url stringlengths 62 66 | repository_url stringclasses 1
value | labels_url stringlengths 76 80 | comments_url stringlengths 71 75 | events_url stringlengths 69 73 | html_url stringlengths 50 56 | id int64 377M 2.15B | node_id stringlengths 18 32 | number int64 1 29.2k | title stringlengths 1 487 | user dict | labels list | state stringclasses 2
values | locked bool 2
classes | assignee dict | assignees list | comments list | created_at int64 1.54k 1.71k | updated_at int64 1.54k 1.71k | closed_at int64 1.54k 1.71k ⌀ | author_association stringclasses 4
values | active_lock_reason stringclasses 2
values | body stringlengths 0 234k ⌀ | reactions dict | timeline_url stringlengths 71 75 | state_reason stringclasses 3
values | draft bool 2
classes | pull_request dict |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
https://api.github.com/repos/huggingface/transformers/issues/9429 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9429/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9429/comments | https://api.github.com/repos/huggingface/transformers/issues/9429/events | https://github.com/huggingface/transformers/issues/9429 | 780,132,135 | MDU6SXNzdWU3ODAxMzIxMzU= | 9,429 | Apache Hadoop (HDFS) File Loading from_pretrained | {
"login": "anninterpreter",
"id": 44267622,
"node_id": "MDQ6VXNlcjQ0MjY3NjIy",
"avatar_url": "https://avatars.githubusercontent.com/u/44267622?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/anninterpreter",
"html_url": "https://github.com/anninterpreter",
"followers_url": "https://api.gi... | [] | closed | false | null | [] | [
"This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md)... | 1,609 | 1,619 | 1,619 | NONE | null | # 🚀 Feature request
Loading configuration file for **transformers.AutoConfig** and **transformers.AutoModelForSequenceClassification** using the function **from_pretrained** by giving the HDFS file path
## Motivation
In case of file that is not locally available, the library utilizes the **get_from_cache** func... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9429/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/huggingface/transformers/issues/9429/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9428 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9428/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9428/comments | https://api.github.com/repos/huggingface/transformers/issues/9428/events | https://github.com/huggingface/transformers/pull/9428 | 780,063,242 | MDExOlB1bGxSZXF1ZXN0NTQ5OTg1MDg4 | 9,428 | Improve documentation coverage for Herbert | {
"login": "Qbiwan",
"id": 69753975,
"node_id": "MDQ6VXNlcjY5NzUzOTc1",
"avatar_url": "https://avatars.githubusercontent.com/u/69753975?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Qbiwan",
"html_url": "https://github.com/Qbiwan",
"followers_url": "https://api.github.com/users/Qbiwan/fo... | [] | closed | false | null | [] | [
"Still not letting me assign you @sgugger :("
] | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | # What does this PR do?
<!--
Congratulations! You've made it this far! You're not quite done yet though.
Once merged, your PR is going to appear in the release notes with the title you set, so make sure it's a great title that fully reflects the extent of your awesome contribution.
Then, please replace this w... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9428/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9428/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9428",
"html_url": "https://github.com/huggingface/transformers/pull/9428",
"diff_url": "https://github.com/huggingface/transformers/pull/9428.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9428.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9427 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9427/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9427/comments | https://api.github.com/repos/huggingface/transformers/issues/9427/events | https://github.com/huggingface/transformers/pull/9427 | 779,957,558 | MDExOlB1bGxSZXF1ZXN0NTQ5ODg3MDUw | 9,427 | Improve documentation coverage for Phobert | {
"login": "Qbiwan",
"id": 69753975,
"node_id": "MDQ6VXNlcjY5NzUzOTc1",
"avatar_url": "https://avatars.githubusercontent.com/u/69753975?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Qbiwan",
"html_url": "https://github.com/Qbiwan",
"followers_url": "https://api.github.com/users/Qbiwan/fo... | [] | closed | false | null | [] | [
"Can't pin you for review @sgugger, so tagging you!"
] | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | # What does this PR do?
<!--
Congratulations! You've made it this far! You're not quite done yet though.
Once merged, your PR is going to appear in the release notes with the title you set, so make sure it's a great title that fully reflects the extent of your awesome contribution.
Then, please replace this w... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9427/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9427/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9427",
"html_url": "https://github.com/huggingface/transformers/pull/9427",
"diff_url": "https://github.com/huggingface/transformers/pull/9427.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9427.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9426 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9426/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9426/comments | https://api.github.com/repos/huggingface/transformers/issues/9426/events | https://github.com/huggingface/transformers/issues/9426 | 779,804,482 | MDU6SXNzdWU3Nzk4MDQ0ODI= | 9,426 | Is it possible to export a pytorch .pt file after finetuning a model? | {
"login": "farazk86",
"id": 33456896,
"node_id": "MDQ6VXNlcjMzNDU2ODk2",
"avatar_url": "https://avatars.githubusercontent.com/u/33456896?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/farazk86",
"html_url": "https://github.com/farazk86",
"followers_url": "https://api.github.com/users/far... | [] | closed | false | null | [] | [
"Just rename your `.bin` to `.pt`",
"Hi @julien-c \r\n\r\nSorry to come back to this but I am having trouble with this. As the .bin file also has a much larger accompanying ``optimizer`` file that I assume holds the weights.\r\n\r\nI am trying to deploy a fine tuned model to google cloud. And even when using a cu... | 1,609 | 1,610 | 1,609 | NONE | null | Hi,
Considering how I am having troubles with the tflite interpreter at issue https://github.com/huggingface/transformers/issues/9392 . I was wondering if I will have better luck trying Pytorch mobile, since the base models are pytorch to begin with.
