "https://github.com/jiwonjoung",
"followers_url": "https://api.github.com/users/jiwonjoung/followers",
"following_url": "https://api.github.com/users/jiwonjoung/following{/other_user}",
"gists_url": "https://api.github.com/users/jiwonjoung/gists{/gist_id}",
"starred_url": "https://api.github.com/users/jiwonjoung/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/jiwonjoung/subscriptions",
"organizations_url": "https://api.github.com/users/jiwonjoung/orgs",
"repos_url": "https://api.github.com/users/jiwonjoung/repos",
"events_url": "https://api.github.com/users/jiwonjoung/events{/privacy}",
"received_events_url": "https://api.github.com/users/jiwonjoung/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 | 2022-06-30T16:22:21 | 2022-09-29T04:39:25 | 2022-09-29T04:39:25 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | I can't seem to get nemo to recognize biomegatron as a pretrained model in the config file. For example, if I put:
`config.model.language_model.pretrained_model_name = 'biomegatron345m_biovocab_30k_cased'` in my config file, as per (https://github.com/NVIDIA/NeMo/blob/main/tutorials/nlp/Token_Classification-BioMegatron.ipynb). I get this error:
```
OSError: biomegatron345m_biovocab_30k_cased is not a local folder and is not a valid model identifier listed on 'https://huggingface.co/models'
If this is a private repository, make sure to pass a token having permission to this repo with `use_auth_token` or log in with `huggingface-cli login` and pass `use_auth_token=True`.
```
And upon scrolling through the models on huggingface.co/models, I do not see a biomegatron model listed. | {
"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/4475/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/4475/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/4474 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4474/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4474/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4474/events | https://github.com/NVIDIA/NeMo/pull/4474 | 1,290,264,890 | PR_kwDOC_bI7s46pfrt | 4,474 | Weighted bucketing | {
"login": "tbartley94",
"id": 90423858,
"node_id": "MDQ6VXNlcjkwNDIzODU4",
"avatar_url": "https://avatars.githubusercontent.com/u/90423858?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/tbartley94",
"html_url": "https://github.com/tbartley94",
"followers_url": "https://api.github.com/users/tbartley94/followers",
"following_url": "https://api.github.com/users/tbartley94/following{/other_user}",
"gists_url": "https://api.github.com/users/tbartley94/gists{/gist_id}",
"starred_url": "https://api.github.com/users/tbartley94/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/tbartley94/subscriptions",
"organizations_url": "https://api.github.com/users/tbartley94/orgs",
"repos_url": "https://api.github.com/users/tbartley94/repos",
"events_url": "https://api.github.com/users/tbartley94/events{/privacy}",
"received_events_url": "https://api.github.com/users/tbartley94/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | null | 2022-06-30T15:01:14 | 2022-07-08T17:53:48 | 2022-07-08T17:53:48 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4474",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4474",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4474.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4474.patch",
"merged_at": "2022-07-08T17:53:47"
} | # What does this PR do ?
- Adds to ASR bucketing config to add weights to target buckets. To be used for upsampling specific datasets.
**Collection**: [Note which collection this PR will affect]
- ASR
# Changelog
- `collections/asr/data/audio_to_text_dataset.py` `get_tarred_dataset` function now checks if dataset config feature `bucketing_weights` is passed as a list in config. (Can be for either dataset but assumed for training in practice.)
- Added type checking to `get_tarred_dataset` to check that passed values are all ints. (Returns `ValueError` if not.)
- using `bucketing_weights`, the dataloader will append a bucket manifest x number of times, with x being the provided weight value.
# Usage
Given a set of bucketed datasets, add to a hydra config file the `bucket-weights` attribute (e.g.):
```
train_ds:
manifest_filepath: [[bucket1manifest],[bucket1manifest],[bucket3manifest],[bucket4manifest]]
bucketing_weights: [weight1, weight2, weight3, weight4]
```
Where `bucketing_weights` is a list of `int` of same length as number of buckets. For each weight `i`, the corresponding manifest will be appended to the dataset `i` number of times. For instance, given `bucketing_weights: [3,2,1,2]`, the resulting training set will contain 3 instance of the first bucket, 2 of the second, 1 of the third, and 2 of the final.
