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/8172 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8172/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8172/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8172/events | https://github.com/NVIDIA/NeMo/issues/8172 | 2,080,785,902 | I_kwDOC_bI7s58BkXu | 8,172 | speech_to_text_buffered_infer: How to use timestamps | {
"login": "MadMaxNice",
"id": 22424847,
"node_id": "MDQ6VXNlcjIyNDI0ODQ3",
"avatar_url": "https://avatars.githubusercontent.com/u/22424847?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/MadMaxNice",
"html_url": "https://github.com/MadMaxNice",
"followers_url": "https://api.github.com/users/MadMaxNice/followers",
"following_url": "https://api.github.com/users/MadMaxNice/following{/other_user}",
"gists_url": "https://api.github.com/users/MadMaxNice/gists{/gist_id}",
"starred_url": "https://api.github.com/users/MadMaxNice/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/MadMaxNice/subscriptions",
"organizations_url": "https://api.github.com/users/MadMaxNice/orgs",
"repos_url": "https://api.github.com/users/MadMaxNice/repos",
"events_url": "https://api.github.com/users/MadMaxNice/events{/privacy}",
"received_events_url": "https://api.github.com/users/MadMaxNice/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"Sadly those two do not support time stamp calculation. You can use a CTC model with NeMo Forced Aligners to get accurate timestamps",
"It's still not supported, as far as a I can tell.\n\nWhat's the current recommendation to use all (or most) features of https://github.com/NVIDIA/NeMo/blob/main/examples/asr/transcribe_speech.py when limited in VRAM?"
] | 2024-01-14T16:01:22 | 2024-11-26T15:21:27 | 2024-01-15T07:02:16 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | Hello!
How can timestamps be used when using **speech_to_text_buffered_infer_ctc.py** or **speech_to_text_buffered_infer_rnnt.py** ?
Apparently compute_timestamps is not used in the code
Thank you | {
"login": "MadMaxNice",
"id": 22424847,
"node_id": "MDQ6VXNlcjIyNDI0ODQ3",
"avatar_url": "https://avatars.githubusercontent.com/u/22424847?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/MadMaxNice",
"html_url": "https://github.com/MadMaxNice",
"followers_url": "https://api.github.com/users/MadMaxNice/followers",
"following_url": "https://api.github.com/users/MadMaxNice/following{/other_user}",
"gists_url": "https://api.github.com/users/MadMaxNice/gists{/gist_id}",
"starred_url": "https://api.github.com/users/MadMaxNice/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/MadMaxNice/subscriptions",
"organizations_url": "https://api.github.com/users/MadMaxNice/orgs",
"repos_url": "https://api.github.com/users/MadMaxNice/repos",
"events_url": "https://api.github.com/users/MadMaxNice/events{/privacy}",
"received_events_url": "https://api.github.com/users/MadMaxNice/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8172/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/8172/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/8171 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8171/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8171/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8171/events | https://github.com/NVIDIA/NeMo/issues/8171 | 2,080,779,360 | I_kwDOC_bI7s58Bixg | 8,171 | Speaker Diarization - Extracting speaker embeddings for labels from files with multiple speakers from Cluster Diarizer models | {
"login": "sunraymoonbeam",
"id": 98078901,
"node_id": "U_kgDOBdiQtQ",
"avatar_url": "https://avatars.githubusercontent.com/u/98078901?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sunraymoonbeam",
"html_url": "https://github.com/sunraymoonbeam",
"followers_url": "https://api.github.com/users/sunraymoonbeam/followers",
"following_url": "https://api.github.com/users/sunraymoonbeam/following{/other_user}",
"gists_url": "https://api.github.com/users/sunraymoonbeam/gists{/gist_id}",
"starred_url": "https://api.github.com/users/sunraymoonbeam/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/sunraymoonbeam/subscriptions",
"organizations_url": "https://api.github.com/users/sunraymoonbeam/orgs",
"repos_url": "https://api.github.com/users/sunraymoonbeam/repos",
"events_url": "https://api.github.com/users/sunraymoonbeam/events{/privacy}",
"received_events_url": "https://api.github.com/users/sunraymoonbeam/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4585431311,
"node_id": "LA_kwDOC_bI7s8AAAABEVAhDw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale",
"name": "stale",
"color": "ededed",
"default": false,
"description": null
}
] | closed | false | {
"login": "tango4j",
"id": 13211483,
"node_id": "MDQ6VXNlcjEzMjExNDgz",
"avatar_url": "https://avatars.githubusercontent.com/u/13211483?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/tango4j",
"html_url": "https://github.com/tango4j",
"followers_url": "https://api.github.com/users/tango4j/followers",
"following_url": "https://api.github.com/users/tango4j/following{/other_user}",
"gists_url": "https://api.github.com/users/tango4j/gists{/gist_id}",
"starred_url": "https://api.github.com/users/tango4j/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/tango4j/subscriptions",
"organizations_url": "https://api.github.com/users/tango4j/orgs",
"repos_url": "https://api.github.com/users/tango4j/repos",
"events_url": "https://api.github.com/users/tango4j/events{/privacy}",
"received_events_url": "https://api.github.com/users/tango4j/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"login": "tango4j",
"id": 13211483,
"node_id": "MDQ6VXNlcjEzMjExNDgz",
"avatar_url": "https://avatars.githubusercontent.com/u/13211483?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/tango4j",
"html_url": "https://github.com/tango4j",
"followers_url": "https://api.github.com/users/tango4j/followers",
"following_url": "https://api.github.com/users/tango4j/following{/other_user}",
"gists_url": "https://api.github.com/users/tango4j/gists{/gist_id}",
"starred_url": "https://api.github.com/users/tango4j/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/tango4j/subscriptions",
"organizations_url": "https://api.github.com/users/tango4j/orgs",
"repos_url": "https://api.github.com/users/tango4j/repos",
"events_url": "https://api.github.com/users/tango4j/events{/privacy}",
"received_events_url": "https://api.github.com/users/tango4j/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
},
{
"login": "nithinraok",
"id": 19668129,
"node_id": "MDQ6VXNlcjE5NjY4MTI5",
"avatar_url": "https://avatars.githubusercontent.com/u/19668129?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/nithinraok",
"html_url": "https://github.com/nithinraok",
"followers_url": "https://api.github.com/users/nithinraok/followers",
"following_url": "https://api.github.com/users/nithinraok/following{/other_user}",
"gists_url": "https://api.github.com/users/nithinraok/gists{/gist_id}",
"starred_url": "https://api.github.com/users/nithinraok/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/nithinraok/subscriptions",
"organizations_url": "https://api.github.com/users/nithinraok/orgs",
"repos_url": "https://api.github.com/users/nithinraok/repos",
"events_url": "https://api.github.com/users/nithinraok/events{/privacy}",
"received_events_url": "https://api.github.com/users/nithinraok/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
}
] | null | [
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"This issue was closed because it has been inactive for 7 days since being marked as stale.",
"the same question",
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"Hi @sunraymoonbeam @maxpain \r\nSorry for missing this question and not answering this in timely manner. \r\nMulti-scale diarization decoder (MSDD) model only supports `diar_infer_telephonic.yaml`.\r\nSo you cannot use `diar_infer_meeting.yaml` settings.\r\n\r\n- Question 1\r\nget_cluster_avg_embs_model() function is as the name suggests, it is designed for model training.\r\nget_cluster_avg_embs() is for inference case, so for your purpose I think using `get_cluster_avg_embs()` function makes more sense. \r\n\r\n- Question 2\r\nMSDD inference requires the Clustering Diarizer results, so MSDD inference takes more time than clustering diarizer.\r\nAlso, you will get different results depending on scale length, scale weights when you run clustering diarizer. \r\nCheck if you use the same Titanet model, the same scale lengths and scale weights. Then it would generate the same results.\r\n\r\n- Question 3 \r\nIf you want to use speaker embeddings for speaker verification, definitely go with the longer scales.\r\nHowever, to get the best results, extract TitaNet speaker embeddings from 5\\~10 seconds of audio.\r\nI recommend you to avoid using the averaged embeddings from diarization process (these are two short),\r\nand I suggest to use TitaNet stand alone model to generate speaker embedding from 5~10 seconds of audio from the final diarization results.",
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"As @sunraymoonbeam mentioned get_cluster_avg_embs returns emb_sess_avg_dict, a dictionary containing speaker mapping information and cluster-average speaker embedding vectors. The emb_sess_avg_dict keys appear to correspond to the scales (0, 1, 2, etc.), and the associated values are embeddings with a shape of (192, num_of_speakers). I'm also little confused by why the number of identified speakers in the prediction RTTM doesn't match the specified num_of_speakers, perhaps you can provide some insight into this!\n\nI am trying to find a match with my own speaker embeddings so I can assign a \"speaker\" to that embeddings. Why does the identified speakers in the prediction RTTM not match the specified num_of_speakers?"
] | 2024-01-14T15:43:32 | 2024-12-12T08:44:56 | 2024-08-14T23:59:10 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | Hello,
I'm currently engaged in diarization for meetings domains using the **ClusteringDiarizer** model from the pre-trained pipeline, along with the meeting configuration (_diar_infer_meeting.yaml_), which has proven quite effective for my use case, just relying on listening by ear. Thanks for creating such a great library! My goal is to acquire cluster-averaged speaker embeddings for each speaker based on the clustering results. However, I'm facing challenges in finding a direct method to retrieve these embeddings from the **ClusteringDiarizer** module.
I came across a related issue (#6965) for **EncDecDiarLabelModel** (MSDD) models, where it was recommended to obtain the [ms_avg_embs](https://github.com/NVIDIA/NeMo/blob/5a5c77aab6a2568987a19ad78299c3c57bcc9766/nemo/collections/asr/models/msdd_models.py#L417C24-L417C79) variable from the [MSDD forward function](https://github.com/NVIDIA/NeMo/blob/5a5c77aab6a2568987a19ad78299c3c57bcc9766/nemo/collections/asr/models/msdd_models.py#L565), which utilizes the _get_cluster_avg_embs_model_ function.
### Problem with MSDD model
Thus, I tried to use the MSDD model instead of the ClusteringDiarizer model instead. However, I am facing some challenges as the MSDD model for the meeting domain has not been released (#6748). When attempting to load the MSDD model with the meeting configuration (diar_infer_meeting.yaml) and the "diar_msdd_telephonic" model path, I encounter a size mismatch error after running diarize(). This is likely due to differences in parameters and configurations between domains. I could train a MSDD model on a meeting domain corpus with the correct configurations but that would take some time and effort
### Workaround with get_cluster_avg_embs method
As a workaround, I employed the following approach:
```
msdd_model.clustering_embedding.prepare_cluster_embs_infer()
msdd_model.clustering_embedding.emb_sess_test_dict
```
I opted for the [_prepare_cluster_embs_infer()_](https://github.com/NVIDIA/NeMo/blob/5a5c77aab6a2568987a19ad78299c3c57bcc9766/nemo/collections/asr/models/msdd_models.py#L694C1-L701C10) method, which internally calls the [_run_clustering_diarizer()_](https://github.com/NVIDIA/NeMo/blob/5a5c77aab6a2568987a19ad78299c3c57bcc9766/nemo/collections/asr/models/msdd_models.py#L825C14-L825C14) method, and, in turn, uses the [**_get_cluster_avg_embs_**](https://github.com/NVIDIA/NeMo/blob/5a5c77aab6a2568987a19ad78299c3c57bcc9766/nemo/collections/asr/models/msdd_models.py#L764) method. I then accessed msdd_model.clustering_embedding.emb_sess_test_dict to retrieve cluster average embeddings. Essentially, I am running the clustering diarizer step with the MSDD model without feeding the output into the neural diarizer model.
I have 3 questions:
### Question 1: get_cluster_avg_embs vs get_cluster_avg_embs_model method: Correct Approach?
I would like your thoughts on this approach for obtaining cluster average embeddings and whether it is the correct strategy. I'm unclear about the distinctions between [**_get_cluster_avg_embs_**](https://github.com/NVIDIA/NeMo/blob/5a5c77aab6a2568987a19ad78299c3c57bcc9766/nemo/collections/asr/models/msdd_models.py#L764) and [**_get_cluster_avg_embs_model_**](https://github.com/NVIDIA/NeMo/blob/5a5c77aab6a2568987a19ad78299c3c57bcc9766/nemo/collections/asr/models/msdd_models.py#L384). The former is used by prepare_cluster_embs_infer(), and the latter is used in the MSDD forward function. Do they essentially return the same information? The documentation mentions that **_get_cluster_avg_embs_** returns _emb_sess_avg_dict_, a dictionary containing speaker mapping information and cluster-average speaker embedding vectors. The emb_sess_avg_dict keys appear to correspond to the scales (0, 1, 2, etc.), and the associated values are embeddings with a shape of (192, num_of_speakers). I'm also little confused by why the number of identified speakers in the prediction RTTM doesn't match the specified num_of_speakers, perhaps you can provide some insight into this!
### Question 2: Obtaining cluster averaged embeddings from the ClusteringDiarizer model instead of MSDD
I'm also curious if there's a more direct way to retrieve embeddings from the clustering diarizer module without relying on this workaround with the MSDD model. The outcomes from the clustering diarizer module, using both diarize() and the MSDD model's .prepare_cluster_embs_infer(), exhibit slight differences (4 vs 5 speakers, the statistics were roughly the same but it seemed that two clusters got merged where one of the cluster was very small in size), which is quite confusing to me as I expected them to utilize the same methods. The MSDD model also seems to run much slower then the ClusteringDiarizer Model.
### Question 3: Best representation of embeddings for each speaker
Lastly, I'm seeking guidance on obtaining the most reliable representation of the representative embeddings for each speaker, considering multiple scales. My assumption is that a straightforward approach would be to calculate the average of all the embeddings for each scale. But it is mentioned that decisions are made ultimately on the base scale....if I would like to compare embeddings for speaker recognition or visualise them on a plot, is average or taking the base scale better?
Thank you once again!
Zack | {
"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/8171/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/8171/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/8170 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8170/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8170/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8170/events | https://github.com/NVIDIA/NeMo/pull/8170 | 2,080,487,639 | PR_kwDOC_bI7s5kAvT5 | 8,170 | Fix FSDP SHARP integration | {
"login": "janEbert",
"id": 12694897,
"node_id": "MDQ6VXNlcjEyNjk0ODk3",
"avatar_url": "https://avatars.githubusercontent.com/u/12694897?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/janEbert",
"html_url": "https://github.com/janEbert",
"followers_url": "https://api.github.com/users/janEbert/followers",
"following_url": "https://api.github.com/users/janEbert/following{/other_user}",
"gists_url": "https://api.github.com/users/janEbert/gists{/gist_id}",
"starred_url": "https://api.github.com/users/janEbert/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/janEbert/subscriptions",
"organizations_url": "https://api.github.com/users/janEbert/orgs",
"repos_url": "https://api.github.com/users/janEbert/repos",
"events_url": "https://api.github.com/users/janEbert/events{/privacy}",
"received_events_url": "https://api.github.com/users/janEbert/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4585431311,
"node_id": "LA_kwDOC_bI7s8AAAABEVAhDw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale",
"name": "stale",
"color": "ededed",
"default": false,
"description": null
},
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"This PR is stale because it has been open for 14 days with no activity. Remove stale label or comment or update or this will be closed in 7 days.",
"Bla.",
"This PR is stale because it has been open for 14 days with no activity. Remove stale label or comment or update or this will be closed in 7 days.",
"Bla.",
"This PR is stale because it has been open for 14 days with no activity. Remove stale label or comment or update or this will be closed in 7 days.",
"Bla.",
"This PR is stale because it has been open for 14 days with no activity. Remove stale label or comment or update or this will be closed in 7 days.",
"Bla.",
"This PR is stale because it has been open for 14 days with no activity. Remove stale label or comment or update or this will be closed in 7 days.",
"Bla.",
"This PR is stale because it has been open for 14 days with no activity. Remove stale label or comment or update or this will be closed in 7 days.",
"Bla.",
"Independently fixed in dd69c7a7d."
] | 2024-01-13T21:50:47 | 2024-05-29T14:17:24 | 2024-05-29T12:06:10 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8170",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8170",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8170.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8170.patch",
"merged_at": null
} | The attribute was never implemented.
# What does this PR do ?
For NeMo NLP, the FSDP sharding strategy cannot be used currently because the `sharp` attribute for the `NLPFSDPStrategy` was not actually implemented in https://github.com/NVIDIA/NeMo/pull/7793, even though it is queried.
**Collection**:
- NLP
# Changelog
- Implement missing `NLPFSDPStrategy.sharp` attribute.
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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.
# Additional Information
Related to #7793 (pull request)
| {
"login": "janEbert",
"id": 12694897,
"node_id": "MDQ6VXNlcjEyNjk0ODk3",
"avatar_url": "https://avatars.githubusercontent.com/u/12694897?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/janEbert",
"html_url": "https://github.com/janEbert",
"followers_url": "https://api.github.com/users/janEbert/followers",
"following_url": "https://api.github.com/users/janEbert/following{/other_user}",
"gists_url": "https://api.github.com/users/janEbert/gists{/gist_id}",
"starred_url": "https://api.github.com/users/janEbert/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/janEbert/subscriptions",
"organizations_url": "https://api.github.com/users/janEbert/orgs",
"repos_url": "https://api.github.com/users/janEbert/repos",
"events_url": "https://api.github.com/users/janEbert/events{/privacy}",
"received_events_url": "https://api.github.com/users/janEbert/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8170/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/8170/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8169 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8169/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8169/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8169/events | https://github.com/NVIDIA/NeMo/pull/8169 | 2,080,468,285 | PR_kwDOC_bI7s5kAroy | 8,169 | remove more deprecated files | {
"login": "arendu",
"id": 108822655,
"node_id": "U_kgDOBnyAfw",
"avatar_url": "https://avatars.githubusercontent.com/u/108822655?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/arendu",
"html_url": "https://github.com/arendu",
"followers_url": "https://api.github.com/users/arendu/followers",
"following_url": "https://api.github.com/users/arendu/following{/other_user}",
"gists_url": "https://api.github.com/users/arendu/gists{/gist_id}",
"starred_url": "https://api.github.com/users/arendu/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/arendu/subscriptions",
"organizations_url": "https://api.github.com/users/arendu/orgs",
"repos_url": "https://api.github.com/users/arendu/repos",
"events_url": "https://api.github.com/users/arendu/events{/privacy}",
"received_events_url": "https://api.github.com/users/arendu/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"jenkins"
] | 2024-01-13T20:41:00 | 2024-01-18T02:53:45 | 2024-01-18T02:53:44 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8169",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8169",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8169.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8169.patch",
"merged_at": "2024-01-18T02:53:44"
} | # What does this PR do ?
removes deprecated scripts for peft t5 tuning.
**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
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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": "cuichenx",
"id": 43478052,
"node_id": "MDQ6VXNlcjQzNDc4MDUy",
"avatar_url": "https://avatars.githubusercontent.com/u/43478052?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/cuichenx",
"html_url": "https://github.com/cuichenx",
"followers_url": "https://api.github.com/users/cuichenx/followers",
"following_url": "https://api.github.com/users/cuichenx/following{/other_user}",
"gists_url": "https://api.github.com/users/cuichenx/gists{/gist_id}",
"starred_url": "https://api.github.com/users/cuichenx/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/cuichenx/subscriptions",
"organizations_url": "https://api.github.com/users/cuichenx/orgs",
"repos_url": "https://api.github.com/users/cuichenx/repos",
"events_url": "https://api.github.com/users/cuichenx/events{/privacy}",
"received_events_url": "https://api.github.com/users/cuichenx/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8169/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/8169/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8168 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8168/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8168/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8168/events | https://github.com/NVIDIA/NeMo/issues/8168 | 2,080,252,923 | I_kwDOC_bI7s57_iP7 | 8,168 | Error in exporting the speaker verification model to ONNX | {
"login": "csukuangfj",
"id": 5284924,
"node_id": "MDQ6VXNlcjUyODQ5MjQ=",
"avatar_url": "https://avatars.githubusercontent.com/u/5284924?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/csukuangfj",
"html_url": "https://github.com/csukuangfj",
"followers_url": "https://api.github.com/users/csukuangfj/followers",
"following_url": "https://api.github.com/users/csukuangfj/following{/other_user}",
"gists_url": "https://api.github.com/users/csukuangfj/gists{/gist_id}",
"starred_url": "https://api.github.com/users/csukuangfj/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/csukuangfj/subscriptions",
"organizations_url": "https://api.github.com/users/csukuangfj/orgs",
"repos_url": "https://api.github.com/users/csukuangfj/repos",
"events_url": "https://api.github.com/users/csukuangfj/events{/privacy}",
"received_events_url": "https://api.github.com/users/csukuangfj/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 | [
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"This issue was closed because it has been inactive for 7 days since being marked as stale.",
"it is not stale.",
"reopen"
] | 2024-01-13T11:36:05 | 2024-02-21T02:16:30 | 2024-02-21T01:44:03 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | The following code
```python3
import torch
import nemo.collections.asr as nemo_asr
speaker_model = nemo_asr.models.EncDecSpeakerLabelModel.from_pretrained(model_name="ecapa_tdnn")
with torch.no_grad():
speaker_model.eval()
speaker_model.export('model.onnx')
```
fails to export the model using the latest master and it throws the following error logs:
```
[NeMo I 2024-01-13 11:35:35 features:289] PADDING: 16
[NeMo I 2024-01-13 11:35:36 save_restore_connector:249] Model EncDecSpeakerLabelModel was successfully restored from /root/.cache/torch/NeMo/NeMo_1.21.0rc0/ecapa_tdnn/3e0c5c4731b176aeb70c29a74d800c81/ecapa_tdnn.nemo.
---------------------------------------------------------------------------
TypeError Traceback (most recent call last)
[<ipython-input-2-3f0cb43be965>](https://localhost:8080/#) in <cell line: 6>()
6 with torch.no_grad():
7 speaker_model.eval()
----> 8 speaker_model.export('model.onnx')
2 frames
[/usr/local/lib/python3.10/dist-packages/nemo/utils/export_utils.py](https://localhost:8080/#) in parse_input_example(input_example)
116
117 def parse_input_example(input_example):
--> 118 input_list = list(input_example)
119 input_dict = {}
120 # process possible kwargs
TypeError: 'NoneType' object is not iterable
```
---
Please use the following colab notebook to reproduce it
https://colab.research.google.com/drive/1hQumtSeKdDMCupcUHgrEIcX67c3AW1bj?usp=sharing | {
"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/8168/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/8168/timeline | null | not_planned | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/8167 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8167/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8167/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8167/events | https://github.com/NVIDIA/NeMo/pull/8167 | 2,079,986,486 | PR_kwDOC_bI7s5j_H79 | 8,167 | ASR Transcription Refactor | {
"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
} | [
{
"id": 4811884691,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR",
"name": "ASR",
"color": "F3B33E",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"jenkins",
"jenkins",
"jenkins",
"jenkins",
"jenkins"
] | 2024-01-13T02:13:22 | 2024-02-06T21:12:14 | 2024-02-06T21:12:13 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8167",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8167",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8167.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8167.patch",
"merged_at": "2024-02-06T21:12:13"
} | # What does this PR do ?
Refactors the Transcribe API to be more configurable and reduce the duplication throughout multiple models
Note: The spec is optional to some classes since they are higher level models that operate on entire NeMo Models and as such would require a complete override of all aspects of this API.
**Collection**: [ASR]
# Changelog
- New `TranscriptionMixin` mixin class that holds the responsibility for providing unified transcription semantics to all classes that implement its abstract methods.
- Removal of `transcribe()` as a function of `ASRModel`.
- Reason: ASRModel has become a defacto base class for all models under the speech to text paradigm, which includes classification models, diarizers, SLU, and AST. However the term "transcribe" usually is reserved for basic ASR
- Subclasses must now chose to opt into transcription via the `TranscriptionMixin`
- Refactor all major classes that utilize the transcription API
- Add support for tensor (numpy ndarray, torch tensor) as input to transcribe() function. Can also accept list of numpy arrays to perform batch inference.
# Breaking Changes
The API mostly tries to be backward compatible, with the following exceptions:
- The first argument is no longer called `path2audio_files` and is now instead just `audio`. This may break cases where keyword argument is used to call `model.transcribe()`. This is because this API aims to unify audio filepath via list and pytorch / numpy tensor as input for single file transcription.
- Deprecation and removal of `logprobs` argument. You must use `return_hypothesis` argument now and access the logprobs via the `alignments` variable of the Hypothesis dataclass.
# Usage
## General Transcription
`model.transcribe()` remains the general API after refactor.
## Generator-based transcription
Additionall you can now treat the transcription engine as a generator (though you must explicitly pass a dataclass for the config of the transition engine in this case).
This might be useful in cases where lists of hundreds of thousands of samples are provided to the `audio` argument. Previously, all results would be hosted on CPU RAM (including full logprobs or hypothesis objects) and if the ram is insufficient it would OOM and the process would be killed.
Generator-based inference will make this a constant memory cost, both for the GPU and CPU.
```python
import nemo.collections.asr as nemo_asr
model = nemo_asr.models.ASRModel.from_pretrained("nvidia/stt_en_conformer_transducer_large")
config = nemo_asr.models.TranscribeConfig(batch_size=32)
generator = model.transcribe_generator(audio, config)
for processed_outputs in generator:
# process a batch of 32 results (or less if last batch does not contain 32 elements)
....
```
## Audio tensor inference
NeMo ASR now supports numpy or torch tensors as input to the transcribe function (also includes the ability to use list of tensors for batched inference).
Note: All numpy tensors are cast to torch.Tensor() (CPU) internally, then moved automatically to device of the model that called transcribe(). Movement to correct device is automatic for both numpy and pytorch tensors.
The API is uniform with filepath based inference:
```python
import torch
import soundfile as sf
from nemo.collections.asr.models import ASRModel
model = ASRModel.from_pretrained("stt_en_conformer_ctc_small")
model.eval()
# Load audio file
audio_file = os.path.join(test_data_dir, "asr", "train", "an4", "wav", "an46-mmap-b.wav")
audio, sr = sf.read(audio_file, dtype='float32')
audio_file_2 = os.path.join(test_data_dir, "asr", "train", "an4", "wav", "an104-mrcb-b.wav")
audio_2, sr = sf.read(audio_file_2, dtype='float32')
# Mix second audio to torch.tensor()
audio_2 = torch.from_numpy(audio_2)
# Numpy array + torch tensor mixed tensor input (for batched inference)
outputs = model.transcribe([audio, audio_2], batch_size=2)
```
# 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
| {
"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/8167/reactions",
"total_count": 2,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 2,
"eyes": 0
} | https://api.github.com/repos/NVIDIA/NeMo/issues/8167/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8166 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8166/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8166/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8166/events | https://github.com/NVIDIA/NeMo/pull/8166 | 2,079,809,980 | PR_kwDOC_bI7s5j-ijJ | 8,166 | Github Actions CICD init | {
"login": "pablo-garay",
"id": 7166088,
"node_id": "MDQ6VXNlcjcxNjYwODg=",
"avatar_url": "https://avatars.githubusercontent.com/u/7166088?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/pablo-garay",
"html_url": "https://github.com/pablo-garay",
"followers_url": "https://api.github.com/users/pablo-garay/followers",
"following_url": "https://api.github.com/users/pablo-garay/following{/other_user}",
"gists_url": "https://api.github.com/users/pablo-garay/gists{/gist_id}",
"starred_url": "https://api.github.com/users/pablo-garay/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/pablo-garay/subscriptions",
"organizations_url": "https://api.github.com/users/pablo-garay/orgs",
"repos_url": "https://api.github.com/users/pablo-garay/repos",
"events_url": "https://api.github.com/users/pablo-garay/events{/privacy}",
"received_events_url": "https://api.github.com/users/pablo-garay/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"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 | [] | 2024-01-12T22:23:56 | 2024-03-19T00:47:58 | 2024-03-19T00:47:57 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8166",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8166",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8166.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8166.patch",
"merged_at": "2024-03-19T00:47:57"
} | Signed-off-by: Pablo Garay <[email protected]>
# What does this PR do ?
Github Actions CICD init
| {
"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/8166/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/8166/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8165 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8165/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8165/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8165/events | https://github.com/NVIDIA/NeMo/pull/8165 | 2,079,803,551 | PR_kwDOC_bI7s5j-hL0 | 8,165 | Github Actions CICD init | {
"login": "pablo-garay",
"id": 7166088,
"node_id": "MDQ6VXNlcjcxNjYwODg=",
"avatar_url": "https://avatars.githubusercontent.com/u/7166088?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/pablo-garay",
"html_url": "https://github.com/pablo-garay",
"followers_url": "https://api.github.com/users/pablo-garay/followers",
"following_url": "https://api.github.com/users/pablo-garay/following{/other_user}",
"gists_url": "https://api.github.com/users/pablo-garay/gists{/gist_id}",
"starred_url": "https://api.github.com/users/pablo-garay/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/pablo-garay/subscriptions",
"organizations_url": "https://api.github.com/users/pablo-garay/orgs",
"repos_url": "https://api.github.com/users/pablo-garay/repos",
"events_url": "https://api.github.com/users/pablo-garay/events{/privacy}",
"received_events_url": "https://api.github.com/users/pablo-garay/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"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 | [] | 2024-01-12T22:16:23 | 2024-01-12T22:25:21 | 2024-01-12T22:24:11 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8165",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8165",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8165.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8165.patch",
"merged_at": null
} | # What does this PR do ?
Github Actions CICD init
| {
"login": "pablo-garay",
"id": 7166088,
"node_id": "MDQ6VXNlcjcxNjYwODg=",
"avatar_url": "https://avatars.githubusercontent.com/u/7166088?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/pablo-garay",
"html_url": "https://github.com/pablo-garay",
"followers_url": "https://api.github.com/users/pablo-garay/followers",
"following_url": "https://api.github.com/users/pablo-garay/following{/other_user}",
"gists_url": "https://api.github.com/users/pablo-garay/gists{/gist_id}",
"starred_url": "https://api.github.com/users/pablo-garay/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/pablo-garay/subscriptions",
"organizations_url": "https://api.github.com/users/pablo-garay/orgs",
"repos_url": "https://api.github.com/users/pablo-garay/repos",
"events_url": "https://api.github.com/users/pablo-garay/events{/privacy}",
"received_events_url": "https://api.github.com/users/pablo-garay/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8165/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/8165/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8164 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8164/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8164/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8164/events | https://github.com/NVIDIA/NeMo/pull/8164 | 2,079,709,976 | PR_kwDOC_bI7s5j-N46 | 8,164 | Revert "adding OnlineSampleMapping" | {
"login": "pablo-garay",
"id": 7166088,
"node_id": "MDQ6VXNlcjcxNjYwODg=",
"avatar_url": "https://avatars.githubusercontent.com/u/7166088?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/pablo-garay",
"html_url": "https://github.com/pablo-garay",
"followers_url": "https://api.github.com/users/pablo-garay/followers",
"following_url": "https://api.github.com/users/pablo-garay/following{/other_user}",
"gists_url": "https://api.github.com/users/pablo-garay/gists{/gist_id}",
"starred_url": "https://api.github.com/users/pablo-garay/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/pablo-garay/subscriptions",
"organizations_url": "https://api.github.com/users/pablo-garay/orgs",
"repos_url": "https://api.github.com/users/pablo-garay/repos",
"events_url": "https://api.github.com/users/pablo-garay/events{/privacy}",
"received_events_url": "https://api.github.com/users/pablo-garay/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | {
"login": "pablo-garay",
"id": 7166088,
"node_id": "MDQ6VXNlcjcxNjYwODg=",
"avatar_url": "https://avatars.githubusercontent.com/u/7166088?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/pablo-garay",
"html_url": "https://github.com/pablo-garay",
"followers_url": "https://api.github.com/users/pablo-garay/followers",
"following_url": "https://api.github.com/users/pablo-garay/following{/other_user}",
"gists_url": "https://api.github.com/users/pablo-garay/gists{/gist_id}",
"starred_url": "https://api.github.com/users/pablo-garay/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/pablo-garay/subscriptions",
"organizations_url": "https://api.github.com/users/pablo-garay/orgs",
"repos_url": "https://api.github.com/users/pablo-garay/repos",
"events_url": "https://api.github.com/users/pablo-garay/events{/privacy}",
"received_events_url": "https://api.github.com/users/pablo-garay/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"login": "pablo-garay",
"id": 7166088,
"node_id": "MDQ6VXNlcjcxNjYwODg=",
"avatar_url": "https://avatars.githubusercontent.com/u/7166088?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/pablo-garay",
"html_url": "https://github.com/pablo-garay",
"followers_url": "https://api.github.com/users/pablo-garay/followers",
"following_url": "https://api.github.com/users/pablo-garay/following{/other_user}",
"gists_url": "https://api.github.com/users/pablo-garay/gists{/gist_id}",
"starred_url": "https://api.github.com/users/pablo-garay/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/pablo-garay/subscriptions",
"organizations_url": "https://api.github.com/users/pablo-garay/orgs",
"repos_url": "https://api.github.com/users/pablo-garay/repos",
"events_url": "https://api.github.com/users/pablo-garay/events{/privacy}",
"received_events_url": "https://api.github.com/users/pablo-garay/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
}
] | null | [
"jenkins",
"Is there a CI test which uses context parallel? @pablo-garay @ericharper \r\ncc @michalivne ",
"jenkins",
"@xrennvidia yes but w/o CI its hard to debug/iterate on the PR.\r\n",
"jenkins"
] | 2024-01-12T20:59:14 | 2024-01-18T02:31:56 | 2024-01-13T23:41:20 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8164",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8164",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8164.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8164.patch",
"merged_at": "2024-01-13T23:41:20"
} | Reverts NVIDIA/NeMo#8137
Nemo pipeline is broken (finetune tests failing) due to this. Hence revert.
More context/details:
- two gpt3 fine-tunning tests failed on EOS (finetune.gpt3.126m_tp1_pp1_1node_100steps_squad and finetune.gpt3.400m_improved_tp1_pp1_1node_100steps_squad). Check out the error log, it’s not the fault of context parallelism. The error comes from dataset, which is introduced by this PR https://github.com/NVIDIA/NeMo/pull/8137
[10:43](https://nvidia.slack.com/archives/D062J0U5PJ8/p1705084988269489)
Above problematic PR was merged after my context parallelism PR. So for sanity check, I also ran a pipeline with the NeMo commit ID of context parallelism merging (i.e., without above problematic PR)
you can see those two fine-tuning test passed, so I think context parallelism code works fine | {
"login": "pablo-garay",
"id": 7166088,
"node_id": "MDQ6VXNlcjcxNjYwODg=",
"avatar_url": "https://avatars.githubusercontent.com/u/7166088?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/pablo-garay",
"html_url": "https://github.com/pablo-garay",
"followers_url": "https://api.github.com/users/pablo-garay/followers",
"following_url": "https://api.github.com/users/pablo-garay/following{/other_user}",
"gists_url": "https://api.github.com/users/pablo-garay/gists{/gist_id}",
"starred_url": "https://api.github.com/users/pablo-garay/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/pablo-garay/subscriptions",
"organizations_url": "https://api.github.com/users/pablo-garay/orgs",
"repos_url": "https://api.github.com/users/pablo-garay/repos",
"events_url": "https://api.github.com/users/pablo-garay/events{/privacy}",
"received_events_url": "https://api.github.com/users/pablo-garay/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8164/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/8164/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8163 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8163/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8163/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8163/events | https://github.com/NVIDIA/NeMo/pull/8163 | 2,079,508,277 | PR_kwDOC_bI7s5j9gsc | 8,163 | Upgrade to DLFW PyTorch 23.12 | {
"login": "ericharper",
"id": 11999610,
"node_id": "MDQ6VXNlcjExOTk5NjEw",
"avatar_url": "https://avatars.githubusercontent.com/u/11999610?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ericharper",
"html_url": "https://github.com/ericharper",
"followers_url": "https://api.github.com/users/ericharper/followers",
"following_url": "https://api.github.com/users/ericharper/following{/other_user}",
"gists_url": "https://api.github.com/users/ericharper/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ericharper/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ericharper/subscriptions",
"organizations_url": "https://api.github.com/users/ericharper/orgs",
"repos_url": "https://api.github.com/users/ericharper/repos",
"events_url": "https://api.github.com/users/ericharper/events{/privacy}",
"received_events_url": "https://api.github.com/users/ericharper/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 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 | [
"jenkins",
"jenkins"
] | 2024-01-12T19:02:06 | 2024-01-18T20:26:36 | 2024-01-18T20:26:35 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8163",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8163",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8163.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8163.patch",
"merged_at": "2024-01-18T20:26:35"
} | # What does this PR do ?
Updates the base pytorch container
**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
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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/8163/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/8163/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8162 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8162/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8162/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8162/events | https://github.com/NVIDIA/NeMo/pull/8162 | 2,079,389,716 | PR_kwDOC_bI7s5j9GLo | 8,162 | Add random_seed argument to generate | {
"login": "Kipok",
"id": 2354422,
"node_id": "MDQ6VXNlcjIzNTQ0MjI=",
"avatar_url": "https://avatars.githubusercontent.com/u/2354422?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Kipok",
"html_url": "https://github.com/Kipok",
"followers_url": "https://api.github.com/users/Kipok/followers",
"following_url": "https://api.github.com/users/Kipok/following{/other_user}",
"gists_url": "https://api.github.com/users/Kipok/gists{/gist_id}",
"starred_url": "https://api.github.com/users/Kipok/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/Kipok/subscriptions",
"organizations_url": "https://api.github.com/users/Kipok/orgs",
"repos_url": "https://api.github.com/users/Kipok/repos",
"events_url": "https://api.github.com/users/Kipok/events{/privacy}",
"received_events_url": "https://api.github.com/users/Kipok/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"jenkins",
"jenkins",
"jenkins"
] | 2024-01-12T17:53:52 | 2024-01-22T19:07:58 | 2024-01-22T19:07:58 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8162",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8162",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8162.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8162.patch",
"merged_at": "2024-01-22T19:07:58"
} | # What does this PR do ?
Add random_seed argument to generate. By default it does not set anything, so backward compatibility is preserved.