But to use the pytorch converter I need the saved ``.pt`` file.... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9426/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9426/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9425 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9425/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9425/comments | https://api.github.com/repos/huggingface/transformers/issues/9425/events | https://github.com/huggingface/transformers/issues/9425 | 779,562,303 | MDU6SXNzdWU3Nzk1NjIzMDM= | 9,425 | [utils/get_modified_files.py] fails with a few PR checkout tools | {
"login": "stas00",
"id": 10676103,
"node_id": "MDQ6VXNlcjEwNjc2MTAz",
"avatar_url": "https://avatars.githubusercontent.com/u/10676103?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/stas00",
"html_url": "https://github.com/stas00",
"followers_url": "https://api.github.com/users/stas00/fo... | [] | closed | false | {
"login": "stas00",
"id": 10676103,
"node_id": "MDQ6VXNlcjEwNjc2MTAz",
"avatar_url": "https://avatars.githubusercontent.com/u/10676103?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/stas00",
"html_url": "https://github.com/stas00",
"followers_url": "https://api.github.com/users/stas00/fo... | [
{
"login": "stas00",
"id": 10676103,
"node_id": "MDQ6VXNlcjEwNjc2MTAz",
"avatar_url": "https://avatars.githubusercontent.com/u/10676103?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/stas00",
"html_url": "https://github.com/stas00",
"followers_url": "https://api.github... | [] | 1,609 | 1,610 | 1,610 | CONTRIBUTOR | null | I have noticed that when using [gh cli](https://github.com/cli/cli) to checkout a pr
```
git merge-base --fork-point master
```
fails, which breaks `utils/get_modified_files.py`
e.g.:
```
gh pr checkout 9423
python utils/get_modified_files.py
Traceback (most recent call last):
File "utils/get_modified_fil... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9425/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9425/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9424 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9424/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9424/comments | https://api.github.com/repos/huggingface/transformers/issues/9424/events | https://github.com/huggingface/transformers/pull/9424 | 779,494,475 | MDExOlB1bGxSZXF1ZXN0NTQ5NDU5NDA1 | 9,424 | improve readme text to private models/versioning/api | {
"login": "clmnt",
"id": 821155,
"node_id": "MDQ6VXNlcjgyMTE1NQ==",
"avatar_url": "https://avatars.githubusercontent.com/u/821155?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/clmnt",
"html_url": "https://github.com/clmnt",
"followers_url": "https://api.github.com/users/clmnt/followers"... | [] | closed | false | null | [] | [
"👍 "
] | 1,609 | 1,609 | 1,609 | MEMBER | null | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9424/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9424/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9424",
"html_url": "https://github.com/huggingface/transformers/pull/9424",
"diff_url": "https://github.com/huggingface/transformers/pull/9424.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9424.patch",
"merged_at": 1... | |
https://api.github.com/repos/huggingface/transformers/issues/9423 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9423/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9423/comments | https://api.github.com/repos/huggingface/transformers/issues/9423/events | https://github.com/huggingface/transformers/pull/9423 | 779,461,294 | MDExOlB1bGxSZXF1ZXN0NTQ5NDI5MDM2 | 9,423 | Upgrade styler to better handle lists | {
"login": "sgugger",
"id": 35901082,
"node_id": "MDQ6VXNlcjM1OTAxMDgy",
"avatar_url": "https://avatars.githubusercontent.com/u/35901082?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sgugger",
"html_url": "https://github.com/sgugger",
"followers_url": "https://api.github.com/users/sgugge... | [] | closed | false | null | [] | [
"I suppose I need to stick to the common bullet format, as it couldn't handle this `\\d\\) ` style of bullets. Leading to this rewrite:\r\n```\r\n-1) Optimizer State Partitioning (stage 1)\r\n-2) Add Gradient Partitioning (stage 2)\r\n+1) Optimizer State Partitioning (stage 1) 2) Add Gradient Partitioning (stage 2)... | 1,609 | 1,610 | 1,609 | COLLABORATOR | null | # What does this PR do?
This PR upgrades the doc styling script to automatically add new lines before lists. This makes the script more robust as it will avoid reformatting those lists and make them appear properly once sphinx has done its thing.
In passing a few badly formatted docstrings/doc pages are fixed, ju... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9423/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9423/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9423",
"html_url": "https://github.com/huggingface/transformers/pull/9423",
"diff_url": "https://github.com/huggingface/transformers/pull/9423.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9423.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9422 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9422/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9422/comments | https://api.github.com/repos/huggingface/transformers/issues/9422/events | https://github.com/huggingface/transformers/issues/9422 | 779,286,834 | MDU6SXNzdWU3NzkyODY4MzQ= | 9,422 | [Announcement] Changing model type of Barthez | {
"login": "patrickvonplaten",
"id": 23423619,
"node_id": "MDQ6VXNlcjIzNDIzNjE5",
"avatar_url": "https://avatars.githubusercontent.com/u/23423619?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/patrickvonplaten",
"html_url": "https://github.com/patrickvonplaten",
"followers_url": "https://... | [] | closed | false | null | [] | [
"Applied the change",
"Hi @patrickvonplaten. Sorry for the late reply.\r\n\r\nActually I tested the model with `BartForConditionalGeneration` and everything was working well. On the other hand, after the modification I am getting the following error: \r\n```\r\nUnrecognized configuration class for this kind of Au... | 1,609 | 1,618 | 1,618 | MEMBER | null | We are currently undergoing some major refactoring of Bart-like models as shown in: https://github.com/huggingface/transformers/pull/9343.
After the refactoring, the Barthez models would not work anymore with the `AutoModel` and `AutoModelForSeq2SeqLM` classes because Barthez actually corresponds more to the mbart ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9422/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9422/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9421 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9421/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9421/comments | https://api.github.com/repos/huggingface/transformers/issues/9421/events | https://github.com/huggingface/transformers/pull/9421 | 779,268,737 | MDExOlB1bGxSZXF1ZXN0NTQ5MjU0OTE2 | 9,421 | Store transformers version info when saving the model | {
"login": "JetRunner",
"id": 22514219,
"node_id": "MDQ6VXNlcjIyNTE0MjE5",
"avatar_url": "https://avatars.githubusercontent.com/u/22514219?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/JetRunner",
"html_url": "https://github.com/JetRunner",
"followers_url": "https://api.github.com/users/... | [] | closed | false | null | [] | [] | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | # What does this PR do?
<!--
Congratulations! You've made it this far! You're not quite done yet though.
Once merged, your PR is going to appear in the release notes with the title you set, so make sure it's a great title that fully reflects the extent of your awesome contribution.