# Before your PR is "Ready for review"
**Pre checks**:
- [Y] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md)
- [N] Did you write any new necessary tests?
- [N] Did you add or update any necessary documentation?
- [N] 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**:
- [Y] 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": "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/4474/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/4474/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4472 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4472/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4472/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4472/events | https://github.com/NVIDIA/NeMo/pull/4472 | 1,290,085,502 | PR_kwDOC_bI7s46o42D | 4,472 | Allow Noam lr scheduler to run for more than max_steps | {
"login": "alancucki",
"id": 16889482,
"node_id": "MDQ6VXNlcjE2ODg5NDgy",
"avatar_url": "https://avatars.githubusercontent.com/u/16889482?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/alancucki",
"html_url": "https://github.com/alancucki",
"followers_url": "https://api.github.com/users/alancucki/followers",
"following_url": "https://api.github.com/users/alancucki/following{/other_user}",
"gists_url": "https://api.github.com/users/alancucki/gists{/gist_id}",
"starred_url": "https://api.github.com/users/alancucki/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/alancucki/subscriptions",
"organizations_url": "https://api.github.com/users/alancucki/orgs",
"repos_url": "https://api.github.com/users/alancucki/repos",
"events_url": "https://api.github.com/users/alancucki/events{/privacy}",
"received_events_url": "https://api.github.com/users/alancucki/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | null | 2022-06-30T12:44:52 | 2022-07-07T08:19:03 | 2022-07-07T08:19:02 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4472",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4472",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4472.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4472.patch",
"merged_at": "2022-07-07T08:19:02"
} | Signed-off-by: Adrian Lancucki <[email protected]>
# What does this PR do ?
Allows to train for more than `max_steps` without lowering lr to `min_lr`.
Noam lr scheduler should decay lr indefinitely. The current implementation puts a hard limit at `max_steps` - the number of intended training epochs - after which it returns `min_lr`. This is contrary to the definition in the paper. Upon resuming from a ckpt `max_steps` are loaded from the ckpt, and if the user wishes to train for more epochs, they will unexpectedly get `min_lr`.
**Collection**: [Note which collection this PR will affect]
N/A
# 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": "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/4472/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/4472/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/4471 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/4471/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/4471/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/4471/events | https://github.com/NVIDIA/NeMo/pull/4471 | 1,290,007,381 | PR_kwDOC_bI7s46on49 | 4,471 | [TTS] Fix dataset parameter typo on tacotron2 example yaml | {
"login": "saarus72",
"id": 22793884,
"node_id": "MDQ6VXNlcjIyNzkzODg0",
"avatar_url": "https://avatars.githubusercontent.com/u/22793884?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/saarus72",
"html_url": "https://github.com/saarus72",
"followers_url": "https://api.github.com/users/saarus72/followers",
"following_url": "https://api.github.com/users/saarus72/following{/other_user}",
"gists_url": "https://api.github.com/users/saarus72/gists{/gist_id}",
"starred_url": "https://api.github.com/users/saarus72/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/saarus72/subscriptions",
"organizations_url": "https://api.github.com/users/saarus72/orgs",
"repos_url": "https://api.github.com/users/saarus72/repos",
"events_url": "https://api.github.com/users/saarus72/events{/privacy}",
"received_events_url": "https://api.github.com/users/saarus72/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | 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 | 2022-06-30T11:40:17 | 2022-07-07T06:09:31 | 2022-07-07T06:09:31 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/4471",
"html_url": "https://github.com/NVIDIA/NeMo/pull/4471",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/4471.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/4471.patch",
"merged_at": "2022-07-07T06:09:31"
} | # What does this PR do ?
This PR fixes dataset parameter typo on tacotron2 example yaml
**Collection**: [TTS]
# Changelog
- Change "train_dataset" to "validation_datasets" on tacotron2.yaml conf example.
# 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? -> I checked that locally
- [ ] Did you add or update any necessary documentation? -> No documentation needed
- [ ] Does the PR affect components that are optional to install? (Ex: Numba, Pynini, Apex etc) -> It does not
- [ ] Reviewer: Does the PR have correct import guards for all optional libraries?
**PR Type**:
- [x] 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": "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/4471/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/4471/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.