**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
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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": "Kipok",
"id": 2354422,
"node_id": "MDQ6VXNlcjIzNTQ0MjI=",
"avatar_url": "https://avatars.githubusercontent.com/u/2354422?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Kipok",
"html_url": "https://github.com/Kipok",
"followers_url": "https://api.github.com/users/Kipok/followers",
"following_url": "https://api.github.com/users/Kipok/following{/other_user}",
"gists_url": "https://api.github.com/users/Kipok/gists{/gist_id}",
"starred_url": "https://api.github.com/users/Kipok/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/Kipok/subscriptions",
"organizations_url": "https://api.github.com/users/Kipok/orgs",
"repos_url": "https://api.github.com/users/Kipok/repos",
"events_url": "https://api.github.com/users/Kipok/events{/privacy}",
"received_events_url": "https://api.github.com/users/Kipok/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8162/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/8162/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8161 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8161/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8161/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8161/events | https://github.com/NVIDIA/NeMo/issues/8161 | 2,079,258,558 | I_kwDOC_bI7s577ve- | 8,161 | TUTORIAL BUG ONLY: Missing model.eval() in the SapBERT implementations | {
"login": "LiuRossa",
"id": 108694783,
"node_id": "U_kgDOBnqM_w",
"avatar_url": "https://avatars.githubusercontent.com/u/108694783?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/LiuRossa",
"html_url": "https://github.com/LiuRossa",
"followers_url": "https://api.github.com/users/LiuRossa/followers",
"following_url": "https://api.github.com/users/LiuRossa/following{/other_user}",
"gists_url": "https://api.github.com/users/LiuRossa/gists{/gist_id}",
"starred_url": "https://api.github.com/users/LiuRossa/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/LiuRossa/subscriptions",
"organizations_url": "https://api.github.com/users/LiuRossa/orgs",
"repos_url": "https://api.github.com/users/LiuRossa/repos",
"events_url": "https://api.github.com/users/LiuRossa/events{/privacy}",
"received_events_url": "https://api.github.com/users/LiuRossa/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 | [
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"This issue was closed because it has been inactive for 7 days since being marked as stale."
] | 2024-01-12T16:42:29 | 2024-02-19T01:44:21 | 2024-02-19T01:44:21 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | **Describe the bug**
TUTORIAL BUG ONLY: Missing model.eval() in the SapBERT implementations.
There is a mistake causing inconsistent results in the NLP tutorials for at least two notebooks:
- In stable branch Nemo/tutorials/nlp/Entity_Linking_Medical.ipynb
- In v1.0.2 branch Nemo/tutorials/nlp/Entity_Linking_Medical.ipynb, which is referred by NeMo official tutorial https://docs.nvidia.com/deeplearning/nemo/user-guide/docs/en/main/nlp/entity_linking.html and SapBERT author in https://github.com/cambridgeltl/sapbert?tab=readme-ov-file
and should also be in the stable branch NeMo/examples/nlp/entity_linking in every .py file who use *model.forward(...)* function.
After the training part the model is not set to eval mode and thus the Dropout remains active.
This could be quite misleading for anyone who whishes to implement those methods.
**Steps/Code to reproduce bug**
1) Run the tutorial code for the aforementionned notebooks in Colab for instance (any setting).
2) Run the evaluation part twice: you get different results.
**Wrong results printed in Notebook**
Once run the block Model Evaluation, will see:
| Model| 1|5|
| ------ | ------ | ------ |
| BERT Base Baseline | 0.700000 |1.000000|
| BERT + SAP | 0.800000 |1.000000|
Run this block at twice, will see:
| Model| 1|5|
| ------ | ------ | ------ |
| BERT Base Baseline | 0.700000 |1.000000|
| BERT + SAP | 0.700000 |0.800000|
The second line can be any other proportion possible.
In Entity Linking via Nearest Neighbor Search,
The similar score of SapBERT and most similar concepts is always changing even if the model is always the same one.
**Expected behavior**
The evaluation of the model should give consistent results.
**One solution suggestion**
Set model mode to *eval* each time at evaluation and restore the initial mode at the end of evaluation.
Replace the function *get_embeddings(model, dataloader)* by:
```
# Helper function to get data embeddings
def get_embeddings(model, dataloader):
mode = model.training
embeddings, cids = [], []
with torch.no_grad():
model.eval()
for batch in tqdm(dataloader):
input_ids, token_type_ids, attention_mask, batch_cids = batch
batch_embeddings = model.forward(input_ids=input_ids.to(device),
token_type_ids=token_type_ids.to(device),
attention_mask=attention_mask.to(device))
# Accumulate index embeddings and their corresponding IDs
embeddings.extend(batch_embeddings.cpu().detach().numpy())
cids.extend(batch_cids)
# set mode back to its original value
model.train(mode=mode)
return embeddings, cids
```
It resolve the problem and the output of SapBERT is fixed.
**Additional comment**
We think this bug always exist as this project is created. And the author has observed this mistaking result. But he wrote
````
This evaluation set contains very little data, and no serious conclusions should be drawn about model performance. Top 1 accuracy should be between 0.7 and 1.0 for both models and top 5 accuracy should be between 0.8 and 1.0.
````
in the section of Model Evaluation.
Actually, after done the debug, the result (on our side) is
| Model| 1|5|
| ------ | ------ | ------ |
| BERT Base Baseline | 0.700000 |1.000000|
| BERT + SAP | 0.900000 |1.000000|
BERT + SAP acts always better than BERT Base Baseline even thought with the tiny example dataset.
**Environment overview **
- Environment location: Colab
- Method of NeMo install: pip install
```
BRANCH = 'r1.22.0'
!python -m pip install git+https://github.com/NVIDIA/NeMo.git@$BRANCH#egg=nemo_toolkit[all]
```
**Environment details**
- OS version: Linux-6.1.58+-x86_64-with-glibc2.35
- PyTorch version: 2.1.0+cu121
- Python version: 3.10.12
**Additional context**
Add any other context about the problem here.
Example: GPU model | {
"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/8161/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/8161/timeline | null | not_planned | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/8160 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8160/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8160/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8160/events | https://github.com/NVIDIA/NeMo/issues/8160 | 2,078,419,740 | I_kwDOC_bI7s574isc | 8,160 | Quartznet onnx model can't run on onnxruntime(CUDA backend) benchmark | {
"login": "wwiwush",
"id": 58901796,
"node_id": "MDQ6VXNlcjU4OTAxNzk2",
"avatar_url": "https://avatars.githubusercontent.com/u/58901796?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/wwiwush",
"html_url": "https://github.com/wwiwush",
"followers_url": "https://api.github.com/users/wwiwush/followers",
"following_url": "https://api.github.com/users/wwiwush/following{/other_user}",
"gists_url": "https://api.github.com/users/wwiwush/gists{/gist_id}",
"starred_url": "https://api.github.com/users/wwiwush/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/wwiwush/subscriptions",
"organizations_url": "https://api.github.com/users/wwiwush/orgs",
"repos_url": "https://api.github.com/users/wwiwush/repos",
"events_url": "https://api.github.com/users/wwiwush/events{/privacy}",
"received_events_url": "https://api.github.com/users/wwiwush/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 | [
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"This issue was closed because it has been inactive for 7 days since being marked as stale."
] | 2024-01-12T09:24:54 | 2024-02-19T01:44:23 | 2024-02-19T01:44:23 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | The develop ENV is as follows:
cuda 11.8
cudnn 8.2.4
onnxruntime(cuda backend) 1.16.3
When I run my exported Quartznet onnx model on onnxruntime(CUDA backend) benchmark, the following error occurs:
[E:onnxruntime:benchmark, cuda_call.cc:116 CudaCall] CUDNN failure 3: CUDNN_STATUS_BAD_PARAM ; GPU=0 ; file=/onnxruntime_src/onnxruntime/contrib_ops/cuda/fused_conv.cc ; line=85 ; expr=cudnnAddTensor(cudnnHandle, &alpha, Base::s_.z_tensor, Base::s_.z_data, &alpha, Base::s_.y_tensor, Base::s_.y_data)
[E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running FusedConv node. Name:'Conv_41' Status Message: CUDNN failure 3: CUDNN_STATUS_BAD_PARAM ; GPU=0 ;file=/onnxruntime_src/onnxruntime/contrib_ops/cuda/fused_conv.cc ; line=85 ; expr=cudnnAddTensor(cudnnHandle, &alpha, Base::s_.z_tensor, Base::s_.z_data, &alpha, Base::s_.y_tensor, Base::s_.y_data);
terminate called after throwing an instance of 'Ort::Exception'
what(): Non-zero status code returned while running FusedConv node. Name:'Conv_41' Status Message: CUDNN failure 3: CUDNN_STATUS_BAD_PARAM ; GPU=0 ;file=/onnxruntime_src/onnxruntime/contrib_ops/cuda/fused_conv.cc ; line=85 ; expr=cudnnAddTensor(cudnnHandle, &alpha, Base::s_.z_tensor, Base::s_.z_data, &alpha, Base::s_.y_tensor, Base::s_.y_data);
Thank you very much for help!
| {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers",
"following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}",
"gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}",
"starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions",
"organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs",
"repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos",
"events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}",
"received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events",
"type": "Bot",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8160/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/8160/timeline | null | not_planned | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/8159 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8159/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8159/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8159/events | https://github.com/NVIDIA/NeMo/issues/8159 | 2,078,060,109 | I_kwDOC_bI7s573K5N | 8,159 | Knowledge distillation support for Nemo ASR models | {
"login": "nabil6391",
"id": 21167899,
"node_id": "MDQ6VXNlcjIxMTY3ODk5",
"avatar_url": "https://avatars.githubusercontent.com/u/21167899?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/nabil6391",
"html_url": "https://github.com/nabil6391",
"followers_url": "https://api.github.com/users/nabil6391/followers",
"following_url": "https://api.github.com/users/nabil6391/following{/other_user}",
"gists_url": "https://api.github.com/users/nabil6391/gists{/gist_id}",
"starred_url": "https://api.github.com/users/nabil6391/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/nabil6391/subscriptions",
"organizations_url": "https://api.github.com/users/nabil6391/orgs",
"repos_url": "https://api.github.com/users/nabil6391/repos",
"events_url": "https://api.github.com/users/nabil6391/events{/privacy}",
"received_events_url": "https://api.github.com/users/nabil6391/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4585431311,
"node_id": "LA_kwDOC_bI7s8AAAABEVAhDw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale",
"name": "stale",
"color": "ededed",
"default": false,
"description": null
}
] | closed | false | {
"login": "okuchaiev",
"id": 22035961,
"node_id": "MDQ6VXNlcjIyMDM1OTYx",
"avatar_url": "https://avatars.githubusercontent.com/u/22035961?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/okuchaiev",
"html_url": "https://github.com/okuchaiev",
"followers_url": "https://api.github.com/users/okuchaiev/followers",
"following_url": "https://api.github.com/users/okuchaiev/following{/other_user}",
"gists_url": "https://api.github.com/users/okuchaiev/gists{/gist_id}",
"starred_url": "https://api.github.com/users/okuchaiev/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/okuchaiev/subscriptions",
"organizations_url": "https://api.github.com/users/okuchaiev/orgs",
"repos_url": "https://api.github.com/users/okuchaiev/repos",
"events_url": "https://api.github.com/users/okuchaiev/events{/privacy}",
"received_events_url": "https://api.github.com/users/okuchaiev/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"login": "okuchaiev",
"id": 22035961,
"node_id": "MDQ6VXNlcjIyMDM1OTYx",
"avatar_url": "https://avatars.githubusercontent.com/u/22035961?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/okuchaiev",
"html_url": "https://github.com/okuchaiev",
"followers_url": "https://api.github.com/users/okuchaiev/followers",
"following_url": "https://api.github.com/users/okuchaiev/following{/other_user}",
"gists_url": "https://api.github.com/users/okuchaiev/gists{/gist_id}",
"starred_url": "https://api.github.com/users/okuchaiev/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/okuchaiev/subscriptions",
"organizations_url": "https://api.github.com/users/okuchaiev/orgs",
"repos_url": "https://api.github.com/users/okuchaiev/repos",
"events_url": "https://api.github.com/users/okuchaiev/events{/privacy}",
"received_events_url": "https://api.github.com/users/okuchaiev/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
}
] | null | [
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"I'm also interested in this topic. any update?",
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"This issue was closed because it has been inactive for 7 days since being marked as stale."
] | 2024-01-12T04:49:16 | 2024-03-29T01:44:06 | 2024-03-29T01:44:05 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | **Is your feature request related to a problem? Please describe.**
I am currently working with Nemo FastConformer Large model, but would like to deploy a smaller model on mobile devices. Instead of training from scratch, I would like to perform KD
**Describe the solution you'd like**
I would greatly appreciate it if the Nemo ASR team could implement Knowledge Distillation (KD) support or maybe guide me how to do it with a FastConformer Model :
1. Maybe an Easier API support for KD within the Nemo framework.
2. Documentation and examples demonstrating how to use KD for model compression.
3. Tools or guidelines for effectively transferring knowledge from large to small models.
**Describe alternatives you've considered**
As an alternative, I have considered manually implementing KD techniques using pytorch lightening as described in #1996. However I did not get the full picture.
Thank you for your consideration
| {
"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/8159/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/8159/timeline | null | not_planned | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/8158 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8158/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8158/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8158/events | https://github.com/NVIDIA/NeMo/pull/8158 | 2,077,917,291 | PR_kwDOC_bI7s5j4DYC | 8,158 | Update CONTRIBUTING.md | {
"login": "kit1980",
"id": 420184,
"node_id": "MDQ6VXNlcjQyMDE4NA==",
"avatar_url": "https://avatars.githubusercontent.com/u/420184?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/kit1980",
"html_url": "https://github.com/kit1980",
"followers_url": "https://api.github.com/users/kit1980/followers",
"following_url": "https://api.github.com/users/kit1980/following{/other_user}",
"gists_url": "https://api.github.com/users/kit1980/gists{/gist_id}",
"starred_url": "https://api.github.com/users/kit1980/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/kit1980/subscriptions",
"organizations_url": "https://api.github.com/users/kit1980/orgs",
"repos_url": "https://api.github.com/users/kit1980/repos",
"events_url": "https://api.github.com/users/kit1980/events{/privacy}",
"received_events_url": "https://api.github.com/users/kit1980/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | [] | 2024-01-12T01:47:07 | 2024-01-12T01:51:02 | 2024-01-12T01:51:02 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | true | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8158",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8158",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8158.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8158.patch",
"merged_at": null
} | # What does this PR do ?
Add a one line overview of what this PR aims to accomplish.
**Collection**: [Note which collection this PR will affect]
# Changelog
- Add specific line by line info of high level changes in this PR.
# Usage
* You can potentially add a usage example below
```python
# Add a code snippet demonstrating how to use this
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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": "kit1980",
"id": 420184,
"node_id": "MDQ6VXNlcjQyMDE4NA==",
"avatar_url": "https://avatars.githubusercontent.com/u/420184?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/kit1980",
"html_url": "https://github.com/kit1980",
"followers_url": "https://api.github.com/users/kit1980/followers",
"following_url": "https://api.github.com/users/kit1980/following{/other_user}",
"gists_url": "https://api.github.com/users/kit1980/gists{/gist_id}",
"starred_url": "https://api.github.com/users/kit1980/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/kit1980/subscriptions",
"organizations_url": "https://api.github.com/users/kit1980/orgs",
"repos_url": "https://api.github.com/users/kit1980/repos",
"events_url": "https://api.github.com/users/kit1980/events{/privacy}",
"received_events_url": "https://api.github.com/users/kit1980/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8158/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/8158/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8157 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8157/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8157/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8157/events | https://github.com/NVIDIA/NeMo/pull/8157 | 2,077,819,077 | PR_kwDOC_bI7s5j3tuT | 8,157 | MPT to NeMo conversion | {
"login": "ethanhe42",
"id": 10027339,
"node_id": "MDQ6VXNlcjEwMDI3MzM5",
"avatar_url": "https://avatars.githubusercontent.com/u/10027339?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ethanhe42",
"html_url": "https://github.com/ethanhe42",
"followers_url": "https://api.github.com/users/ethanhe42/followers",
"following_url": "https://api.github.com/users/ethanhe42/following{/other_user}",
"gists_url": "https://api.github.com/users/ethanhe42/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ethanhe42/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ethanhe42/subscriptions",
"organizations_url": "https://api.github.com/users/ethanhe42/orgs",
"repos_url": "https://api.github.com/users/ethanhe42/repos",
"events_url": "https://api.github.com/users/ethanhe42/events{/privacy}",
"received_events_url": "https://api.github.com/users/ethanhe42/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 | [
"@yaoyu-33 ",
"This PR is stale because it has been open for 14 days with no activity. Remove stale label or comment or update or this will be closed in 7 days.",
"This PR was closed because it has been inactive for 7 days since being marked as stale.",
"This PR is stale because it has been open for 14 days with no activity. Remove stale label or comment or update or this will be closed in 7 days.",
"This PR was closed because it has been inactive for 7 days since being marked as stale."
] | 2024-01-12T00:25:07 | 2024-03-10T01:44:57 | 2024-03-10T01:44:57 | MEMBER | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8157",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8157",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8157.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8157.patch",
"merged_at": null
} | This script converts mpt-7b to mcore (transformer engine).
two things need to be aware:
1. layernorm instead of Low precision layernorm is used
2. no bias layernorm is not supported by transformer engine as of now
to use
```
python scripts/checkpoint_converters/convert_mpt_7b_hf_to_nemo.py
```
text generation:
```
python examples/nlp/language_modeling/megatron_gpt_eval.py gpt_model_file=mpt_7b_mcore.nemo inference.greedy=True trainer.devices=1 trainer.num_nodes=1 tensor_model_parallel_size=1 pipeline_model_parallel_size=1 prompts="[query: how much protein should a female eat]"
```
example outputs:
```
[{'sentences': ['query: how much protein should a female eat?\nI am a female, 5\'4" and weigh about 130 lbs. I am a vegetarian and I am trying to lose weight. I'], 'tokens': [['<|endoftext|>', 'query', ':', 'Ġhow', 'Ġmuch', 'Ġprotein', 'Ġshould', 'Ġa', 'Ġfemale', 'Ġeat', '?', 'Ċ', 'I', 'Ġam', 'Ġa', 'Ġfemale', ',', 'Ġ5', "'", '4', '"', 'Ġand', 'Ġweigh', 'Ġabout', 'Ġ130', 'Ġlbs', '.', 'ĠI', 'Ġam', 'Ġa', 'Ġvegetarian', 'Ġand', 'ĠI', 'Ġam', 'Ġtrying', 'Ġto', 'Ġlose', 'Ġweight', '.', 'ĠI']], 'logprob': None, 'full_logprob': None, 'token_ids': [[0, 7267, 27, 849, 1199, 2601, 943, 247, 5343, 6008, 32, 187, 42, 717, 247, 5343, 13, 608, 8, 21, 3, 285, 14357, 670, 11084, 38818, 15, 309, 717, 247, 39203, 285, 309, 717, 2820, 281, 7168, 2801, 15, 309]], 'offsets': [[0, 0, 5, 6, 10, 15, 23, 30, 32, 39, 43, 44, 45, 46, 49, 51, 58, 59, 61, 62, 63, 64, 68, 74, 80, 84, 88, 89, 91, 94, 96, 107, 111, 113, 116, 123, 126, 131, 138, 139]]}]
```
This PR depends on https://github.com/NVIDIA/Megatron-LM/pull/668 | {
"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/8157/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/8157/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8156 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8156/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8156/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8156/events | https://github.com/NVIDIA/NeMo/pull/8156 | 2,077,624,196 | PR_kwDOC_bI7s5j3DAJ | 8,156 | Update dependencies | {
"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
} | [
{
"id": 4811884691,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR",
"name": "ASR",
"color": "F3B33E",
"default": false,
"description": ""
},
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"jenkins"
] | 2024-01-11T21:30:22 | 2024-01-12T21:40:18 | 2024-01-12T21:40:17 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8156",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8156",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8156.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8156.patch",
"merged_at": "2024-01-12T21:40:17"
} | # What does this PR do ?
Updates dependencies for NeMo
Fixes #8118
**Collection**: [Core, Common]
# Changelog
- Remove Numpy upperbound
- Remove pydantic pin < 2
# 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": "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/8156/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/8156/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8155 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8155/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8155/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8155/events | https://github.com/NVIDIA/NeMo/pull/8155 | 2,077,296,165 | PR_kwDOC_bI7s5j17BQ | 8,155 | Mixtral to NeMo conversion script. | {
"login": "akoumpa",
"id": 153118171,
"node_id": "U_kgDOCSBl2w",
"avatar_url": "https://avatars.githubusercontent.com/u/153118171?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/akoumpa",
"html_url": "https://github.com/akoumpa",
"followers_url": "https://api.github.com/users/akoumpa/followers",
"following_url": "https://api.github.com/users/akoumpa/following{/other_user}",
"gists_url": "https://api.github.com/users/akoumpa/gists{/gist_id}",
"starred_url": "https://api.github.com/users/akoumpa/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/akoumpa/subscriptions",
"organizations_url": "https://api.github.com/users/akoumpa/orgs",
"repos_url": "https://api.github.com/users/akoumpa/repos",
"events_url": "https://api.github.com/users/akoumpa/events{/privacy}",
"received_events_url": "https://api.github.com/users/akoumpa/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
},
{
"id": 4840216855,
"node_id": "LA_kwDOC_bI7s8AAAABIH_ZFw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/CI",
"name": "CI",
"color": "1D76DB",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"Could you provide the convert NeMo to huggingface script? ",
"@cdj0311 will try to have the NeMo to HF by next week",
"jenkins",
"jenkins",
"jenkins",
"jenkins",
"jenkins",
"jenkins",
"jenkins",
"jenkins",
"Hi @ericharper. I browsed CodeQL suggestions in this pull request. This is a good example when **Flake8** is better than CodeQL: CodeQL fails to detect undefined variables (a simple issue), see `head_num` [here](https://github.com/NVIDIA/NeMo/blob/6865c39d85c4f3b120a02fe74024ad10d5fab286/scripts/nlp_language_modeling/convert_nemo_mixtral_to_hf.py#L124), while Flake8 catches them. I strongly suggest to start using Flake8 complementary to CodeQL in CI. ",
"Another example are the unused imports `from megatron.core.utils import init_method_normal, scaled_init_method_normal` [here](https://github.com/NVIDIA/NeMo/blob/main/nemo/collections/nlp/models/language_modeling/megatron_gpt_model.py#L94). This file is frequently edited btw. Possibly CodeQL misses this due to import guards, but I'm not sure. cc @ericharper "
] | 2024-01-11T17:54:53 | 2024-02-12T12:35:04 | 2024-01-27T00:49:27 | MEMBER | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8155",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8155",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8155.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8155.patch",
"merged_at": "2024-01-27T00:49:27"
} | # 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
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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.
# 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/8155/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/8155/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8154 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8154/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8154/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8154/events | https://github.com/NVIDIA/NeMo/pull/8154 | 2,077,283,801 | PR_kwDOC_bI7s5j14QP | 8,154 | Update README.rst | {
"login": "fayejf",
"id": 36722593,
"node_id": "MDQ6VXNlcjM2NzIyNTkz",
"avatar_url": "https://avatars.githubusercontent.com/u/36722593?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/fayejf",
"html_url": "https://github.com/fayejf",
"followers_url": "https://api.github.com/users/fayejf/followers",
"following_url": "https://api.github.com/users/fayejf/following{/other_user}",
"gists_url": "https://api.github.com/users/fayejf/gists{/gist_id}",
"starred_url": "https://api.github.com/users/fayejf/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/fayejf/subscriptions",
"organizations_url": "https://api.github.com/users/fayejf/orgs",
"repos_url": "https://api.github.com/users/fayejf/repos",
"events_url": "https://api.github.com/users/fayejf/events{/privacy}",
"received_events_url": "https://api.github.com/users/fayejf/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | [] | 2024-01-11T17:48:10 | 2024-01-11T17:52:24 | 2024-01-11T17:52:23 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8154",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8154",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8154.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8154.patch",
"merged_at": "2024-01-11T17:52:23"
} | # 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
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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/8154/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/8154/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8153 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8153/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8153/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8153/events | https://github.com/NVIDIA/NeMo/pull/8153 | 2,076,279,660 | PR_kwDOC_bI7s5jyXIu | 8,153 | Avoid NFS consistency issues during ckpt dir creation | {
"login": "mikolajblaz",
"id": 9979479,
"node_id": "MDQ6VXNlcjk5Nzk0Nzk=",
"avatar_url": "https://avatars.githubusercontent.com/u/9979479?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/mikolajblaz",
"html_url": "https://github.com/mikolajblaz",
"followers_url": "https://api.github.com/users/mikolajblaz/followers",
"following_url": "https://api.github.com/users/mikolajblaz/following{/other_user}",
"gists_url": "https://api.github.com/users/mikolajblaz/gists{/gist_id}",
"starred_url": "https://api.github.com/users/mikolajblaz/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/mikolajblaz/subscriptions",
"organizations_url": "https://api.github.com/users/mikolajblaz/orgs",
"repos_url": "https://api.github.com/users/mikolajblaz/repos",
"events_url": "https://api.github.com/users/mikolajblaz/events{/privacy}",
"received_events_url": "https://api.github.com/users/mikolajblaz/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4585431311,
"node_id": "LA_kwDOC_bI7s8AAAABEVAhDw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale",
"name": "stale",
"color": "ededed",
"default": false,
"description": null
},
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"This PR is stale because it has been open for 14 days with no activity. Remove stale label or comment or update or this will be closed in 7 days.",
"@ericharper can you review?",
"This PR is stale because it has been open for 14 days with no activity. Remove stale label or comment or update or this will be closed in 7 days.",
"jenkins",
"This PR is stale because it has been open for 14 days with no activity. Remove stale label or comment or update or this will be closed in 7 days.",
"This PR was closed because it has been inactive for 7 days since being marked as stale."
] | 2024-01-11T10:44:34 | 2024-03-10T01:44:59 | 2024-03-10T01:44:58 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8153",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8153",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8153.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8153.patch",
"merged_at": null
} | # What does this PR do ?
On some NFS setups, the root checkpoint directory created on rank 0 was not visible to other ranks even 30 seconds after creation, probably due to NFS caching. This PR solves it by calling `makedirs(..., exist_ok=True)` on all ranks.
Related nvbug: #4440262
**Collection**: NLP
# Changelog
- Create root checkpoint directory on all ranks instead of rank 0 only
# Usage
* You can potentially add a usage example below
```python
# Add a code snippet demonstrating how to use this
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# Before your PR is "Ready for review"
**Pre checks**:
- [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md)
- [ ] Did you write any new necessary tests?
- [ ] Did you add or update any necessary documentation?
- [ ] Does the PR affect components that are optional to install? (Ex: Numba, Pynini, Apex etc)
- [ ] Reviewer: Does the PR have correct import guards for all optional libraries?
**PR Type**:
- [ ] New Feature
- [x] Bugfix
- [ ] Documentation
If you haven't finished some of the above items you can still open "Draft" PR.
## Who can review?
Anyone in the community is free to review the PR once the checks have passed.
[Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas.
# Additional Information
* Related to # (issue)
| {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers",
"following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}",
"gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}",
"starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions",
"organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs",
"repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos",
"events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}",
"received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events",
"type": "Bot",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8153/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/8153/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8152 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8152/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8152/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8152/events | https://github.com/NVIDIA/NeMo/pull/8152 | 2,075,829,827 | PR_kwDOC_bI7s5jw0wo | 8,152 | Fix a bug caused by a command line option typo for `–local-rank` in megatron_ckpt_to_nemo.py | {
"login": "suhara",
"id": 32640,
"node_id": "MDQ6VXNlcjMyNjQw",
"avatar_url": "https://avatars.githubusercontent.com/u/32640?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/suhara",
"html_url": "https://github.com/suhara",
"followers_url": "https://api.github.com/users/suhara/followers",
"following_url": "https://api.github.com/users/suhara/following{/other_user}",
"gists_url": "https://api.github.com/users/suhara/gists{/gist_id}",
"starred_url": "https://api.github.com/users/suhara/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/suhara/subscriptions",
"organizations_url": "https://api.github.com/users/suhara/orgs",
"repos_url": "https://api.github.com/users/suhara/repos",
"events_url": "https://api.github.com/users/suhara/events{/privacy}",
"received_events_url": "https://api.github.com/users/suhara/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4585431311,
"node_id": "LA_kwDOC_bI7s8AAAABEVAhDw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale",
"name": "stale",
"color": "ededed",
"default": false,
"description": null
},
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"This PR is stale because it has been open for 14 days with no activity. Remove stale label or comment or update or this will be closed in 7 days.",
"This PR was closed because it has been inactive for 7 days since being marked as stale."
] | 2024-01-11T06:08:37 | 2024-02-03T01:42:58 | 2024-02-03T01:42:58 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8152",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8152",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8152.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8152.patch",
"merged_at": null
} | # What does this PR do ?
This PR fixes a command line argument typo in `megatron_ckpt_to_nemo.py`, which causes an error when `torch.distributed.launch --nproc_per_node=n (n>1)` is used. It also adds information about necessary changes in `hparams.yaml` as a comment.
**PR Type**:
- [ ] New Feature
- [x] Bugfix
- [ ] Documentation
# Details
When I used `torch.distributed.launch --nproc_node=n` following the example in the comment. I got the following error. Apparently, the command line argument has to be `--local-rank` not `--local_rank`.
> If your script expects `--local-rank` argument to be set, please
change it to read from `os.environ['LOCAL_RANK']` instead. See
```
# python -m torch.distributed.launch --nproc_per_node=4 ${NEMO_HOME_DIR}/examples/nlp/language_modeling/megatron_ckpt_to_nemo.py --checkpoint_folder ${CKPT_DIR} --checkpoint_nam
e ${CKPT_NAME} --nemo_file_path ${OUTPUT_NAME} --tensor_model_parallel_size 4 --pipeline_model_parallel_size 1 --gpus_per_node 8 --precision bf16-mixed --model_type gpt --hparams_file ${HPARAMS_F
ILEPATH}
/usr/local/lib/python3.10/dist-packages/torch/distributed/launch.py:181: FutureWarning: The module torch.distributed.launch is deprecated
and will be removed in future. Use torchrun.
Note that --use-env is set by default in torchrun.
If your script expects `--local-rank` argument to be set, please
change it to read from `os.environ['LOCAL_RANK']` instead. See
https://pytorch.org/docs/stable/distributed.html#launch-utility for
further instructions
warnings.warn(
[2024-01-11 03:38:10,336] torch.distributed.run: [WARNING]
[2024-01-11 03:38:10,336] torch.distributed.run: [WARNING] *****************************************
[2024-01-11 03:38:10,336] torch.distributed.run: [WARNING] Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed.
[2024-01-11 03:38:10,336] torch.distributed.run: [WARNING] *****************************************
[NeMo W 2024-01-11 03:38:18 nemo_logging:349] /usr/local/lib/python3.10/dist-packages/pandas/core/computation/expressions.py:20: UserWarning: Pandas requires version '2.7.3' or newer of 'numexpr' (version '2.7.2' currently install
ed).
from pandas.core.computation.check import NUMEXPR_INSTALLED
usage: megatron_ckpt_to_nemo.py [-h] --checkpoint_folder CHECKPOINT_FOLDER --checkpoint_name CHECKPOINT_NAME [--hparams_file HPARAMS_FILE] --nemo_file_path NEMO_FILE_PATH --gpus_per_node GPUS_PER_NODE
--tensor_model_parallel_size TENSOR_MODEL_PARALLEL_SIZE --pipeline_model_parallel_size PIPELINE_MODEL_PARALLEL_SIZE [--pipeline_model_parallel_split_rank PIPELINE_MODEL_PARALLEL_SPLIT_RANK]
--model_type {gpt,sft,t5,bert,nmt,bart,retro} [--local_rank LOCAL_RANK] [--bcp] [--precision {32-true,16-mixed,bf16-mixed}]
megatron_ckpt_to_nemo.py: error: unrecognized arguments: --local-rank=2
[2024-01-11 03:38:25,357] torch.distributed.elastic.multiprocessing.api: [WARNING] Sending process 2521 closing signal SIGTERM
[2024-01-11 03:38:25,357] torch.distributed.elastic.multiprocessing.api: [WARNING] Sending process 2522 closing signal SIGTERM
[2024-01-11 03:38:25,357] torch.distributed.elastic.multiprocessing.api: [WARNING] Sending process 2524 closing signal SIGTERM
[2024-01-11 03:38:25,420] torch.distributed.elastic.multiprocessing.api: [ERROR] failed (exitcode: 2) local_rank: 2 (pid: 2523) of binary: /usr/bin/python
Traceback (most recent call last):
File "/usr/lib/python3.10/runpy.py", line 196, in _run_module_as_main
return _run_code(code, main_globals, None,
File "/usr/lib/python3.10/runpy.py", line 86, in _run_code
exec(code, run_globals)
File "/usr/local/lib/python3.10/dist-packages/torch/distributed/launch.py", line 196, in <module>
main()
File "/usr/local/lib/python3.10/dist-packages/torch/distributed/launch.py", line 192, in main
launch(args)
File "/usr/local/lib/python3.10/dist-packages/torch/distributed/launch.py", line 177, in launch
run(args)
File "/usr/local/lib/python3.10/dist-packages/torch/distributed/run.py", line 797, in run
elastic_launch(
File "/usr/local/lib/python3.10/dist-packages/torch/distributed/launcher/api.py", line 134, in __call__
return launch_agent(self._config, self._entrypoint, list(args))
File "/usr/local/lib/python3.10/dist-packages/torch/distributed/launcher/api.py", line 264, in launch_agent
raise ChildFailedError(
torch.distributed.elastic.multiprocessing.errors.ChildFailedError:
============================================================
/opt/NeMo//examples/nlp/language_modeling/megatron_ckpt_to_nemo.py FAILED
------------------------------------------------------------
Failures:
<NO_OTHER_FAILURES>
------------------------------------------------------------
Root Cause (first observed failure):
[0]:
time : 2024-01-11_03:38:25
host : dgxa0335p.av4.bm.infra.nvda.ai
rank : 2 (local_rank: 2)
exitcode : 2 (pid: 2523)
error_file: <N/A>
traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html
```
| {
"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/8152/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/8152/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8151 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8151/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8151/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8151/events | https://github.com/NVIDIA/NeMo/pull/8151 | 2,075,406,130 | PR_kwDOC_bI7s5jvX7z | 8,151 | Improve documentation gpt_training.rst | {
"login": "ethanhe42",
"id": 10027339,
"node_id": "MDQ6VXNlcjEwMDI3MzM5",
"avatar_url": "https://avatars.githubusercontent.com/u/10027339?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ethanhe42",
"html_url": "https://github.com/ethanhe42",
"followers_url": "https://api.github.com/users/ethanhe42/followers",
"following_url": "https://api.github.com/users/ethanhe42/following{/other_user}",
"gists_url": "https://api.github.com/users/ethanhe42/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ethanhe42/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ethanhe42/subscriptions",
"organizations_url": "https://api.github.com/users/ethanhe42/orgs",
"repos_url": "https://api.github.com/users/ethanhe42/repos",
"events_url": "https://api.github.com/users/ethanhe42/events{/privacy}",
"received_events_url": "https://api.github.com/users/ethanhe42/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4585431311,
"node_id": "LA_kwDOC_bI7s8AAAABEVAhDw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale",
"name": "stale",
"color": "ededed",
"default": false,
"description": null
},
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"@yaoyu-33 ",
"This PR is stale because it has been open for 14 days with no activity. Remove stale label or comment or update or this will be closed in 7 days.",
"This PR was closed because it has been inactive for 7 days since being marked as stale.",
"/cicd"
] | 2024-01-10T23:42:41 | 2024-05-06T22:10:52 | 2024-02-09T01:43:35 | MEMBER | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8151",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8151",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8151.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8151.patch",
"merged_at": null
} | **PR Type**:
- [ ] New Feature
- [ ] Bugfix
- [x] Documentation
| {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers",
"following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}",
"gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}",
"starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions",
"organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs",
"repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos",
"events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}",
"received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events",
"type": "Bot",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8151/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/8151/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8150 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8150/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8150/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8150/events | https://github.com/NVIDIA/NeMo/pull/8150 | 2,075,278,733 | PR_kwDOC_bI7s5ju7rD | 8,150 | add first_val_step to mcore scheduler | {
"login": "JimmyZhang12",
"id": 67203904,
"node_id": "MDQ6VXNlcjY3MjAzOTA0",
"avatar_url": "https://avatars.githubusercontent.com/u/67203904?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/JimmyZhang12",
"html_url": "https://github.com/JimmyZhang12",
"followers_url": "https://api.github.com/users/JimmyZhang12/followers",
"following_url": "https://api.github.com/users/JimmyZhang12/following{/other_user}",
"gists_url": "https://api.github.com/users/JimmyZhang12/gists{/gist_id}",
"starred_url": "https://api.github.com/users/JimmyZhang12/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/JimmyZhang12/subscriptions",
"organizations_url": "https://api.github.com/users/JimmyZhang12/orgs",
"repos_url": "https://api.github.com/users/JimmyZhang12/repos",
"events_url": "https://api.github.com/users/JimmyZhang12/events{/privacy}",
"received_events_url": "https://api.github.com/users/JimmyZhang12/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
},
{
"id": 4840216855,
"node_id": "LA_kwDOC_bI7s8AAAABIH_ZFw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/CI",
"name": "CI",
"color": "1D76DB",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"jenkins",
"@ericharper @athitten \r\nCan you help review this PR?",
"jenkins",
"jenkins",
"jenkins",
"jenkins",
"jenkins",
"jenkins",
"jenkins",
"jenkins",
"jenkins"
] | 2024-01-10T21:55:40 | 2024-01-25T18:43:50 | 2024-01-25T18:43:49 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8150",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8150",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8150.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8150.patch",
"merged_at": "2024-01-25T18:43:49"
} | # 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
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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/8150/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/8150/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8149 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8149/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8149/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8149/events | https://github.com/NVIDIA/NeMo/pull/8149 | 2,074,489,611 | PR_kwDOC_bI7s5jsOwU | 8,149 | switch to mcore dataset [with FIM support] | {
"login": "dimapihtar",
"id": 37850217,
"node_id": "MDQ6VXNlcjM3ODUwMjE3",
"avatar_url": "https://avatars.githubusercontent.com/u/37850217?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/dimapihtar",
"html_url": "https://github.com/dimapihtar",
"followers_url": "https://api.github.com/users/dimapihtar/followers",
"following_url": "https://api.github.com/users/dimapihtar/following{/other_user}",
"gists_url": "https://api.github.com/users/dimapihtar/gists{/gist_id}",
"starred_url": "https://api.github.com/users/dimapihtar/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/dimapihtar/subscriptions",
"organizations_url": "https://api.github.com/users/dimapihtar/orgs",
"repos_url": "https://api.github.com/users/dimapihtar/repos",
"events_url": "https://api.github.com/users/dimapihtar/events{/privacy}",
"received_events_url": "https://api.github.com/users/dimapihtar/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 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 | [
"jenkins",
"jenkins",
"jenkins",
"jenkins",
"jenkins",
"jenkins",
"jenkins",
"jenkins",
"jenkins"
] | 2024-01-10T14:24:38 | 2024-01-25T22:30:36 | 2024-01-25T22:30:35 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8149",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8149",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8149.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8149.patch",
"merged_at": "2024-01-25T22:30:35"
} | # What does this PR do ?