Then, please replace this w... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9421/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9421/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9421",
"html_url": "https://github.com/huggingface/transformers/pull/9421",
"diff_url": "https://github.com/huggingface/transformers/pull/9421.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9421.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9420 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9420/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9420/comments | https://api.github.com/repos/huggingface/transformers/issues/9420/events | https://github.com/huggingface/transformers/issues/9420 | 779,245,705 | MDU6SXNzdWU3NzkyNDU3MDU= | 9,420 | Transformer models for semantic parsing | {
"login": "ayushjain1144",
"id": 28894174,
"node_id": "MDQ6VXNlcjI4ODk0MTc0",
"avatar_url": "https://avatars.githubusercontent.com/u/28894174?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ayushjain1144",
"html_url": "https://github.com/ayushjain1144",
"followers_url": "https://api.githu... | [] | closed | false | null | [] | [
"Hi @ayushjain1144 \r\n\r\nThat's an interesting question, would be better if you ask it on the [forum](https://discuss.huggingface.co/) "
] | 1,609 | 1,609 | 1,609 | NONE | null | Hi! Thank you for your awesome work!
I want to perform semantic parsing. Unfortunately, I couldn't find any examples on hugging face repo for that. Could you please let me know how I should proceed? I suppose I could use a Seq2Seq EncoderDecoder model like BERT2BERT and finetune it for semantic parsing. Or do you th... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9420/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9420/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9419 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9419/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9419/comments | https://api.github.com/repos/huggingface/transformers/issues/9419/events | https://github.com/huggingface/transformers/pull/9419 | 779,221,520 | MDExOlB1bGxSZXF1ZXN0NTQ5MjEyNDcy | 9,419 | New serving | {
"login": "jplu",
"id": 959590,
"node_id": "MDQ6VXNlcjk1OTU5MA==",
"avatar_url": "https://avatars.githubusercontent.com/u/959590?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jplu",
"html_url": "https://github.com/jplu",
"followers_url": "https://api.github.com/users/jplu/followers",
... | [] | closed | false | null | [] | [
"Humm looks like the quick test I have added to make sure that a saved model can be properly created is a bit too long for at least one of the models. @LysandreJik any idea how I can figure out for which model?",
"Ok, the test `test_saved_model_creation` is skipped if it needs more than 30sec to be executed for a... | 1,609 | 1,610 | 1,610 | CONTRIBUTOR | null | # What does this PR do?
This PR proposes a new way to create a saved model to be properly served via TF Serving. The logic behind is to create a `serving` that will be used to create the expected saved model with a proper input signature. Currently the saved models are very limited:
- input sequence length limite... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9419/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/huggingface/transformers/issues/9419/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9419",
"html_url": "https://github.com/huggingface/transformers/pull/9419",
"diff_url": "https://github.com/huggingface/transformers/pull/9419.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9419.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9418 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9418/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9418/comments | https://api.github.com/repos/huggingface/transformers/issues/9418/events | https://github.com/huggingface/transformers/pull/9418 | 778,892,447 | MDExOlB1bGxSZXF1ZXN0NTQ4OTE1NDY1 | 9,418 | New TF embeddings (cleaner and faster) | {
"login": "jplu",
"id": 959590,
"node_id": "MDQ6VXNlcjk1OTU5MA==",
"avatar_url": "https://avatars.githubusercontent.com/u/959590?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jplu",
"html_url": "https://github.com/jplu",
"followers_url": "https://api.github.com/users/jplu/followers",
... | [] | closed | false | null | [] | [
"I like this PR in general! \r\n\r\nJust wondering about two things:\r\n\r\n1) Do we need this `get_config` function?\r\n2) Not a huge fan of the `Add()` keras layer...does this really improve performance much?",
"Good point @LysandreJik! Basically here most of the models share the similar embedding computation t... | 1,609 | 1,611 | 1,611 | CONTRIBUTOR | null | # What does this PR do?
This PR propose a better implementation of the embedding layer for the BERT-Like TF models. Another benefit of this cleaning is a better computational performance:
```
model = TFBertForMaskedLM.from_pretrained("bert-base-cased")
cProfile.run("model(model.dummy_inputs)")
# current master... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9418/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9418/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9418",
"html_url": "https://github.com/huggingface/transformers/pull/9418",
"diff_url": "https://github.com/huggingface/transformers/pull/9418.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9418.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9417 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9417/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9417/comments | https://api.github.com/repos/huggingface/transformers/issues/9417/events | https://github.com/huggingface/transformers/issues/9417 | 778,812,412 | MDU6SXNzdWU3Nzg4MTI0MTI= | 9,417 | shift_tokens_right in BART, FSMT incompatible with DataCollatorForLanguageModelling | {
"login": "jethrokuan",
"id": 1667473,
"node_id": "MDQ6VXNlcjE2Njc0NzM=",
"avatar_url": "https://avatars.githubusercontent.com/u/1667473?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jethrokuan",
"html_url": "https://github.com/jethrokuan",
"followers_url": "https://api.github.com/users... | [] | closed | false | null | [] | [
"Hey @jethrokuan, we've merged a big BART PR yesterday just as a heads up, I think this might solve this problem for Bart -> could you check again?",
"@patrickvonplaten I think it does solve this problem: code looks good and runs fine, model results not great, but possibly a mistake of mine. Thanks!"
] | 1,609 | 1,610 | 1,610 | CONTRIBUTOR | null | ## Environment info
- `transformers` version: 4.2.0dev0
- Platform: Linux-4.14.81.bm.15-amd64-x86_64-with-debian-9.11
- Python version: 3.7.9
- PyTorch version (GPU?): 1.7.1 (False)
- Tensorflow version (GPU?): not installed (NA)
- Using GPU in script?: No
- Using distributed or parallel set-up in script?: No
... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9417/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9417/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9416 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9416/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9416/comments | https://api.github.com/repos/huggingface/transformers/issues/9416/events | https://github.com/huggingface/transformers/issues/9416 | 778,773,835 | MDU6SXNzdWU3Nzg3NzM4MzU= | 9,416 | Why was DataCollatorForNextSentencePrediction removed ? | {
"login": "dwarfer7634",
"id": 19330059,
"node_id": "MDQ6VXNlcjE5MzMwMDU5",
"avatar_url": "https://avatars.githubusercontent.com/u/19330059?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/dwarfer7634",
"html_url": "https://github.com/dwarfer7634",
"followers_url": "https://api.github.com/... | [] | closed | false | {
"login": "sgugger",
"id": 35901082,
"node_id": "MDQ6VXNlcjM1OTAxMDgy",
"avatar_url": "https://avatars.githubusercontent.com/u/35901082?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sgugger",
"html_url": "https://github.com/sgugger",
"followers_url": "https://api.github.com/users/sgugge... | [
{
"login": "sgugger",
"id": 35901082,
"node_id": "MDQ6VXNlcjM1OTAxMDgy",
"avatar_url": "https://avatars.githubusercontent.com/u/35901082?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sgugger",
"html_url": "https://github.com/sgugger",
"followers_url": "https://api.git... | [
"This class is not necessary anymore: it was the same as `DataCollatorForLanguageModeling` while keeping the `nsp_labels` but `DataCollatorForLanguageModeling` will keep any extra things (like `nsp_labels`) you pass to it. So you can just replace it with `DataCollatorForLanguageModeling`.",
"Thank you for your q... | 1,609 | 1,609 | 1,609 | NONE | null | # 🚀 Feature request
I want to ask a question about why DataCollatorForNextSentencePrediction was removed.