1. Switches to megatron core dataset.
2. Adds FIM support.
3. Adds starcoder config.
**Collection**: NLP
# Changelog
- Added additional functions in oreder to use mcore ds.
- Changed NeMo ds classes to mcore ds classes (megatron_gpt_model.py).
- Created gpt_fim_dataset.py with GPTFIMDatasetConfig and GPTFIMDataset classes in order to use FIM techinque.
- Added new config for starcoder models (megatron_starcoder_config.yaml).
# Usage
Example of FIM usage. This part should be added to the data section in the model config:
```yaml
add_fim: True # Set to True to use FIM
fim:
# fill in the middle
rate: 0.5 # Probability to convert a training sample into a "Fill-in-the-Middle" format. Must be between 0 and 1
spm_rate: 0.5 # Probability that the a FIM sample uses the SPM format over the PSM format
split_sample: null # String around which to split the sample for FIM. If None (default), FIM is applied on the sample-level
fragment_rate: 0.5 # Rate of FIM on each fragment when fim_split_sample is not None
no_prefix: null # Do not apply FIM to fragments that start with this prefix
extra_tokens:
prefix: "<fim_prefix>"
middle: "<fim_middle>"
suffix: "<fim_suffix>"
pad: "<fim_pad>"
eod: "<|endoftext|>"
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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": "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/8149/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/8149/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8148 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8148/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8148/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8148/events | https://github.com/NVIDIA/NeMo/pull/8148 | 2,073,751,891 | PR_kwDOC_bI7s5jps9o | 8,148 | Added sample cpu_offloading switch to YAML | {
"login": "sanandaraj5597",
"id": 55270934,
"node_id": "MDQ6VXNlcjU1MjcwOTM0",
"avatar_url": "https://avatars.githubusercontent.com/u/55270934?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sanandaraj5597",
"html_url": "https://github.com/sanandaraj5597",
"followers_url": "https://api.github.com/users/sanandaraj5597/followers",
"following_url": "https://api.github.com/users/sanandaraj5597/following{/other_user}",
"gists_url": "https://api.github.com/users/sanandaraj5597/gists{/gist_id}",
"starred_url": "https://api.github.com/users/sanandaraj5597/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/sanandaraj5597/subscriptions",
"organizations_url": "https://api.github.com/users/sanandaraj5597/orgs",
"repos_url": "https://api.github.com/users/sanandaraj5597/repos",
"events_url": "https://api.github.com/users/sanandaraj5597/events{/privacy}",
"received_events_url": "https://api.github.com/users/sanandaraj5597/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"Jenkins",
"jenkins",
"jenkins",
"jenkins",
"jenkins"
] | 2024-01-10T07:20:49 | 2024-01-24T17:53:35 | 2024-01-24T17:53:34 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8148",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8148",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8148.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8148.patch",
"merged_at": "2024-01-24T17:53:34"
} | # What does this PR do ?
Add a sample to enable cpu offloading in megatron_gpt_config.yaml | {
"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/8148/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/8148/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8147 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8147/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8147/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8147/events | https://github.com/NVIDIA/NeMo/issues/8147 | 2,073,589,598 | I_kwDOC_bI7s57mHde | 8,147 | Can be any improvement from using learned preprocessors? | {
"login": "ArtyomZemlyak",
"id": 42967602,
"node_id": "MDQ6VXNlcjQyOTY3NjAy",
"avatar_url": "https://avatars.githubusercontent.com/u/42967602?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ArtyomZemlyak",
"html_url": "https://github.com/ArtyomZemlyak",
"followers_url": "https://api.github.com/users/ArtyomZemlyak/followers",
"following_url": "https://api.github.com/users/ArtyomZemlyak/following{/other_user}",
"gists_url": "https://api.github.com/users/ArtyomZemlyak/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ArtyomZemlyak/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ArtyomZemlyak/subscriptions",
"organizations_url": "https://api.github.com/users/ArtyomZemlyak/orgs",
"repos_url": "https://api.github.com/users/ArtyomZemlyak/repos",
"events_url": "https://api.github.com/users/ArtyomZemlyak/events{/privacy}",
"received_events_url": "https://api.github.com/users/ArtyomZemlyak/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4585431311,
"node_id": "LA_kwDOC_bI7s8AAAABEVAhDw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale",
"name": "stale",
"color": "ededed",
"default": false,
"description": null
}
] | closed | false | {
"login": "okuchaiev",
"id": 22035961,
"node_id": "MDQ6VXNlcjIyMDM1OTYx",
"avatar_url": "https://avatars.githubusercontent.com/u/22035961?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/okuchaiev",
"html_url": "https://github.com/okuchaiev",
"followers_url": "https://api.github.com/users/okuchaiev/followers",
"following_url": "https://api.github.com/users/okuchaiev/following{/other_user}",
"gists_url": "https://api.github.com/users/okuchaiev/gists{/gist_id}",
"starred_url": "https://api.github.com/users/okuchaiev/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/okuchaiev/subscriptions",
"organizations_url": "https://api.github.com/users/okuchaiev/orgs",
"repos_url": "https://api.github.com/users/okuchaiev/repos",
"events_url": "https://api.github.com/users/okuchaiev/events{/privacy}",
"received_events_url": "https://api.github.com/users/okuchaiev/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"login": "okuchaiev",
"id": 22035961,
"node_id": "MDQ6VXNlcjIyMDM1OTYx",
"avatar_url": "https://avatars.githubusercontent.com/u/22035961?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/okuchaiev",
"html_url": "https://github.com/okuchaiev",
"followers_url": "https://api.github.com/users/okuchaiev/followers",
"following_url": "https://api.github.com/users/okuchaiev/following{/other_user}",
"gists_url": "https://api.github.com/users/okuchaiev/gists{/gist_id}",
"starred_url": "https://api.github.com/users/okuchaiev/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/okuchaiev/subscriptions",
"organizations_url": "https://api.github.com/users/okuchaiev/orgs",
"repos_url": "https://api.github.com/users/okuchaiev/repos",
"events_url": "https://api.github.com/users/okuchaiev/events{/privacy}",
"received_events_url": "https://api.github.com/users/okuchaiev/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
}
] | null | [
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"This issue was closed because it has been inactive for 7 days since being marked as stale."
] | 2024-01-10T05:00:09 | 2024-02-17T01:43:33 | 2024-02-17T01:43:33 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | Duplicated: https://github.com/NVIDIA/NeMo/discussions/8133
I think about preprocessors like in this repo: https://github.com/KinWaiCheuk/nnAudio
Maybe some improvements (speed of training or inference, quality) can be reached from using learned preprocessors? (and also they can be learned also then ASR model trained) | {
"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/8147/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/8147/timeline | null | not_planned | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/8146 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8146/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8146/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8146/events | https://github.com/NVIDIA/NeMo/pull/8146 | 2,073,356,523 | PR_kwDOC_bI7s5joYMa | 8,146 | Remove typo | {
"login": "ericharper",
"id": 11999610,
"node_id": "MDQ6VXNlcjExOTk5NjEw",
"avatar_url": "https://avatars.githubusercontent.com/u/11999610?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ericharper",
"html_url": "https://github.com/ericharper",
"followers_url": "https://api.github.com/users/ericharper/followers",
"following_url": "https://api.github.com/users/ericharper/following{/other_user}",
"gists_url": "https://api.github.com/users/ericharper/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ericharper/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ericharper/subscriptions",
"organizations_url": "https://api.github.com/users/ericharper/orgs",
"repos_url": "https://api.github.com/users/ericharper/repos",
"events_url": "https://api.github.com/users/ericharper/events{/privacy}",
"received_events_url": "https://api.github.com/users/ericharper/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 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 | [] | 2024-01-10T00:00:03 | 2024-01-10T00:00:36 | 2024-01-10T00:00:35 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8146",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8146",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8146.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8146.patch",
"merged_at": "2024-01-10T00:00:35"
} | # 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
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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/8146/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/8146/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8145 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8145/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8145/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8145/events | https://github.com/NVIDIA/NeMo/pull/8145 | 2,073,336,868 | PR_kwDOC_bI7s5joT7v | 8,145 | fix: numba.*_num_threads resets torch num_threads #8141 | {
"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": 4811884691,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR",
"name": "ASR",
"color": "F3B33E",
"default": false,
"description": ""
}
] | closed | false | {
"login": "titu1994",
"id": 3048602,
"node_id": "MDQ6VXNlcjMwNDg2MDI=",
"avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/titu1994",
"html_url": "https://github.com/titu1994",
"followers_url": "https://api.github.com/users/titu1994/followers",
"following_url": "https://api.github.com/users/titu1994/following{/other_user}",
"gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}",
"starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/titu1994/subscriptions",
"organizations_url": "https://api.github.com/users/titu1994/orgs",
"repos_url": "https://api.github.com/users/titu1994/repos",
"events_url": "https://api.github.com/users/titu1994/events{/privacy}",
"received_events_url": "https://api.github.com/users/titu1994/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"login": "titu1994",
"id": 3048602,
"node_id": "MDQ6VXNlcjMwNDg2MDI=",
"avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/titu1994",
"html_url": "https://github.com/titu1994",
"followers_url": "https://api.github.com/users/titu1994/followers",
"following_url": "https://api.github.com/users/titu1994/following{/other_user}",
"gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}",
"starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/titu1994/subscriptions",
"organizations_url": "https://api.github.com/users/titu1994/orgs",
"repos_url": "https://api.github.com/users/titu1994/repos",
"events_url": "https://api.github.com/users/titu1994/events{/privacy}",
"received_events_url": "https://api.github.com/users/titu1994/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
}
] | null | [
"Could you sign the commits in PR by following the DCOBot? ",
"jenkins ",
"jenkins",
"Hmm core dumped in numba unit-tests. I think ill try running them one more time in case it was a fluke, and then well need to debug more"
] | 2024-01-09T23:35:20 | 2024-01-12T15:22:23 | 2024-01-11T18:56:58 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8145",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8145",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8145.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8145.patch",
"merged_at": "2024-01-11T18:56:58"
} | # What does this PR do ?
Temporary fix until https://github.com/numba/numba/issues/9387 gets resolved.
**Collection**: [ASR]
# Changelog
- wrap rnnt numba num_threads check and change to protect torch num_threads from being reset.
# Usage
* You can potentially add a usage example below
```python
# Add a code snippet demonstrating how to use this
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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/8145/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/8145/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8144 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8144/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8144/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8144/events | https://github.com/NVIDIA/NeMo/pull/8144 | 2,073,264,833 | PR_kwDOC_bI7s5joEY_ | 8,144 | Fix TP>1 issue for conversion script | {
"login": "cuichenx",
"id": 43478052,
"node_id": "MDQ6VXNlcjQzNDc4MDUy",
"avatar_url": "https://avatars.githubusercontent.com/u/43478052?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/cuichenx",
"html_url": "https://github.com/cuichenx",
"followers_url": "https://api.github.com/users/cuichenx/followers",
"following_url": "https://api.github.com/users/cuichenx/following{/other_user}",
"gists_url": "https://api.github.com/users/cuichenx/gists{/gist_id}",
"starred_url": "https://api.github.com/users/cuichenx/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/cuichenx/subscriptions",
"organizations_url": "https://api.github.com/users/cuichenx/orgs",
"repos_url": "https://api.github.com/users/cuichenx/repos",
"events_url": "https://api.github.com/users/cuichenx/events{/privacy}",
"received_events_url": "https://api.github.com/users/cuichenx/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"jenkins"
] | 2024-01-09T22:18:43 | 2024-01-11T21:50:21 | 2024-01-11T21:50:20 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8144",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8144",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8144.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8144.patch",
"merged_at": "2024-01-11T21:50:20"
} | # What does this PR do ?
Fix issue where nemo -> hf conversion script crashes for models trained with TP > 1
**Collection**: NLP
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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 #7770 #7922
| {
"login": "cuichenx",
"id": 43478052,
"node_id": "MDQ6VXNlcjQzNDc4MDUy",
"avatar_url": "https://avatars.githubusercontent.com/u/43478052?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/cuichenx",
"html_url": "https://github.com/cuichenx",
"followers_url": "https://api.github.com/users/cuichenx/followers",
"following_url": "https://api.github.com/users/cuichenx/following{/other_user}",
"gists_url": "https://api.github.com/users/cuichenx/gists{/gist_id}",
"starred_url": "https://api.github.com/users/cuichenx/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/cuichenx/subscriptions",
"organizations_url": "https://api.github.com/users/cuichenx/orgs",
"repos_url": "https://api.github.com/users/cuichenx/repos",
"events_url": "https://api.github.com/users/cuichenx/events{/privacy}",
"received_events_url": "https://api.github.com/users/cuichenx/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8144/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/8144/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8143 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8143/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8143/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8143/events | https://github.com/NVIDIA/NeMo/pull/8143 | 2,073,146,793 | PR_kwDOC_bI7s5jnqUq | 8,143 | updating default transcribe behavior to en lowercase for Canary Models | {
"login": "krishnacpuvvada",
"id": 93558329,
"node_id": "U_kgDOBZOWOQ",
"avatar_url": "https://avatars.githubusercontent.com/u/93558329?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/krishnacpuvvada",
"html_url": "https://github.com/krishnacpuvvada",
"followers_url": "https://api.github.com/users/krishnacpuvvada/followers",
"following_url": "https://api.github.com/users/krishnacpuvvada/following{/other_user}",
"gists_url": "https://api.github.com/users/krishnacpuvvada/gists{/gist_id}",
"starred_url": "https://api.github.com/users/krishnacpuvvada/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/krishnacpuvvada/subscriptions",
"organizations_url": "https://api.github.com/users/krishnacpuvvada/orgs",
"repos_url": "https://api.github.com/users/krishnacpuvvada/repos",
"events_url": "https://api.github.com/users/krishnacpuvvada/events{/privacy}",
"received_events_url": "https://api.github.com/users/krishnacpuvvada/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811884691,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR",
"name": "ASR",
"color": "F3B33E",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [] | 2024-01-09T20:48:38 | 2024-01-09T21:16:44 | 2024-01-09T21:16:44 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8143",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8143",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8143.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8143.patch",
"merged_at": "2024-01-09T21:16:44"
} | # What does this PR do ?
Updated default transcribe behavior to be ASR-English-lowercase for canary models. Model will default to this if provided with a list of audio paths (paths2audio_files) with out specifying additional task details.
**Collection**: ASR
# 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": "krishnacpuvvada",
"id": 93558329,
"node_id": "U_kgDOBZOWOQ",
"avatar_url": "https://avatars.githubusercontent.com/u/93558329?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/krishnacpuvvada",
"html_url": "https://github.com/krishnacpuvvada",
"followers_url": "https://api.github.com/users/krishnacpuvvada/followers",
"following_url": "https://api.github.com/users/krishnacpuvvada/following{/other_user}",
"gists_url": "https://api.github.com/users/krishnacpuvvada/gists{/gist_id}",
"starred_url": "https://api.github.com/users/krishnacpuvvada/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/krishnacpuvvada/subscriptions",
"organizations_url": "https://api.github.com/users/krishnacpuvvada/orgs",
"repos_url": "https://api.github.com/users/krishnacpuvvada/repos",
"events_url": "https://api.github.com/users/krishnacpuvvada/events{/privacy}",
"received_events_url": "https://api.github.com/users/krishnacpuvvada/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8143/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/8143/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8142 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8142/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8142/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8142/events | https://github.com/NVIDIA/NeMo/pull/8142 | 2,073,060,713 | PR_kwDOC_bI7s5jnXTE | 8,142 | Fix layer_norm_1p Apex call to match Apex API | {
"login": "ShriyaPalsamudram",
"id": 69161273,
"node_id": "MDQ6VXNlcjY5MTYxMjcz",
"avatar_url": "https://avatars.githubusercontent.com/u/69161273?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ShriyaPalsamudram",
"html_url": "https://github.com/ShriyaPalsamudram",
"followers_url": "https://api.github.com/users/ShriyaPalsamudram/followers",
"following_url": "https://api.github.com/users/ShriyaPalsamudram/following{/other_user}",
"gists_url": "https://api.github.com/users/ShriyaPalsamudram/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ShriyaPalsamudram/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ShriyaPalsamudram/subscriptions",
"organizations_url": "https://api.github.com/users/ShriyaPalsamudram/orgs",
"repos_url": "https://api.github.com/users/ShriyaPalsamudram/repos",
"events_url": "https://api.github.com/users/ShriyaPalsamudram/events{/privacy}",
"received_events_url": "https://api.github.com/users/ShriyaPalsamudram/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"This is a duplicate: https://github.com/NVIDIA/NeMo/pull/8129"
] | 2024-01-09T19:54:10 | 2024-01-10T14:40:08 | 2024-01-10T14:39:57 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8142",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8142",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8142.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8142.patch",
"merged_at": null
} | # What does this PR do ?
Apex _fast_layer_norm expects required memory_efficient parameter to be passed when it is called: https://github.com/NVIDIA/apex/blob/master/apex/contrib/layer_norm/layer_norm.py#L37
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
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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?
- [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**:
- [ ] 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": "ShriyaPalsamudram",
"id": 69161273,
"node_id": "MDQ6VXNlcjY5MTYxMjcz",
"avatar_url": "https://avatars.githubusercontent.com/u/69161273?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ShriyaPalsamudram",
"html_url": "https://github.com/ShriyaPalsamudram",
"followers_url": "https://api.github.com/users/ShriyaPalsamudram/followers",
"following_url": "https://api.github.com/users/ShriyaPalsamudram/following{/other_user}",
"gists_url": "https://api.github.com/users/ShriyaPalsamudram/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ShriyaPalsamudram/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ShriyaPalsamudram/subscriptions",
"organizations_url": "https://api.github.com/users/ShriyaPalsamudram/orgs",
"repos_url": "https://api.github.com/users/ShriyaPalsamudram/repos",
"events_url": "https://api.github.com/users/ShriyaPalsamudram/events{/privacy}",
"received_events_url": "https://api.github.com/users/ShriyaPalsamudram/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8142/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/8142/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8141 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8141/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8141/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8141/events | https://github.com/NVIDIA/NeMo/issues/8141 | 2,072,218,939 | I_kwDOC_bI7s57g407 | 8,141 | [ASR] hybrid models ignore OMP_NUM_THREADS/torch.set_num_threads() setting | {
"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 | null | [] | null | [
"Very odd. \r\n\r\nHybrid code doesn't explicitly set num threads anywhere. Transcribe() is also never called during training. \r\n\r\nWhat about num workers on data loader, that's the only place I can think of that spawns workers threads. ",
"There are quite a few locations, where `num_threads()` is updated (not `torch` explicitly, but `numba`), and many of them check `multiprocessing.cpu_count()`, but the real source of the issue here is Numba. More precisely In the case of fastconformer_hybrid it is this section\r\nhttps://github.com/NVIDIA/NeMo/blob/76a712a323c70e5701a730048627509a11336424/nemo/collections/asr/parts/numba/rnnt_loss/utils/cuda_utils/gpu_rnnt.py#L85-L89\r\n\r\nHere the call becomes `numba.set_num_threads(1)`, which resets the torch num_threads. It turns out that Numba resets the torch num_thread even when just reading the Numba num_threads. This can easily be checked with the following snippet:\r\n``` python\r\n$ srun --mem=48G -c32 --container-image=./nemo_head.sqfs --pty bash\r\nroot@x:/workspace/nemo# nproc\r\n32\r\nroot@x:/workspace/nemo# export OMP_NUM_THREADS=8\r\nroot@x:/workspace/nemo# nproc\r\n8\r\nroot@x:/workspace/nemo# python3\r\nPython 3.10.12 (main, Jun 11 2023, 05:26:28) [GCC 11.4.0] on linux\r\nType \"help\", \"copyright\", \"credits\" or \"license\" for more information.\r\n>>> import multiprocessing as mp, numba, torch\r\n>>> mp.cpu_count()\r\n256\r\n\r\n>>> torch.get_num_threads()\r\n8\r\n\r\n>>> print(torch.__config__.parallel_info())\r\nATen/Parallel:\r\n at::get_num_threads() : 8\r\n at::get_num_interop_threads() : 128\r\nOpenMP 201511 (a.k.a. OpenMP 4.5)\r\n omp_get_max_threads() : 8\r\nIntel(R) oneAPI Math Kernel Library Version 2021.1-Product Build 20201104 for Intel(R) 64 architecture applications\r\n mkl_get_max_threads() : 8\r\nIntel(R) MKL-DNN v2.7.3 (Git Hash N/A)\r\nstd::thread::hardware_concurrency() : 256\r\nEnvironment variables:\r\n OMP_NUM_THREADS : 8\r\n MKL_NUM_THREADS : [not set]\r\nATen parallel backend: OpenMP\r\n\r\n>>> numba.get_num_threads()\r\n32\r\n\r\n>>> torch.get_num_threads()\r\n32\r\n\r\n>>> print(torch.__config__.parallel_info())\r\nATen/Parallel:\r\n at::get_num_threads() : 32\r\n at::get_num_interop_threads() : 128\r\nOpenMP 201511 (a.k.a. OpenMP 4.5)\r\n omp_get_max_threads() : 32\r\nIntel(R) oneAPI Math Kernel Library Version 2021.1-Product Build 20201104 for Intel(R) 64 architecture applications\r\n mkl_get_max_threads() : 32\r\nIntel(R) MKL-DNN v2.7.3 (Git Hash N/A)\r\nstd::thread::hardware_concurrency() : 256\r\nEnvironment variables:\r\n OMP_NUM_THREADS : 8\r\n MKL_NUM_THREADS : [not set]\r\nATen parallel backend: OpenMP\r\n\r\n>>> numba.set_num_threads(2)\r\n>>> numba.get_num_threads()\r\n2\r\n\r\n>>> torch.get_num_threads()\r\n32\r\n\r\n>>> torch.__version__\r\n'2.1.0a0+29c30b1'\r\n\r\n>>> numba.__version__\r\n'0.57.1+1.gc785c8f1f'\r\n\r\n>>> exit()\r\nroot@x:/workspace/nemo# exit\r\nexit\r\n```\r\n\r\nI have opened an issue on numba https://github.com/numba/numba/issues/9387. \r\n",
"Until numba fixes the issue one possible solution is to first read the torch num_threads, and reset them back post numba.set_num_threads(). ",
"This isn't specific to hybrid models, I dunno why it shows up only there. In either case, I'll send a PR for temporary fix. \r\n\r\nBasically before I call get or set numba threads, I should cache pytorch threads then set that explicitly after numba thread set ?",
"I have opened a PR, if there are other locations and you prefer to fix by yourself, I can close it.",
"Perfect, let's get yours passing the tests and merged ",
"Adding this here just for completnes; found it while researching something else. Lightning 2.1.0 changelog mentions:\r\n> - If not set by the user, Lightning will set `OMP_NUM_THREADS` to `num_cpus / num_processes` when launching subprocesses (e.g. when DDP is used) to avoid system overload for CPU-intensive tasks ([#18677](https://github.com/Lightning-AI/lightning/pull/18677))\r\n",
"Hmm actually were seeing some segfaults randomly in NeMo ASR and it might be due to the numba fix PR (I saw the segfault first on that pr but thought it was random fluke). \r\n\r\nWe might have to revert that pr for the time being and depend on numba to fix it properly ",
"Hmmm, this is strange. I haven't seen anything like this. During training or inference? \r\n\r\nWe use pyxis/enroot, which by default injects `OMP_NUM_THREADS=1` following PyTorch >1.9.0, which does so if the environment variable is not set. Lightning 2.10, will on the other hand set the env (if not previously declared) as mentioned earlier. I have run quick experiments with `OMP_NUM_THREADS=num_cpus / num_processes`, but at least on first impression the difference seems negligible. \r\n\r\nThe good thing is that issues with oversubscription (due to numba resetting the value to `min(NUMBA_NUM_THREADS, os.sched_getaffinity(), os.num_cpu())`) start arising only when no cpu binding is used. This holds at least on most Linux platforms, but `os.sched_getaffinity()` is not available on Windows and OsX. There's a comment in torch dataloader about this https://github.com/pytorch/pytorch/blob/763ddb396df4bc14791fbf9149d46d5713a699df/torch/utils/data/dataloader.py#L501-L512\r\n\r\nAs Numba maintainers pointed out, one approach might be to ensure `NUMBA_NUM_THREADS` is set to match `OMP_NUM_THREADS`, as this is the value to which the global OMP num_threads (and accordingly also torch num_threads) will be reset to on the first numba call. With pyxis/enroot the injection is simple, but it is not general for all use cases (e.g. docker, ...). \r\n\r\nYet another approach could be to go the route of Pytorch/Lightning and declare both environment variables (if not already set) during NeMo startup (e.g. in exp_manager). ",
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"@titu1994 have you noticed any more segfaults? ",
"No thankfully none in 1.23 branch and main. Seems to be very random results during that one-two week period ",
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"This issue was closed because it has been inactive for 7 days since being marked as stale."
] | 2024-01-09T11:50:16 | 2024-03-29T01:44:07 | 2024-03-29T01:44:06 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | Tested on a SLURM 23.11 cluster with Pyxis/Enroot 0.16.1/3.4.1 running NGC containers `nemo:23.03`, `nemo:23.08`, `ea-bignlp/ga-participants/nemofw-training:23.11`, and a custom container built from main branch.
When running single-node multi-gpu training in exclusive mode or without CPU binding, but with `OMP_NUM_THREADS` set to nCPUs/tasks or 1, the hybrid models (conformer_hybrid, or fastconformer_hybrid) ignore this setting and start too many threads which leads to CPU oversubscription. This results in a performance drop of ~2x. See also https://pytorch.org/docs/stable/notes/multiprocessing.html#avoid-cpu-oversubscription.
More specifically this happens both when running as SLURM sbatch
```bash
#SBATCH --nodes=1
#SBATCH --exclusive
#SBATCH --mem=0
#SBATCH --ntasks-per-node=8
#SBATCH --gpus-per-node=8
srun <train_command>
```
or as docker command
```bash
docker run --rm \
--net=host --uts=host --ipc=host --security-opt=seccomp=unconfined \
--ulimit=stack=67108864 --ulimit=memlock=-1 \
<train_command>
```
In SLURM the workaround is to use cpu binding and explicitly set the CPU count per task (directive `#SBATCH --cpus-per-task`). For docker I haven't found a workaround.
Compared to fastconformer_ctc, fastconformer_transducer, conformer_ctc, the issue is present only in the case of hybrid models (conformer or fastconformer). On a DGX-A100 with 256 vCPU and 8 GPU this can be seen by the following vCPU count (as reported by bash command `nproc` ran as part of the `<train_command>`) and number of main processes' thread counts (as reported by [btop](https://github.com/aristocratos/btop) ran on the host). Note that in all cases python calls `os.cpu_count()` and `multiprocessing.cpu_count()` ran from within the `<train_command>` but prior to actual training return 256, and `torch.__config__.parallel_info()` returns the correct values (as `nproc`).
conformer_ctc
- exclusive mode, cpus-per-task not set, OMP_NUM_THREADS env not set [vCPU=192; nthreads: 153]
- exclusive mode, cpus-per-task not set, OMP_NUM_THREADS=24 env set [vCPU=24, nthreads: 49]
- exclusive mode, set cpus-per-tasks on srun, OMP_NUM_THREADS=32 env set [vCPU=32, nthreads: 57]
conformer_hybrid
- exclusive mode, cpus-per-task not set, OMP_NUM_THREADS env not set [vCPU=192; nthreads: 217]
- **exclusive mode, cpus-per-task not set, OMP_NUM_THREADS=24 env set [vCPU=24, nthreads: 217]**
- exclusive mode, set cpus-per-tasks on srun, OMP_NUM_THREADS=32 env set [vCPU=32, nthreads: 57]
fastconformer_ctc
- exclusive mode, cpus-per-task not set, OMP_NUM_THREADS env not set [vCPU=192; nthreads: 153]
- exclusive mode, cpus-per-task not set, OMP_NUM_THREADS=24 env set [vCPU=24, nthreads: 49]
- exclusive mode, set cpus-per-tasks on srun, OMP_NUM_THREADS=32 env set [vCPU=32, nthreads: 57]
fastconformer_transducer
- exclusive mode, cpus-per-task not set, OMP_NUM_THREADS env not set [vCPU=192; nthreads: 153]
- exclusive mode, cpus-per-task not set, OMP_NUM_THREADS=24 env set [vCPU=24, nthreads: 49]
- exclusive mode, set cpus-per-tasks on srun, OMP_NUM_THREADS=32 env set [vCPU=32, nthreads: 57]
fastconformer_hybrid
- exclusive mode, cpus-per-task not set, OMP_NUM_THREADS env not set [vCPU=192; nthreads: 217]
- **exclusive mode, cpus-per-task not set, OMP_NUM_THREADS=24 env set [vCPU=24, nthreads: 217]**
- exclusive mode, set cpus-per-tasks on srun, OMP_NUM_THREADS=32 env set [vCPU=32, nthreads: 57]
Unfortunately I wasn't able to find the exact location where this happens. I may be completely wrong, but one possible cause could be that it is related to decoding, or someplace where `torch.set_num_threads()` gets reset based on the value of `*.cpu_count()`.
| {
"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/8141/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/8141/timeline | null | not_planned | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/8140 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8140/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8140/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8140/events | https://github.com/NVIDIA/NeMo/pull/8140 | 2,071,726,231 | PR_kwDOC_bI7s5jizGY | 8,140 | Add knobs for swiglu and rope fusion | {
"login": "lhb8125",
"id": 16619280,
"node_id": "MDQ6VXNlcjE2NjE5Mjgw",
"avatar_url": "https://avatars.githubusercontent.com/u/16619280?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lhb8125",
"html_url": "https://github.com/lhb8125",
"followers_url": "https://api.github.com/users/lhb8125/followers",
"following_url": "https://api.github.com/users/lhb8125/following{/other_user}",
"gists_url": "https://api.github.com/users/lhb8125/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lhb8125/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lhb8125/subscriptions",
"organizations_url": "https://api.github.com/users/lhb8125/orgs",
"repos_url": "https://api.github.com/users/lhb8125/repos",
"events_url": "https://api.github.com/users/lhb8125/events{/privacy}",
"received_events_url": "https://api.github.com/users/lhb8125/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"jenkins",
"Actually, should we default `apply_rope_fusion` to True or False?\r\nBefore this PR it would be been \"False\". Are there risks to changing the default behavior to \"True\" now?",
"> Actually, should we default `apply_rope_fusion` to True or False? Before this PR it would be been \"False\". Are there risks to changing the default behavior to \"True\" now?\r\n\r\nI think we should keep it as true at most cases since rope is a separated operator and well fused in apex. MCore will check if the fused implementation is included in apex, or the config will be set to false.",
"> > Actually, should we default `apply_rope_fusion` to True or False? Before this PR it would be been \"False\". Are there risks to changing the default behavior to \"True\" now?\r\n> \r\n> I think we should keep it as true at most cases since rope is a separated operator and well fused in apex. MCore will check if the fused implementation is included in apex, or the config will be set to false.\r\n\r\nCool, thanks!",
"jenkins",
"The changes have been enabled in another cleaner PR: https://github.com/NVIDIA/NeMo/pull/8184.\r\nClose it now."
] | 2024-01-09T06:41:31 | 2024-01-17T05:57:01 | 2024-01-17T05:57:01 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8140",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8140",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8140.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8140.patch",
"merged_at": null
} | # What does this PR do ?
Add knob for rope fusion
Add knob for swiglu fusion
Change knob for bias_gelu fusion
The PR is related to the PR of MLM: https://gitlab-master.nvidia.com/ADLR/megatron-lm/-/merge_requests/984
**Collection**: [NLP]
# Changelog
- Add specific line by line info of high level changes in this PR.
# Usage
* set `model.apply_rope_fusion: true` to enable rope fusion
* set `model.bias_activation_fusion: true` to enable bias+activation fusion, which works for both gelu and swiglu type.
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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.
Add @ericharper @erhoo82 for review.
[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": "lhb8125",
"id": 16619280,
"node_id": "MDQ6VXNlcjE2NjE5Mjgw",
"avatar_url": "https://avatars.githubusercontent.com/u/16619280?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lhb8125",
"html_url": "https://github.com/lhb8125",
"followers_url": "https://api.github.com/users/lhb8125/followers",
"following_url": "https://api.github.com/users/lhb8125/following{/other_user}",
"gists_url": "https://api.github.com/users/lhb8125/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lhb8125/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lhb8125/subscriptions",
"organizations_url": "https://api.github.com/users/lhb8125/orgs",
"repos_url": "https://api.github.com/users/lhb8125/repos",
"events_url": "https://api.github.com/users/lhb8125/events{/privacy}",
"received_events_url": "https://api.github.com/users/lhb8125/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8140/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/8140/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8139 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8139/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8139/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8139/events | https://github.com/NVIDIA/NeMo/pull/8139 | 2,071,717,789 | PR_kwDOC_bI7s5jixQR | 8,139 | G2P CTC - allow setting vocab_file from config | {
"login": "ppisljar",
"id": 13629809,
"node_id": "MDQ6VXNlcjEzNjI5ODA5",
"avatar_url": "https://avatars.githubusercontent.com/u/13629809?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ppisljar",
"html_url": "https://github.com/ppisljar",
"followers_url": "https://api.github.com/users/ppisljar/followers",
"following_url": "https://api.github.com/users/ppisljar/following{/other_user}",
"gists_url": "https://api.github.com/users/ppisljar/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ppisljar/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ppisljar/subscriptions",
"organizations_url": "https://api.github.com/users/ppisljar/orgs",
"repos_url": "https://api.github.com/users/ppisljar/repos",
"events_url": "https://api.github.com/users/ppisljar/events{/privacy}",
"received_events_url": "https://api.github.com/users/ppisljar/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4418253970,
"node_id": "LA_kwDOC_bI7s8AAAABB1k0kg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/TTS",
"name": "TTS",
"color": "49AF28",
"default": false,
"description": ""
},
{
"id": 4585431311,
"node_id": "LA_kwDOC_bI7s8AAAABEVAhDw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale",
"name": "stale",
"color": "ededed",
"default": false,
"description": null
}
] | closed | false | null | [] | null | [
"jenkins",
"@titu1994 what should i do to move this forward ? thank you.",
"G2P would be @rlangman \r\nPlus, we cannot write to file inside of training step or val step. File writing must be done offline somewhere else cause ddp will introduce too many cases to it. ",
"file writing already happened here, i didnt change anything about it, except i skip file writing if dict is provided, so i guess for DDP this should work if file is provided and not generated, so i think this should be an improvement for tht case as well",
"jenkins",
"jenkins",
"This PR is stale because it has been open for 14 days with no activity. Remove stale label or comment or update or this will be closed in 7 days.",
"This PR was closed because it has been inactive for 7 days since being marked as stale.",
"i hope you can reopen this, i was still waiting for the review, as far as i am aware this is ready to merge",
"jenkins",
"jenkins",
"@ppisljar [this test](https://github.com/NVIDIA/NeMo/blob/main/Jenkinsfile#L1216) is still failing ",
"This PR is stale because it has been open for 14 days with no activity. Remove stale label or comment or update or this will be closed in 7 days.",
"This PR was closed because it has been inactive for 7 days since being marked as stale."
] | 2024-01-09T06:33:48 | 2024-04-13T01:39:20 | 2024-04-13T01:39:19 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8139",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8139",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8139.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8139.patch",
"merged_at": null
} |
# What does this PR do ?
ctc G2P model always generates the vocab_file which is only suitable for english. This allows you to set custom vocab_file thru config.
**Collection**: TTS/G2p
# Changelog
- allow passing custom vocab file to grapheme_tokenizer
| {
"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/8139/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/8139/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8138 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8138/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8138/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8138/events | https://github.com/NVIDIA/NeMo/pull/8138 | 2,071,417,158 | PR_kwDOC_bI7s5jhxVb | 8,138 | remove deprecated scripts | {
"login": "arendu",
"id": 108822655,
"node_id": "U_kgDOBnyAfw",
"avatar_url": "https://avatars.githubusercontent.com/u/108822655?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/arendu",
"html_url": "https://github.com/arendu",
"followers_url": "https://api.github.com/users/arendu/followers",
"following_url": "https://api.github.com/users/arendu/following{/other_user}",
"gists_url": "https://api.github.com/users/arendu/gists{/gist_id}",
"starred_url": "https://api.github.com/users/arendu/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/arendu/subscriptions",
"organizations_url": "https://api.github.com/users/arendu/orgs",
"repos_url": "https://api.github.com/users/arendu/repos",
"events_url": "https://api.github.com/users/arendu/events{/privacy}",
"received_events_url": "https://api.github.com/users/arendu/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"jenkins",
"jenkins"
] | 2024-01-09T00:08:39 | 2024-01-10T17:01:09 | 2024-01-10T17:01:08 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8138",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8138",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8138.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8138.patch",
"merged_at": "2024-01-10T17:01:08"
} | # What does this PR do ?
Removes deprecated scripts.
**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
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# Before your PR is "Ready for review"
**Pre checks**:
- [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md)
- [ ] Did you write any new necessary tests?
- [ ] Did you add or update any necessary documentation?
- [ ] Does the PR affect components that are optional to install? (Ex: Numba, Pynini, Apex etc)
- [ ] Reviewer: Does the PR have correct import guards for all optional libraries?
**PR Type**:
- [ ] New Feature
- [ ] Bugfix
- [ ] Documentation
If you haven't finished some of the above items you can still open "Draft" PR.
## Who can review?
Anyone in the community is free to review the PR once the checks have passed.
[Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas.
# Additional Information
* Related to # (issue)
| {
"login": "arendu",
"id": 108822655,
"node_id": "U_kgDOBnyAfw",
"avatar_url": "https://avatars.githubusercontent.com/u/108822655?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/arendu",
"html_url": "https://github.com/arendu",
"followers_url": "https://api.github.com/users/arendu/followers",
"following_url": "https://api.github.com/users/arendu/following{/other_user}",
"gists_url": "https://api.github.com/users/arendu/gists{/gist_id}",
"starred_url": "https://api.github.com/users/arendu/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/arendu/subscriptions",
"organizations_url": "https://api.github.com/users/arendu/orgs",
"repos_url": "https://api.github.com/users/arendu/repos",
"events_url": "https://api.github.com/users/arendu/events{/privacy}",
"received_events_url": "https://api.github.com/users/arendu/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8138/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/8138/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8137 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8137/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8137/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8137/events | https://github.com/NVIDIA/NeMo/pull/8137 | 2,071,408,348 | PR_kwDOC_bI7s5jhvdc | 8,137 | adding OnlineSampleMapping | {
"login": "arendu",
"id": 108822655,
"node_id": "U_kgDOBnyAfw",
"avatar_url": "https://avatars.githubusercontent.com/u/108822655?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/arendu",
"html_url": "https://github.com/arendu",
"followers_url": "https://api.github.com/users/arendu/followers",
"following_url": "https://api.github.com/users/arendu/following{/other_user}",
"gists_url": "https://api.github.com/users/arendu/gists{/gist_id}",
"starred_url": "https://api.github.com/users/arendu/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/arendu/subscriptions",
"organizations_url": "https://api.github.com/users/arendu/orgs",
"repos_url": "https://api.github.com/users/arendu/repos",
"events_url": "https://api.github.com/users/arendu/events{/privacy}",
"received_events_url": "https://api.github.com/users/arendu/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"jenkins",
"jenkins\r\n",
"jenkins"
] | 2024-01-08T23:57:31 | 2024-01-11T15:22:59 | 2024-01-11T15:22:58 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8137",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8137",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8137.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8137.patch",
"merged_at": "2024-01-11T15:22:58"
} | # 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
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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": "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/8137/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/8137/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8136 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8136/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8136/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8136/events | https://github.com/NVIDIA/NeMo/pull/8136 | 2,071,401,067 | PR_kwDOC_bI7s5jht3Z | 8,136 | Token count and sequence length logging for MegatronGPTSFTModel | {
"login": "vysarge",
"id": 10239441,
"node_id": "MDQ6VXNlcjEwMjM5NDQx",
"avatar_url": "https://avatars.githubusercontent.com/u/10239441?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/vysarge",
"html_url": "https://github.com/vysarge",
"followers_url": "https://api.github.com/users/vysarge/followers",
"following_url": "https://api.github.com/users/vysarge/following{/other_user}",
"gists_url": "https://api.github.com/users/vysarge/gists{/gist_id}",
"starred_url": "https://api.github.com/users/vysarge/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/vysarge/subscriptions",
"organizations_url": "https://api.github.com/users/vysarge/orgs",
"repos_url": "https://api.github.com/users/vysarge/repos",
"events_url": "https://api.github.com/users/vysarge/events{/privacy}",
"received_events_url": "https://api.github.com/users/vysarge/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"@cuichenx / @ericharper can you review?",
"jenkins",
"jenkins",
"jenkins"
] | 2024-01-08T23:47:46 | 2024-03-12T22:36:44 | 2024-01-14T23:34:10 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8136",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8136",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8136.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8136.patch",
"merged_at": "2024-01-14T23:34:10"
} | # What does this PR do ?
Adds an option to enable progress bar logging for sequence length (including padding tokens) and real token count for each train step with MegatronGPTSFTModel to enable accurate calculation of tokens/sec and related metrics.
**Collection**: nlp/language_modeling
# Changelog
- Alters GPTSFTDataset and GPTSFTPackedDataset to pass out a list of token counts as part of the batch
- Alters MegatronGPTSFTModel to expect, average, and log that list of token counts at the beginning of the step if enabled via config
# Usage
```
python megatron_gpt_peft_tuning.py \
... \
++model.log_token_counts=true
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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.
| {
"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/8136/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/8136/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8135 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8135/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8135/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8135/events | https://github.com/NVIDIA/NeMo/pull/8135 | 2,070,917,311 | PR_kwDOC_bI7s5jgDn_ | 8,135 | langid bug fix | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers",
"following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}",
"gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}",
"starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions",
"organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs",
"repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos",
"events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}",
"received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events",
"type": "Bot",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4556669155,
"node_id": "LA_kwDOC_bI7s8AAAABD5lA4w",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/cherry-pick",
"name": "cherry-pick",
"color": "ededed",
"default": false,
"description": null
},
{
"id": 4585431311,
"node_id": "LA_kwDOC_bI7s8AAAABEVAhDw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale",
"name": "stale",
"color": "ededed",
"default": false,
"description": null
},
{
"id": 4811884691,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR",
"name": "ASR",
"color": "F3B33E",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"jenkins",
"This PR is stale because it has been open for 14 days with no activity. Remove stale label or comment or update or this will be closed in 7 days.",
"This PR was closed because it has been inactive for 7 days since being marked as stale."
] | 2024-01-08T17:26:04 | 2024-02-04T01:44:49 | 2024-02-04T01:44:49 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8135",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8135",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8135.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8135.patch",
"merged_at": null
} | # What does this PR do ?
Bug fix for overwritten audio for num_segments>1
**Collection**: ASR
# Usage
* Can be used with num_segments>1
```python
lang_id_model.get_label(path2audio_file, segment_duration=20.0, num_segments = 3, random_seed: int = None)
```
**PR Type**:
- [ ] New Feature
- [ V] Bugfix
- [ ] Documentation
| {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers",
"following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}",
"gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}",
"starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions",
"organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs",
"repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos",
"events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}",
"received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events",
"type": "Bot",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8135/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/8135/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8134 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8134/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8134/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8134/events | https://github.com/NVIDIA/NeMo/pull/8134 | 2,070,640,187 | PR_kwDOC_bI7s5jfGyJ | 8,134 | langid bug fix | {
"login": "karpnv",
"id": 1645775,
"node_id": "MDQ6VXNlcjE2NDU3NzU=",
"avatar_url": "https://avatars.githubusercontent.com/u/1645775?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/karpnv",
"html_url": "https://github.com/karpnv",
"followers_url": "https://api.github.com/users/karpnv/followers",
"following_url": "https://api.github.com/users/karpnv/following{/other_user}",
"gists_url": "https://api.github.com/users/karpnv/gists{/gist_id}",
"starred_url": "https://api.github.com/users/karpnv/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/karpnv/subscriptions",
"organizations_url": "https://api.github.com/users/karpnv/orgs",
"repos_url": "https://api.github.com/users/karpnv/repos",
"events_url": "https://api.github.com/users/karpnv/events{/privacy}",
"received_events_url": "https://api.github.com/users/karpnv/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811884691,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR",
"name": "ASR",
"color": "F3B33E",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"jenkins"
] | 2024-01-08T15:02:33 | 2024-01-08T17:25:28 | 2024-01-08T17:25:27 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8134",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8134",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8134.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8134.patch",
"merged_at": "2024-01-08T17:25:27"
} | # What does this PR do ?
Bug fix for overwritten audio for num_segments>1
**Collection**: ASR
# Usage
* Can be used with num_segments>1
```python
lang_id_model.get_label(path2audio_file, segment_duration=20.0, num_segments = 3, random_seed: int = None)
```
**PR Type**:
- [ ] New Feature
- [ V] 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/8134/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/8134/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8131 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8131/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8131/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8131/events | https://github.com/NVIDIA/NeMo/pull/8131 | 2,068,021,332 | PR_kwDOC_bI7s5jWUNN | 8,131 | Improve PEFT UX | {
"login": "cuichenx",
"id": 43478052,
"node_id": "MDQ6VXNlcjQzNDc4MDUy",
"avatar_url": "https://avatars.githubusercontent.com/u/43478052?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/cuichenx",
"html_url": "https://github.com/cuichenx",
"followers_url": "https://api.github.com/users/cuichenx/followers",
"following_url": "https://api.github.com/users/cuichenx/following{/other_user}",
"gists_url": "https://api.github.com/users/cuichenx/gists{/gist_id}",
"starred_url": "https://api.github.com/users/cuichenx/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/cuichenx/subscriptions",
"organizations_url": "https://api.github.com/users/cuichenx/orgs",
"repos_url": "https://api.github.com/users/cuichenx/repos",
"events_url": "https://api.github.com/users/cuichenx/events{/privacy}",
"received_events_url": "https://api.github.com/users/cuichenx/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"jenkins"
] | 2024-01-05T20:47:12 | 2024-01-08T20:04:44 | 2024-01-08T20:04:43 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8131",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8131",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8131.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8131.patch",
"merged_at": "2024-01-08T20:04:43"
} | # What does this PR do ?
Two improvements of PEFT UX
1. Correct layer_selection logging messages in the case of more than one adapters.
2. Make output (label, ground truth) field optional for test data, for use cases where user only wants to generate text with gpt_peft_eval and not calculate any metrics.
- Note that using the default config file, `data.test_ds.label_key` and `data.test_ds.prompt_template` still reference an "output" field. These do not need to be changed. This PR simply ignores a key error if `data.test_ds.label_key` does not exist in the test data file.
**Collection**: NLP
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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
- [x] User Experience Improvement
If you haven't finished some of the above items you can still open "Draft" PR.
## Who can review?
Anyone in the community is free to review the PR once the checks have passed.
[Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas.
# Additional Information
* Related to # (issue)
| {
"login": "arendu",
"id": 108822655,
"node_id": "U_kgDOBnyAfw",
"avatar_url": "https://avatars.githubusercontent.com/u/108822655?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/arendu",
"html_url": "https://github.com/arendu",
"followers_url": "https://api.github.com/users/arendu/followers",
"following_url": "https://api.github.com/users/arendu/following{/other_user}",
"gists_url": "https://api.github.com/users/arendu/gists{/gist_id}",
"starred_url": "https://api.github.com/users/arendu/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/arendu/subscriptions",
"organizations_url": "https://api.github.com/users/arendu/orgs",
"repos_url": "https://api.github.com/users/arendu/repos",
"events_url": "https://api.github.com/users/arendu/events{/privacy}",
"received_events_url": "https://api.github.com/users/arendu/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8131/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/8131/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8130 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8130/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8130/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8130/events | https://github.com/NVIDIA/NeMo/pull/8130 | 2,067,891,001 | PR_kwDOC_bI7s5jV3cA | 8,130 | add war fix for sync issues | {
"login": "gshennvm",
"id": 119401249,
"node_id": "U_kgDOBx3rIQ",
"avatar_url": "https://avatars.githubusercontent.com/u/119401249?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/gshennvm",
"html_url": "https://github.com/gshennvm",
"followers_url": "https://api.github.com/users/gshennvm/followers",
"following_url": "https://api.github.com/users/gshennvm/following{/other_user}",
"gists_url": "https://api.github.com/users/gshennvm/gists{/gist_id}",
"starred_url": "https://api.github.com/users/gshennvm/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/gshennvm/subscriptions",
"organizations_url": "https://api.github.com/users/gshennvm/orgs",
"repos_url": "https://api.github.com/users/gshennvm/repos",
"events_url": "https://api.github.com/users/gshennvm/events{/privacy}",
"received_events_url": "https://api.github.com/users/gshennvm/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"jenkins"
] | 2024-01-05T18:58:40 | 2024-01-05T23:49:12 | 2024-01-05T23:49:07 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8130",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8130",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8130.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8130.patch",
"merged_at": "2024-01-05T23:49:07"
} | # What does this PR do ?
this is a war around a deeper issue where checkpointing is not guarded by pytorch distributed calls. So sometimes it removes checkpoints that don't exist and rmtree errors | {
"login": "gshennvm",
"id": 119401249,
"node_id": "U_kgDOBx3rIQ",
"avatar_url": "https://avatars.githubusercontent.com/u/119401249?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/gshennvm",
"html_url": "https://github.com/gshennvm",
"followers_url": "https://api.github.com/users/gshennvm/followers",
"following_url": "https://api.github.com/users/gshennvm/following{/other_user}",
"gists_url": "https://api.github.com/users/gshennvm/gists{/gist_id}",
"starred_url": "https://api.github.com/users/gshennvm/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/gshennvm/subscriptions",
"organizations_url": "https://api.github.com/users/gshennvm/orgs",
"repos_url": "https://api.github.com/users/gshennvm/repos",
"events_url": "https://api.github.com/users/gshennvm/events{/privacy}",
"received_events_url": "https://api.github.com/users/gshennvm/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8130/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/8130/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8129 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8129/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8129/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8129/events | https://github.com/NVIDIA/NeMo/pull/8129 | 2,067,746,564 | PR_kwDOC_bI7s5jVW_O | 8,129 | Use latest apex internal API | {
"login": "jbaczek",
"id": 45043825,
"node_id": "MDQ6VXNlcjQ1MDQzODI1",
"avatar_url": "https://avatars.githubusercontent.com/u/45043825?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jbaczek",
"html_url": "https://github.com/jbaczek",
"followers_url": "https://api.github.com/users/jbaczek/followers",
"following_url": "https://api.github.com/users/jbaczek/following{/other_user}",
"gists_url": "https://api.github.com/users/jbaczek/gists{/gist_id}",
"starred_url": "https://api.github.com/users/jbaczek/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/jbaczek/subscriptions",
"organizations_url": "https://api.github.com/users/jbaczek/orgs",
"repos_url": "https://api.github.com/users/jbaczek/repos",
"events_url": "https://api.github.com/users/jbaczek/events{/privacy}",
"received_events_url": "https://api.github.com/users/jbaczek/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [] | 2024-01-05T17:27:15 | 2024-01-16T16:08:14 | 2024-01-16T16:08:13 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8129",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8129",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8129.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8129.patch",
"merged_at": "2024-01-16T16:08:13"
} | # What does this PR do ?
Apex _fast_layer_norm added mandatory argument `memory_efficient`. This PR fixes call to this function to accommodate to the change.
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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
## Who can review?
@mikolajblaz can you review this one? I can see that you introduced this line 17 months ago. Can we move to the APEX implementation entirely as suggested in the TODO?
# Additional Information
* Related to # (issue)
| {
"login": "layalir",
"id": 31741533,
"node_id": "MDQ6VXNlcjMxNzQxNTMz",
"avatar_url": "https://avatars.githubusercontent.com/u/31741533?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/layalir",
"html_url": "https://github.com/layalir",
"followers_url": "https://api.github.com/users/layalir/followers",
"following_url": "https://api.github.com/users/layalir/following{/other_user}",
"gists_url": "https://api.github.com/users/layalir/gists{/gist_id}",
"starred_url": "https://api.github.com/users/layalir/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/layalir/subscriptions",
"organizations_url": "https://api.github.com/users/layalir/orgs",
"repos_url": "https://api.github.com/users/layalir/repos",
"events_url": "https://api.github.com/users/layalir/events{/privacy}",
"received_events_url": "https://api.github.com/users/layalir/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8129/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/8129/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8127 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8127/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8127/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8127/events | https://github.com/NVIDIA/NeMo/pull/8127 | 2,066,503,247 | PR_kwDOC_bI7s5jRLt3 | 8,127 | Final multimodal PR with our recent developments on MM side | {
"login": "yaoyu-33",
"id": 54727607,
"node_id": "MDQ6VXNlcjU0NzI3NjA3",
"avatar_url": "https://avatars.githubusercontent.com/u/54727607?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/yaoyu-33",
"html_url": "https://github.com/yaoyu-33",
"followers_url": "https://api.github.com/users/yaoyu-33/followers",
"following_url": "https://api.github.com/users/yaoyu-33/following{/other_user}",
"gists_url": "https://api.github.com/users/yaoyu-33/gists{/gist_id}",
"starred_url": "https://api.github.com/users/yaoyu-33/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/yaoyu-33/subscriptions",
"organizations_url": "https://api.github.com/users/yaoyu-33/orgs",
"repos_url": "https://api.github.com/users/yaoyu-33/repos",
"events_url": "https://api.github.com/users/yaoyu-33/events{/privacy}",
"received_events_url": "https://api.github.com/users/yaoyu-33/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"jenkins",
"jenkins",
"jenkins",
"jenkins",
"jenkins",
"jenkins",
"jenkins",
"jenkins",
"jenkins",
"jenkins",
"jenkins",
"jenkins",
"jenkins"
] | 2024-01-04T23:45:09 | 2024-01-20T01:59:24 | 2024-01-20T01:59:24 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8127",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8127",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8127.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8127.patch",
"merged_at": "2024-01-20T01:59:24"
} | # What does this PR do ?
multimodal PR with our recent developments on MM side after we start the open sourcing process. Mainly update in Neva/Llava and stable diffusion
**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
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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/8127/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/8127/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8126 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8126/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8126/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8126/events | https://github.com/NVIDIA/NeMo/pull/8126 | 2,066,487,423 | PR_kwDOC_bI7s5jRIQm | 8,126 | Fixes NVIDIA/apex installation to not erroneously install the `install` pkg | {
"login": "terrykong",
"id": 7576060,
"node_id": "MDQ6VXNlcjc1NzYwNjA=",
"avatar_url": "https://avatars.githubusercontent.com/u/7576060?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/terrykong",
"html_url": "https://github.com/terrykong",
"followers_url": "https://api.github.com/users/terrykong/followers",
"following_url": "https://api.github.com/users/terrykong/following{/other_user}",
"gists_url": "https://api.github.com/users/terrykong/gists{/gist_id}",
"starred_url": "https://api.github.com/users/terrykong/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/terrykong/subscriptions",
"organizations_url": "https://api.github.com/users/terrykong/orgs",
"repos_url": "https://api.github.com/users/terrykong/repos",
"events_url": "https://api.github.com/users/terrykong/events{/privacy}",
"received_events_url": "https://api.github.com/users/terrykong/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"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 | [] | 2024-01-04T23:26:07 | 2024-01-08T21:30:11 | 2024-01-08T21:17:39 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8126",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8126",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8126.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8126.patch",
"merged_at": "2024-01-08T21:17:39"
} | # What does this PR do ?
> Fixes NVIDIA/apex installation to not erroneously install the `install` pkg
**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
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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?
- [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**:
- [ ] New Feature
- [x] Bugfix
- [ ] Documentation
If you haven't finished some of the above items you can still open "Draft" PR.
## Who can review?
Anyone in the community is free to review the PR once the checks have passed.
[Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas.
# Additional Information
* Related to # (issue)
| {
"login": "ericharper",
"id": 11999610,
"node_id": "MDQ6VXNlcjExOTk5NjEw",
"avatar_url": "https://avatars.githubusercontent.com/u/11999610?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ericharper",
"html_url": "https://github.com/ericharper",
"followers_url": "https://api.github.com/users/ericharper/followers",
"following_url": "https://api.github.com/users/ericharper/following{/other_user}",
"gists_url": "https://api.github.com/users/ericharper/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ericharper/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ericharper/subscriptions",
"organizations_url": "https://api.github.com/users/ericharper/orgs",
"repos_url": "https://api.github.com/users/ericharper/repos",
"events_url": "https://api.github.com/users/ericharper/events{/privacy}",
"received_events_url": "https://api.github.com/users/ericharper/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8126/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/8126/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8125 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8125/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8125/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8125/events | https://github.com/NVIDIA/NeMo/pull/8125 | 2,066,257,776 | PR_kwDOC_bI7s5jQWGZ | 8,125 | Update post | {
"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 | [] | 2024-01-04T20:01:05 | 2024-01-04T20:01:57 | 2024-01-04T20:01:56 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8125",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8125",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8125.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8125.patch",
"merged_at": "2024-01-04T20:01:56"
} | # What does this PR do ?
Reorder content
**Collection**: [ASR]
# Before your PR is "Ready for review"
**Pre checks**:
- [x] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md)
- [ ] Did you write any new necessary tests?
- [ ] Did you add or update any necessary documentation?
- [ ] Does the PR affect components that are optional to install? (Ex: Numba, Pynini, Apex etc)
- [ ] Reviewer: Does the PR have correct import guards for all optional libraries?
**PR Type**:
- [ ] New Feature
- [ ] Bugfix
- [x] Documentation
| {
"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/8125/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/8125/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8124 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8124/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8124/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8124/events | https://github.com/NVIDIA/NeMo/pull/8124 | 2,065,905,418 | PR_kwDOC_bI7s5jPKW1 | 8,124 | Fix Adapter for MCore models | {
"login": "cuichenx",
"id": 43478052,
"node_id": "MDQ6VXNlcjQzNDc4MDUy",
"avatar_url": "https://avatars.githubusercontent.com/u/43478052?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/cuichenx",
"html_url": "https://github.com/cuichenx",
"followers_url": "https://api.github.com/users/cuichenx/followers",
"following_url": "https://api.github.com/users/cuichenx/following{/other_user}",
"gists_url": "https://api.github.com/users/cuichenx/gists{/gist_id}",
"starred_url": "https://api.github.com/users/cuichenx/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/cuichenx/subscriptions",
"organizations_url": "https://api.github.com/users/cuichenx/orgs",
"repos_url": "https://api.github.com/users/cuichenx/repos",
"events_url": "https://api.github.com/users/cuichenx/events{/privacy}",
"received_events_url": "https://api.github.com/users/cuichenx/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"jenkins"
] | 2024-01-04T15:55:07 | 2024-01-05T17:41:18 | 2024-01-05T17:41:17 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8124",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8124",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8124.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8124.patch",
"merged_at": "2024-01-05T17:41:17"
} | # What does this PR do ?
Adapter for MCore models was broken due to upstream changes in Megatron-LM. This PR updates the adapter code paths in NeMo to accommodate the upstream changes.
Tested on Llama2-7b on Squad dataset and accuracy matches before
**Collection**: NLP
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# Before your PR is "Ready for review"
**Pre checks**:
- [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md)
- [ ] Did you write any new necessary tests?
- [ ] Did you add or update any necessary documentation?
- [ ] Does the PR affect components that are optional to install? (Ex: Numba, Pynini, Apex etc)
- [ ] Reviewer: Does the PR have correct import guards for all optional libraries?
**PR Type**:
- [ ] New Feature
- [x] Bugfix
- [ ] Documentation
If you haven't finished some of the above items you can still open "Draft" PR.
## Who can review?
Anyone in the community is free to review the PR once the checks have passed.
[Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas.
# Additional Information
* Related to # (issue)
| {
"login": "arendu",
"id": 108822655,
"node_id": "U_kgDOBnyAfw",
"avatar_url": "https://avatars.githubusercontent.com/u/108822655?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/arendu",
"html_url": "https://github.com/arendu",
"followers_url": "https://api.github.com/users/arendu/followers",
"following_url": "https://api.github.com/users/arendu/following{/other_user}",
"gists_url": "https://api.github.com/users/arendu/gists{/gist_id}",
"starred_url": "https://api.github.com/users/arendu/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/arendu/subscriptions",
"organizations_url": "https://api.github.com/users/arendu/orgs",
"repos_url": "https://api.github.com/users/arendu/repos",
"events_url": "https://api.github.com/users/arendu/events{/privacy}",
"received_events_url": "https://api.github.com/users/arendu/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8124/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/8124/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8122 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8122/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8122/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8122/events | https://github.com/NVIDIA/NeMo/pull/8122 | 2,065,366,639 | PR_kwDOC_bI7s5jNU4I | 8,122 | README edit to change Apple Silicon install instructions (to fix a break introduced by pytorch 2) | {
"login": "stephenmcconnachie",
"id": 16166531,
"node_id": "MDQ6VXNlcjE2MTY2NTMx",
"avatar_url": "https://avatars.githubusercontent.com/u/16166531?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/stephenmcconnachie",
"html_url": "https://github.com/stephenmcconnachie",
"followers_url": "https://api.github.com/users/stephenmcconnachie/followers",
"following_url": "https://api.github.com/users/stephenmcconnachie/following{/other_user}",
"gists_url": "https://api.github.com/users/stephenmcconnachie/gists{/gist_id}",
"starred_url": "https://api.github.com/users/stephenmcconnachie/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/stephenmcconnachie/subscriptions",
"organizations_url": "https://api.github.com/users/stephenmcconnachie/orgs",
"repos_url": "https://api.github.com/users/stephenmcconnachie/repos",
"events_url": "https://api.github.com/users/stephenmcconnachie/events{/privacy}",
"received_events_url": "https://api.github.com/users/stephenmcconnachie/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"git clone and cd'ing into NeMo became obsolete when installing from pypi. Imo those commands can be removed?\r\n```shell\r\ngit clone https://github.com/NVIDIA/NeMo\r\ncd NeMo\r\n```\r\nIn my case, I only get v1.21 installed from pypi. With `pip install 'nemo-toolkit[asr]==1.22.0'` I again run into an triton dependency issue:\r\n`ERROR: Could not find a version that satisfies the requirement triton (from nemo-toolkit) (from versions: none)`",
"According to [this comment](https://github.com/openai/triton/issues/1057#issuecomment-1500691831), macOS is not supported by triton at all.\r\nWhat eventually _did_ work for me was removing `triton` from the requirements altogether and install from the local checkout.\r\n`$ pip install '.[asr]'`"
] | 2024-01-04T10:12:43 | 2024-02-10T15:48:16 | 2024-01-04T17:27:51 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8122",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8122",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8122.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8122.patch",
"merged_at": "2024-01-04T17:27:51"
} | # What does this PR do ?
This PR aims to correct a small error in the Apple Silcon installation instructions in README, where a triton dependency is problematic due to pytorch 2. The use of `/reinstall.sh` doesn't work, so using `pip install` instead. Also added a note about MacBook, to clarify that only asr toolkit is supported.
# Additional Information
* Related to #8116
| {
"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/8122/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/8122/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8121 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8121/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8121/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8121/events | https://github.com/NVIDIA/NeMo/pull/8121 | 2,064,876,795 | PR_kwDOC_bI7s5jLtAS | 8,121 | Draft:add fp8 weight caching to mcore_gpt | {
"login": "JimmyZhang12",
"id": 67203904,
"node_id": "MDQ6VXNlcjY3MjAzOTA0",
"avatar_url": "https://avatars.githubusercontent.com/u/67203904?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/JimmyZhang12",
"html_url": "https://github.com/JimmyZhang12",
"followers_url": "https://api.github.com/users/JimmyZhang12/followers",
"following_url": "https://api.github.com/users/JimmyZhang12/following{/other_user}",
"gists_url": "https://api.github.com/users/JimmyZhang12/gists{/gist_id}",
"starred_url": "https://api.github.com/users/JimmyZhang12/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/JimmyZhang12/subscriptions",
"organizations_url": "https://api.github.com/users/JimmyZhang12/orgs",
"repos_url": "https://api.github.com/users/JimmyZhang12/repos",
"events_url": "https://api.github.com/users/JimmyZhang12/events{/privacy}",
"received_events_url": "https://api.github.com/users/JimmyZhang12/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4585431311,
"node_id": "LA_kwDOC_bI7s8AAAABEVAhDw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale",
"name": "stale",
"color": "ededed",
"default": false,
"description": null
},
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"@erhoo82 can you please review?",
"Can you add the case handling the validation step as well? At validation, we need to cache the fp8 w/t at the first micro-batch in the first global batch, and no need to cache in the following global batches. I think you can use `model.training` to tell if the current micro-batch belongs to the training or validation routine. Ned to tell if the current global-batch is the first global-batch after a training step routine.\r\n\r\nThis is the reference. \r\nhttps://github.com/NVIDIA/NeMo/blob/main/nemo/collections/nlp/modules/common/megatron/transformer.py#L1543-L1544",
"I tested validation and it seems ok; the global batch size is the same between validation and training, so the is_first_microbatch logic is unaffected, and validation and training both share `get_forward_output_and_loss_func`\r\nThe case where the global batch size is changed should be handled by this [code](https://github.com/NVIDIA/NeMo/blob/99fa120c637bafd7ec364ff2159661ec0784a4e0/nemo/collections/nlp/models/language_modeling/megatron_gpt_model.py#L890): ",
"In validation, we only need to cache the fp8 w/t at `the first global-batch` + `the first micro-batch`. Did you check if fp8 w/t is not cached in the other first micro-batch in the later global-batch?",
"Good point thanks, I added a fix for validation ",
"This PR is stale because it has been open for 14 days with no activity. Remove stale label or comment or update or this will be closed in 7 days.",
"This PR was closed because it has been inactive for 7 days since being marked as stale."
] | 2024-01-04T02:01:11 | 2024-02-01T01:45:00 | 2024-02-01T01:44:59 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | true | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8121",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8121",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8121.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8121.patch",
"merged_at": null
} | # What does this PR do ?
Add a one line overview of what this PR aims to accomplish.
**Collection**: [Note which collection this PR will affect]
# Changelog
- Add specific line by line info of high level changes in this PR.
# Usage
* You can potentially add a usage example below
```python
# Add a code snippet demonstrating how to use this
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# Before your PR is "Ready for review"
**Pre checks**:
- [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md)
- [ ] Did you write any new necessary tests?
- [ ] Did you add or update any necessary documentation?
- [ ] Does the PR affect components that are optional to install? (Ex: Numba, Pynini, Apex etc)
- [ ] Reviewer: Does the PR have correct import guards for all optional libraries?
**PR Type**:
- [ ] New Feature
- [ ] Bugfix
- [ ] Documentation
If you haven't finished some of the above items you can still open "Draft" PR.
## Who can review?
Anyone in the community is free to review the PR once the checks have passed.
[Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas.
# Additional Information
* Related to # (issue)
| {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers",
"following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}",
"gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}",
"starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions",
"organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs",
"repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos",
"events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}",
"received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events",
"type": "Bot",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8121/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/8121/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8120 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8120/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8120/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8120/events | https://github.com/NVIDIA/NeMo/pull/8120 | 2,064,611,880 | PR_kwDOC_bI7s5jK1KJ | 8,120 | fix gated_linear_unit config | {
"login": "JimmyZhang12",
"id": 67203904,
"node_id": "MDQ6VXNlcjY3MjAzOTA0",
"avatar_url": "https://avatars.githubusercontent.com/u/67203904?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/JimmyZhang12",
"html_url": "https://github.com/JimmyZhang12",
"followers_url": "https://api.github.com/users/JimmyZhang12/followers",
"following_url": "https://api.github.com/users/JimmyZhang12/following{/other_user}",
"gists_url": "https://api.github.com/users/JimmyZhang12/gists{/gist_id}",
"starred_url": "https://api.github.com/users/JimmyZhang12/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/JimmyZhang12/subscriptions",
"organizations_url": "https://api.github.com/users/JimmyZhang12/orgs",
"repos_url": "https://api.github.com/users/JimmyZhang12/repos",
"events_url": "https://api.github.com/users/JimmyZhang12/events{/privacy}",
"received_events_url": "https://api.github.com/users/JimmyZhang12/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"jenkins",
"jenkins"
] | 2024-01-03T20:32:55 | 2024-01-04T21:53:22 | 2024-01-04T21:53:22 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8120",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8120",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8120.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8120.patch",
"merged_at": null
} | # What does this PR do ?
Add a one line overview of what this PR aims to accomplish.
**Collection**: [Note which collection this PR will affect]
# Changelog
- Add specific line by line info of high level changes in this PR.
# Usage
* You can potentially add a usage example below
```python
# Add a code snippet demonstrating how to use this
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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": "JimmyZhang12",
"id": 67203904,
"node_id": "MDQ6VXNlcjY3MjAzOTA0",
"avatar_url": "https://avatars.githubusercontent.com/u/67203904?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/JimmyZhang12",
"html_url": "https://github.com/JimmyZhang12",
"followers_url": "https://api.github.com/users/JimmyZhang12/followers",
"following_url": "https://api.github.com/users/JimmyZhang12/following{/other_user}",
"gists_url": "https://api.github.com/users/JimmyZhang12/gists{/gist_id}",
"starred_url": "https://api.github.com/users/JimmyZhang12/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/JimmyZhang12/subscriptions",
"organizations_url": "https://api.github.com/users/JimmyZhang12/orgs",
"repos_url": "https://api.github.com/users/JimmyZhang12/repos",
"events_url": "https://api.github.com/users/JimmyZhang12/events{/privacy}",
"received_events_url": "https://api.github.com/users/JimmyZhang12/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8120/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/8120/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8119 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8119/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8119/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8119/events | https://github.com/NVIDIA/NeMo/pull/8119 | 2,064,306,918 | PR_kwDOC_bI7s5jJzT1 | 8,119 | Add missing audio_filepath validation for Canary | {
"login": "pzelasko",
"id": 15930688,
"node_id": "MDQ6VXNlcjE1OTMwNjg4",
"avatar_url": "https://avatars.githubusercontent.com/u/15930688?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/pzelasko",
"html_url": "https://github.com/pzelasko",
"followers_url": "https://api.github.com/users/pzelasko/followers",
"following_url": "https://api.github.com/users/pzelasko/following{/other_user}",
"gists_url": "https://api.github.com/users/pzelasko/gists{/gist_id}",
"starred_url": "https://api.github.com/users/pzelasko/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/pzelasko/subscriptions",
"organizations_url": "https://api.github.com/users/pzelasko/orgs",
"repos_url": "https://api.github.com/users/pzelasko/repos",
"events_url": "https://api.github.com/users/pzelasko/events{/privacy}",
"received_events_url": "https://api.github.com/users/pzelasko/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4847373924,
"node_id": "LA_kwDOC_bI7s8AAAABIO0OZA",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/common",
"name": "common",
"color": "c5def5",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [] | 2024-01-03T16:21:54 | 2024-01-03T17:01:45 | 2024-01-03T17:01:45 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8119",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8119",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8119.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8119.patch",
"merged_at": "2024-01-03T17:01:45"
} | # 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
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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": "krishnacpuvvada",
"id": 93558329,
"node_id": "U_kgDOBZOWOQ",
"avatar_url": "https://avatars.githubusercontent.com/u/93558329?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/krishnacpuvvada",
"html_url": "https://github.com/krishnacpuvvada",
"followers_url": "https://api.github.com/users/krishnacpuvvada/followers",
"following_url": "https://api.github.com/users/krishnacpuvvada/following{/other_user}",
"gists_url": "https://api.github.com/users/krishnacpuvvada/gists{/gist_id}",
"starred_url": "https://api.github.com/users/krishnacpuvvada/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/krishnacpuvvada/subscriptions",
"organizations_url": "https://api.github.com/users/krishnacpuvvada/orgs",
"repos_url": "https://api.github.com/users/krishnacpuvvada/repos",
"events_url": "https://api.github.com/users/krishnacpuvvada/events{/privacy}",
"received_events_url": "https://api.github.com/users/krishnacpuvvada/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8119/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/8119/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8118 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8118/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8118/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8118/events | https://github.com/NVIDIA/NeMo/issues/8118 | 2,064,183,569 | I_kwDOC_bI7s57CPER | 8,118 | Remove Pydantic Version Constraint Due to Inflect Update | {
"login": "Joost385",
"id": 61097554,
"node_id": "MDQ6VXNlcjYxMDk3NTU0",
"avatar_url": "https://avatars.githubusercontent.com/u/61097554?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Joost385",
"html_url": "https://github.com/Joost385",
"followers_url": "https://api.github.com/users/Joost385/followers",
"following_url": "https://api.github.com/users/Joost385/following{/other_user}",
"gists_url": "https://api.github.com/users/Joost385/gists{/gist_id}",
"starred_url": "https://api.github.com/users/Joost385/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/Joost385/subscriptions",
"organizations_url": "https://api.github.com/users/Joost385/orgs",
"repos_url": "https://api.github.com/users/Joost385/repos",
"events_url": "https://api.github.com/users/Joost385/events{/privacy}",
"received_events_url": "https://api.github.com/users/Joost385/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | {
"login": "titu1994",
"id": 3048602,
"node_id": "MDQ6VXNlcjMwNDg2MDI=",
"avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/titu1994",
"html_url": "https://github.com/titu1994",
"followers_url": "https://api.github.com/users/titu1994/followers",
"following_url": "https://api.github.com/users/titu1994/following{/other_user}",
"gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}",
"starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/titu1994/subscriptions",
"organizations_url": "https://api.github.com/users/titu1994/orgs",
"repos_url": "https://api.github.com/users/titu1994/repos",
"events_url": "https://api.github.com/users/titu1994/events{/privacy}",
"received_events_url": "https://api.github.com/users/titu1994/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"login": "titu1994",
"id": 3048602,
"node_id": "MDQ6VXNlcjMwNDg2MDI=",
"avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/titu1994",
"html_url": "https://github.com/titu1994",
"followers_url": "https://api.github.com/users/titu1994/followers",
"following_url": "https://api.github.com/users/titu1994/following{/other_user}",
"gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}",
"starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/titu1994/subscriptions",
"organizations_url": "https://api.github.com/users/titu1994/orgs",
"repos_url": "https://api.github.com/users/titu1994/repos",
"events_url": "https://api.github.com/users/titu1994/events{/privacy}",
"received_events_url": "https://api.github.com/users/titu1994/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
}
] | null | [
"I'll unpin it tomorrow, seems it should be fine. "
] | 2024-01-03T15:01:47 | 2024-01-12T21:40:18 | 2024-01-12T21:40:18 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | Currently, pydantic is pinned to version 1, which leads to conflicts with common frameworks such as fastapi.
https://github.com/NVIDIA/NeMo/blob/d3c589646b453f0395f19c2eef1ed958d048c6c7/requirements/requirements_common.txt#L4
It seems that inflect now supports pydantic 2, see https://github.com/jaraco/inflect/issues/187. Their requirements now list `pydantic >= 1.9.1`. Given this update, is it possible to remove the pydantic version constraint without further 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/8118/reactions",
"total_count": 1,
"+1": 1,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/NVIDIA/NeMo/issues/8118/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/8117 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8117/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8117/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8117/events | https://github.com/NVIDIA/NeMo/pull/8117 | 2,064,049,952 | PR_kwDOC_bI7s5jI7is | 8,117 | Added marathi ITN support in documentation | {
"login": "ChinmayPatil11",
"id": 72211393,
"node_id": "MDQ6VXNlcjcyMjExMzkz",
"avatar_url": "https://avatars.githubusercontent.com/u/72211393?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ChinmayPatil11",
"html_url": "https://github.com/ChinmayPatil11",
"followers_url": "https://api.github.com/users/ChinmayPatil11/followers",
"following_url": "https://api.github.com/users/ChinmayPatil11/following{/other_user}",
"gists_url": "https://api.github.com/users/ChinmayPatil11/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ChinmayPatil11/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ChinmayPatil11/subscriptions",
"organizations_url": "https://api.github.com/users/ChinmayPatil11/orgs",
"repos_url": "https://api.github.com/users/ChinmayPatil11/repos",
"events_url": "https://api.github.com/users/ChinmayPatil11/events{/privacy}",
"received_events_url": "https://api.github.com/users/ChinmayPatil11/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4585431311,
"node_id": "LA_kwDOC_bI7s8AAAABEVAhDw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale",
"name": "stale",
"color": "ededed",
"default": false,
"description": null
},
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"This PR is stale because it has been open for 14 days with no activity. Remove stale label or comment or update or this will be closed in 7 days.",
"Commenting to remove stale tag",
"This PR is stale because it has been open for 14 days with no activity. Remove stale label or comment or update or this will be closed in 7 days.",
"This PR was closed because it has been inactive for 7 days since being marked as stale."
] | 2024-01-03T13:35:27 | 2024-02-16T01:44:12 | 2024-02-16T01:44:11 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8117",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8117",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8117.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8117.patch",
"merged_at": null
} | # What does this PR do ?
Added marathi ITN support to the documentation in wfst_text_normalization.rst. This corresponds to the pr https://github.com/NVIDIA/NeMo-text-processing/pull/134 in nemo_text_processing which adds ITN support for Marathi language.
# Changelog
- Added marathi ITN support to the given file NeMo/docs/source/nlp/text_normalization/wfst/wfst_text_normalization.rst .