That class was implement in the pull request down below.
https://github.com/huggingface/transformers/pull/6572
It was so useful for me.
But, this feature is not included in the latest version.
Do anyone kn... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9416/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/huggingface/transformers/issues/9416/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9415 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9415/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9415/comments | https://api.github.com/repos/huggingface/transformers/issues/9415/events | https://github.com/huggingface/transformers/issues/9415 | 778,598,584 | MDU6SXNzdWU3Nzg1OTg1ODQ= | 9,415 | About Multi GPU | {
"login": "HyeyeonKoo",
"id": 43692697,
"node_id": "MDQ6VXNlcjQzNjkyNjk3",
"avatar_url": "https://avatars.githubusercontent.com/u/43692697?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/HyeyeonKoo",
"html_url": "https://github.com/HyeyeonKoo",
"followers_url": "https://api.github.com/use... | [] | closed | false | null | [] | [
"Hi there, those kinds of questions should be asked on the [forums](https://discuss.huggingface.co/) as we keep issues for bugs and feature requests only, so closing this.\r\n\r\nFor a quick answer (move the discussion to the forums for a longer discussion!), it's normal that each GPU takes a slightly different tim... | 1,609 | 1,609 | 1,609 | NONE | null | ## Environment info
- `transformers` version: 3.5.0
- Platform: Linux-3.10.0-514.el7.x86_64-x86_64-with-centos-7.3.1611-Core
- Python version: 3.6.4
- PyTorch version (GPU?): 1.7.0
- Using GPU in script?: Y
- Using distributed or parallel set-up in script?: Y
### Who can help
@LysandreJik, @sgugger
## In... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9415/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9415/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9414 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9414/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9414/comments | https://api.github.com/repos/huggingface/transformers/issues/9414/events | https://github.com/huggingface/transformers/pull/9414 | 778,596,629 | MDExOlB1bGxSZXF1ZXN0NTQ4NjU0OTk1 | 9,414 | Fix link to Evaluate TAPAS Notebook | {
"login": "mrm8488",
"id": 3653789,
"node_id": "MDQ6VXNlcjM2NTM3ODk=",
"avatar_url": "https://avatars.githubusercontent.com/u/3653789?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/mrm8488",
"html_url": "https://github.com/mrm8488",
"followers_url": "https://api.github.com/users/mrm8488/... | [] | closed | false | null | [] | [] | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | # What does this PR do?
<!--
Congratulations! You've made it this far! You're not quite done yet though.
Once merged, your PR is going to appear in the release notes with the title you set, so make sure it's a great title that fully reflects the extent of your awesome contribution.
Then, please replace this w... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9414/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9414/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9414",
"html_url": "https://github.com/huggingface/transformers/pull/9414",
"diff_url": "https://github.com/huggingface/transformers/pull/9414.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9414.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9413 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9413/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9413/comments | https://api.github.com/repos/huggingface/transformers/issues/9413/events | https://github.com/huggingface/transformers/pull/9413 | 778,595,312 | MDExOlB1bGxSZXF1ZXN0NTQ4NjUzOTAy | 9,413 | Fix link to Notebook to fine-tune TAPAS | {
"login": "mrm8488",
"id": 3653789,
"node_id": "MDQ6VXNlcjM2NTM3ODk=",
"avatar_url": "https://avatars.githubusercontent.com/u/3653789?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/mrm8488",
"html_url": "https://github.com/mrm8488",
"followers_url": "https://api.github.com/users/mrm8488/... | [] | closed | false | null | [] | [] | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | # What does this PR do?
<!--
Congratulations! You've made it this far! You're not quite done yet though.
Once merged, your PR is going to appear in the release notes with the title you set, so make sure it's a great title that fully reflects the extent of your awesome contribution.
Then, please replace this w... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9413/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9413/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9413",
"html_url": "https://github.com/huggingface/transformers/pull/9413",
"diff_url": "https://github.com/huggingface/transformers/pull/9413.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9413.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9412 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9412/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9412/comments | https://api.github.com/repos/huggingface/transformers/issues/9412/events | https://github.com/huggingface/transformers/pull/9412 | 778,574,153 | MDExOlB1bGxSZXF1ZXN0NTQ4NjM1ODEy | 9,412 | [model parallel] add experimental warning | {
"login": "stas00",
"id": 10676103,
"node_id": "MDQ6VXNlcjEwNjc2MTAz",
"avatar_url": "https://avatars.githubusercontent.com/u/10676103?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/stas00",
"html_url": "https://github.com/stas00",
"followers_url": "https://api.github.com/users/stas00/fo... | [] | closed | false | null | [] | [
"Thanks!"
] | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | This PR documents that model parallelism is experimental and can change at any moment, so that we are not committing to any APIs until we sorted this out and it appears to be stable.
This in particular applies to the device map which is far from being sorted out.
@sgugger | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9412/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9412/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9412",
"html_url": "https://github.com/huggingface/transformers/pull/9412",
"diff_url": "https://github.com/huggingface/transformers/pull/9412.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9412.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9411 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9411/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9411/comments | https://api.github.com/repos/huggingface/transformers/issues/9411/events | https://github.com/huggingface/transformers/pull/9411 | 778,540,542 | MDExOlB1bGxSZXF1ZXN0NTQ4NjA4MDMw | 9,411 | [examples/text-classification] Fix a bug for using own regression dataset | {
"login": "forest1988",
"id": 2755894,
"node_id": "MDQ6VXNlcjI3NTU4OTQ=",
"avatar_url": "https://avatars.githubusercontent.com/u/2755894?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/forest1988",
"html_url": "https://github.com/forest1988",
"followers_url": "https://api.github.com/users... | [] | closed | false | null | [] | [
"@sgugger \r\nThank you for checking and merging this PR!"
] | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | # What does this PR do?
This PR is to fix https://github.com/huggingface/transformers/issues/9393
Fix a bug in `run_glue.py` so that it can be used for our own dataset of regression tasks.
close #9393
## Before submitting
- [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9411/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9411/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9411",
"html_url": "https://github.com/huggingface/transformers/pull/9411",
"diff_url": "https://github.com/huggingface/transformers/pull/9411.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9411.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9410 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9410/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9410/comments | https://api.github.com/repos/huggingface/transformers/issues/9410/events | https://github.com/huggingface/transformers/issues/9410 | 778,508,425 | MDU6SXNzdWU3Nzg1MDg0MjU= | 9,410 | `pip install -e .[dev]` in Python 3.9.1+ fails because `jaxlib==0.1.55` cannot be found | {
"login": "forest1988",
"id": 2755894,
"node_id": "MDQ6VXNlcjI3NTU4OTQ=",
"avatar_url": "https://avatars.githubusercontent.com/u/2755894?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/forest1988",
"html_url": "https://github.com/forest1988",
"followers_url": "https://api.github.com/users... | [] | closed | false | null | [] | [
"I retried to install `transformers [dev]` with Python 3.9.1. \r\nThe latest `git tag` in the cloned repository is `v4.2.1`.\r\n\r\nI assumed that the same error would occur, but in this time it failed in installing `tensorflow`. \r\n\r\n``` sh\r\n****@**** $ conda create -n transformers-py39-dev\r\nCollecting pack... | 1,609 | 1,617 | 1,610 | CONTRIBUTOR | null | ## Environment info
- `transformers` version: 4.2.0dev0 (the error is during the installation)
- Platform: Linux-4.15.0-123-generic-x86_64-with-glibc2.10
- Python version: 3.9.1 (the error occurs) -> 3.8.0 (the error does not occur)
- PyTorch version (GPU?): 1.7.1 (True)
- Tensorflow version (GPU?): 2.4.0 (... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9410/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/huggingface/transformers/issues/9410/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9409 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9409/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9409/comments | https://api.github.com/repos/huggingface/transformers/issues/9409/events | https://github.com/huggingface/transformers/pull/9409 | 778,491,047 | MDExOlB1bGxSZXF1ZXN0NTQ4NTY5MTI5 | 9,409 | [trainer] group fp16 args together | {
"login": "stas00",
"id": 10676103,
"node_id": "MDQ6VXNlcjEwNjc2MTAz",
"avatar_url": "https://avatars.githubusercontent.com/u/10676103?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/stas00",
"html_url": "https://github.com/stas00",
"followers_url": "https://api.github.com/users/stas00/fo... | [] | closed | false | null | [] | [
"Thanks!"
] | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | this PR proposes a purely cosmetic change that puts all the fp16 args together - so they are easier to manage/read
@sgugger
| {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9409/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9409/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9409",
"html_url": "https://github.com/huggingface/transformers/pull/9409",
"diff_url": "https://github.com/huggingface/transformers/pull/9409.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9409.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9408 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9408/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9408/comments | https://api.github.com/repos/huggingface/transformers/issues/9408/events | https://github.com/huggingface/transformers/issues/9408 | 778,452,777 | MDU6SXNzdWU3Nzg0NTI3Nzc= | 9,408 | [autoformatters] wrapping destroying items/lists | {
"login": "stas00",
"id": 10676103,
"node_id": "MDQ6VXNlcjEwNjc2MTAz",
"avatar_url": "https://avatars.githubusercontent.com/u/10676103?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/stas00",
"html_url": "https://github.com/stas00",
"followers_url": "https://api.github.com/users/stas00/fo... | [] | closed | false | null | [] | [
"The styling script does indeed break your list in this instance, which is kind of a bug that is a feature instead. Let me explain. If you use sphinx to convert this docstring to HTML, here is the result that it will produce:\r\n![image](https://user-images.githubusercontent.com/35901082/103658226-ef827800-4f38-11e... | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | # 🚀 Feature request
Would it be possible to make the auto-wrappers respect items/lists?
e.g. ended up with:
```
"""
number of training steps is either 1. args.max_steps if --max_steps > 1 2. else derive from dataset if we can
get its size
"""
```
not only it's broken, it's ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9408/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9408/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9407 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9407/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9407/comments | https://api.github.com/repos/huggingface/transformers/issues/9407/events | https://github.com/huggingface/transformers/pull/9407 | 778,419,886 | MDExOlB1bGxSZXF1ZXN0NTQ4NTEwNjk0 | 9,407 | Allow example to use a revision and work with private models | {
"login": "sgugger",
"id": 35901082,
"node_id": "MDQ6VXNlcjM1OTAxMDgy",
"avatar_url": "https://avatars.githubusercontent.com/u/35901082?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sgugger",
"html_url": "https://github.com/sgugger",
"followers_url": "https://api.github.com/users/sgugge... | [] | closed | false | null | [] | [
"+1 on the `model_revision` part\r\n\r\nOn the `use_auth_token`, I was thinking we could try to implement auto-use of the token _iff_ the model is not public (i.e. send the token for non-existent models and private models, as the server doesn't make a difference for those two cases when unauthorized – you get a 404... | 1,609 | 1,609 | 1,609 | COLLABORATOR | null | # What does this PR do?
This PR adds the ability to:
- pick a particular revision for a model checkpoint
- use private models when the user is logged in
with the example scripts.
Just did `run_glue` as a proof of concept, will duplicate to all examples and the new example template if this suits everyone.
| {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9407/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9407/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9407",
"html_url": "https://github.com/huggingface/transformers/pull/9407",
"diff_url": "https://github.com/huggingface/transformers/pull/9407.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9407.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9406 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9406/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9406/comments | https://api.github.com/repos/huggingface/transformers/issues/9406/events | https://github.com/huggingface/transformers/issues/9406 | 778,378,201 | MDU6SXNzdWU3NzgzNzgyMDE= | 9,406 | Unable to train xlnet with tensorflow | {
"login": "nicolas-ferland",
"id": 76973180,
"node_id": "MDQ6VXNlcjc2OTczMTgw",
"avatar_url": "https://avatars.githubusercontent.com/u/76973180?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/nicolas-ferland",
"html_url": "https://github.com/nicolas-ferland",
"followers_url": "https://api... | [] | closed | false | null | [] | [
"Hello! Can you try with master instead of the old `2.0.0` release? In order to know if the problem is still here or not.",
"By \"with master\", do you mean installed from source?\r\n\r\ngit clone https://github.com/huggingface/transformers.git\r\ncd transformers\r\npip install -e .",
"After installing from sou... | 1,609 | 1,619 | 1,619 | NONE | null | ## Environment info
- `transformers` version: '2.0.0'
- Platform: jupyter notebook
- Python version: 3.7.6
- PyTorch version (GPU?):
- Tensorflow version (GPU?): 2.1.0 GPU
- Using GPU in script?: Yes
- Using distributed or parallel set-up in script?: No
### Who can help
@sgugger, @TevenLeScao, @jplu
## In... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9406/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9406/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9405 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9405/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9405/comments | https://api.github.com/repos/huggingface/transformers/issues/9405/events | https://github.com/huggingface/transformers/issues/9405 | 778,370,772 | MDU6SXNzdWU3NzgzNzA3NzI= | 9,405 | Retrieval Collapse when fine-tuning RAG | {
"login": "JamesDeAntonis",
"id": 33379057,
"node_id": "MDQ6VXNlcjMzMzc5MDU3",
"avatar_url": "https://avatars.githubusercontent.com/u/33379057?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/JamesDeAntonis",
"html_url": "https://github.com/JamesDeAntonis",
"followers_url": "https://api.gi... | [] | closed | false | null | [] | [
"Maybe @patrickvonplaten or @lhoestq can chime in here.",
"It may very well be that this feature is not implemented. @ola13, @lhoestq do you have more insight here maybe?",
"The gradient does propagate to the question encoder. What configuration did you use ?",
"We used the default configuration and 'compres... | 1,609 | 1,682 | 1,620 | CONTRIBUTOR | null | ## Environment info
<!-- You can run the command `transformers-cli env` and copy-and-paste its output below.
Don't forget to fill out the missing fields in that output! -->
- `transformers` version: latest production version
- Platform:
- Python version: 3.8
- PyTorch version (GPU?): 1.7
- Tensorflow... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9405/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9405/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9404 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9404/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9404/comments | https://api.github.com/repos/huggingface/transformers/issues/9404/events | https://github.com/huggingface/transformers/pull/9404 | 778,341,957 | MDExOlB1bGxSZXF1ZXN0NTQ4NDQ1NDA4 | 9,404 | Add head_mask/decoder_head_mask for BART | {
"login": "stancld",
"id": 46073029,
"node_id": "MDQ6VXNlcjQ2MDczMDI5",
"avatar_url": "https://avatars.githubusercontent.com/u/46073029?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/stancld",
"html_url": "https://github.com/stancld",
"followers_url": "https://api.github.com/users/stancl... | [] | closed | false | null | [] | [
"Dear @patrickvonplaten and the rest of HuggingFace group.\r\n\r\nI implemented the concept of `head_mask` from BERT into BART so that the internal of decoder-encoder-like models can be studied as well. However, as this is my very first attempt to contribute to such a large-scale open-source project, I have been a ... | 1,609 | 1,610 | 1,610 | CONTRIBUTOR | null | Description:
This PR adds `head_mask` and `decoder_head_mask` for BART PyTorch implementation according to BERT implementation.
Motivation:
According to HuggingFace's websites "There is a growing field of study concerned with investigating the inner working of large-scale transformers like BERT (that some call “BE... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9404/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/huggingface/transformers/issues/9404/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9404",
"html_url": "https://github.com/huggingface/transformers/pull/9404",
"diff_url": "https://github.com/huggingface/transformers/pull/9404.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9404.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/9403 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9403/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9403/comments | https://api.github.com/repos/huggingface/transformers/issues/9403/events | https://github.com/huggingface/transformers/pull/9403 | 778,336,461 | MDExOlB1bGxSZXF1ZXN0NTQ4NDQwNjM5 | 9,403 | added head_mask/decoder_head_mask for BART | {
"login": "stancld",
"id": 46073029,
"node_id": "MDQ6VXNlcjQ2MDczMDI5",
"avatar_url": "https://avatars.githubusercontent.com/u/46073029?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/stancld",
"html_url": "https://github.com/stancld",
"followers_url": "https://api.github.com/users/stancl... | [] | closed | false | null | [] | [] | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | Description:
This PR adds `head_mask` and `decoder_head_mask` for BART PyTorch implementation according to BERT implementation.
Motivation:
According to HuggingFace's websites "There is a growing field of study concerned with investigating the inner working of large-scale transformers like BERT (that some call “BE... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9403/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9403/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9403",
"html_url": "https://github.com/huggingface/transformers/pull/9403",
"diff_url": "https://github.com/huggingface/transformers/pull/9403.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9403.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/9402 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9402/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9402/comments | https://api.github.com/repos/huggingface/transformers/issues/9402/events | https://github.com/huggingface/transformers/pull/9402 | 778,154,833 | MDExOlB1bGxSZXF1ZXN0NTQ4Mjk0MjU2 | 9,402 | Bump notebook from 6.1.4 to 6.1.5 in /examples/research_projects/lxmert | {
"login": "dependabot[bot]",
"id": 49699333,
"node_id": "MDM6Qm90NDk2OTkzMzM=",
"avatar_url": "https://avatars.githubusercontent.com/in/29110?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/dependabot%5Bbot%5D",
"html_url": "https://github.com/apps/dependabot",
"followers_url": "https://a... | [
{
"id": 1905493434,
"node_id": "MDU6TGFiZWwxOTA1NDkzNDM0",
"url": "https://api.github.com/repos/huggingface/transformers/labels/dependencies",
"name": "dependencies",
"color": "0366d6",
"default": false,
"description": "Pull requests that update a dependency file"
}
] | closed | false | null | [] | [
"Thanks dependabot!"
] | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | Bumps [notebook](https://github.com/jupyter/jupyterhub) from 6.1.4 to 6.1.5.
<details>
<summary>Commits</summary>
<ul>
<li>See full diff in <a href="https://github.com/jupyter/jupyterhub/commits">compare view</a></li>
</ul>
</details>
<br />
[![Dependabot compatibility score](https://dependabot-badges.githubapp.com/b... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9402/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9402/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9402",
"html_url": "https://github.com/huggingface/transformers/pull/9402",
"diff_url": "https://github.com/huggingface/transformers/pull/9402.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9402.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9401 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9401/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9401/comments | https://api.github.com/repos/huggingface/transformers/issues/9401/events | https://github.com/huggingface/transformers/pull/9401 | 778,133,316 | MDExOlB1bGxSZXF1ZXN0NTQ4Mjc3MDE4 | 9,401 | Put back LXMert example | {
"login": "sgugger",
"id": 35901082,
"node_id": "MDQ6VXNlcjM1OTAxMDgy",
"avatar_url": "https://avatars.githubusercontent.com/u/35901082?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sgugger",
"html_url": "https://github.com/sgugger",
"followers_url": "https://api.github.com/users/sgugge... | [] | closed | false | null | [] | [
"Anyway, I doubt Lxmert is currently not supported for customized/ personal datasets (nlp+images, images are harder to prepare). See issues relevant to feature extraction in [Lxmert](https://github.com/airsplay/lxmert#faster-r-cnn-feature-extraction), for example, [issue#79](https://github.com/airsplay/lxmert/issue... | 1,609 | 1,609 | 1,609 | COLLABORATOR | null | # What does this PR do?
During the example reorganization, LXMert seems to have slipped into the cracks and got accidentally deleted. This PR puts it back.
Fixes #9309 | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9401/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9401/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9401",
"html_url": "https://github.com/huggingface/transformers/pull/9401",
"diff_url": "https://github.com/huggingface/transformers/pull/9401.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9401.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9400 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9400/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9400/comments | https://api.github.com/repos/huggingface/transformers/issues/9400/events | https://github.com/huggingface/transformers/issues/9400 | 778,098,563 | MDU6SXNzdWU3NzgwOTg1NjM= | 9,400 | Generate Function - Manual decoder_input_ids Error (Bart, Pegasus) | {
"login": "marcoabrate",
"id": 43387597,
"node_id": "MDQ6VXNlcjQzMzg3NTk3",
"avatar_url": "https://avatars.githubusercontent.com/u/43387597?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/marcoabrate",
"html_url": "https://github.com/marcoabrate",
"followers_url": "https://api.github.com/... | [] | closed | false | null | [] | [] | 1,609 | 1,610 | 1,610 | NONE | null | ## Environment info
<!-- You can run the command `transformers-cli env` and copy-and-paste its output below.
Don't forget to fill out the missing fields in that output! -->
- `transformers` version: 4.1.1
- Platform: Google Colab
- Python version: 3.6.9
### Who can help
@patrickvonplaten
## T... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9400/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9400/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9399 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9399/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9399/comments | https://api.github.com/repos/huggingface/transformers/issues/9399/events | https://github.com/huggingface/transformers/issues/9399 | 778,094,416 | MDU6SXNzdWU3NzgwOTQ0MTY= | 9,399 | How to use Longformer for summarization | {
"login": "chetanambi",
"id": 37707687,
"node_id": "MDQ6VXNlcjM3NzA3Njg3",
"avatar_url": "https://avatars.githubusercontent.com/u/37707687?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/chetanambi",
"html_url": "https://github.com/chetanambi",
"followers_url": "https://api.github.com/use... | [] | closed | false | null | [] | [
"Perhaps this will help: https://huggingface.co/patrickvonplaten/longformer2roberta-cnn_dailymail-fp16",
"`LongformerEncoderDecoder` will be added to the lib once #9278 is merged! It can be used for summarization or any other seq2seq task.",
"@patil-suraj, @christianversloot I tried this model [longformer2rober... | 1,609 | 1,619 | 1,619 | NONE | null | Hi - Do you have a sample code on how to use Longformer for summarization tasks. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9399/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9399/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9398 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9398/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9398/comments | https://api.github.com/repos/huggingface/transformers/issues/9398/events | https://github.com/huggingface/transformers/issues/9398 | 778,085,869 | MDU6SXNzdWU3NzgwODU4Njk= | 9,398 | trainer.predict() returns different values from model.logits | {
"login": "connectlym",
"id": 25912288,
"node_id": "MDQ6VXNlcjI1OTEyMjg4",
"avatar_url": "https://avatars.githubusercontent.com/u/25912288?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/connectlym",
"html_url": "https://github.com/connectlym",
"followers_url": "https://api.github.com/use... | [] | closed | false | null | [] | [
"Hi,\r\nI have the same problem different results using `model()` vs `trainer.predict()`.",
"> Hi,\r\n> I have the same problem different results using `model()` vs `trainer.predict()`.\r\n\r\nThanks for replying. I assumed this feature was made for other usages so I ended up using `model()`. Anyway, I would like... | 1,609 | 1,618 | 1,610 | NONE | null | Hi dear authors!
When I was using my **fine-tuned bert model** to do the sequence classification task, I found the values returned by `trainer.predict(test_dataset)` were very different from what I got from `model(**test_encodings)`. I did not find messages describing what the `predictions` actually are in the docu... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9398/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9398/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9397 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9397/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9397/comments | https://api.github.com/repos/huggingface/transformers/issues/9397/events | https://github.com/huggingface/transformers/issues/9397 | 778,023,255 | MDU6SXNzdWU3NzgwMjMyNTU= | 9,397 | CUDA runtime error during benchmarking | {
"login": "serkansulun",
"id": 26304981,
"node_id": "MDQ6VXNlcjI2MzA0OTgx",
"avatar_url": "https://avatars.githubusercontent.com/u/26304981?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/serkansulun",
"html_url": "https://github.com/serkansulun",
"followers_url": "https://api.github.com/... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"This issue has been automatically marked as stale and been closed because it has not had recent activity. Thank you for your contributions.\n\nIf you think this still needs to be addressed please comment on this thread."