# 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 add or update any necessary documentation?
**PR Type**:
- [x] New Feature
- [x] Documentation
If you haven't finished some of the above items you can still open "Draft" PR.
## Who can review?
Anyone in the community is free to review the PR once the checks have passed. | {
"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/8117/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/8117/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8116 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8116/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8116/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8116/events | https://github.com/NVIDIA/NeMo/issues/8116 | 2,063,766,824 | I_kwDOC_bI7s57ApUo | 8,116 | Installation error on M1 Macbook - No matching distribution found for triton | {
"login": "stephenmcconnachie",
"id": 16166531,
"node_id": "MDQ6VXNlcjE2MTY2NTMx",
"avatar_url": "https://avatars.githubusercontent.com/u/16166531?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/stephenmcconnachie",
"html_url": "https://github.com/stephenmcconnachie",
"followers_url": "https://api.github.com/users/stephenmcconnachie/followers",
"following_url": "https://api.github.com/users/stephenmcconnachie/following{/other_user}",
"gists_url": "https://api.github.com/users/stephenmcconnachie/gists{/gist_id}",
"starred_url": "https://api.github.com/users/stephenmcconnachie/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/stephenmcconnachie/subscriptions",
"organizations_url": "https://api.github.com/users/stephenmcconnachie/orgs",
"repos_url": "https://api.github.com/users/stephenmcconnachie/repos",
"events_url": "https://api.github.com/users/stephenmcconnachie/events{/privacy}",
"received_events_url": "https://api.github.com/users/stephenmcconnachie/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 | [
"Triton dependency comes from pytorch 2.0 - you can try to install pytorch CPU first and then install NeMo instead of installing everything at the same time ",
"Thank you! I used conda to install, like this:\r\n\r\n\r\n\r\nBut still getting this from ./reinstall.sh\r\n\r\n\r\n\r\nI'll dig into it a bit more.",
"Btw only the asr toolkit can be guaranteed installed on MacBook. Don't use reinstall.sh, use pip install nemo_toolkit[asr] ",
"Those instructions were fine for pytorch 1.x but triton complicates things ",
"Thank you so much for your help - that worked with `pip install 'nemo_toolkit[asr]'`"
] | 2024-01-03T10:53:24 | 2024-01-04T09:48:30 | 2024-01-04T09:48:13 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | I am installing NeMo on M1 Macbook, following the installation guide for M1:
https://github.com/NVIDIA/NeMo?tab=readme-ov-file#mac-computers-with-apple-silicon
All works fine in Conda environment, up to running ./reinstall.sh. After a few successful packages installed I get this error:
```
INFO: pip is looking at multiple versions of nemo-toolkit to determine which version is compatible with other requirements. This could take a while.
ERROR: Could not find a version that satisfies the requirement triton (from nemo-toolkit) (from versions: none)
ERROR: No matching distribution found for triton
```
The triton compatibility statement on the GitHub readme says Linux only:
https://github.com/openai/triton#compatibility
Is there a way to work around this in the M1 Mac installation?
- OS version: 14.1.2 (23B92)
- PyTorch version: 2.1.2
- Python version: 3.11.0
| {
"login": "stephenmcconnachie",
"id": 16166531,
"node_id": "MDQ6VXNlcjE2MTY2NTMx",
"avatar_url": "https://avatars.githubusercontent.com/u/16166531?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/stephenmcconnachie",
"html_url": "https://github.com/stephenmcconnachie",
"followers_url": "https://api.github.com/users/stephenmcconnachie/followers",
"following_url": "https://api.github.com/users/stephenmcconnachie/following{/other_user}",
"gists_url": "https://api.github.com/users/stephenmcconnachie/gists{/gist_id}",
"starred_url": "https://api.github.com/users/stephenmcconnachie/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/stephenmcconnachie/subscriptions",
"organizations_url": "https://api.github.com/users/stephenmcconnachie/orgs",
"repos_url": "https://api.github.com/users/stephenmcconnachie/repos",
"events_url": "https://api.github.com/users/stephenmcconnachie/events{/privacy}",
"received_events_url": "https://api.github.com/users/stephenmcconnachie/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8116/reactions",
"total_count": 2,
"+1": 2,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/NVIDIA/NeMo/issues/8116/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/8115 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8115/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8115/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8115/events | https://github.com/NVIDIA/NeMo/issues/8115 | 2,063,288,803 | I_kwDOC_bI7s56-0nj | 8,115 | Finetuning ASR models and change the vocabulary error！！！ | {
"login": "h310558606",
"id": 38098690,
"node_id": "MDQ6VXNlcjM4MDk4Njkw",
"avatar_url": "https://avatars.githubusercontent.com/u/38098690?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/h310558606",
"html_url": "https://github.com/h310558606",
"followers_url": "https://api.github.com/users/h310558606/followers",
"following_url": "https://api.github.com/users/h310558606/following{/other_user}",
"gists_url": "https://api.github.com/users/h310558606/gists{/gist_id}",
"starred_url": "https://api.github.com/users/h310558606/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/h310558606/subscriptions",
"organizations_url": "https://api.github.com/users/h310558606/orgs",
"repos_url": "https://api.github.com/users/h310558606/repos",
"events_url": "https://api.github.com/users/h310558606/events{/privacy}",
"received_events_url": "https://api.github.com/users/h310558606/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 | [
"That's more of a warning than an error. The actual crash is because of the OSErrorv- network is unreachable. HF wpe requires connecting to the HF servers even for local models, you need to use the environment flag TRANSFORMERS_OFFLINE=1 (or set that in os.environ before importing HF). \r\n\r\nOtherwise you can create and use a sentence piece tokenizer instead (recommended) ",
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"This issue was closed because it has been inactive for 7 days since being marked as stale."
] | 2024-01-03T05:59:24 | 2024-02-10T01:43:01 | 2024-02-10T01:43:00 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | My code:
....
model = nemo_asr.models.ASRModel.from_pretrained("stt_en_citrinet_512")
model.change_vocabulary(
new_tokenizer_dir="./tokenizers/an4/tokenizer_spe_unigram_v32/",
new_tokenizer_type="wpe"
)
**error:
You tried to register an artifact under config key=tokenizer.vocab_path but an artifact for it has already been registered**.
| {
"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/8115/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/8115/timeline | null | not_planned | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/8114 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8114/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8114/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8114/events | https://github.com/NVIDIA/NeMo/pull/8114 | 2,063,162,821 | PR_kwDOC_bI7s5jF2Xn | 8,114 | Add NeMo Parakeet post | {
"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 | [] | 2024-01-03T02:29:48 | 2024-01-03T20:30:50 | 2024-01-03T20:30:49 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8114",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8114",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8114.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8114.patch",
"merged_at": "2024-01-03T20:30:49"
} | # What does this PR do ?
Adds the content for Parakeet post.
**Collection**: [ASR]
# Changelog
- Add parakeet post
- Update template to refer to 2024
# Before your PR is "Ready for review"
**Pre checks**:
- [x] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md)
- [ ] Did you write any new necessary tests?
- [ ] Did you add or update any necessary documentation?
- [ ] Does the PR affect components that are optional to install? (Ex: Numba, Pynini, Apex etc)
- [ ] Reviewer: Does the PR have correct import guards for all optional libraries?
**PR Type**:
- [x] New Feature
- [ ] Bugfix
- [ ] Documentation | {
"login": "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/8114/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/8114/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8113 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8113/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8113/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8113/events | https://github.com/NVIDIA/NeMo/pull/8113 | 2,063,038,866 | PR_kwDOC_bI7s5jFclY | 8,113 | Fix lora merge script | {
"login": "cuichenx",
"id": 43478052,
"node_id": "MDQ6VXNlcjQzNDc4MDUy",
"avatar_url": "https://avatars.githubusercontent.com/u/43478052?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/cuichenx",
"html_url": "https://github.com/cuichenx",
"followers_url": "https://api.github.com/users/cuichenx/followers",
"following_url": "https://api.github.com/users/cuichenx/following{/other_user}",
"gists_url": "https://api.github.com/users/cuichenx/gists{/gist_id}",
"starred_url": "https://api.github.com/users/cuichenx/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/cuichenx/subscriptions",
"organizations_url": "https://api.github.com/users/cuichenx/orgs",
"repos_url": "https://api.github.com/users/cuichenx/repos",
"events_url": "https://api.github.com/users/cuichenx/events{/privacy}",
"received_events_url": "https://api.github.com/users/cuichenx/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"jenkins\r\n",
"jenkins"
] | 2024-01-02T22:45:41 | 2024-01-03T18:04:40 | 2024-01-03T18:04:39 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8113",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8113",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8113.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8113.patch",
"merged_at": "2024-01-03T18:04:39"
} | # What does this PR do ?
- Fix typo in "mcore" argument location
- Remove dependency on MegatronGPTLoRAModel, which has been deprecated since #7308
**Collection**: NLP
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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": "cuichenx",
"id": 43478052,
"node_id": "MDQ6VXNlcjQzNDc4MDUy",
"avatar_url": "https://avatars.githubusercontent.com/u/43478052?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/cuichenx",
"html_url": "https://github.com/cuichenx",
"followers_url": "https://api.github.com/users/cuichenx/followers",
"following_url": "https://api.github.com/users/cuichenx/following{/other_user}",
"gists_url": "https://api.github.com/users/cuichenx/gists{/gist_id}",
"starred_url": "https://api.github.com/users/cuichenx/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/cuichenx/subscriptions",
"organizations_url": "https://api.github.com/users/cuichenx/orgs",
"repos_url": "https://api.github.com/users/cuichenx/repos",
"events_url": "https://api.github.com/users/cuichenx/events{/privacy}",
"received_events_url": "https://api.github.com/users/cuichenx/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8113/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/8113/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8112 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8112/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8112/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8112/events | https://github.com/NVIDIA/NeMo/pull/8112 | 2,063,023,172 | PR_kwDOC_bI7s5jFZSs | 8,112 | Fix AST eval | {
"login": "stevehuang52",
"id": 105218074,
"node_id": "U_kgDOBkWAGg",
"avatar_url": "https://avatars.githubusercontent.com/u/105218074?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/stevehuang52",
"html_url": "https://github.com/stevehuang52",
"followers_url": "https://api.github.com/users/stevehuang52/followers",
"following_url": "https://api.github.com/users/stevehuang52/following{/other_user}",
"gists_url": "https://api.github.com/users/stevehuang52/gists{/gist_id}",
"starred_url": "https://api.github.com/users/stevehuang52/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/stevehuang52/subscriptions",
"organizations_url": "https://api.github.com/users/stevehuang52/orgs",
"repos_url": "https://api.github.com/users/stevehuang52/repos",
"events_url": "https://api.github.com/users/stevehuang52/events{/privacy}",
"received_events_url": "https://api.github.com/users/stevehuang52/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811884691,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR",
"name": "ASR",
"color": "F3B33E",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"jenkins",
"jenkins",
"jenkins"
] | 2024-01-02T22:23:24 | 2024-01-10T17:04:18 | 2024-01-10T17:04:17 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8112",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8112",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8112.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8112.patch",
"merged_at": "2024-01-10T17:04:17"
} | Fix bleu calculation for ASR evaluator | {
"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/8112/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/8112/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8111 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8111/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8111/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8111/events | https://github.com/NVIDIA/NeMo/pull/8111 | 2,063,022,527 | PR_kwDOC_bI7s5jFZKA | 8,111 | Multi codec support (nemo audio codec), CTC loss and context patterns | {
"login": "shehzeen",
"id": 7191931,
"node_id": "MDQ6VXNlcjcxOTE5MzE=",
"avatar_url": "https://avatars.githubusercontent.com/u/7191931?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/shehzeen",
"html_url": "https://github.com/shehzeen",
"followers_url": "https://api.github.com/users/shehzeen/followers",
"following_url": "https://api.github.com/users/shehzeen/following{/other_user}",
"gists_url": "https://api.github.com/users/shehzeen/gists{/gist_id}",
"starred_url": "https://api.github.com/users/shehzeen/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/shehzeen/subscriptions",
"organizations_url": "https://api.github.com/users/shehzeen/orgs",
"repos_url": "https://api.github.com/users/shehzeen/repos",
"events_url": "https://api.github.com/users/shehzeen/events{/privacy}",
"received_events_url": "https://api.github.com/users/shehzeen/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
},
{
"id": 4811884691,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR",
"name": "ASR",
"color": "F3B33E",
"default": false,
"description": ""
},
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"This PR is stale because it has been open for 14 days with no activity. Remove stale label or comment or update or this will be closed in 7 days.",
"This PR was closed because it has been inactive for 7 days since being marked as stale."
] | 2024-01-02T22:22:35 | 2024-01-24T01:47:05 | 2024-01-24T01:47:05 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8111",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8111",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8111.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8111.patch",
"merged_at": null
} | # What does this PR do ?
Add a one line overview of what this PR aims to accomplish.
**Collection**: [Note which collection this PR will affect]
# Changelog
- Add specific line by line info of high level changes in this PR.
# Usage
* You can potentially add a usage example below
```python
# Add a code snippet demonstrating how to use this
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# Before your PR is "Ready for review"
**Pre checks**:
- [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md)
- [ ] Did you write any new necessary tests?
- [ ] Did you add or update any necessary documentation?
- [ ] Does the PR affect components that are optional to install? (Ex: Numba, Pynini, Apex etc)
- [ ] Reviewer: Does the PR have correct import guards for all optional libraries?
**PR Type**:
- [ ] New Feature
- [ ] Bugfix
- [ ] Documentation
If you haven't finished some of the above items you can still open "Draft" PR.
## Who can review?
Anyone in the community is free to review the PR once the checks have passed.
[Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas.
# Additional Information
* Related to # (issue)
| {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers",
"following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}",
"gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}",
"starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions",
"organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs",
"repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos",
"events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}",
"received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events",
"type": "Bot",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8111/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/8111/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8110 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8110/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8110/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8110/events | https://github.com/NVIDIA/NeMo/pull/8110 | 2,062,747,079 | PR_kwDOC_bI7s5jEeS2 | 8,110 | transcribe fn for Canary models | {
"login": "krishnacpuvvada",
"id": 93558329,
"node_id": "U_kgDOBZOWOQ",
"avatar_url": "https://avatars.githubusercontent.com/u/93558329?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/krishnacpuvvada",
"html_url": "https://github.com/krishnacpuvvada",
"followers_url": "https://api.github.com/users/krishnacpuvvada/followers",
"following_url": "https://api.github.com/users/krishnacpuvvada/following{/other_user}",
"gists_url": "https://api.github.com/users/krishnacpuvvada/gists{/gist_id}",
"starred_url": "https://api.github.com/users/krishnacpuvvada/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/krishnacpuvvada/subscriptions",
"organizations_url": "https://api.github.com/users/krishnacpuvvada/orgs",
"repos_url": "https://api.github.com/users/krishnacpuvvada/repos",
"events_url": "https://api.github.com/users/krishnacpuvvada/events{/privacy}",
"received_events_url": "https://api.github.com/users/krishnacpuvvada/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811884691,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR",
"name": "ASR",
"color": "F3B33E",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [] | 2024-01-02T17:45:17 | 2024-01-09T03:44:37 | 2024-01-02T18:10:11 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8110",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8110",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8110.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8110.patch",
"merged_at": "2024-01-02T18:10:11"
} | # What does this PR do ?
Add a one line overview of what this PR aims to accomplish.
**Collection**: ASR
# 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
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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": "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/8110/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/8110/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8109 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8109/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8109/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8109/events | https://github.com/NVIDIA/NeMo/issues/8109 | 2,062,384,155 | I_kwDOC_bI7s567Xwb | 8,109 | Implementation of Disfluencies in TextToSpeech | {
"login": "rodrigoGA",
"id": 2334523,
"node_id": "MDQ6VXNlcjIzMzQ1MjM=",
"avatar_url": "https://avatars.githubusercontent.com/u/2334523?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/rodrigoGA",
"html_url": "https://github.com/rodrigoGA",
"followers_url": "https://api.github.com/users/rodrigoGA/followers",
"following_url": "https://api.github.com/users/rodrigoGA/following{/other_user}",
"gists_url": "https://api.github.com/users/rodrigoGA/gists{/gist_id}",
"starred_url": "https://api.github.com/users/rodrigoGA/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/rodrigoGA/subscriptions",
"organizations_url": "https://api.github.com/users/rodrigoGA/orgs",
"repos_url": "https://api.github.com/users/rodrigoGA/repos",
"events_url": "https://api.github.com/users/rodrigoGA/events{/privacy}",
"received_events_url": "https://api.github.com/users/rodrigoGA/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": 4418253970,
"node_id": "LA_kwDOC_bI7s8AAAABB1k0kg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/TTS",
"name": "TTS",
"color": "49AF28",
"default": false,
"description": ""
},
{
"id": 4585431311,
"node_id": "LA_kwDOC_bI7s8AAAABEVAhDw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale",
"name": "stale",
"color": "ededed",
"default": false,
"description": null
}
] | closed | false | null | [] | null | [
"Any update on this?",
"Hi @rodrigoGA, thanks for your interest in NeMo TTS toolkit. If sticking to our current phoneme-based TTS models, such as FastPitch, you have to add new IPA dictionary entries for those filler words, and better to have paired filler speech as well. \r\n\r\nSo far as I know, we haven't added support for disfluencies synthesis yet, unless the training corpus has filler speech/text pairs and corresponding phonemes. \r\n\r\nIf you have filler speech/text pairs, and can't figure out canonical phonemes for filler words, you may try to use grapheme-based tokenizer for FastPitch. ",
"Hi, thanks for the detailed response. It's really interesting and useful to consider support for disfluency synthesis. I hope this feature will be considered for future training versions.\r\n",
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"This issue was closed because it has been inactive for 7 days since being marked as stale."
] | 2024-01-02T12:56:11 | 2024-03-04T01:46:50 | 2024-03-04T01:46:50 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | I am working on a project that involves creating dialogues that sound spontaneous and natural. A key feature I would like to implement is the use of disfluencies (such as “Mm-hmm”) during moments when calculations are being made or there are pauses in the dialogue.
So far, I have experimented with using phonemes to create these disfluencies, but the results have not been satisfactory. It's possible that I haven't found the correct IPA pronunciation for these expressions. (If you have any IPA pronunciation suggestions to try, they would be greatly appreciated).
My questions are as follows:
- Is the integration of disfluencies in dialogue currently supported?
- If possible, could you recommend a strategy or approach to implement these disfluencies more naturally?
| {
"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/8109/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/8109/timeline | null | not_planned | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/8108 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8108/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8108/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8108/events | https://github.com/NVIDIA/NeMo/pull/8108 | 2,061,884,446 | PR_kwDOC_bI7s5jBlp1 | 8,108 | Pre-generate cu_seqlens argmin and max_seqlen to remove host-to-device sync | {
"login": "erhoo82",
"id": 15861259,
"node_id": "MDQ6VXNlcjE1ODYxMjU5",
"avatar_url": "https://avatars.githubusercontent.com/u/15861259?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/erhoo82",
"html_url": "https://github.com/erhoo82",
"followers_url": "https://api.github.com/users/erhoo82/followers",
"following_url": "https://api.github.com/users/erhoo82/following{/other_user}",
"gists_url": "https://api.github.com/users/erhoo82/gists{/gist_id}",
"starred_url": "https://api.github.com/users/erhoo82/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/erhoo82/subscriptions",
"organizations_url": "https://api.github.com/users/erhoo82/orgs",
"repos_url": "https://api.github.com/users/erhoo82/repos",
"events_url": "https://api.github.com/users/erhoo82/events{/privacy}",
"received_events_url": "https://api.github.com/users/erhoo82/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"jenkins",
"jenkins\r\n",
"jenkins",
"@ericharper\r\nPlease review this PR.",
"Just waiting on internal CI tests. They look good so far. We'll most likely merge first thing tomorrow.",
"jenkins",
"jenkins\r\n",
"jenkins",
"jenkins",
"jenkins",
"jenkins"
] | 2024-01-02T02:33:26 | 2024-01-18T23:00:47 | 2024-01-18T23:00:47 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8108",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8108",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8108.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8108.patch",
"merged_at": "2024-01-18T23:00:47"
} | …e data copies
# What does this PR do ?
At SFT, pre-generate `cu_seqlens_argmin` and `max_seqlen` to skip CPU-side computation on the main process.
This removes host-to-device sync, which is performance-critical when each kernel execution is short.
# 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
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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/8108/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/8108/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8107 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8107/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8107/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8107/events | https://github.com/NVIDIA/NeMo/pull/8107 | 2,061,757,205 | PR_kwDOC_bI7s5jBLcT | 8,107 | [pre-commit.ci] pre-commit suggestions | {
"login": "pre-commit-ci[bot]",
"id": 66853113,
"node_id": "MDM6Qm90NjY4NTMxMTM=",
"avatar_url": "https://avatars.githubusercontent.com/in/68672?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/pre-commit-ci%5Bbot%5D",
"html_url": "https://github.com/apps/pre-commit-ci",
"followers_url": "https://api.github.com/users/pre-commit-ci%5Bbot%5D/followers",
"following_url": "https://api.github.com/users/pre-commit-ci%5Bbot%5D/following{/other_user}",
"gists_url": "https://api.github.com/users/pre-commit-ci%5Bbot%5D/gists{/gist_id}",
"starred_url": "https://api.github.com/users/pre-commit-ci%5Bbot%5D/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/pre-commit-ci%5Bbot%5D/subscriptions",
"organizations_url": "https://api.github.com/users/pre-commit-ci%5Bbot%5D/orgs",
"repos_url": "https://api.github.com/users/pre-commit-ci%5Bbot%5D/repos",
"events_url": "https://api.github.com/users/pre-commit-ci%5Bbot%5D/events{/privacy}",
"received_events_url": "https://api.github.com/users/pre-commit-ci%5Bbot%5D/received_events",
"type": "Bot",
"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 | [
"This PR is stale because it has been open for 14 days with no activity. Remove stale label or comment or update or this will be closed in 7 days.",
"This PR was closed because it has been inactive for 7 days since being marked as stale."
] | 2024-01-01T20:25:52 | 2024-01-23T01:47:21 | 2024-01-23T01:47:21 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8107",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8107",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8107.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8107.patch",
"merged_at": null
} | <!--pre-commit.ci start-->
updates:
- [github.com/pre-commit/pre-commit-hooks: v4.3.0 → v4.5.0](https://github.com/pre-commit/pre-commit-hooks/compare/v4.3.0...v4.5.0)
- [github.com/PyCQA/isort: 5.12.0 → 5.13.2](https://github.com/PyCQA/isort/compare/5.12.0...5.13.2)
- [github.com/psf/black: 19.10b0 → 23.12.1](https://github.com/psf/black/compare/19.10b0...23.12.1)
<!--pre-commit.ci end--> | {
"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/8107/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/8107/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8105 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8105/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8105/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8105/events | https://github.com/NVIDIA/NeMo/issues/8105 | 2,060,870,667 | I_kwDOC_bI7s561mQL | 8,105 | ImportError: cannot import name 'compute_metrics_per_sample' from 'nemo.collections.asr.parts.utils.transcribe_utils' (/DATA/anaconda3/envs/nemo/lib/python3.10/site-packages/nemo/collections/asr/parts/utils/transcribe_utils.py) | {
"login": "mukherjeesougata",
"id": 95704630,
"node_id": "U_kgDOBbRWNg",
"avatar_url": "https://avatars.githubusercontent.com/u/95704630?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/mukherjeesougata",
"html_url": "https://github.com/mukherjeesougata",
"followers_url": "https://api.github.com/users/mukherjeesougata/followers",
"following_url": "https://api.github.com/users/mukherjeesougata/following{/other_user}",
"gists_url": "https://api.github.com/users/mukherjeesougata/gists{/gist_id}",
"starred_url": "https://api.github.com/users/mukherjeesougata/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/mukherjeesougata/subscriptions",
"organizations_url": "https://api.github.com/users/mukherjeesougata/orgs",
"repos_url": "https://api.github.com/users/mukherjeesougata/repos",
"events_url": "https://api.github.com/users/mukherjeesougata/events{/privacy}",
"received_events_url": "https://api.github.com/users/mukherjeesougata/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 | [
"You need a more recent NeMo version, 1.21 of 1.22. ",
"Or you can switch branch to 1.19 and then use the script from that branch ",
"> You need a more recent NeMo version, 1.21 of 1.22.\r\n\r\nWhat is the command line for checking nemo version?",
"```\r\nimport nemo\r\nprint(nemo.__version__)\r\n```",
"> You need a more recent NeMo version, 1.21 of 1.22.\r\n\r\nYes I have checked the nemo version is 1.21.0 but still it is giving the [same](https://github.com/NVIDIA/NeMo/issues/8105#issue-2060870667) error.",
"I meant use the version of the script from branch r1.21.0 on GitHub - https://github.com/NVIDIA/NeMo/blob/r1.21.0/examples/asr/transcribe_speech.py\r\n\r\nThis one corresponds to the NeMo installed version of 1.21.0. \r\n\r\nIf you use main branch script (upcoming release of 1.22.0), with installed NeMo 1.21 - it will generally not work. ",
"Since you specifically looking to calculate scores, you'll need the r1.21.0 version of the eval script as well - https://github.com/NVIDIA/NeMo/blob/r1.21.0/examples/asr/speech_to_text_eval.py",
"> I meant use the version of the script from branch r1.21.0 on GitHub - https://github.com/NVIDIA/NeMo/blob/r1.21.0/examples/asr/transcribe_speech.py\r\n> \r\n> This one corresponds to the NeMo installed version of 1.21.0.\r\n> \r\n> If you use main branch script (upcoming release of 1.22.0), with installed NeMo 1.21 - it will generally not work.\r\n\r\nI have directly downloaded [this](https://github.com/NVIDIA/NeMo/blob/r1.21.0/examples/asr/speech_to_text_eval.py) script from the r1.21.0 branch and replaced the original one but it is again showing the following error:-\r\n```\r\nTraceback (most recent call last):\r\n File \"/home/conformer/NeMo/examples/asr/speech_to_text_eval.py\", line 185, in <module>\r\n main() # noqa pylint: disable=no-value-for-parameter\r\n File \"/opt/conda/lib/python3.8/site-packages/nemo/core/config/hydra_runner.py\", line 126, in wrapper\r\n _run_hydra(\r\n File \"/opt/conda/lib/python3.8/site-packages/hydra/_internal/utils.py\", line 394, in _run_hydra\r\n _run_app(\r\n File \"/opt/conda/lib/python3.8/site-packages/hydra/_internal/utils.py\", line 457, in _run_app\r\n run_and_report(\r\n File \"/opt/conda/lib/python3.8/site-packages/hydra/_internal/utils.py\", line 223, in run_and_report\r\n raise ex\r\n File \"/opt/conda/lib/python3.8/site-packages/hydra/_internal/utils.py\", line 220, in run_and_report\r\n return func()\r\n File \"/opt/conda/lib/python3.8/site-packages/hydra/_internal/utils.py\", line 458, in <lambda>\r\n lambda: hydra.run(\r\n File \"/opt/conda/lib/python3.8/site-packages/hydra/_internal/hydra.py\", line 105, in run\r\n cfg = self.compose_config(\r\n File \"/opt/conda/lib/python3.8/site-packages/hydra/_internal/hydra.py\", line 594, in compose_config\r\n cfg = self.config_loader.load_configuration(\r\n File \"/opt/conda/lib/python3.8/site-packages/hydra/_internal/config_loader_impl.py\", line 142, in load_configuration\r\n return self._load_configuration_impl(\r\n File \"/opt/conda/lib/python3.8/site-packages/hydra/_internal/config_loader_impl.py\", line 244, in _load_configuration_impl\r\n parsed_overrides, caching_repo = self._parse_overrides_and_create_caching_repo(\r\n File \"/opt/conda/lib/python3.8/site-packages/hydra/_internal/config_loader_impl.py\", line 228, in _parse_overrides_and_create_caching_repo\r\n parsed_overrides = parser.parse_overrides(overrides=overrides)\r\n File \"/opt/conda/lib/python3.8/site-packages/hydra/core/override_parser/overrides_parser.py\", line 96, in parse_overrides\r\n raise OverrideParseException(\r\nhydra.errors.OverrideParseException: LexerNoViableAltException: \r\n ^\r\nSee https://hydra.cc/docs/1.2/advanced/override_grammar/basic for details\r\n```\r\nIs this happening due to the python version? Here python3.8 is being used.",
"I have also tried with python 3.10 version after downloading [this](https://github.com/NVIDIA/NeMo/blob/r1.21.0/examples/asr/speech_to_text_eval.py) script from the r1.21.0 branch and replaced the previous one but it is again showing the following error:-\r\n```\r\nTraceback (most recent call last):\r\n File \"/DATA/Sougata/BTech/NeMo/examples/asr/speech_to_text_eval.py\", line 75, in <module>\r\n from nemo.collections.asr.parts.utils.transcribe_utils import (\r\nImportError: cannot import name 'compute_metrics_per_sample' from 'nemo.collections.asr.parts.utils.transcribe_utils' (/DATA/anaconda3/envs/nemo/lib/python3.10/site-packages/nemo/collections/asr/parts/utils/transcribe_utils.py)\r\n```\r\n",
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"This issue was closed because it has been inactive for 7 days since being marked as stale."
] | 2023-12-30T22:15:47 | 2024-02-15T01:44:08 | 2024-02-15T01:44:08 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | I am using [this](https://github.com/NVIDIA/NeMo/blob/main/examples/asr/speech_to_text_eval.py) script to decode a conformer-CTC finetuned using the following bash script:-
```
CUDA_VISIBLE_DEVICES=0 python /DATA/Sougata/BTech/NeMo/examples/asr/speech_to_text_eval.py \
model_path=${model_path} \
dataset_manifest=${dataset_manifest} \
output_filename=${output_filename} \
batch_size=32 \
amp=True \
use_cer=True
```
I am getting the following error:-
```
Traceback (most recent call last):
File "/home/conformer/NeMo/examples/asr/speech_to_text_eval.py", line 75, in <module>
from nemo.collections.asr.parts.utils.transcribe_utils import (
ImportError: cannot import name 'compute_metrics_per_sample' from 'nemo.collections.asr.parts.utils.transcribe_utils' (/opt/conda/lib/python3.8/site-packages/nemo/collections/asr/parts/utils/transcribe_utils.py)
```
After following a similar [issue](https://github.com/NVIDIA/NeMo/issues/7104#issuecomment-1650726632) I have installed ``` nemo_toolkit[asr]==1.19.1``` but still it is showing the same error.
Can anybody kindly please help me in resolving this 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/8105/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/8105/timeline | null | not_planned | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/8104 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8104/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8104/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8104/events | https://github.com/NVIDIA/NeMo/issues/8104 | 2,060,699,951 | I_kwDOC_bI7s5608kv | 8,104 | Add a progress bar when loading model from pretrained | {
"login": "EurFelux",
"id": 59059173,
"node_id": "MDQ6VXNlcjU5MDU5MTcz",
"avatar_url": "https://avatars.githubusercontent.com/u/59059173?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/EurFelux",
"html_url": "https://github.com/EurFelux",
"followers_url": "https://api.github.com/users/EurFelux/followers",
"following_url": "https://api.github.com/users/EurFelux/following{/other_user}",
"gists_url": "https://api.github.com/users/EurFelux/gists{/gist_id}",
"starred_url": "https://api.github.com/users/EurFelux/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/EurFelux/subscriptions",
"organizations_url": "https://api.github.com/users/EurFelux/orgs",
"repos_url": "https://api.github.com/users/EurFelux/repos",
"events_url": "https://api.github.com/users/EurFelux/events{/privacy}",
"received_events_url": "https://api.github.com/users/EurFelux/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": 4585431311,
"node_id": "LA_kwDOC_bI7s8AAAABEVAhDw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale",
"name": "stale",
"color": "ededed",
"default": false,
"description": null
}
] | closed | false | {
"login": "okuchaiev",
"id": 22035961,
"node_id": "MDQ6VXNlcjIyMDM1OTYx",
"avatar_url": "https://avatars.githubusercontent.com/u/22035961?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/okuchaiev",
"html_url": "https://github.com/okuchaiev",
"followers_url": "https://api.github.com/users/okuchaiev/followers",
"following_url": "https://api.github.com/users/okuchaiev/following{/other_user}",
"gists_url": "https://api.github.com/users/okuchaiev/gists{/gist_id}",
"starred_url": "https://api.github.com/users/okuchaiev/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/okuchaiev/subscriptions",
"organizations_url": "https://api.github.com/users/okuchaiev/orgs",
"repos_url": "https://api.github.com/users/okuchaiev/repos",
"events_url": "https://api.github.com/users/okuchaiev/events{/privacy}",
"received_events_url": "https://api.github.com/users/okuchaiev/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"login": "okuchaiev",
"id": 22035961,
"node_id": "MDQ6VXNlcjIyMDM1OTYx",
"avatar_url": "https://avatars.githubusercontent.com/u/22035961?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/okuchaiev",
"html_url": "https://github.com/okuchaiev",
"followers_url": "https://api.github.com/users/okuchaiev/followers",
"following_url": "https://api.github.com/users/okuchaiev/following{/other_user}",
"gists_url": "https://api.github.com/users/okuchaiev/gists{/gist_id}",
"starred_url": "https://api.github.com/users/okuchaiev/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/okuchaiev/subscriptions",
"organizations_url": "https://api.github.com/users/okuchaiev/orgs",
"repos_url": "https://api.github.com/users/okuchaiev/repos",
"events_url": "https://api.github.com/users/okuchaiev/events{/privacy}",
"received_events_url": "https://api.github.com/users/okuchaiev/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
}
] | null | [
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"This issue was closed because it has been inactive for 7 days since being marked as stale."
] | 2023-12-30T10:01:24 | 2024-02-08T01:43:41 | 2024-02-08T01:43:41 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | I'm attempting to download a pre-trained model from the NGC source. Unfortunately, my network connection is not performing optimally. The absence of a download progress bar makes it difficult for me to estimate the remaining time for completion, resulting in frustration and uncertainty during the download.
I suggest incorporating a progress bar into the pre-training model download process. This feature should display both the download progress and speed, providing users with real-time feedback and enhancing the overall downloading experience. | {
"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/8104/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/8104/timeline | null | not_planned | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/8103 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8103/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8103/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8103/events | https://github.com/NVIDIA/NeMo/pull/8103 | 2,060,476,355 | PR_kwDOC_bI7s5i9PFZ | 8,103 | Speechllm tts paarth - Alignment CTC loss, custom number of codebooks (dac) and other changes | {
"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
} | [
{
"id": 4418253970,
"node_id": "LA_kwDOC_bI7s8AAAABB1k0kg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/TTS",
"name": "TTS",
"color": "49AF28",
"default": false,
"description": ""
},
{
"id": 4585431311,
"node_id": "LA_kwDOC_bI7s8AAAABEVAhDw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale",
"name": "stale",
"color": "ededed",
"default": false,
"description": null
},
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"This PR is stale because it has been open for 14 days with no activity. Remove stale label or comment or update or this will be closed in 7 days.",
"This PR was closed because it has been inactive for 7 days since being marked as stale."
] | 2023-12-30T00:27:55 | 2024-01-21T01:49:39 | 2024-01-21T01:49:38 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8103",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8103",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8103.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8103.patch",
"merged_at": null
} | # What does this PR do ?
Add a one line overview of what this PR aims to accomplish.
**Collection**: [Note which collection this PR will affect]
# Changelog
- Add specific line by line info of high level changes in this PR.
# Usage
* You can potentially add a usage example below
```python
# Add a code snippet demonstrating how to use this
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# Before your PR is "Ready for review"
**Pre checks**:
- [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md)
- [ ] Did you write any new necessary tests?
- [ ] Did you add or update any necessary documentation?
- [ ] Does the PR affect components that are optional to install? (Ex: Numba, Pynini, Apex etc)
- [ ] Reviewer: Does the PR have correct import guards for all optional libraries?
**PR Type**:
- [ ] New Feature
- [ ] Bugfix
- [ ] Documentation
If you haven't finished some of the above items you can still open "Draft" PR.
## Who can review?
Anyone in the community is free to review the PR once the checks have passed.
[Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas.
# Additional Information
* Related to # (issue)
| {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers",
"following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}",
"gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}",
"starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions",
"organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs",
"repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos",
"events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}",
"received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events",
"type": "Bot",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8103/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/8103/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8102 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8102/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8102/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8102/events | https://github.com/NVIDIA/NeMo/pull/8102 | 2,060,402,510 | PR_kwDOC_bI7s5i9APl | 8,102 | Support transcoding audio formats when saving tarred datasets (FLAC, OPUS) | {
"login": "pzelasko",
"id": 15930688,
"node_id": "MDQ6VXNlcjE1OTMwNjg4",
"avatar_url": "https://avatars.githubusercontent.com/u/15930688?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/pzelasko",
"html_url": "https://github.com/pzelasko",
"followers_url": "https://api.github.com/users/pzelasko/followers",
"following_url": "https://api.github.com/users/pzelasko/following{/other_user}",
"gists_url": "https://api.github.com/users/pzelasko/gists{/gist_id}",
"starred_url": "https://api.github.com/users/pzelasko/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/pzelasko/subscriptions",
"organizations_url": "https://api.github.com/users/pzelasko/orgs",
"repos_url": "https://api.github.com/users/pzelasko/repos",
"events_url": "https://api.github.com/users/pzelasko/events{/privacy}",
"received_events_url": "https://api.github.com/users/pzelasko/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"jenkins",
"jenkins"
] | 2023-12-29T21:09:13 | 2024-01-03T18:36:32 | 2024-01-03T18:36:32 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8102",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8102",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8102.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8102.patch",
"merged_at": "2024-01-03T18:36:32"
} | # What does this PR do ?
Supports saving audio into compressed formats when creating a tarred dataset.
**Collection**: Any speech collection
# Changelog
- Add option `--force_codec [name]` to `scripts/speech_recognition/convert_to_tarred_audio_dataset.py`
- Acceptable values: `wav`, `flac`, `opus`, possibly others supported by libsndfile
- When `--force_codec` is provided, and the audio file extension is different than codec's name, we'll read the audio and transcode it in memory, then save to tar; otherwise the existing code path is selected
- Change the default to save files as `opus`. Soundfile uses 28kbps for OPUS which is adequate for podcasts/audiobooks according to https://wiki.xiph.org/Opus_Recommended_Settings -- the result is ~17x smaller than FLAC and 26x smaller than WAVE (current default).
# Usage
```bash
python scripts/speech_recognition/convert_to_tarred_audio_dataset.py --manifest_path nemo_manifest.json --num_shards 4 --force_codec opus --workers 4 --max_duration 30
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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": "pzelasko",
"id": 15930688,
"node_id": "MDQ6VXNlcjE1OTMwNjg4",
"avatar_url": "https://avatars.githubusercontent.com/u/15930688?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/pzelasko",
"html_url": "https://github.com/pzelasko",
"followers_url": "https://api.github.com/users/pzelasko/followers",
"following_url": "https://api.github.com/users/pzelasko/following{/other_user}",
"gists_url": "https://api.github.com/users/pzelasko/gists{/gist_id}",
"starred_url": "https://api.github.com/users/pzelasko/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/pzelasko/subscriptions",
"organizations_url": "https://api.github.com/users/pzelasko/orgs",
"repos_url": "https://api.github.com/users/pzelasko/repos",
"events_url": "https://api.github.com/users/pzelasko/events{/privacy}",
"received_events_url": "https://api.github.com/users/pzelasko/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8102/reactions",
"total_count": 1,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 1,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/NVIDIA/NeMo/issues/8102/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8101 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8101/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8101/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8101/events | https://github.com/NVIDIA/NeMo/pull/8101 | 2,060,389,412 | PR_kwDOC_bI7s5i89iB | 8,101 | Fix for #8100: Audio with filename containing spaces generate invalid rttm files | {
"login": "gaspardpetit",
"id": 9883156,
"node_id": "MDQ6VXNlcjk4ODMxNTY=",
"avatar_url": "https://avatars.githubusercontent.com/u/9883156?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/gaspardpetit",
"html_url": "https://github.com/gaspardpetit",
"followers_url": "https://api.github.com/users/gaspardpetit/followers",
"following_url": "https://api.github.com/users/gaspardpetit/following{/other_user}",
"gists_url": "https://api.github.com/users/gaspardpetit/gists{/gist_id}",
"starred_url": "https://api.github.com/users/gaspardpetit/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/gaspardpetit/subscriptions",
"organizations_url": "https://api.github.com/users/gaspardpetit/orgs",
"repos_url": "https://api.github.com/users/gaspardpetit/repos",
"events_url": "https://api.github.com/users/gaspardpetit/events{/privacy}",
"received_events_url": "https://api.github.com/users/gaspardpetit/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": 4811884691,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR",
"name": "ASR",
"color": "F3B33E",
"default": false,
"description": ""
},
{
"id": 4847373924,
"node_id": "LA_kwDOC_bI7s8AAAABIO0OZA",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/common",
"name": "common",
"color": "c5def5",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"This PR is stale because it has been open for 14 days with no activity. Remove stale label or comment or update or this will be closed in 7 days.",
"This PR is stale because it has been open for 14 days with no activity. Remove stale label or comment or update or this will be closed in 7 days.",
"This PR was closed because it has been inactive for 7 days since being marked as stale."
] | 2023-12-29T20:40:22 | 2024-02-13T01:43:58 | 2024-02-13T01:43:57 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8101",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8101",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8101.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8101.patch",
"merged_at": null
} | # What does this PR do ?
Resolves issue #8100. Ensure that the unique_name saved and expected to be found in rttm is uri-compliant and does not contain spaces even if the original audio file contained spaces.
**Collection**: [Note which collection this PR will affect]
# Changelog
- RTTM filenames are now uri encoded
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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.
# Additional Information
* Related to # 8100
| {
"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/8101/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/8101/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8100 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8100/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8100/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8100/events | https://github.com/NVIDIA/NeMo/issues/8100 | 2,060,384,983 | I_kwDOC_bI7s56zvrX | 8,100 | Audio with filename containing spaces generate invalid rttm files | {
"login": "gaspardpetit",
"id": 9883156,
"node_id": "MDQ6VXNlcjk4ODMxNTY=",
"avatar_url": "https://avatars.githubusercontent.com/u/9883156?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/gaspardpetit",
"html_url": "https://github.com/gaspardpetit",
"followers_url": "https://api.github.com/users/gaspardpetit/followers",
"following_url": "https://api.github.com/users/gaspardpetit/following{/other_user}",
"gists_url": "https://api.github.com/users/gaspardpetit/gists{/gist_id}",
"starred_url": "https://api.github.com/users/gaspardpetit/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/gaspardpetit/subscriptions",
"organizations_url": "https://api.github.com/users/gaspardpetit/orgs",
"repos_url": "https://api.github.com/users/gaspardpetit/repos",
"events_url": "https://api.github.com/users/gaspardpetit/events{/privacy}",
"received_events_url": "https://api.github.com/users/gaspardpetit/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 | [
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"This issue was closed because it has been inactive for 7 days since being marked as stale."
] | 2023-12-29T20:30:52 | 2024-02-05T01:44:56 | 2024-02-05T01:44:55 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | **Describe the bug**
NeMo uses the file name without extension as the unique id for rttm file generation. RTTM parsing code typically expect a name that is uri-compliant, without spaces.
The typical code used in several places is:
```
os.path.splitext(os.path.basename(sample.audio_file))[0]
```
When the file name contains space, this generates a unique_id with a space. Since the RTTM file format relies on space-separated fields, this later causes parsing error.
**Steps/Code to reproduce bug**
For example, when using the following manifest file:
```
{
"audio_filepath": "mono vocals.wav",
"label": "infer",
"text": "-",
"num_speakers": 5,
}
```
The following code:
```
neural_diarizer = NeuralDiarizer(cfg=self.config)
neural_diarizer.diarize()
```
Results in the following exception:
```
[NeMo I 2023-12-29 15:20:38 msdd_models:938] Loading cluster label file from /data/nemo/outputs/speaker_outputs/subsegments_scale4_cluster.label
Traceback (most recent call last):
File "/data/nemo/services/nemo/diarization.py", line 224, in <module>
main()
File "/data/nemo/services/nemo/diarization.py", line 155, in _diarize_neural
neural_diarizer .diarize()
File "/data/nemo/.venv/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "/data/nemo/.venv/lib/python3.10/site-packages/nemo/collections/asr/models/msdd_models.py", line 1180, in diarize
self.clustering_embedding.prepare_cluster_embs_infer()
File "/data/nemo/.venv/lib/python3.10/site-packages/nemo/collections/asr/models/msdd_models.py", line 699, in prepare_cluster_embs_infer
self.emb_sess_test_dict, self.emb_seq_test, self.clus_test_label_dict, _ = self.run_clustering_diarizer(
File "/data/nemo/.venv/lib/python3.10/site-packages/nemo/collections/asr/models/msdd_models.py", line 881, in run_clustering_diarizer
emb_sess_avg_dict, base_clus_label_dict = self.get_cluster_avg_embs(
File "/data/nemo/.venv/lib/python3.10/site-packages/nemo/collections/asr/models/msdd_models.py", line 797, in get_cluster_avg_embs
output_clus_label_dict, emb_dim = self.get_base_clus_label_dict(clus_labels, emb_scale_seq_dict)
File "/data/nemo/.venv/lib/python3.10/site-packages/nemo/collections/asr/models/msdd_models.py", line 759, in get_base_clus_label_dict
stt, end = [round(float(x), 2) for x in line.split()[1:3]]
File "/data/nemo/.venv/lib/python3.10/site-packages/nemo/collections/asr/models/msdd_models.py", line 759, in <listcomp>
stt, end = [round(float(x), 2) for x in line.split()[1:3]]
ValueError: could not convert string to float: 'vocals'
```
**Expected behavior**
I cannot find consensus on the RTTM format, however, pyannote refers to the name a uri, and the code within NeMo expects space separated fields. I would expect the unique name generated from the filename to be uri-compliant.
**Environment overview (please complete the following information)**
- Python 3.10
- Ubuntu 22.04
| {
"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/8100/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/8100/timeline | null | not_planned | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/8099 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8099/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8099/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8099/events | https://github.com/NVIDIA/NeMo/pull/8099 | 2,060,350,392 | PR_kwDOC_bI7s5i81f7 | 8,099 | Reconfigure limit_val_batches only for int | {
"login": "athitten",
"id": 47577437,
"node_id": "MDQ6VXNlcjQ3NTc3NDM3",
"avatar_url": "https://avatars.githubusercontent.com/u/47577437?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/athitten",
"html_url": "https://github.com/athitten",
"followers_url": "https://api.github.com/users/athitten/followers",
"following_url": "https://api.github.com/users/athitten/following{/other_user}",
"gists_url": "https://api.github.com/users/athitten/gists{/gist_id}",
"starred_url": "https://api.github.com/users/athitten/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/athitten/subscriptions",
"organizations_url": "https://api.github.com/users/athitten/orgs",
"repos_url": "https://api.github.com/users/athitten/repos",
"events_url": "https://api.github.com/users/athitten/events{/privacy}",
"received_events_url": "https://api.github.com/users/athitten/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"jenkins",
"jenkins",
"jenkins"
] | 2023-12-29T19:24:49 | 2024-01-02T22:48:17 | 2024-01-02T22:48:16 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8099",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8099",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8099.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8099.patch",
"merged_at": "2024-01-02T22:48:16"
} | # 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
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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": "athitten",
"id": 47577437,
"node_id": "MDQ6VXNlcjQ3NTc3NDM3",
"avatar_url": "https://avatars.githubusercontent.com/u/47577437?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/athitten",
"html_url": "https://github.com/athitten",
"followers_url": "https://api.github.com/users/athitten/followers",
"following_url": "https://api.github.com/users/athitten/following{/other_user}",
"gists_url": "https://api.github.com/users/athitten/gists{/gist_id}",
"starred_url": "https://api.github.com/users/athitten/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/athitten/subscriptions",
"organizations_url": "https://api.github.com/users/athitten/orgs",
"repos_url": "https://api.github.com/users/athitten/repos",
"events_url": "https://api.github.com/users/athitten/events{/privacy}",
"received_events_url": "https://api.github.com/users/athitten/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8099/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/8099/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8098 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8098/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8098/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8098/events | https://github.com/NVIDIA/NeMo/issues/8098 | 2,060,070,212 | I_kwDOC_bI7s56yi1E | 8,098 | Read files BytesIO format | {
"login": "Andrey-Nikitin",
"id": 15768855,
"node_id": "MDQ6VXNlcjE1NzY4ODU1",
"avatar_url": "https://avatars.githubusercontent.com/u/15768855?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Andrey-Nikitin",
"html_url": "https://github.com/Andrey-Nikitin",
"followers_url": "https://api.github.com/users/Andrey-Nikitin/followers",
"following_url": "https://api.github.com/users/Andrey-Nikitin/following{/other_user}",
"gists_url": "https://api.github.com/users/Andrey-Nikitin/gists{/gist_id}",
"starred_url": "https://api.github.com/users/Andrey-Nikitin/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/Andrey-Nikitin/subscriptions",
"organizations_url": "https://api.github.com/users/Andrey-Nikitin/orgs",
"repos_url": "https://api.github.com/users/Andrey-Nikitin/repos",
"events_url": "https://api.github.com/users/Andrey-Nikitin/events{/privacy}",
"received_events_url": "https://api.github.com/users/Andrey-Nikitin/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4585431311,
"node_id": "LA_kwDOC_bI7s8AAAABEVAhDw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale",
"name": "stale",
"color": "ededed",
"default": false,
"description": null
}
] | closed | false | {
"login": "okuchaiev",
"id": 22035961,
"node_id": "MDQ6VXNlcjIyMDM1OTYx",
"avatar_url": "https://avatars.githubusercontent.com/u/22035961?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/okuchaiev",
"html_url": "https://github.com/okuchaiev",
"followers_url": "https://api.github.com/users/okuchaiev/followers",
"following_url": "https://api.github.com/users/okuchaiev/following{/other_user}",
"gists_url": "https://api.github.com/users/okuchaiev/gists{/gist_id}",
"starred_url": "https://api.github.com/users/okuchaiev/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/okuchaiev/subscriptions",
"organizations_url": "https://api.github.com/users/okuchaiev/orgs",
"repos_url": "https://api.github.com/users/okuchaiev/repos",
"events_url": "https://api.github.com/users/okuchaiev/events{/privacy}",
"received_events_url": "https://api.github.com/users/okuchaiev/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"login": "okuchaiev",
"id": 22035961,
"node_id": "MDQ6VXNlcjIyMDM1OTYx",
"avatar_url": "https://avatars.githubusercontent.com/u/22035961?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/okuchaiev",
"html_url": "https://github.com/okuchaiev",
"followers_url": "https://api.github.com/users/okuchaiev/followers",
"following_url": "https://api.github.com/users/okuchaiev/following{/other_user}",
"gists_url": "https://api.github.com/users/okuchaiev/gists{/gist_id}",
"starred_url": "https://api.github.com/users/okuchaiev/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/okuchaiev/subscriptions",
"organizations_url": "https://api.github.com/users/okuchaiev/orgs",
"repos_url": "https://api.github.com/users/okuchaiev/repos",
"events_url": "https://api.github.com/users/okuchaiev/events{/privacy}",
"received_events_url": "https://api.github.com/users/okuchaiev/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
}
] | null | [
"We currently do not support any input other than file path. However, subclassing and overriding the transcribe function should be fine, usually does not cause wer degradation. ",
"> We currently do not support any input other than file path. However, subclassing and overriding the transcribe function should be fine, usually does not cause wer degradation.\r\n\r\n\r\nHow Nemo reads sound files by default? DALI, librosa?\r\n",
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"This issue was closed because it has been inactive for 7 days since being marked as stale."
] | 2023-12-29T14:07:00 | 2024-02-16T01:44:14 | 2024-02-16T01:44:13 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | **Is your feature request related to a problem? Please describe.**
I have audio files in BytesIO, how can I feed them into transcribe if this function expects file_path?
**Describe the solution you'd like**
The transcribe method expects not only file_path but also BytesIO
**Describe alternatives you've considered**
I made a custom EncDecCTCModelBPE and read librosa files, but the accuracy of the transcription is reduced in contrast to the basic EncDecCTCModelBPE.
**Additional context**
-
| {
"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/8098/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/8098/timeline | null | not_planned | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/8096 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8096/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8096/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8096/events | https://github.com/NVIDIA/NeMo/issues/8096 | 2,059,305,565 | I_kwDOC_bI7s56voJd | 8,096 | [ASR] Add Paraformer model | {
"login": "ArtyomZemlyak",
"id": 42967602,
"node_id": "MDQ6VXNlcjQyOTY3NjAy",
"avatar_url": "https://avatars.githubusercontent.com/u/42967602?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ArtyomZemlyak",
"html_url": "https://github.com/ArtyomZemlyak",
"followers_url": "https://api.github.com/users/ArtyomZemlyak/followers",
"following_url": "https://api.github.com/users/ArtyomZemlyak/following{/other_user}",
"gists_url": "https://api.github.com/users/ArtyomZemlyak/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ArtyomZemlyak/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ArtyomZemlyak/subscriptions",
"organizations_url": "https://api.github.com/users/ArtyomZemlyak/orgs",
"repos_url": "https://api.github.com/users/ArtyomZemlyak/repos",
"events_url": "https://api.github.com/users/ArtyomZemlyak/events{/privacy}",
"received_events_url": "https://api.github.com/users/ArtyomZemlyak/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": 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": "titu1994",
"id": 3048602,
"node_id": "MDQ6VXNlcjMwNDg2MDI=",
"avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/titu1994",
"html_url": "https://github.com/titu1994",
"followers_url": "https://api.github.com/users/titu1994/followers",
"following_url": "https://api.github.com/users/titu1994/following{/other_user}",
"gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}",
"starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/titu1994/subscriptions",
"organizations_url": "https://api.github.com/users/titu1994/orgs",
"repos_url": "https://api.github.com/users/titu1994/repos",
"events_url": "https://api.github.com/users/titu1994/events{/privacy}",
"received_events_url": "https://api.github.com/users/titu1994/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"login": "titu1994",
"id": 3048602,
"node_id": "MDQ6VXNlcjMwNDg2MDI=",
"avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/titu1994",
"html_url": "https://github.com/titu1994",
"followers_url": "https://api.github.com/users/titu1994/followers",
"following_url": "https://api.github.com/users/titu1994/following{/other_user}",
"gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}",
"starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/titu1994/subscriptions",
"organizations_url": "https://api.github.com/users/titu1994/orgs",
"repos_url": "https://api.github.com/users/titu1994/repos",
"events_url": "https://api.github.com/users/titu1994/events{/privacy}",
"received_events_url": "https://api.github.com/users/titu1994/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
},
{
"login": "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 | [
"I've read this paper before, so some of my comments are below. \r\n\r\n1) MWER is a pain to train with and implement, plus there's no public efficient implementation as far as I can see. I'd rather stick to rnnt loss or even CTC. 2\r\n\r\n2) Continuous Integrate and Fire is a novel concept but has not gained much traction in the 4 years since it's paper in 2019. I've personally experimented a bit with it in some branch, and found that while it works fine, it's wer is inferior to RNNT. Maybe there are new variants to CIF that surpass RNNT\r\n\r\n3) NeMo support for AED models will come soon* (no release date for the time being). \r\n\r\n4) The papers core contribution seems to be RTF at compatible accuracy to AR models. This is very good but for RTF - Fast Conformer already gets very good RTF (0.0003x on long audio files ~ 3687 times real time factor). So encoder level optimizations already surpass this paper with CTC. With RNNT, that RTF is something like 0.009 which is on par with this paper. I'm somewhat confident that rnnt wer would be competitive as well with this model, while also supporting long audio inference and other tasks like speech translation. \r\n\r\nSo it seems para former would be a middle ground between CTC and RNNT in terms of accuracy and speed both. \r\n\r\nStill, I need to reread the paper to see if I'm missing something crucial. \r\n\r\nJust a note, there are just my personal comments, the team will need to discuss whether the model will be added to NeMo or not. \r\n\r\nIf you're up for it, we will gladly welcome contributions to add this model too !",
"@titu1994 Thank you very much for your detailed answer!\r\nYes, I also had doubts about how high-quality Paraformer could actually be.\r\n\r\nIn our team we already use `Fast Conformer (CTC or HAT)` (xlarge for Knowledge Distillation and medium for production) and its speed is quite good compared to the regular `Conformer`, and the quality practically does not drop.\r\n\r\nBut in search of some improvements, it would be interesting to try alternative architectures. And one of them is `Paraformer` (due to the fact that it is `NAR`) and probably even its `CIF`.\r\n\r\nThere are also several interesting articles on `CIF`, using which maybe can be achievied interesting or better results:\r\n\r\n 1. https://arxiv.org/pdf/1905.11235.pdf - standard article (CIF: CONTINUOUS INTEGRATE-AND-FIRE FOR END-TO-END SPEECH RECOGNITION) (https://github.com/MingLunHan/CIF-PyTorch)\r\n 2. https://github.com/MingLunHan/CIF-ColDec (add selective context to improve decoding result)\r\n 3. https://github.com/MingLunHan/CIF-HieraDist (transfer of knowledge from a PLM model to an ASR model, at the linguistic and acoustic level) (in this article Branchformer shows a good summary - you can take a look at it)\r\n 4. https://arxiv.org/pdf/2307.14132.pdf (CIF-T: A NOVEL CIF-BASED TRANSDUCER ARCHITECTURE FOR AUTOMATIC SPEECH RECOGNITION) (good article with interesting additions) (here, by the way, paraformer does not show very good results)\r\n\r\n> NeMo support for `AED` models will come soon* (no release date for the time being).\r\n- Its be very good! ",
"Very interesting papers, especially the CIF-T. Will read them after the long weekend. \r\n\r\nMy thought was pretty much the same, very hard to beat RNNT without sacrificing something (speed or quality). \r\n\r\nAEDs, despite being slow and having hallucination issues, enable lots of sophisticated capability which rnnt alone can probably handle but with poor results. So imo it makes sense to trade off some of the benefits of RNNT for those new capabilities. \r\n\r\nCIF, to me at least, just falls in the middle of the niche already dominated by CTC and RNNT, that is efficient and highly accurate ASR. AEDs fill another niche - advanced capabilities that aren't efficient with monotonic alignment learning losses. ",
"What about [AIF](https://arxiv.org/pdf/2311.11353.pdf) ? ",
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"This issue was closed because it has been inactive for 7 days since being marked as stale."
] | 2023-12-29T10:46:22 | 2024-02-10T01:43:03 | 2024-02-10T01:43:02 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | **Is your feature request related to a problem? Please describe.**
What needs to be done for adding `Paraformer` model to `NeMo`? Or maybe its already in work/plans?
**Describe the solution you'd like**
1. Add `CIF` implementation as module (good if it can be modular and usable for all/most ASR models)
2. `MAE loss` for `predictor` of `Paraformer` (can we use already existed `MaskedMAELoss` from `TTS`?)
3. `MWER loss` as main loss
4. `Cross-Entropy loss` as main loss
5. `AED` or `Transformer` decoder module (and its can be useful and for another models) (or can we use `nemo/collections/asr/modules/transformer/transformer_decoders.py` ?) (and can be added integration to any HuggingFace transformer decoder model?)
6. `Sampler` module
7. And maybe add another paraformer layers, if they implementation differ from conformer (need deep check)
What do you think?
**Describe alternatives you've considered**
wenet or funasr integration maybe
**Additional context**
[Paper arxiv PDF](https://arxiv.org/pdf/2206.08317.pdf)
[FunASR orig implementation](https://github.com/alibaba-damo-academy/FunASR/blob/main/funasr/models/e2e_asr_paraformer.py)
[wenet implementation](https://github.com/wenet-e2e/wenet/tree/main/wenet/paraformer)
| {
"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/8096/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/8096/timeline | null | not_planned | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/8095 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8095/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8095/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8095/events | https://github.com/NVIDIA/NeMo/pull/8095 | 2,058,831,741 | PR_kwDOC_bI7s5i5-PS | 8,095 | Canary: option to limit the number of open streams | {
"login": "pzelasko",
"id": 15930688,
"node_id": "MDQ6VXNlcjE1OTMwNjg4",
"avatar_url": "https://avatars.githubusercontent.com/u/15930688?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/pzelasko",
"html_url": "https://github.com/pzelasko",
"followers_url": "https://api.github.com/users/pzelasko/followers",
"following_url": "https://api.github.com/users/pzelasko/following{/other_user}",
"gists_url": "https://api.github.com/users/pzelasko/gists{/gist_id}",
"starred_url": "https://api.github.com/users/pzelasko/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/pzelasko/subscriptions",
"organizations_url": "https://api.github.com/users/pzelasko/orgs",
"repos_url": "https://api.github.com/users/pzelasko/repos",
"events_url": "https://api.github.com/users/pzelasko/events{/privacy}",
"received_events_url": "https://api.github.com/users/pzelasko/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4847373924,
"node_id": "LA_kwDOC_bI7s8AAAABIO0OZA",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/common",
"name": "common",
"color": "c5def5",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [] | 2023-12-28T21:25:16 | 2023-12-29T00:18:04 | 2023-12-29T00:18:04 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8095",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8095",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8095.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8095.patch",
"merged_at": "2023-12-29T00:18:04"
} | # 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
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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": "krishnacpuvvada",
"id": 93558329,
"node_id": "U_kgDOBZOWOQ",
"avatar_url": "https://avatars.githubusercontent.com/u/93558329?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/krishnacpuvvada",
"html_url": "https://github.com/krishnacpuvvada",
"followers_url": "https://api.github.com/users/krishnacpuvvada/followers",
"following_url": "https://api.github.com/users/krishnacpuvvada/following{/other_user}",
"gists_url": "https://api.github.com/users/krishnacpuvvada/gists{/gist_id}",
"starred_url": "https://api.github.com/users/krishnacpuvvada/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/krishnacpuvvada/subscriptions",
"organizations_url": "https://api.github.com/users/krishnacpuvvada/orgs",
"repos_url": "https://api.github.com/users/krishnacpuvvada/repos",
"events_url": "https://api.github.com/users/krishnacpuvvada/events{/privacy}",
"received_events_url": "https://api.github.com/users/krishnacpuvvada/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8095/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/8095/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8094 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8094/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8094/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8094/events | https://github.com/NVIDIA/NeMo/pull/8094 | 2,058,504,452 | PR_kwDOC_bI7s5i44hi | 8,094 | Change the megatron config lr scheduler default and fix to change partitions script | {
"login": "shan18",
"id": 12576178,
"node_id": "MDQ6VXNlcjEyNTc2MTc4",
"avatar_url": "https://avatars.githubusercontent.com/u/12576178?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/shan18",
"html_url": "https://github.com/shan18",
"followers_url": "https://api.github.com/users/shan18/followers",
"following_url": "https://api.github.com/users/shan18/following{/other_user}",
"gists_url": "https://api.github.com/users/shan18/gists{/gist_id}",
"starred_url": "https://api.github.com/users/shan18/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/shan18/subscriptions",
"organizations_url": "https://api.github.com/users/shan18/orgs",
"repos_url": "https://api.github.com/users/shan18/repos",
"events_url": "https://api.github.com/users/shan18/events{/privacy}",
"received_events_url": "https://api.github.com/users/shan18/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"jenkins"
] | 2023-12-28T14:28:47 | 2023-12-29T17:38:18 | 2023-12-29T17:38:18 | MEMBER | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8094",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8094",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8094.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8094.patch",
"merged_at": "2023-12-29T17:38:17"
} | Two fixes:
* Change the default: `optim.sched.constant_steps=0`
* Resolving the issue where the change partition script throws an error when `sequence_parallel=true` in config when converting the model to TP=1. | {
"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/8094/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/8094/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8093 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8093/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8093/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8093/events | https://github.com/NVIDIA/NeMo/issues/8093 | 2,058,087,104 | I_kwDOC_bI7s56q-rA | 8,093 | Privately reporting a Vulnerability to NVIDIA/NeMo | {
"login": "Sim4n6",
"id": 13036531,
"node_id": "MDQ6VXNlcjEzMDM2NTMx",
"avatar_url": "https://avatars.githubusercontent.com/u/13036531?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Sim4n6",
"html_url": "https://github.com/Sim4n6",
"followers_url": "https://api.github.com/users/Sim4n6/followers",
"following_url": "https://api.github.com/users/Sim4n6/following{/other_user}",
"gists_url": "https://api.github.com/users/Sim4n6/gists{/gist_id}",
"starred_url": "https://api.github.com/users/Sim4n6/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/Sim4n6/subscriptions",
"organizations_url": "https://api.github.com/users/Sim4n6/orgs",
"repos_url": "https://api.github.com/users/Sim4n6/repos",
"events_url": "https://api.github.com/users/Sim4n6/events{/privacy}",
"received_events_url": "https://api.github.com/users/Sim4n6/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"The report has been submitted in private : https://github.com/NVIDIA/NeMo/security/advisories/GHSA-x392-p65g-4rxx",
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"Well, it is not in a stale state. But rather I've been waiting for a CVE to be assigned and the report to be published."
] | 2023-12-28T07:21:00 | 2024-03-28T06:08:39 | 2024-03-28T06:08:39 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | Hey ,
I have identified a vulnerability in the latest codebase NVIDIA/NeMo.
Could you please enable the GitHub feature for [private vulnerability reporting](https://docs.github.com/en/code-security/security-advisories/working-with-repository-security-advisories/configuring-private-vulnerability-reporting-for-a-repository), or in case NVIDIA Corporation has a bug bounty program in HackerOne platform for instance, I would like to report that.
Thank you
@sim4n6 | {
"login": "Sim4n6",
"id": 13036531,
"node_id": "MDQ6VXNlcjEzMDM2NTMx",
"avatar_url": "https://avatars.githubusercontent.com/u/13036531?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Sim4n6",
"html_url": "https://github.com/Sim4n6",
"followers_url": "https://api.github.com/users/Sim4n6/followers",
"following_url": "https://api.github.com/users/Sim4n6/following{/other_user}",
"gists_url": "https://api.github.com/users/Sim4n6/gists{/gist_id}",
"starred_url": "https://api.github.com/users/Sim4n6/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/Sim4n6/subscriptions",
"organizations_url": "https://api.github.com/users/Sim4n6/orgs",
"repos_url": "https://api.github.com/users/Sim4n6/repos",
"events_url": "https://api.github.com/users/Sim4n6/events{/privacy}",
"received_events_url": "https://api.github.com/users/Sim4n6/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8093/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/8093/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/8092 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8092/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8092/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8092/events | https://github.com/NVIDIA/NeMo/issues/8092 | 2,057,892,650 | I_kwDOC_bI7s56qPMq | 8,092 | Llama SFT - Single Dataset File Not Working Even With ${CONCAT_SAMPLING_PROBS}="[1]" | {
"login": "taozhang9527",
"id": 141076017,
"node_id": "U_kgDOCGimMQ",
"avatar_url": "https://avatars.githubusercontent.com/u/141076017?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/taozhang9527",
"html_url": "https://github.com/taozhang9527",
"followers_url": "https://api.github.com/users/taozhang9527/followers",
"following_url": "https://api.github.com/users/taozhang9527/following{/other_user}",
"gists_url": "https://api.github.com/users/taozhang9527/gists{/gist_id}",
"starred_url": "https://api.github.com/users/taozhang9527/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/taozhang9527/subscriptions",
"organizations_url": "https://api.github.com/users/taozhang9527/orgs",
"repos_url": "https://api.github.com/users/taozhang9527/repos",
"events_url": "https://api.github.com/users/taozhang9527/events{/privacy}",
"received_events_url": "https://api.github.com/users/taozhang9527/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 | [
"Tried the latest docker image: `nemofw-training:23.11`. Still failed, but the error message is kind of different. \r\n\r\n```\r\n[NeMo W 2024-01-04 04:44:03 megatron_base_model:821] The model: MegatronGPTSFTModel() does not have field.name: context_parallel_size in its cfg. Add this key to cfg or config_mapping to ma\r\nke to make it configurable.\r\n[NeMo W 2024-01-04 04:44:03 megatron_base_model:821] The model: MegatronGPTSFTModel() does not have field.name: expert_model_parallel_size in its cfg. Add this key to cfg or config_mapping\r\nto make to make it configurable.\r\n[NeMo W 2024-01-04 04:44:03 megatron_base_model:821] The model: MegatronGPTSFTModel() does not have field.name: tp_comm_overlap in its cfg. Add this key to cfg or config_mapping to make to\r\nmake it configurable.\r\n[NeMo W 2024-01-04 04:44:03 megatron_base_model:821] The model: MegatronGPTSFTModel() does not have field.name: tp_comm_split_ag in its cfg. Add this key to cfg or config_mapping to make to\r\n make it configurable.\r\n[NeMo W 2024-01-04 04:44:03 megatron_base_model:821] The model: MegatronGPTSFTModel() does not have field.name: tp_comm_split_rs in its cfg. Add this key to cfg or config_mapping to make to\r\n make it configurable.\r\n[NeMo W 2024-01-04 04:44:03 megatron_base_model:821] The model: MegatronGPTSFTModel() does not have field.name: tp_comm_bulk_wgrad in its cfg. Add this key to cfg or config_mapping to make\r\nto make it configurable.\r\n[NeMo W 2024-01-04 04:44:03 megatron_base_model:821] The model: MegatronGPTSFTModel() does not have field.name: tp_comm_bulk_dgrad in its cfg. Add this key to cfg or config_mapping to make\r\nto make it configurable.\r\n[NeMo W 2024-01-04 04:44:03 megatron_base_model:821] The model: MegatronGPTSFTModel() does not have field.name: finalize_model_grads_func in its cfg. Add this key to cfg or config_mapping t\r\no make to make it configurable.\r\n[NeMo W 2024-01-04 04:44:03 megatron_base_model:821] The model: MegatronGPTSFTModel() does not have field.name: barrier_with_L1_time in its cfg. Add this key to cfg or config_mapping to mak\r\ne to make it configurable.\r\n[NeMo W 2024-01-04 04:44:03 megatron_gpt_model:1554] apply_query_key_layer_scaling is only enabled when using FP16, setting it to False and setting NVTE_APPLY_QK_LAYER_SCALING=0\r\n[NeMo W 2024-01-04 04:44:03 megatron_gpt_model:1619] The model: MegatronGPTSFTModel() does not have field.name: num_moe_experts in its cfg. Add this key to cfg or config_mapping to make to\r\nmake it configurable.\r\n[NeMo W 2024-01-04 04:44:03 megatron_base_model:821] The model: MegatronGPTSFTModel() does not have field.name: barrier_with_L1_time in its cfg. Add this key to cfg or config_mapping to mak\r\ne to make it configurable.\r\n[NeMo W 2024-01-04 04:44:03 megatron_gpt_model:1554] apply_query_key_layer_scaling is only enabled when using FP16, setting it to False and setting NVTE_APPLY_QK_LAYER_SCALING=0\r\n[NeMo W 2024-01-04 04:44:03 megatron_gpt_model:1619] The model: MegatronGPTSFTModel() does not have field.name: num_moe_experts in its cfg. Add this key to cfg or config_mapping to make to\r\nmake it configurable.\r\n[NeMo W 2024-01-04 04:44:03 megatron_gpt_model:1619] The model: MegatronGPTSFTModel() does not have field.name: fp8_wgrad in its cfg. Add this key to cfg or config_mapping to make to make i\r\nt configurable.\r\n[NeMo W 2024-01-04 04:44:03 megatron_gpt_model:1619] The model: MegatronGPTSFTModel() does not have field.name: clone_scatter_output_in_embedding in its cfg. Add this key to cfg or config_m\r\napping to make to make it configurable.\r\n[NeMo W 2024-01-04 04:44:03 nemo_logging:349] /usr/local/lib/python3.10/dist-packages/transformer_engine/pytorch/module/base.py:643: UserWarning: To guarantee overlapping TP and SP collecti\r\nves with the backwardGEMMs, set environment variable CUDA_DEVICE_MAX_CONNECTIONS = 1\r\n warnings.warn(\r\n\r\nInitializing distributed: GLOBAL_RANK: 0, MEMBER: 1/8\r\n----------------------------------------------------------------------------------------------------\r\ndistributed_backend=nccl\r\nAll distributed processes registered. Starting with 8 processes\r\n----------------------------------------------------------------------------------------------------\r\n\r\nInitializing distributed: GLOBAL_RANK: 5, MEMBER: 6/8\r\nInitializing distributed: GLOBAL_RANK: 2, MEMBER: 3/8\r\nInitializing distributed: GLOBAL_RANK: 7, MEMBER: 8/8\r\nInitializing distributed: GLOBAL_RANK: 1, MEMBER: 2/8\r\nInitializing distributed: GLOBAL_RANK: 2, MEMBER: 3/8\r\nInitializing distributed: GLOBAL_RANK: 7, MEMBER: 8/8\r\nInitializing distributed: GLOBAL_RANK: 1, MEMBER: 2/8\r\nInitializing distributed: GLOBAL_RANK: 6, MEMBER: 7/8\r\nInitializing distributed: GLOBAL_RANK: 4, MEMBER: 5/8\r\nInitializing distributed: GLOBAL_RANK: 3, MEMBER: 4/8\r\nError executing job with overrides: ['trainer.precision=bf16', 'trainer.devices=8', 'trainer.num_nodes=1', 'trainer.val_check_interval=0.1', 'trainer.max_steps=50', 'model.restore_from_path\r\n=/workspace/llama2-7b.nemo', 'model.micro_batch_size=1', 'model.global_batch_size=128', 'model.tensor_model_parallel_size=8', 'model.pipeline_model_parallel_size=1', 'model.megatron_amp_O2=\r\nTrue', 'model.sequence_parallel=True', 'model.activations_checkpoint_granularity=selective', 'model.activations_checkpoint_method=uniform', 'model.optim.name=distributed_fused_adam', 'model\r\n.optim.lr=5e-6', 'model.answer_only_loss=True', 'model.data.train_ds.file_names=', 'model.data.validation_ds.file_names=', 'model.data.test_ds.file_names=', 'model.data.train_ds.concat_samp\r\nling_probabilities=[1]', 'model.data.train_ds.max_seq_length=2048', 'model.data.validation_ds.max_seq_length=2048', 'model.data.train_ds.micro_batch_size=1', 'model.data.train_ds.global_bat\r\nch_size=128', 'model.data.validation_ds.micro_batch_size=1', 'model.data.validation_ds.global_batch_size=128', 'model.data.test_ds.micro_batch_size=1', 'model.data.test_ds.global_batch_size\r\n=256', 'model.data.train_ds.num_workers=0', 'model.data.validation_ds.num_workers=0', 'model.data.test_ds.num_workers=0', 'model.data.validation_ds.metric.name=loss', 'model.data.test_ds.me\r\ntric.name=loss', 'exp_manager.create_wandb_logger=False', 'exp_manager.explicit_log_dir=/results', 'exp_manager.resume_if_exists=True', 'exp_manager.resume_ignore_no_checkpoint=True', 'exp_\r\nmanager.create_checkpoint_callback=True', 'exp_manager.checkpoint_callback_params.monitor=validation_loss', 'exp_manager.checkpoint_callback_params.save_best_model=False', 'exp_manager.chec\r\nkpoint_callback_params.save_nemo_on_train_end=True', '++cluster_type=BCP']\r\nTraceback (most recent call last):\r\n File \"/opt/NeMo/examples/nlp/language_modeling/tuning/megatron_gpt_sft.py\", line 218, in main\r\n model = load_from_nemo(MegatronGPTSFTModel, cfg, trainer, gpt_cfg, modify_confg_fn=_modify_config)\r\n File \"/opt/NeMo/examples/nlp/language_modeling/tuning/megatron_gpt_sft.py\", line 113, in load_from_nemo\r\n model = cls.restore_from(\r\n File \"/opt/NeMo/nemo/collections/nlp/models/nlp_model.py\", line 465, in restore_from\r\n return super().restore_from(\r\n File \"/opt/NeMo/nemo/core/classes/modelPT.py\", line 442, in restore_from\r\n instance = cls._save_restore_connector.restore_from(\r\n File \"/opt/NeMo/nemo/collections/nlp/parts/nlp_overrides.py\", line 742, in restore_from\r\n checkpoint = dist_checkpointing.load(\r\n File \"/opt/megatron-lm/megatron/core/dist_checkpointing/serialization.py\", line 75, in load\r\n sharded_objects, sharded_state_dict = load_sharded_objects(sharded_state_dict, checkpoint_dir)\r\n File \"/opt/megatron-lm/megatron/core/dist_checkpointing/serialization.py\", line 129, in load_sharded_objects\r\n return dict_list_map_inplace(load_sharded_object, sharded_objects), sharded_state_dict\r\n File \"/opt/megatron-lm/megatron/core/dist_checkpointing/dict_utils.py\", line 170, in dict_list_map_inplace\r\n x[k] = dict_list_map_inplace(f, v)\r\n File \"/opt/megatron-lm/megatron/core/dist_checkpointing/dict_utils.py\", line 170, in dict_list_map_inplace\r\n x[k] = dict_list_map_inplace(f, v)\r\n File \"/opt/megatron-lm/megatron/core/dist_checkpointing/dict_utils.py\", line 174, in dict_list_map_inplace\r\n return f(x)\r\n File \"/opt/megatron-lm/megatron/core/dist_checkpointing/serialization.py\", line 126, in load_sharded_object\r\n loaded_obj = torch.load(load_path)\r\n File \"/usr/local/lib/python3.10/dist-packages/torch/serialization.py\", line 988, in load\r\n with _open_file_like(f, 'rb') as opened_file:\r\n File \"/usr/local/lib/python3.10/dist-packages/torch/serialization.py\", line 437, in _open_file_like\r\n return _open_file(name_or_buffer, mode)\r\n File \"/usr/local/lib/python3.10/dist-packages/torch/serialization.py\", line 418, in __init__\r\n super().__init__(open(name, mode))\r\nFileNotFoundError: [Errno 2] No such file or directory: '/tmp/tmpg9bo8424/model_weights/model.module.decoder.layers.mlp.linear_fc1._extra_state/shard_0_32.pt'\r\nSet the environment variable HYDRA_FULL_ERROR=1 for a complete stack trace.\r\n[2024-01-04 04:44:32,206] torch.distributed.elastic.multiprocessing.api: [WARNING] Sending process 1890 closing signal SIGTERM\r\n[2024-01-04 04:44:32,207] torch.distributed.elastic.multiprocessing.api: [WARNING] Sending process 1891 closing signal SIGTERM\r\n[2024-01-04 04:44:32,207] torch.distributed.elastic.multiprocessing.api: [WARNING] Sending process 1892 closing signal SIGTERM\r\n[2024-01-04 04:44:32,208] torch.distributed.elastic.multiprocessing.api: [WARNING] Sending process 1893 closing signal SIGTERM\r\n[2024-01-04 04:44:32,209] torch.distributed.elastic.multiprocessing.api: [WARNING] Sending process 1894 closing signal SIGTERM\r\n[2024-01-04 04:44:32,209] torch.distributed.elastic.multiprocessing.api: [WARNING] Sending process 1895 closing signal SIGTERM\r\n[2024-01-04 04:44:32,210] torch.distributed.elastic.multiprocessing.api: [WARNING] Sending process 1896 closing signal SIGTERM\r\n[2024-01-04 04:44:32,325] torch.distributed.elastic.multiprocessing.api: [ERROR] failed (exitcode: 1) local_rank: 0 (pid: 1889) of binary: /usr/bin/python\r\nTraceback (most recent call last):\r\n File \"/usr/local/bin/torchrun\", line 8, in <module>\r\n sys.exit(main())\r\n File \"/usr/local/lib/python3.10/dist-packages/torch/distributed/elastic/multiprocessing/errors/__init__.py\", line 346, in wrapper\r\n return f(*args, **kwargs)\r\n File \"/usr/local/lib/python3.10/dist-packages/torch/distributed/run.py\", line 806, in main\r\n run(args)\r\n File \"/usr/local/lib/python3.10/dist-packages/torch/distributed/run.py\", line 797, in run\r\n elastic_launch(\r\n File \"/usr/local/lib/python3.10/dist-packages/torch/distributed/launcher/api.py\", line 134, in __call__\r\n return launch_agent(self._config, self._entrypoint, list(args))\r\n File \"/usr/local/lib/python3.10/dist-packages/torch/distributed/launcher/api.py\", line 264, in launch_agent\r\n raise ChildFailedError(\r\ntorch.distributed.elastic.multiprocessing.errors.ChildFailedError:\r\n============================================================\r\n/opt/NeMo/examples/nlp/language_modeling/tuning/megatron_gpt_sft.py FAILED\r\n------------------------------------------------------------\r\nFailures:\r\n <NO_OTHER_FAILURES>\r\n------------------------------------------------------------\r\nRoot Cause (first observed failure):\r\n[0]:\r\n time : 2024-01-04_04:44:32\r\n host : node073.cm.cluster\r\n rank : 0 (local_rank: 0)\r\n exitcode : 1 (pid: 1889)\r\n error_file: <N/A>\r\n traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html\r\n============================================================\r\n```\r\n",
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"This issue was closed because it has been inactive for 7 days since being marked as stale."
] | 2023-12-28T01:34:08 | 2024-02-13T01:44:00 | 2024-02-13T01:43:59 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | **Describe the bug**
I am following [the official instructions](https://docs.nvidia.com/nemo-framework/user-guide/latest/playbooks/llama2sft.html) to fine-tune Llama-7b-hf model on a 8xH100 machine. It gave me the following error even I set the ${CONCAT_SAMPLING_PROBS}="[1]".
```
File "/usr/local/lib/python3.10/dist-packages/pytorch_lightning/strategies/launchers/subprocess_script.py", line 93, in launch
return function(*args, **kwargs)
File "/usr/local/lib/python3.10/dist-packages/pytorch_lightning/trainer/trainer.py", line 571, in _fit_impl
self._run(model, ckpt_path=ckpt_path)
File "/usr/local/lib/python3.10/dist-packages/pytorch_lightning/trainer/trainer.py", line 941, in _run
call._call_setup_hook(self) # allow user to setup lightning_module in accelerator environment
File "/usr/local/lib/python3.10/dist-packages/pytorch_lightning/trainer/call.py", line 86, in _call_setup_hook
_call_lightning_module_hook(trainer, "setup", stage=fn)
File "/usr/local/lib/python3.10/dist-packages/pytorch_lightning/trainer/call.py", line 145, in _call_lightning_module_hook
output = fn(*args, **kwargs)
File "/opt/NeMo/nemo/collections/nlp/models/language_modeling/megatron_gpt_sft_model.py", line 184, in setup
self.build_train_valid_test_datasets(stage=stage)
File "/opt/NeMo/nemo/collections/nlp/models/language_modeling/megatron_gpt_sft_model.py", line 754, in build_train_valid_test_datasets
self._validation_ds = self._build_dataset(self.cfg.data.validation_ds, is_train=False)
File "/opt/NeMo/nemo/collections/nlp/models/language_modeling/megatron_gpt_sft_model.py", line 218, in _build_dataset
raise ValueError(f"SFT train/validation datasets must be provided as a list of individual JSONL files.")
**ValueError: SFT train/validation datasets must be provided as a list of individual JSONL files.**
```
**Steps/Code to reproduce bug**
`export ${CONCAT_SAMPLING_PROBS}="[1]"`
```
torchrun --nproc_per_node=8 \
/opt/NeMo/examples/nlp/language_modeling/tuning/megatron_gpt_sft.py \
trainer.precision=bf16 \
trainer.devices=8 \
trainer.num_nodes=1 \
trainer.val_check_interval=0.1 \
trainer.max_steps=50 \
model.restore_from_path=${MODEL} \
model.micro_batch_size=1 \
model.global_batch_size=128 \
model.tensor_model_parallel_size=${TP_SIZE} \
model.pipeline_model_parallel_size=${PP_SIZE} \
model.megatron_amp_O2=True \
model.sequence_parallel=True \
model.activations_checkpoint_granularity=selective \
model.activations_checkpoint_method=uniform \
model.optim.name=distributed_fused_adam \
model.optim.lr=5e-6 \
model.answer_only_loss=True \
model.data.train_ds.file_names=${TRAIN_DS} \
model.data.validation_ds.file_names=${VALID_DS} \
model.data.test_ds.file_names=${TEST_DS} \
model.data.train_ds.concat_sampling_probabilities=${CONCAT_SAMPLING_PROBS} \
model.data.train_ds.max_seq_length=2048 \
model.data.validation_ds.max_seq_length=2048 \
model.data.train_ds.micro_batch_size=1 \
model.data.train_ds.global_batch_size=128 \
model.data.validation_ds.micro_batch_size=1 \
model.data.validation_ds.global_batch_size=128 \
model.data.test_ds.micro_batch_size=1 \
model.data.test_ds.global_batch_size=256 \
model.data.train_ds.num_workers=0 \
model.data.validation_ds.num_workers=0 \
model.data.test_ds.num_workers=0 \
model.data.validation_ds.metric.name=loss \
model.data.test_ds.metric.name=loss \
exp_manager.create_wandb_logger=False \
exp_manager.explicit_log_dir=/results \
exp_manager.resume_if_exists=True \
exp_manager.resume_ignore_no_checkpoint=True \
exp_manager.create_checkpoint_callback=True \
exp_manager.checkpoint_callback_params.monitor=validation_loss \
exp_manager.checkpoint_callback_params.save_best_model=False \
exp_manager.checkpoint_callback_params.save_nemo_on_train_end=True \
++cluster_type=BCP
```
It seems the following code in `/megatron_gpt_sft_model.py` is always triggered.
```
def _build_dataset(self, data_cfg, is_train=True):
datasets = []
# Determine if we are using a single dataset or a list of datasets.
is_list_config = isinstance(data_cfg.file_names, ListConfig)
if not is_list_config:
raise ValueError(f"SFT train/validation datasets must be provided as a list of individual JSONL files.")
```
**Environment overview
I am using docker image `nvcr.io/ea-bignlp/ga-participants/nemofw-training:23.08.03`. OS is `Ubuntu 20.04.6.LTS`.
| {
"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/8092/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/8092/timeline | null | not_planned | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/8091 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8091/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8091/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8091/events | https://github.com/NVIDIA/NeMo/pull/8091 | 2,057,694,920 | PR_kwDOC_bI7s5i2LE- | 8,091 | Transformer decoder with conditioning for canary | {
"login": "krishnacpuvvada",
"id": 93558329,
"node_id": "U_kgDOBZOWOQ",
"avatar_url": "https://avatars.githubusercontent.com/u/93558329?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/krishnacpuvvada",
"html_url": "https://github.com/krishnacpuvvada",
"followers_url": "https://api.github.com/users/krishnacpuvvada/followers",
"following_url": "https://api.github.com/users/krishnacpuvvada/following{/other_user}",
"gists_url": "https://api.github.com/users/krishnacpuvvada/gists{/gist_id}",
"starred_url": "https://api.github.com/users/krishnacpuvvada/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/krishnacpuvvada/subscriptions",
"organizations_url": "https://api.github.com/users/krishnacpuvvada/orgs",
"repos_url": "https://api.github.com/users/krishnacpuvvada/repos",
"events_url": "https://api.github.com/users/krishnacpuvvada/events{/privacy}",
"received_events_url": "https://api.github.com/users/krishnacpuvvada/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811884691,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR",
"name": "ASR",
"color": "F3B33E",
"default": false,
"description": ""
},
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
},
{
"id": 4847373924,
"node_id": "LA_kwDOC_bI7s8AAAABIO0OZA",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/common",
"name": "common",
"color": "c5def5",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [] | 2023-12-27T19:07:59 | 2023-12-27T19:42:34 | 2023-12-27T19:42:34 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8091",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8091",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8091.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8091.patch",
"merged_at": "2023-12-27T19:42:34"
} | # What does this PR do ?
Transformer decoder with conditioning for canary
**Collection**: ASR
# 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
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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": "pzelasko",
"id": 15930688,
"node_id": "MDQ6VXNlcjE1OTMwNjg4",
"avatar_url": "https://avatars.githubusercontent.com/u/15930688?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/pzelasko",
"html_url": "https://github.com/pzelasko",
"followers_url": "https://api.github.com/users/pzelasko/followers",
"following_url": "https://api.github.com/users/pzelasko/following{/other_user}",
"gists_url": "https://api.github.com/users/pzelasko/gists{/gist_id}",
"starred_url": "https://api.github.com/users/pzelasko/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/pzelasko/subscriptions",
"organizations_url": "https://api.github.com/users/pzelasko/orgs",
"repos_url": "https://api.github.com/users/pzelasko/repos",
"events_url": "https://api.github.com/users/pzelasko/events{/privacy}",
"received_events_url": "https://api.github.com/users/pzelasko/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8091/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/8091/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8090 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8090/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8090/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8090/events | https://github.com/NVIDIA/NeMo/issues/8090 | 2,057,617,666 | I_kwDOC_bI7s56pMEC | 8,090 | EncDecTransfModelBPE model has bugs in transcribe() method | {
"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": 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 | [
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"This issue was closed because it has been inactive for 7 days since being marked as stale."
] | 2023-12-27T17:28:12 | 2024-02-04T01:44:53 | 2024-02-04T01:44:52 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | **Describe the bug**
1. `setting return_hypotheses=True` in EncDecTransfModelBPE.transcribe() will crash because of undefined local variables ([here](https://github.com/NVIDIA/NeMo/blob/main/nemo/collections/asr/models/transformer_bpe_models.py#L261))
2. EncDecTransfModelBPE.transcribe() has different APIs from other ASR models, causing `transcripbe_speech.py` to crash. (to be fixed in #8087 )
| {
"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/8090/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/8090/timeline | null | not_planned | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/8089 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8089/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8089/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8089/events | https://github.com/NVIDIA/NeMo/issues/8089 | 2,057,602,372 | I_kwDOC_bI7s56pIVE | 8,089 | official training logs with loss/wer | {
"login": "orena1",
"id": 8983713,
"node_id": "MDQ6VXNlcjg5ODM3MTM=",
"avatar_url": "https://avatars.githubusercontent.com/u/8983713?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/orena1",
"html_url": "https://github.com/orena1",
"followers_url": "https://api.github.com/users/orena1/followers",
"following_url": "https://api.github.com/users/orena1/following{/other_user}",
"gists_url": "https://api.github.com/users/orena1/gists{/gist_id}",
"starred_url": "https://api.github.com/users/orena1/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/orena1/subscriptions",
"organizations_url": "https://api.github.com/users/orena1/orgs",
"repos_url": "https://api.github.com/users/orena1/repos",
"events_url": "https://api.github.com/users/orena1/events{/privacy}",
"received_events_url": "https://api.github.com/users/orena1/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4585431311,
"node_id": "LA_kwDOC_bI7s8AAAABEVAhDw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/stale",
"name": "stale",
"color": "ededed",
"default": false,
"description": null
}
] | closed | false | {
"login": "okuchaiev",
"id": 22035961,
"node_id": "MDQ6VXNlcjIyMDM1OTYx",
"avatar_url": "https://avatars.githubusercontent.com/u/22035961?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/okuchaiev",
"html_url": "https://github.com/okuchaiev",
"followers_url": "https://api.github.com/users/okuchaiev/followers",
"following_url": "https://api.github.com/users/okuchaiev/following{/other_user}",
"gists_url": "https://api.github.com/users/okuchaiev/gists{/gist_id}",
"starred_url": "https://api.github.com/users/okuchaiev/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/okuchaiev/subscriptions",
"organizations_url": "https://api.github.com/users/okuchaiev/orgs",
"repos_url": "https://api.github.com/users/okuchaiev/repos",
"events_url": "https://api.github.com/users/okuchaiev/events{/privacy}",
"received_events_url": "https://api.github.com/users/okuchaiev/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"login": "okuchaiev",
"id": 22035961,
"node_id": "MDQ6VXNlcjIyMDM1OTYx",
"avatar_url": "https://avatars.githubusercontent.com/u/22035961?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/okuchaiev",
"html_url": "https://github.com/okuchaiev",
"followers_url": "https://api.github.com/users/okuchaiev/followers",
"following_url": "https://api.github.com/users/okuchaiev/following{/other_user}",
"gists_url": "https://api.github.com/users/okuchaiev/gists{/gist_id}",
"starred_url": "https://api.github.com/users/okuchaiev/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/okuchaiev/subscriptions",
"organizations_url": "https://api.github.com/users/okuchaiev/orgs",
"repos_url": "https://api.github.com/users/okuchaiev/repos",
"events_url": "https://api.github.com/users/okuchaiev/events{/privacy}",
"received_events_url": "https://api.github.com/users/okuchaiev/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
}
] | null | [
"Probably won't be possible to share wandb",
"thanks @titu1994 , wandb would be a best case scenario, but even a tensorboard screenshot will be amazingly valuable ",
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"This issue was closed because it has been inactive for 7 days since being marked as stale."
] | 2023-12-27T17:09:32 | 2024-02-13T01:44:02 | 2024-02-13T01:44:00 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | **Is your feature request related to a problem? - Describe the solution you'd like**
Firstly, I want to express my gratitude for the comprehensive documentation provided, which I find immensely valuable.
When initiating training for a new model, distinguishing between a potential bug and normal behavior during the training process can be quite challenging. Having access to continues logging metrics loss/WER (Word Error Rate) from your training sessions would significantly enhance our debugging capabilities and the security of our training procedures. Would it be feasible for you to release or publish these logs?
**Additional context**
Something of that kind would be amaizng:
https://wandb.ai/patrickvonplaten/wav2vec2-pretrained-demo/reports/Wav2Vec2-PreTraining-Demo-Run--VmlldzoxMDk3MjAw?accessToken=oa05s1y57lizo2ocxy3k01g6db1u4pt8m6ur2n8nl4cb0ug02ms2cw313kb8ruch
But even just screenshot of tensorboard will be great | {
"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/8089/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/8089/timeline | null | not_planned | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/8088 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8088/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8088/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8088/events | https://github.com/NVIDIA/NeMo/pull/8088 | 2,057,588,339 | PR_kwDOC_bI7s5i10Lo | 8,088 | Add Bert HF checkpoint converter | {
"login": "yaoyu-33",
"id": 54727607,
"node_id": "MDQ6VXNlcjU0NzI3NjA3",
"avatar_url": "https://avatars.githubusercontent.com/u/54727607?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/yaoyu-33",
"html_url": "https://github.com/yaoyu-33",
"followers_url": "https://api.github.com/users/yaoyu-33/followers",
"following_url": "https://api.github.com/users/yaoyu-33/following{/other_user}",
"gists_url": "https://api.github.com/users/yaoyu-33/gists{/gist_id}",
"starred_url": "https://api.github.com/users/yaoyu-33/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/yaoyu-33/subscriptions",
"organizations_url": "https://api.github.com/users/yaoyu-33/orgs",
"repos_url": "https://api.github.com/users/yaoyu-33/repos",
"events_url": "https://api.github.com/users/yaoyu-33/events{/privacy}",
"received_events_url": "https://api.github.com/users/yaoyu-33/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"jenkins",
"jenkins",
"jenkins",
"jenkins",
"jenkins"
] | 2023-12-27T16:53:03 | 2024-02-01T01:16:17 | 2024-01-31T20:04:15 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8088",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8088",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8088.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8088.patch",
"merged_at": "2024-01-31T20:04:15"
} | # What does this PR do ?
Add BERT HF checkpoint converter and HF Bert Support in NeMo
**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
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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/8088/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/8088/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8087 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8087/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8087/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8087/events | https://github.com/NVIDIA/NeMo/pull/8087 | 2,057,464,681 | PR_kwDOC_bI7s5i1ZGZ | 8,087 | Add text metrics to asr eval | {
"login": "stevehuang52",
"id": 105218074,
"node_id": "U_kgDOBkWAGg",
"avatar_url": "https://avatars.githubusercontent.com/u/105218074?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/stevehuang52",
"html_url": "https://github.com/stevehuang52",
"followers_url": "https://api.github.com/users/stevehuang52/followers",
"following_url": "https://api.github.com/users/stevehuang52/following{/other_user}",
"gists_url": "https://api.github.com/users/stevehuang52/gists{/gist_id}",
"starred_url": "https://api.github.com/users/stevehuang52/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/stevehuang52/subscriptions",
"organizations_url": "https://api.github.com/users/stevehuang52/orgs",
"repos_url": "https://api.github.com/users/stevehuang52/repos",
"events_url": "https://api.github.com/users/stevehuang52/events{/privacy}",
"received_events_url": "https://api.github.com/users/stevehuang52/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811884691,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR",
"name": "ASR",
"color": "F3B33E",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"jenkins",
"jenkins"
] | 2023-12-27T14:51:49 | 2023-12-29T15:34:45 | 2023-12-29T15:34:44 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8087",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8087",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8087.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8087.patch",
"merged_at": "2023-12-29T15:34:44"
} | # What does this PR do ?
Add some text metrics to ASR evaluator to support AST models | {
"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/8087/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/8087/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8086 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8086/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8086/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8086/events | https://github.com/NVIDIA/NeMo/issues/8086 | 2,056,767,681 | I_kwDOC_bI7s56l8jB | 8,086 | GPU memory overhead when calling trainer.fit the second time | {
"login": "acphile",
"id": 24974319,
"node_id": "MDQ6VXNlcjI0OTc0MzE5",
"avatar_url": "https://avatars.githubusercontent.com/u/24974319?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/acphile",
"html_url": "https://github.com/acphile",
"followers_url": "https://api.github.com/users/acphile/followers",
"following_url": "https://api.github.com/users/acphile/following{/other_user}",
"gists_url": "https://api.github.com/users/acphile/gists{/gist_id}",
"starred_url": "https://api.github.com/users/acphile/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/acphile/subscriptions",
"organizations_url": "https://api.github.com/users/acphile/orgs",
"repos_url": "https://api.github.com/users/acphile/repos",
"events_url": "https://api.github.com/users/acphile/events{/privacy}",
"received_events_url": "https://api.github.com/users/acphile/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 | [
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"This issue was closed because it has been inactive for 7 days since being marked as stale."
] | 2023-12-26T23:35:13 | 2024-02-03T01:43:03 | 2024-02-03T01:43:02 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | **Describe the bug**
Reopen the issue https://github.com/NVIDIA/NeMo/issues/7833
**Steps/Code to reproduce bug**
when calling `trainer.fit` the second time to train a NeMo model like:
```
trainer.fit(model)
trainer.fit_loop.epoch_loop.max_steps += 10
trainer.fit(model)
```
The second time would use more GPU memory than the first.
**Steps/Code to reproduce bug**
Create a `MegatronGPTModel` with `megatron_amp_O2=True` or with `distributed_fused_adam` and then use the above code for training.
When we print `torch.cuda.memory_allocated()` at the training step, we will find the GPU memory usage at the second `trainer.fit` is significantly higher.
**Expected behavior**
We expect the second `trainer.fit` should have the similar GPU memory usage to that in the first `trainer.fit`.
**Environment details**
Check the official Nemo v1.20 docker and NeMo v1.22 branch, both have this 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/8086/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/8086/timeline | null | not_planned | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/8085 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8085/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8085/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8085/events | https://github.com/NVIDIA/NeMo/issues/8085 | 2,056,731,639 | I_kwDOC_bI7s56lzv3 | 8,085 | wav2vec training - regression ? | {
"login": "orena1",
"id": 8983713,
"node_id": "MDQ6VXNlcjg5ODM3MTM=",
"avatar_url": "https://avatars.githubusercontent.com/u/8983713?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/orena1",
"html_url": "https://github.com/orena1",
"followers_url": "https://api.github.com/users/orena1/followers",
"following_url": "https://api.github.com/users/orena1/following{/other_user}",
"gists_url": "https://api.github.com/users/orena1/gists{/gist_id}",
"starred_url": "https://api.github.com/users/orena1/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/orena1/subscriptions",
"organizations_url": "https://api.github.com/users/orena1/orgs",
"repos_url": "https://api.github.com/users/orena1/repos",
"events_url": "https://api.github.com/users/orena1/events{/privacy}",
"received_events_url": "https://api.github.com/users/orena1/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 | [
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"This issue was closed because it has been inactive for 7 days since being marked as stale.",
"Not sure why this one was closed? From what I see `wav2vec` does not work\r\n@titu1994 @tbartley94 \r\n",
"@orena1 Sorry, the CI bot closes it if there's no activity. On the docket to look at.",
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"This issue was closed because it has been inactive for 7 days since being marked as stale."
] | 2023-12-26T22:00:40 | 2024-03-31T01:45:32 | 2024-03-31T01:45:32 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | **Describe the bug**
Hi,
I am trying to train wav2vec as in this ticket:
https://github.com/NVIDIA/NeMo/issues/4362
but I get an error, this is the script that I use:
```
python speech_to_text_ctc.py \
--config-path='../conf/wav2vec_ctc' \
--config-name=wav2vecCTC \
model.train_ds.manifest_filepath=/mnt/NVM.SGPU01/oren/libri_mini/train_clean_5.json \
model.train_ds.batch_size=10 \
model.validation_ds.batch_size=10 \
model.validation_ds.manifest_filepath=/mnt/NVM.SGPU01/oren//libri_mini/dev_clean_2.json \
model.tokenizer.dir=/home/oren.amsalem/.cache/torch/hub/torchaudio/tutorial-assets/spm_unigram_1023.model \
```
I get this error:
```
Error executing job with overrides: ['model.train_ds.manifest_filepath=/mnt/NVM.SGPU01/oren/libri_mini/train_clean_5.json', 'model.train_ds.batch_size=10', 'model.validation_ds.batch_size=10', 'model.validation_ds.manifest_filepath=/mnt/NVM.SGPU01/oren//libri_mini/dev_clean_2.json', 'model.tokenizer.dir=/home/oren.amsalem/.cache/torch/hub/torchaudio/tutorial-assets/spm_unigram_1023.model']
Traceback (most recent call last):
File "/home/oren.amsalem/cloned/NeMo/examples/asr/asr_ctc/speech_to_text_ctc.py", line 91, in main
trainer.fit(asr_model)
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/pytorch_lightning/trainer/trainer.py", line 532, in fit
call._call_and_handle_interrupt(
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/pytorch_lightning/trainer/call.py", line 42, in _call_and_handle_interrupt
return trainer.strategy.launcher.launch(trainer_fn, *args, trainer=trainer, **kwargs)
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/pytorch_lightning/strategies/launchers/subprocess_script.py", line 93, in launch
return function(*args, **kwargs)
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/pytorch_lightning/trainer/trainer.py", line 571, in _fit_impl
self._run(model, ckpt_path=ckpt_path)
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/pytorch_lightning/trainer/trainer.py", line 980, in _run
results = self._run_stage()
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/pytorch_lightning/trainer/trainer.py", line 1023, in _run_stage
self.fit_loop.run()
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/pytorch_lightning/loops/fit_loop.py", line 202, in run
self.advance()
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/pytorch_lightning/loops/fit_loop.py", line 355, in advance
self.epoch_loop.run(self._data_fetcher)
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/pytorch_lightning/loops/training_epoch_loop.py", line 133, in run
self.advance(data_fetcher)
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/pytorch_lightning/loops/training_epoch_loop.py", line 219, in advance
batch_output = self.automatic_optimization.run(trainer.optimizers[0], kwargs)
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/pytorch_lightning/loops/optimization/automatic.py", line 188, in run
self._optimizer_step(kwargs.get("batch_idx", 0), closure)
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/pytorch_lightning/loops/optimization/automatic.py", line 266, in _optimizer_step
call._call_lightning_module_hook(
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/pytorch_lightning/trainer/call.py", line 145, in _call_lightning_module_hook
output = fn(*args, **kwargs)
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/pytorch_lightning/core/module.py", line 1270, in optimizer_step
optimizer.step(closure=optimizer_closure)
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/pytorch_lightning/core/optimizer.py", line 161, in step
step_output = self._strategy.optimizer_step(self._optimizer, closure, **kwargs)
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/pytorch_lightning/strategies/ddp.py", line 257, in optimizer_step
optimizer_output = super().optimizer_step(optimizer, closure, model, **kwargs)
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/pytorch_lightning/strategies/strategy.py", line 231, in optimizer_step
return self.precision_plugin.optimizer_step(optimizer, model=model, closure=closure, **kwargs)
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/pytorch_lightning/plugins/precision/precision_plugin.py", line 116, in optimizer_step
return optimizer.step(closure=closure, **kwargs)
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/torch/optim/lr_scheduler.py", line 68, in wrapper
return wrapped(*args, **kwargs)
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/torch/optim/optimizer.py", line 373, in wrapper
out = func(*args, **kwargs)
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/torch/optim/optimizer.py", line 76, in _use_grad
ret = func(self, *args, **kwargs)
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/torch/optim/adamw.py", line 161, in step
loss = closure()
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/pytorch_lightning/plugins/precision/precision_plugin.py", line 103, in _wrap_closure
closure_result = closure()
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/pytorch_lightning/loops/optimization/automatic.py", line 142, in __call__
self._result = self.closure(*args, **kwargs)
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/pytorch_lightning/loops/optimization/automatic.py", line 128, in closure
step_output = self._step_fn()
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/pytorch_lightning/loops/optimization/automatic.py", line 315, in _training_step
training_step_output = call._call_strategy_hook(trainer, "training_step", *kwargs.values())
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/pytorch_lightning/trainer/call.py", line 293, in _call_strategy_hook
output = fn(*args, **kwargs)
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/pytorch_lightning/strategies/ddp.py", line 330, in training_step
return self.model(*args, **kwargs)
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1518, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1527, in _call_impl
return forward_call(*args, **kwargs)
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/torch/nn/parallel/distributed.py", line 1515, in forward
inputs, kwargs = self._pre_forward(*inputs, **kwargs)
File "/mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/torch/nn/parallel/distributed.py", line 1409, in _pre_forward
if torch.is_grad_enabled() and self.reducer._rebuild_buckets():
RuntimeError: It looks like your LightningModule has parameters that were not used in producing the loss returned by training_step. If this is intentional, you must enable the detection of unused parameters in DDP, either by setting the string value `strategy='ddp_find_unused_parameters_true'` or by setting the flag in the strategy with `strategy=DDPStrategy(find_unused_parameters=True)`.
```
I've tried using `accelerator=cpu` but I get an error about the device - here:
https://github.com/NVIDIA/NeMo/pull/8084
When I fix that error as in the pull request I get this error:
```
377.513 Total estimated model params size (MB)
Epoch 0: 0%| | 0/152 [00:00<?, ?it/s]
double free or corruption (out)
Aborted (core dumped)
```
**Environment overview (please complete the following information)**
**Environment details**
torch 2.1.2
python 3.10
| {
"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/8085/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/8085/timeline | null | not_planned | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/8084 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8084/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8084/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8084/events | https://github.com/NVIDIA/NeMo/pull/8084 | 2,056,717,631 | PR_kwDOC_bI7s5iy5tD | 8,084 | fix device setting to allow using accelerator cpu | {
"login": "orena1",
"id": 8983713,
"node_id": "MDQ6VXNlcjg5ODM3MTM=",
"avatar_url": "https://avatars.githubusercontent.com/u/8983713?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/orena1",
"html_url": "https://github.com/orena1",
"followers_url": "https://api.github.com/users/orena1/followers",
"following_url": "https://api.github.com/users/orena1/following{/other_user}",
"gists_url": "https://api.github.com/users/orena1/gists{/gist_id}",
"starred_url": "https://api.github.com/users/orena1/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/orena1/subscriptions",
"organizations_url": "https://api.github.com/users/orena1/orgs",
"repos_url": "https://api.github.com/users/orena1/repos",
"events_url": "https://api.github.com/users/orena1/events{/privacy}",
"received_events_url": "https://api.github.com/users/orena1/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811884691,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR",
"name": "ASR",
"color": "F3B33E",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"jenkins",
"Thanks @titu1994 , see also here:\r\nhttps://github.com/NVIDIA/NeMo/issues/8085",
"jenkins ",
"@titu1994 i know you pushed but just for record I'm not seeing any issues so review is approved",
"Oh thanks for the update, I think I misremembered as this having your approval on slack somehow, turns out it was a different topic different pr. Sorry about that. ",
"do you @titu1994 or @tbartley94 have any idea about this one:\r\nhttps://github.com/NVIDIA/NeMo/issues/8085",
"Hmm, lemme take a look. I'd wager it's some tech debt between PTL versions. (fervor for wav2vec dipped before PTL 2.0 was released, so the code is likely working off a dated namespace.)"
] | 2023-12-26T21:27:33 | 2024-01-03T01:20:54 | 2023-12-31T21:38:34 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8084",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8084",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8084.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8084.patch",
"merged_at": "2023-12-31T21:38:34"
} | # What does this PR do ?
It is currently impossible to run wav2vec with `trainer.accelerator=cpu` if the machine has a gpu.
The padding_mask is already on the device set in `trainer.accelerator` but if the machine has a GPU the DEVICE variable will contain cuda and this error message will be received:
```
File /mnt/NVM/oren/anaconda3_10/envs/nemo/lib/python3.10/site-packages/nemo/collections/asr/modules/wav2vec_modules.py:348, in Wav2VecTransformerEncoder.create_padding_mask(self, length)
345 padding_mask = torch.arange(max_len, device=DEVICE)
347 # Switch to binary for transformer, 1 for valid tokens, 0 for padding
--> 348 padding_mask = (padding_mask.expand(len(length), max_len) < length.unsqueeze(1)).type(torch.uint8)
350 return padding_mask
RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu!
```
**PR Type**:
- [x] Bugfix
# Additional Information
* Related to # (issue)
*
@titu1994, @redoctopus, @jbalam-nv, or @okuchaiev | {
"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/8084/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/8084/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8083 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8083/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8083/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8083/events | https://github.com/NVIDIA/NeMo/issues/8083 | 2,055,995,043 | I_kwDOC_bI7s56i_6j | 8,083 | How to improve ASR speed, Tested on Nvidia T4, L4, A100 give me same speed | {
"login": "Eittipat",
"id": 1196678,
"node_id": "MDQ6VXNlcjExOTY2Nzg=",
"avatar_url": "https://avatars.githubusercontent.com/u/1196678?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Eittipat",
"html_url": "https://github.com/Eittipat",
"followers_url": "https://api.github.com/users/Eittipat/followers",
"following_url": "https://api.github.com/users/Eittipat/following{/other_user}",
"gists_url": "https://api.github.com/users/Eittipat/gists{/gist_id}",
"starred_url": "https://api.github.com/users/Eittipat/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/Eittipat/subscriptions",
"organizations_url": "https://api.github.com/users/Eittipat/orgs",
"repos_url": "https://api.github.com/users/Eittipat/repos",
"events_url": "https://api.github.com/users/Eittipat/events{/privacy}",
"received_events_url": "https://api.github.com/users/Eittipat/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 | [
"Can you check if your are loading the model every time ? ",
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"This issue was closed because it has been inactive for 7 days since being marked as stale."
] | 2023-12-26T03:36:32 | 2024-02-10T01:43:05 | 2024-02-10T01:43:04 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | Hello,
I want to know how to speed up NEMO.
Originally, I ran test with **[stt_zh_citrinet_1024_gamma_0_25](https://catalog.ngc.nvidia.com/orgs/nvidia/teams/nemo/models/stt_zh_citrinet_1024_gamma_0_25)** model with 30 seconds audio clip on Nvidia T4,
It give me around 1 - 2 seconds. It too slow for our requirements.
Then, I ran test on both Nvidia L4 and A100 which is more powerful than T4,
But it still give me same speed (1-2 seconds),
So, How to improve ASR speed? Maybe I did something wrong. Please advise
PS. I'm sure CUDA is enabled, Using CPU give me more than 10 seconds
python = 3.10
nemo_toolkit = 1.21
cuda = 12.1
debian = 11 | {
"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/8083/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/8083/timeline | null | not_planned | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/8082 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8082/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8082/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8082/events | https://github.com/NVIDIA/NeMo/issues/8082 | 2,055,145,119 | I_kwDOC_bI7s56fwaf | 8,082 | Online Diarizer | {
"login": "yoavellinson",
"id": 75246152,
"node_id": "MDQ6VXNlcjc1MjQ2MTUy",
"avatar_url": "https://avatars.githubusercontent.com/u/75246152?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/yoavellinson",
"html_url": "https://github.com/yoavellinson",
"followers_url": "https://api.github.com/users/yoavellinson/followers",
"following_url": "https://api.github.com/users/yoavellinson/following{/other_user}",
"gists_url": "https://api.github.com/users/yoavellinson/gists{/gist_id}",
"starred_url": "https://api.github.com/users/yoavellinson/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/yoavellinson/subscriptions",
"organizations_url": "https://api.github.com/users/yoavellinson/orgs",
"repos_url": "https://api.github.com/users/yoavellinson/repos",
"events_url": "https://api.github.com/users/yoavellinson/events{/privacy}",
"received_events_url": "https://api.github.com/users/yoavellinson/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 | [
"Hi.\r\n\r\nPlease understand that speaker diarization is a type of pattern recognition system, \r\nso it does not have 100% accuracy on any dataset and should not be regarded as a bug. \r\nThere is a chance of errors in terms of speaker labels or speaker counts.\r\n\r\nHowever, you can try changing [self.min_frame_per_spk](https://github.com/NVIDIA/NeMo/blob/274a21b492f337702f83f7b82abca338d6c9c09e/nemo/collections/asr/parts/utils/online_clustering.py#L606C9-L606C26) variable which controls \"minimum required number of segments to count a speaker\" value.\r\nLowering this value (below default 15) will increase the chance of counting more speakers.\r\nCheck out this `min()` operation at [This line](https://github.com/NVIDIA/NeMo/blob/274a21b492f337702f83f7b82abca338d6c9c09e/nemo/collections/asr/parts/utils/online_clustering.py#L724).\r\n\r\nWe implemented this threshold because it is very challenging to count the number of speakers in a short audio (less then 15 sec per speaker)\r\n\r\n \r\n ",
"Hi, Thanks for the insights!\r\nI tried it and still got one speaker only. is there an example config file for online clustering? \r\n\r\nYoav",
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"This issue was closed because it has been inactive for 7 days since being marked as stale."
] | 2023-12-24T15:51:13 | 2024-02-09T01:43:45 | 2024-02-09T01:43:44 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | **Describe the bug**
The online diarizer is not working a excpected. when trying it on a file with two speakers it only outputs one as active.
**Steps/Code to reproduce bug**
creating a class instance of : https://github.com/NVIDIA/NeMo/blob/274a21b492f337702f83f7b82abca338d6c9c09e/nemo/collections/asr/models/online_diarizer.py
and diarize().
There is no wraper code for the class so it might be the case.
**Environment overview (please complete the following information)**
- Environment location: DOCKER
- Method of NeMo install: source
**Environment details**
If NVIDIA docker image is used you don't need to specify these.
Otherwise, please provide:
- OS version : UBUNTU 20.04
- PyTorch version 1.12.1+cu116
- Python version 3.8.10
**Additional context**
Add any other context about the problem here.
Example: GPU model RTX A6000
| {
"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/8082/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/8082/timeline | null | not_planned | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/8080 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8080/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8080/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8080/events | https://github.com/NVIDIA/NeMo/pull/8080 | 2,054,074,150 | PR_kwDOC_bI7s5iqOL4 | 8,080 | update broken links | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers",
"following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}",
"gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}",
"starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions",
"organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs",
"repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos",
"events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}",
"received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events",
"type": "Bot",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4556669155,
"node_id": "LA_kwDOC_bI7s8AAAABD5lA4w",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/cherry-pick",
"name": "cherry-pick",
"color": "ededed",
"default": false,
"description": null
},
{
"id": 4811884691,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR",
"name": "ASR",
"color": "F3B33E",
"default": false,
"description": ""
},
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [] | 2023-12-22T15:24:59 | 2023-12-22T18:39:43 | 2023-12-22T18:39:42 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8080",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8080",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8080.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8080.patch",
"merged_at": "2023-12-22T18:39:42"
} | # What does this PR do ?
Fix Broken Links
**Collection**: Docs
# Changelog
- update rst files
# 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
- [x] Documentation | {
"login": "titu1994",
"id": 3048602,
"node_id": "MDQ6VXNlcjMwNDg2MDI=",
"avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/titu1994",
"html_url": "https://github.com/titu1994",
"followers_url": "https://api.github.com/users/titu1994/followers",
"following_url": "https://api.github.com/users/titu1994/following{/other_user}",
"gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}",
"starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/titu1994/subscriptions",
"organizations_url": "https://api.github.com/users/titu1994/orgs",
"repos_url": "https://api.github.com/users/titu1994/repos",
"events_url": "https://api.github.com/users/titu1994/events{/privacy}",
"received_events_url": "https://api.github.com/users/titu1994/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8080/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/8080/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8079 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8079/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8079/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8079/events | https://github.com/NVIDIA/NeMo/pull/8079 | 2,053,731,270 | PR_kwDOC_bI7s5ipC85 | 8,079 | update broken links | {
"login": "nithinraok",
"id": 19668129,
"node_id": "MDQ6VXNlcjE5NjY4MTI5",
"avatar_url": "https://avatars.githubusercontent.com/u/19668129?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/nithinraok",
"html_url": "https://github.com/nithinraok",
"followers_url": "https://api.github.com/users/nithinraok/followers",
"following_url": "https://api.github.com/users/nithinraok/following{/other_user}",
"gists_url": "https://api.github.com/users/nithinraok/gists{/gist_id}",
"starred_url": "https://api.github.com/users/nithinraok/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/nithinraok/subscriptions",
"organizations_url": "https://api.github.com/users/nithinraok/orgs",
"repos_url": "https://api.github.com/users/nithinraok/repos",
"events_url": "https://api.github.com/users/nithinraok/events{/privacy}",
"received_events_url": "https://api.github.com/users/nithinraok/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811884691,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR",
"name": "ASR",
"color": "F3B33E",
"default": false,
"description": ""
},
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"jenkins"
] | 2023-12-22T10:39:57 | 2023-12-22T15:24:23 | 2023-12-22T15:24:22 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8079",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8079",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8079.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8079.patch",
"merged_at": "2023-12-22T15:24:22"
} | # What does this PR do ?
Fix Broken Links
**Collection**: Docs
# Changelog
- update rst files
# 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
- [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/8079/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/8079/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8078 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8078/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8078/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8078/events | https://github.com/NVIDIA/NeMo/pull/8078 | 2,053,715,098 | PR_kwDOC_bI7s5io_aY | 8,078 | run with non-dev option | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers",
"following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}",
"gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}",
"starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions",
"organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs",
"repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos",
"events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}",
"received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events",
"type": "Bot",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4556669155,
"node_id": "LA_kwDOC_bI7s8AAAABD5lA4w",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/cherry-pick",
"name": "cherry-pick",
"color": "ededed",
"default": false,
"description": null
},
{
"id": 4811884691,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR",
"name": "ASR",
"color": "F3B33E",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"jenkins"
] | 2023-12-22T10:26:41 | 2023-12-23T19:50:59 | 2023-12-23T19:50:58 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8078",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8078",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8078.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8078.patch",
"merged_at": "2023-12-23T19:50:58"
} | # What does this PR do ?
Looks like pip installs run with -e option, don;t load well without restarting.
**Collection**: [TUTORIALS]
# Changelog
- Run without pip editable
# Before your PR is "Ready for review"
**Pre checks**:
- [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md)
- [ ] Did you write any new necessary tests?
- [ ] Did you add or update any necessary documentation?
- [ ] Does the PR affect components that are optional to install? (Ex: Numba, Pynini, Apex etc)
- [ ] Reviewer: Does the PR have correct import guards for all optional libraries?
**PR Type**:
- [ ] New Feature
- [x] Bugfix
- [ ] Documentation | {
"login": "nithinraok",
"id": 19668129,
"node_id": "MDQ6VXNlcjE5NjY4MTI5",
"avatar_url": "https://avatars.githubusercontent.com/u/19668129?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/nithinraok",
"html_url": "https://github.com/nithinraok",
"followers_url": "https://api.github.com/users/nithinraok/followers",
"following_url": "https://api.github.com/users/nithinraok/following{/other_user}",
"gists_url": "https://api.github.com/users/nithinraok/gists{/gist_id}",
"starred_url": "https://api.github.com/users/nithinraok/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/nithinraok/subscriptions",
"organizations_url": "https://api.github.com/users/nithinraok/orgs",
"repos_url": "https://api.github.com/users/nithinraok/repos",
"events_url": "https://api.github.com/users/nithinraok/events{/privacy}",
"received_events_url": "https://api.github.com/users/nithinraok/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8078/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/8078/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8077 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8077/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8077/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8077/events | https://github.com/NVIDIA/NeMo/pull/8077 | 2,053,694,858 | PR_kwDOC_bI7s5io658 | 8,077 | run with non-dev option | {
"login": "nithinraok",
"id": 19668129,
"node_id": "MDQ6VXNlcjE5NjY4MTI5",
"avatar_url": "https://avatars.githubusercontent.com/u/19668129?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/nithinraok",
"html_url": "https://github.com/nithinraok",
"followers_url": "https://api.github.com/users/nithinraok/followers",
"following_url": "https://api.github.com/users/nithinraok/following{/other_user}",
"gists_url": "https://api.github.com/users/nithinraok/gists{/gist_id}",
"starred_url": "https://api.github.com/users/nithinraok/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/nithinraok/subscriptions",
"organizations_url": "https://api.github.com/users/nithinraok/orgs",
"repos_url": "https://api.github.com/users/nithinraok/repos",
"events_url": "https://api.github.com/users/nithinraok/events{/privacy}",
"received_events_url": "https://api.github.com/users/nithinraok/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811884691,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR",
"name": "ASR",
"color": "F3B33E",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [] | 2023-12-22T10:12:41 | 2023-12-22T10:26:19 | 2023-12-22T10:26:18 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8077",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8077",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8077.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8077.patch",
"merged_at": "2023-12-22T10:26:18"
} | # What does this PR do ?
Looks like pip installs run with -e option, don;t load well without restarting.
**Collection**: [TUTORIALS]
# Changelog
- Run without pip editable
# Before your PR is "Ready for review"
**Pre checks**:
- [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md)
- [ ] Did you write any new necessary tests?
- [ ] Did you add or update any necessary documentation?
- [ ] Does the PR affect components that are optional to install? (Ex: Numba, Pynini, Apex etc)
- [ ] Reviewer: Does the PR have correct import guards for all optional libraries?
**PR Type**:
- [ ] New Feature
- [x] Bugfix
- [ ] Documentation | {
"login": "titu1994",
"id": 3048602,
"node_id": "MDQ6VXNlcjMwNDg2MDI=",
"avatar_url": "https://avatars.githubusercontent.com/u/3048602?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/titu1994",
"html_url": "https://github.com/titu1994",
"followers_url": "https://api.github.com/users/titu1994/followers",
"following_url": "https://api.github.com/users/titu1994/following{/other_user}",
"gists_url": "https://api.github.com/users/titu1994/gists{/gist_id}",
"starred_url": "https://api.github.com/users/titu1994/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/titu1994/subscriptions",
"organizations_url": "https://api.github.com/users/titu1994/orgs",
"repos_url": "https://api.github.com/users/titu1994/repos",
"events_url": "https://api.github.com/users/titu1994/events{/privacy}",
"received_events_url": "https://api.github.com/users/titu1994/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8077/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/8077/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8075 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8075/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8075/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8075/events | https://github.com/NVIDIA/NeMo/issues/8075 | 2,053,164,678 | I_kwDOC_bI7s56YM6G | 8,075 | Issues with Tokenizer in TTS Es Multispeaker FastPitch HiFiGAN Model - Problems with Date Tokenization | {
"login": "rodrigoGA",
"id": 2334523,
"node_id": "MDQ6VXNlcjIzMzQ1MjM=",
"avatar_url": "https://avatars.githubusercontent.com/u/2334523?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/rodrigoGA",
"html_url": "https://github.com/rodrigoGA",
"followers_url": "https://api.github.com/users/rodrigoGA/followers",
"following_url": "https://api.github.com/users/rodrigoGA/following{/other_user}",
"gists_url": "https://api.github.com/users/rodrigoGA/gists{/gist_id}",
"starred_url": "https://api.github.com/users/rodrigoGA/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/rodrigoGA/subscriptions",
"organizations_url": "https://api.github.com/users/rodrigoGA/orgs",
"repos_url": "https://api.github.com/users/rodrigoGA/repos",
"events_url": "https://api.github.com/users/rodrigoGA/events{/privacy}",
"received_events_url": "https://api.github.com/users/rodrigoGA/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | {
"login": "rlangman",
"id": 9942053,
"node_id": "MDQ6VXNlcjk5NDIwNTM=",
"avatar_url": "https://avatars.githubusercontent.com/u/9942053?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/rlangman",
"html_url": "https://github.com/rlangman",
"followers_url": "https://api.github.com/users/rlangman/followers",
"following_url": "https://api.github.com/users/rlangman/following{/other_user}",
"gists_url": "https://api.github.com/users/rlangman/gists{/gist_id}",
"starred_url": "https://api.github.com/users/rlangman/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/rlangman/subscriptions",
"organizations_url": "https://api.github.com/users/rlangman/orgs",
"repos_url": "https://api.github.com/users/rlangman/repos",
"events_url": "https://api.github.com/users/rlangman/events{/privacy}",
"received_events_url": "https://api.github.com/users/rlangman/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"login": "rlangman",
"id": 9942053,
"node_id": "MDQ6VXNlcjk5NDIwNTM=",
"avatar_url": "https://avatars.githubusercontent.com/u/9942053?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/rlangman",
"html_url": "https://github.com/rlangman",
"followers_url": "https://api.github.com/users/rlangman/followers",
"following_url": "https://api.github.com/users/rlangman/following{/other_user}",
"gists_url": "https://api.github.com/users/rlangman/gists{/gist_id}",
"starred_url": "https://api.github.com/users/rlangman/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/rlangman/subscriptions",
"organizations_url": "https://api.github.com/users/rlangman/orgs",
"repos_url": "https://api.github.com/users/rlangman/repos",
"events_url": "https://api.github.com/users/rlangman/events{/privacy}",
"received_events_url": "https://api.github.com/users/rlangman/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
}
] | null | [
"A few days ago, a new Spanish tokenizer was uploaded to NGC that works properly, so I would consider this issue closed.\r\n\r\nhttps://catalog.ngc.nvidia.com/orgs/nvidia/teams/riva/models/normalization_es_es/files"
] | 2023-12-21T23:17:52 | 2024-01-02T12:45:04 | 2024-01-02T12:45:04 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | Hello,
I am currently working on integrating a Latin American Spanish text-to-speech (TTS) model into NVIDIA Riva, specifically using the [tts_es_multispeaker_fastpitchhifigan](https://catalog.ngc.nvidia.com/orgs/nvidia/teams/nemo/models/tts_es_multispeaker_fastpitchhifigan) model from NVIDIA NGC. After converting the model to a .riva extension using nemo2riva, I downloaded the tokenizer and verbalizer from Inverse [Normalization ES-US](https://catalog.ngc.nvidia.com/orgs/nvidia/teams/tao/models/inverse_normalization_es_us).
I have deployed it in riva with
```bash
docker run --init -it --rm --gpus '"'"device=0"'"' -v $(pwd):/data -v riva-model-repo:/data-volumen -e "MODEL_DEPLOY_KEY=tlt_encode" --name riva-service-maker nvcr.io/nvidia/riva/riva-speech:2.13.0-servicemaker
riva-build speech_synthesis tts_es_hifigan_ft_fastpitch_multispeaker.rmir:tlt_encode tts_es_fastpitch_multispeaker.riva:tlt_encode tts_es_hifigan_ft_fastpitch_multispeaker.riva:tlt_encode --voice_name Latin-American-Spanish --wfst_tokenizer_model=tokenize_and_classify.far --wfst_verbalizer_model=verbalize.far --sample_rate 44100 --language_code es-US --num_speakers=174 --phone_set=ipa --subvoices 0:0,1:1,2:2,3:3,4:4,5:5,6:6,7:7,8:8,9:9,10:10,11:11,12:12,13:13,14:14,15:15,16:16,17:17,18:18,19:19,20:20,21:21,22:22,23:23,24:24,25:25,26:26,27:27,28:28,29:29,30:30,31:31,32:32,33:33,34:34,35:35,36:36,37:37,38:38,39:39,40:40,41:41,42:42,43:43,44:44,45:45,46:46,47:47,48:48,49:49,50:50,51:51,52:52,53:53,54:54,55:55,56:56,57:57,58:58,59:59,60:60,61:61,62:62,63:63,64:64,65:65,66:66,67:67,68:68,69:69,70:70,71:71,72:72,73:73,74:74,75:75,76:76,77:77,78:78,79:79,80:80,81:81,82:82,83:83,84:84,85:85,86:86,87:87,88:88,89:89,90:90,91:91,92:92,93:93,94:94,95:95,96:96,97:97,98:98,99:99,100:100,101:101,102:102,103:103,104:104,105:105,106:106,107:107,108:108,109:109,110:110,111:111,112:112,113:113,114:114,115:115,116:116,117:117,118:118,119:119,120:120,121:121,122:122,123:123,124:124,125:125,126:126,127:127,128:128,129:129,130:130,131:131,132:132,133:133,134:134,135:135,136:136,137:137,138:138,139:139,140:140,141:141,142:142,143:143,144:144,145:145,146:146,147:147,148:148,149:149,150:150,151:151,152:152,153:153,154:154,155:155,156:156,157:157,158:158,159:159,160:160,161:161,162:162,163:163,164:164,165:165,166:166,167:167,168:168,169:169,170:170,171:171,172:172,173:173
riva-deploy -f tts_es_hifigan_ft_fastpitch_multispeaker.rmir:tlt_encode /data/models
```
However, I've encountered a significant issue: the tokenizer struggles with words that require tokenization, especially dates. Instead of processing and pronouncing these elements correctly, it either skips them or handles them incorrectly. This is in stark contrast to the performance of the default models in riva_quickstart_v2.13.0, where the es-ES model handles such tokenization challenges efficiently.
According to the [NVIDIA NGC](https://catalog.ngc.nvidia.com/orgs/nvidia/teams/tao/models/inverse_normalization_es_us), the text normalizers were built using NeMo, but they do not seem to function correctly for dates, unlike the default models for es-ES.
I would like to inquire:
- Are there any pre-trained text normalizers for Spanish that work well with dates?
- Could the issue with date tokenization be arising from some other aspect of the configuration or implementation?
- Any guidance or suggestions to resolve this issue with date tokenization would be greatly appreciated.
Best regards. | {
"login": "rodrigoGA",
"id": 2334523,
"node_id": "MDQ6VXNlcjIzMzQ1MjM=",
"avatar_url": "https://avatars.githubusercontent.com/u/2334523?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/rodrigoGA",
"html_url": "https://github.com/rodrigoGA",
"followers_url": "https://api.github.com/users/rodrigoGA/followers",
"following_url": "https://api.github.com/users/rodrigoGA/following{/other_user}",
"gists_url": "https://api.github.com/users/rodrigoGA/gists{/gist_id}",
"starred_url": "https://api.github.com/users/rodrigoGA/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/rodrigoGA/subscriptions",
"organizations_url": "https://api.github.com/users/rodrigoGA/orgs",
"repos_url": "https://api.github.com/users/rodrigoGA/repos",
"events_url": "https://api.github.com/users/rodrigoGA/events{/privacy}",
"received_events_url": "https://api.github.com/users/rodrigoGA/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8075/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/8075/timeline | null | completed | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/8074 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8074/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8074/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8074/events | https://github.com/NVIDIA/NeMo/issues/8074 | 2,051,882,454 | I_kwDOC_bI7s56TT3W | 8,074 | RuntimeError: cusolver error: CUSOLVER_STATUS_NOT_INITIALIZED, when calling `cusolverDnCreate(handle) | {
"login": "Swami-Abhinav",
"id": 110811183,
"node_id": "U_kgDOBprYLw",
"avatar_url": "https://avatars.githubusercontent.com/u/110811183?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Swami-Abhinav",
"html_url": "https://github.com/Swami-Abhinav",
"followers_url": "https://api.github.com/users/Swami-Abhinav/followers",
"following_url": "https://api.github.com/users/Swami-Abhinav/following{/other_user}",
"gists_url": "https://api.github.com/users/Swami-Abhinav/gists{/gist_id}",
"starred_url": "https://api.github.com/users/Swami-Abhinav/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/Swami-Abhinav/subscriptions",
"organizations_url": "https://api.github.com/users/Swami-Abhinav/orgs",
"repos_url": "https://api.github.com/users/Swami-Abhinav/repos",
"events_url": "https://api.github.com/users/Swami-Abhinav/events{/privacy}",
"received_events_url": "https://api.github.com/users/Swami-Abhinav/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 | [
"Using `torch.backends.cudnn.enabled = False` before model.diarize() solves the issue temporarily\r\n",
"This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.",
"This issue was closed because it has been inactive for 7 days since being marked as stale."
] | 2023-12-21T07:36:54 | 2024-01-29T01:44:18 | 2024-01-29T01:44:18 | NONE | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | null | null | I am facing
`RuntimeError: cusolver error: CUSOLVER_STATUS_NOT_INITIALIZED, when calling cusolverDnCreate(handle)
`
I am getting this issue while running the msdd_model.diarize() function, I have checked my GPU and it has enough space to run the file.
I saw the same issue in https://github.com/PanQiWei/AutoGPTQ/issues/478, but in my case there is no AutoGPTQ. It is occuring when I run the line msdd_model.diarize().
My Cuda and Pytorch versions are these:
Cuda version 11.7 (Cuda compilation tools, release 11.7, V11.7.99)
Pytorch Version -2.0.1+cu117
A clear and concise description of what the bug is.
Description
I encountered a RuntimeError: cusolver error: CUSOLVER_STATUS_NOT_INITIALIZED when calling cusolverDnCreate(handle) while running the msdd_model.diarize() function.
Steps to Reproduce
from nemo.collections.asr.models import NeuralDiarizer
from nemo.collections.asr.models.msdd_models import create_config
msdd_model = NeuralDiarizer(cfg=create_config(temp_path)).to("cuda")
msdd_model.diarize()
**Expected behavior**
I expected the msdd_model.diarize() function to run without errors.
**Environment overview (please complete the following information)**
Environment Location: JupyterHub
Method of NeMo Install: pip install
Commands Used for Installation: `!pip install --no-build-isolation nemo_toolkit[asr]==1.21.0 --user`
OS Version: Linux 5.4.254-170.358.amzn2.x86_64
PyTorch Version: 2.0.1+cu117
Python Version: 3.10
GPU Model: NVIDIA A10G
**Environment details**
Ensure that the GPU has enough space.
CUDA Version: 11.7 (Cuda compilation tools, release 11.7, V11.7.99)
**Additional context**
Additional Context
I saw a similar issue reported in [GitHub Issue #478](https://github.com/PanQiWei/AutoGPTQ/issues/478), but my case does not involve AutoGPTQ. The error occurs specifically when running the msdd_model.diarize() function.
| {
"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/8074/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/8074/timeline | null | not_planned | false |
https://api.github.com/repos/NVIDIA/NeMo/issues/8073 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8073/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8073/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8073/events | https://github.com/NVIDIA/NeMo/pull/8073 | 2,051,820,999 | PR_kwDOC_bI7s5iige- | 8,073 | update reqs | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers",
"following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}",
"gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}",
"starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions",
"organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs",
"repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos",
"events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}",
"received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events",
"type": "Bot",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4556669155,
"node_id": "LA_kwDOC_bI7s8AAAABD5lA4w",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/cherry-pick",
"name": "cherry-pick",
"color": "ededed",
"default": false,
"description": null
},
{
"id": 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 | [
"jenkins"
] | 2023-12-21T07:05:17 | 2023-12-23T17:51:26 | 2023-12-23T17:51:26 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8073",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8073",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8073.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8073.patch",
"merged_at": "2023-12-23T17:51:26"
} | # What does this PR do ?
Update requirements
**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
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# Before your PR is "Ready for review"
**Pre checks**:
- [ ] Make sure you read and followed [Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md)
- [ ] Did you write any new necessary tests?
- [ ] Did you add or update any necessary documentation?
- [ ] Does the PR affect components that are optional to install? (Ex: Numba, Pynini, Apex etc)
- [ ] Reviewer: Does the PR have correct import guards for all optional libraries?
**PR Type**:
- [ ] New Feature
- [ ] Bugfix
- [ ] Documentation
If you haven't finished some of the above items you can still open "Draft" PR.
## Who can review?
Anyone in the community is free to review the PR once the checks have passed.
[Contributor guidelines](https://github.com/NVIDIA/NeMo/blob/main/CONTRIBUTING.md) contains specific people who can review PRs to various areas.
# Additional Information
* Related to # (issue)
| {
"login": "nithinraok",
"id": 19668129,
"node_id": "MDQ6VXNlcjE5NjY4MTI5",
"avatar_url": "https://avatars.githubusercontent.com/u/19668129?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/nithinraok",
"html_url": "https://github.com/nithinraok",
"followers_url": "https://api.github.com/users/nithinraok/followers",
"following_url": "https://api.github.com/users/nithinraok/following{/other_user}",
"gists_url": "https://api.github.com/users/nithinraok/gists{/gist_id}",
"starred_url": "https://api.github.com/users/nithinraok/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/nithinraok/subscriptions",
"organizations_url": "https://api.github.com/users/nithinraok/orgs",
"repos_url": "https://api.github.com/users/nithinraok/repos",
"events_url": "https://api.github.com/users/nithinraok/events{/privacy}",
"received_events_url": "https://api.github.com/users/nithinraok/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8073/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/8073/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8072 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8072/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8072/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8072/events | https://github.com/NVIDIA/NeMo/pull/8072 | 2,051,447,818 | PR_kwDOC_bI7s5ihQn2 | 8,072 | update reqs | {
"login": "ericharper",
"id": 11999610,
"node_id": "MDQ6VXNlcjExOTk5NjEw",
"avatar_url": "https://avatars.githubusercontent.com/u/11999610?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ericharper",
"html_url": "https://github.com/ericharper",
"followers_url": "https://api.github.com/users/ericharper/followers",
"following_url": "https://api.github.com/users/ericharper/following{/other_user}",
"gists_url": "https://api.github.com/users/ericharper/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ericharper/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ericharper/subscriptions",
"organizations_url": "https://api.github.com/users/ericharper/orgs",
"repos_url": "https://api.github.com/users/ericharper/repos",
"events_url": "https://api.github.com/users/ericharper/events{/privacy}",
"received_events_url": "https://api.github.com/users/ericharper/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 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 | [
"jenkins"
] | 2023-12-20T23:28:39 | 2023-12-21T07:04:53 | 2023-12-21T07:04:52 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8072",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8072",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8072.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8072.patch",
"merged_at": "2023-12-21T07:04:52"
} | # What does this PR do ?
Update requirements
**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
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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/8072/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/8072/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8071 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8071/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8071/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8071/events | https://github.com/NVIDIA/NeMo/pull/8071 | 2,051,430,423 | PR_kwDOC_bI7s5ihM21 | 8,071 | Update HF Transformers version | {
"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 | [
"jenkins"
] | 2023-12-20T23:02:57 | 2023-12-20T23:35:32 | 2023-12-20T23:33:59 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8071",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8071",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8071.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8071.patch",
"merged_at": null
} | # What does this PR do ?
Updates HF version
**Collection**: [Common]
# 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": "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/8071/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/8071/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8070 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8070/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8070/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8070/events | https://github.com/NVIDIA/NeMo/pull/8070 | 2,051,417,205 | PR_kwDOC_bI7s5ihJ3L | 8,070 | fix for missing mcore arg | {
"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
} | [] | closed | false | null | [] | null | [
"jenkins",
"Addressed in #8113 so closing this"
] | 2023-12-20T22:51:13 | 2024-01-02T22:52:26 | 2024-01-02T22:52:26 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8070",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8070",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8070.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8070.patch",
"merged_at": null
} | # What does this PR do ?
Mcore refactor introduced an error when trying to merge.
```
Loading distributed checkpoint with TensorStoreLoadShardedStrategy
[NeMo I 2023-12-13 12:13:12 nlp_overrides:752] Model MegatronGPTModel was successfully restored from /workspace/models/nemo_version/llama2-7b.nemo.
Error executing job with overrides: []
Traceback (most recent call last):
File "/workspace/NeMo/scripts/nlp_language_modeling/merge_lora_weights/merge.py", line 198, in main
lora_model_cfg = MegatronGPTLoRAModel.restore_from(
TypeError: NLPModel.restore_from() got an unexpected keyword argument 'mcore'
```
This PR is a patch for this error.
**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
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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": "cuichenx",
"id": 43478052,
"node_id": "MDQ6VXNlcjQzNDc4MDUy",
"avatar_url": "https://avatars.githubusercontent.com/u/43478052?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/cuichenx",
"html_url": "https://github.com/cuichenx",
"followers_url": "https://api.github.com/users/cuichenx/followers",
"following_url": "https://api.github.com/users/cuichenx/following{/other_user}",
"gists_url": "https://api.github.com/users/cuichenx/gists{/gist_id}",
"starred_url": "https://api.github.com/users/cuichenx/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/cuichenx/subscriptions",
"organizations_url": "https://api.github.com/users/cuichenx/orgs",
"repos_url": "https://api.github.com/users/cuichenx/repos",
"events_url": "https://api.github.com/users/cuichenx/events{/privacy}",
"received_events_url": "https://api.github.com/users/cuichenx/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8070/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/8070/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8069 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8069/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8069/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8069/events | https://github.com/NVIDIA/NeMo/pull/8069 | 2,051,371,861 | PR_kwDOC_bI7s5ig_eE | 8,069 | Asr bleu | {
"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
},
{
"id": 4811884691,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR",
"name": "ASR",
"color": "F3B33E",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"@titu1994 Draft PR to show BLEU calls. Everywhere that the model calls BLEU is where we normally call WER, so can be aliased. The `multi_validation_epoch_end` just needs to be pushed up to `asr_model` (it's an ugly check but is same logic as how it calls WER and can be model agnostic that way.) \r\n\r\nIf we do WER updates like BLEU updates here, all metrics now on can be updated in individual models agnostic of task. (Checked with Torch folks, it doesn't lead to any complications for tensor devices.)\r\n\r\n(Since it's a draft I just based it off wer_fix, so large amount of file changes are unimportant. Just `metrics.bleu` and `models.transformer_bpe`)",
"Hmm it's a bit hard to understand what are the changes here. Let's hold ofd on this until we merge the first one ",
"@titu1994 post wer draft is up. pinging you to check if there's any design/technical problems before I make it more comprehensive throughout.\r\n\r\nMain changes:\r\n- makes BLEU metric\r\n- bleu metric operates like wer metric throughout asr module, so no change needed beyond a flag in config\r\n- made wer and bleu give dict outputs to clean up asr models (will make more comprehensive for the rest of the asr collection metrics.) Checked with PyTorch channel and they say there should be no syncing problems. Running my own tests off this branch later in week to confirm.",
"Forgot about this, sorry. Will review next week ",
"> Forgot about this, sorry. Will review next week\r\n\r\nAll good, I can keep on backburner for a bit. Don't need a full on review, just want to catch if there's any design issues before I go more thoroughly through the codebase to unify metrics.",
"@titu1994 i'm gonna finish up work on this over the week to avoid canary conflicts. Did you have a chance to review before I have to tear up some more of the codebase?",
"This PR is stale because it has been open for 14 days with no activity. Remove stale label or comment or update or this will be closed in 7 days.",
"Closing since implemented in https://github.com/NVIDIA/NeMo/pull/8363"
] | 2023-12-20T22:21:39 | 2024-02-09T17:17:11 | 2024-02-09T17:17:10 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | true | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8069",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8069",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8069.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8069.patch",
"merged_at": null
} | # What does this PR do ?
- Adds BLEU metric for ASR collections.
**Collection**: [Note which collection this PR will affect]
ASR
# Changelog
- Adds BLEU metric for ASR collections
- Adds TransformerDecoding module to make transformer_bpe similar to other ASR modules
- Implements BLEU module in transformer_bpe for demonstration
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# Before your PR is "Ready for review"
**Pre checks**:
- [N ] 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": "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
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8069/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/8069/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8068 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8068/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8068/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8068/events | https://github.com/NVIDIA/NeMo/pull/8068 | 2,051,371,649 | PR_kwDOC_bI7s5ig_bA | 8,068 | Handle in HF AutoTokenizer with pad_token=None | {
"login": "HuiyingLi",
"id": 1331543,
"node_id": "MDQ6VXNlcjEzMzE1NDM=",
"avatar_url": "https://avatars.githubusercontent.com/u/1331543?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/HuiyingLi",
"html_url": "https://github.com/HuiyingLi",
"followers_url": "https://api.github.com/users/HuiyingLi/followers",
"following_url": "https://api.github.com/users/HuiyingLi/following{/other_user}",
"gists_url": "https://api.github.com/users/HuiyingLi/gists{/gist_id}",
"starred_url": "https://api.github.com/users/HuiyingLi/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/HuiyingLi/subscriptions",
"organizations_url": "https://api.github.com/users/HuiyingLi/orgs",
"repos_url": "https://api.github.com/users/HuiyingLi/repos",
"events_url": "https://api.github.com/users/HuiyingLi/events{/privacy}",
"received_events_url": "https://api.github.com/users/HuiyingLi/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4847373924,
"node_id": "LA_kwDOC_bI7s8AAAABIO0OZA",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/common",
"name": "common",
"color": "c5def5",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"This PR is stale because it has been open for 14 days with no activity. Remove stale label or comment or update or this will be closed in 7 days.",
"This PR was closed because it has been inactive for 7 days since being marked as stale.",
"Is it possible to reopen the pr?",
"jenkins",
"@yaoyu-33 Could you review the resolved issue and say jenkins again? Thanks!!",
"jenkins",
"jenkins",
"jenkins"
] | 2023-12-20T22:21:30 | 2024-02-27T17:50:53 | 2024-02-27T17:50:53 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8068",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8068",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8068.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8068.patch",
"merged_at": "2024-02-27T17:50:53"
} | # What does this PR do ?
In the NML data preprocessing script `scripts/nlp_language_modeling/preprocess_data_for_megatron.py`
the builder the pad_id thru:
`pad_id=tokenizer.pad_id if hasattr(tokenizer, "pad_id") else 0`
While HF wrapper AutoTokenizer class has `pad_id` as a property, but doesn't actually have pad_token, which will throw None error.
This pr checks if special token exists before encode to get ID of special token.
**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
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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": "yaoyu-33",
"id": 54727607,
"node_id": "MDQ6VXNlcjU0NzI3NjA3",
"avatar_url": "https://avatars.githubusercontent.com/u/54727607?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/yaoyu-33",
"html_url": "https://github.com/yaoyu-33",
"followers_url": "https://api.github.com/users/yaoyu-33/followers",
"following_url": "https://api.github.com/users/yaoyu-33/following{/other_user}",
"gists_url": "https://api.github.com/users/yaoyu-33/gists{/gist_id}",
"starred_url": "https://api.github.com/users/yaoyu-33/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/yaoyu-33/subscriptions",
"organizations_url": "https://api.github.com/users/yaoyu-33/orgs",
"repos_url": "https://api.github.com/users/yaoyu-33/repos",
"events_url": "https://api.github.com/users/yaoyu-33/events{/privacy}",
"received_events_url": "https://api.github.com/users/yaoyu-33/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8068/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/8068/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8067 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8067/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8067/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8067/events | https://github.com/NVIDIA/NeMo/pull/8067 | 2,051,269,501 | PR_kwDOC_bI7s5igoOm | 8,067 | [Fix] Increase length check tolerance to prevent test failing | {
"login": "anteju",
"id": 108555623,
"node_id": "U_kgDOBnhtZw",
"avatar_url": "https://avatars.githubusercontent.com/u/108555623?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/anteju",
"html_url": "https://github.com/anteju",
"followers_url": "https://api.github.com/users/anteju/followers",
"following_url": "https://api.github.com/users/anteju/following{/other_user}",
"gists_url": "https://api.github.com/users/anteju/gists{/gist_id}",
"starred_url": "https://api.github.com/users/anteju/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/anteju/subscriptions",
"organizations_url": "https://api.github.com/users/anteju/orgs",
"repos_url": "https://api.github.com/users/anteju/repos",
"events_url": "https://api.github.com/users/anteju/events{/privacy}",
"received_events_url": "https://api.github.com/users/anteju/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4811884691,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR",
"name": "ASR",
"color": "F3B33E",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [
"jenkins"
] | 2023-12-20T21:09:12 | 2023-12-20T23:36:39 | 2023-12-20T23:36:39 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8067",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8067",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8067.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8067.patch",
"merged_at": "2023-12-20T23:36:39"
} | # What does this PR do ?
This PR relaxes constrain on the length to prevent intermittent test failures.
**Collection**: ASR tests
# Changelog
- Increased threshold for 1 to 2
# Original test failure
* You can potentially add a usage example below
```python
=========================== short test summary info ============================
FAILED tests/collections/asr/decoding/test_rnnt_decoding.py::TestRNNTDecoding::test_beam_decoding_preserve_alignments[beam_config4] - AssertionError: assert tensor(2) <= 1
+ where tensor(2) = abs((31 - tensor(33)))
```
# Jenkins CI
To run Jenkins, a NeMo User with write access must comment `jenkins` on the PR.
# 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": "anteju",
"id": 108555623,
"node_id": "U_kgDOBnhtZw",
"avatar_url": "https://avatars.githubusercontent.com/u/108555623?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/anteju",
"html_url": "https://github.com/anteju",
"followers_url": "https://api.github.com/users/anteju/followers",
"following_url": "https://api.github.com/users/anteju/following{/other_user}",
"gists_url": "https://api.github.com/users/anteju/gists{/gist_id}",
"starred_url": "https://api.github.com/users/anteju/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/anteju/subscriptions",
"organizations_url": "https://api.github.com/users/anteju/orgs",
"repos_url": "https://api.github.com/users/anteju/repos",
"events_url": "https://api.github.com/users/anteju/events{/privacy}",
"received_events_url": "https://api.github.com/users/anteju/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/issues/8067/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/8067/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8066 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8066/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8066/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8066/events | https://github.com/NVIDIA/NeMo/pull/8066 | 2,051,212,340 | PR_kwDOC_bI7s5igbwn | 8,066 | Various fixes for typos and urls | {
"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
} | [
{
"id": 4418253970,
"node_id": "LA_kwDOC_bI7s8AAAABB1k0kg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/TTS",
"name": "TTS",
"color": "49AF28",
"default": false,
"description": ""
},
{
"id": 4811884691,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR",
"name": "ASR",
"color": "F3B33E",
"default": false,
"description": ""
},
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [] | 2023-12-20T20:24:48 | 2023-12-20T20:30:08 | 2023-12-20T20:30:08 | COLLABORATOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8066",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8066",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8066.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8066.patch",
"merged_at": "2023-12-20T20:30:07"
} | # What does this PR do ?
**Collection**: [ASR, NLP, TTS]
# Changelog
- See PR https://git
# 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": "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/8066/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/8066/timeline | null | null | true |
https://api.github.com/repos/NVIDIA/NeMo/issues/8065 | https://api.github.com/repos/NVIDIA/NeMo | https://api.github.com/repos/NVIDIA/NeMo/issues/8065/labels{/name} | https://api.github.com/repos/NVIDIA/NeMo/issues/8065/comments | https://api.github.com/repos/NVIDIA/NeMo/issues/8065/events | https://github.com/NVIDIA/NeMo/pull/8065 | 2,051,191,204 | PR_kwDOC_bI7s5igXSg | 8,065 | Fix various issues with broken links and bugs | {
"login": "github-actions[bot]",
"id": 41898282,
"node_id": "MDM6Qm90NDE4OTgyODI=",
"avatar_url": "https://avatars.githubusercontent.com/in/15368?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/github-actions%5Bbot%5D",
"html_url": "https://github.com/apps/github-actions",
"followers_url": "https://api.github.com/users/github-actions%5Bbot%5D/followers",
"following_url": "https://api.github.com/users/github-actions%5Bbot%5D/following{/other_user}",
"gists_url": "https://api.github.com/users/github-actions%5Bbot%5D/gists{/gist_id}",
"starred_url": "https://api.github.com/users/github-actions%5Bbot%5D/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/github-actions%5Bbot%5D/subscriptions",
"organizations_url": "https://api.github.com/users/github-actions%5Bbot%5D/orgs",
"repos_url": "https://api.github.com/users/github-actions%5Bbot%5D/repos",
"events_url": "https://api.github.com/users/github-actions%5Bbot%5D/events{/privacy}",
"received_events_url": "https://api.github.com/users/github-actions%5Bbot%5D/received_events",
"type": "Bot",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 4418253970,
"node_id": "LA_kwDOC_bI7s8AAAABB1k0kg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/TTS",
"name": "TTS",
"color": "49AF28",
"default": false,
"description": ""
},
{
"id": 4556669155,
"node_id": "LA_kwDOC_bI7s8AAAABD5lA4w",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/cherry-pick",
"name": "cherry-pick",
"color": "ededed",
"default": false,
"description": null
},
{
"id": 4811884691,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ikw",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/ASR",
"name": "ASR",
"color": "F3B33E",
"default": false,
"description": ""
},
{
"id": 4811886210,
"node_id": "LA_kwDOC_bI7s8AAAABHs-Ogg",
"url": "https://api.github.com/repos/NVIDIA/NeMo/labels/NLP",
"name": "NLP",
"color": "CBF09A",
"default": false,
"description": ""
}
] | closed | false | null | [] | null | [] | 2023-12-20T20:06:33 | 2023-12-20T20:21:31 | 2023-12-20T20:21:27 | CONTRIBUTOR | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | null | false | {
"url": "https://api.github.com/repos/NVIDIA/NeMo/pulls/8065",
"html_url": "https://github.com/NVIDIA/NeMo/pull/8065",
"diff_url": "https://github.com/NVIDIA/NeMo/pull/8065.diff",
"patch_url": "https://github.com/NVIDIA/NeMo/pull/8065.patch",
"merged_at": null
} | # What does this PR do ?
**Collection**: [ASR, NLP, TTS]
# Changelog
Fixes bugs and typos in
- https://nvbugspro.nvidia.com/bug/4413845
- https://nvbugspro.nvidia.com/bug/4411322
- https://nvbugspro.nvidia.com/bug/4410625
- https://nvbugspro.nvidia.com/bug/4407526
- https://nvbugspro.nvidia.com/bug/4405781
- https://nvbugspro.nvidia.com/bug/4426768
# 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": "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/8065/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/8065/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.