] | 1,609 | 1,614 | 1,614 | NONE | null | Running `transformers/examples/benchmarking/run_benchmark.py` with any type of model, with multi-processing gives the following error:
```
1 / 1
THCudaCheck FAIL file=/opt/conda/conda-bld/pytorch_1565272271120/work/aten/src/THC/THCGeneral.cpp line=54 error=3 : initialization error
cuda runtime error (3) : initial... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9397/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9397/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9396 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9396/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9396/comments | https://api.github.com/repos/huggingface/transformers/issues/9396/events | https://github.com/huggingface/transformers/issues/9396 | 777,978,221 | MDU6SXNzdWU3Nzc5NzgyMjE= | 9,396 | run_glue.py with XLNet model on CoLA dataset reaches 0 accuracy | {
"login": "yonatanbitton",
"id": 26148975,
"node_id": "MDQ6VXNlcjI2MTQ4OTc1",
"avatar_url": "https://avatars.githubusercontent.com/u/26148975?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/yonatanbitton",
"html_url": "https://github.com/yonatanbitton",
"followers_url": "https://api.githu... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"This issue has been automatically marked as stale and been closed because it has not had recent activity. Thank you for your contributions.\n\nIf you think this still needs to be addressed please comment on this thread.",
"Hi, did you fix this issue? I also got this problem when using BART-large."
] | 1,609 | 1,642 | 1,614 | NONE | null | ## Environment info
- `transformers` version: 4.1.1
- Platform: Linux
- Python version: 3.6.10
- PyTorch version (GPU?): 1.7.0 (True)
- Tensorflow version (GPU?): not installed (NA)
- Using GPU in script?: Yes
- Using distributed or parallel set-up in script?: No
### Who can help
@TevenLeScao
## Inf... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9396/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9396/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9395 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9395/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9395/comments | https://api.github.com/repos/huggingface/transformers/issues/9395/events | https://github.com/huggingface/transformers/issues/9395 | 777,868,857 | MDU6SXNzdWU3Nzc4Njg4NTc= | 9,395 | wrong output for Bert-larged-uncased | {
"login": "Twsschx",
"id": 45708562,
"node_id": "MDQ6VXNlcjQ1NzA4NTYy",
"avatar_url": "https://avatars.githubusercontent.com/u/45708562?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Twsschx",
"html_url": "https://github.com/Twsschx",
"followers_url": "https://api.github.com/users/Twssch... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"Hi @Twsschx \r\n\r\nIn the newer version of transformers, PyTorch models have outputs that are instances of subclasses of `ModelOutput`,\r\nto access it as a tuple, you can use slicing, or to get a particular tensor, just provide the key to the output class.\r\n\r\ni.e \r\n```python3\r\nlast_hidden_states, pooling... | 1,609 | 1,614 | 1,614 | NONE | null | I am running Bert for the pytorch version:
from transformers import BertConfig, BertTokenizer, BertModel
config_class, model_class, tokenizer_class = (BertConfig, BertModel, BertTokenizer)
transformer_config = config_class.from_pretrained(pretrained_model + "/bert_config.json")
tokenizer = tokenizer_c... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9395/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9395/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9394 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9394/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9394/comments | https://api.github.com/repos/huggingface/transformers/issues/9394/events | https://github.com/huggingface/transformers/pull/9394 | 777,690,098 | MDExOlB1bGxSZXF1ZXN0NTQ3OTE3MjM1 | 9,394 | Simplify marian distillation script | {
"login": "sshleifer",
"id": 6045025,
"node_id": "MDQ6VXNlcjYwNDUwMjU=",
"avatar_url": "https://avatars.githubusercontent.com/u/6045025?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sshleifer",
"html_url": "https://github.com/sshleifer",
"followers_url": "https://api.github.com/users/ss... | [] | closed | false | null | [] | [] | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | Simplify marian distillation script, by adding a suggested MAX_LEN and using finetune.py directly. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9394/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/huggingface/transformers/issues/9394/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9394",
"html_url": "https://github.com/huggingface/transformers/pull/9394",
"diff_url": "https://github.com/huggingface/transformers/pull/9394.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9394.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9393 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9393/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9393/comments | https://api.github.com/repos/huggingface/transformers/issues/9393/events | https://github.com/huggingface/transformers/issues/9393 | 777,660,030 | MDU6SXNzdWU3Nzc2NjAwMzA= | 9,393 | `run_glue.py` fails when using my own dataset of regression task | {
"login": "forest1988",
"id": 2755894,
"node_id": "MDQ6VXNlcjI3NTU4OTQ=",
"avatar_url": "https://avatars.githubusercontent.com/u/2755894?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/forest1988",
"html_url": "https://github.com/forest1988",
"followers_url": "https://api.github.com/users... | [] | closed | false | {
"login": "sgugger",
"id": 35901082,
"node_id": "MDQ6VXNlcjM1OTAxMDgy",
"avatar_url": "https://avatars.githubusercontent.com/u/35901082?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sgugger",
"html_url": "https://github.com/sgugger",
"followers_url": "https://api.github.com/users/sgugge... | [
{
"login": "sgugger",
"id": 35901082,
"node_id": "MDQ6VXNlcjM1OTAxMDgy",
"avatar_url": "https://avatars.githubusercontent.com/u/35901082?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sgugger",
"html_url": "https://github.com/sgugger",
"followers_url": "https://api.git... | [
"This is the correct fix indeed (though we can group this with the previous test with `elif data_args.task_name is None and not is_regression`)! Thanks for flagging this, do you want to open a PR with the fix you found?",
"@sgugger \r\nThank you for checking this issue and giving the comment.\r\nI'd love to open ... | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | ## Environment info
- `transformers` version: 4.1.1
- Platform: Linux-4.15.0-123-generic-x86_64-with-glibc2.10
- Python version: 3.8.3
- PyTorch version (GPU?): 1.7.0 (True)
- Tensorflow version (GPU?): 2.3.1 (True)
- Using GPU in script?: Yes
- Using distributed or parallel set-up in script?: No
### W... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9393/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9393/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9392 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9392/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9392/comments | https://api.github.com/repos/huggingface/transformers/issues/9392/events | https://github.com/huggingface/transformers/issues/9392 | 777,651,721 | MDU6SXNzdWU3Nzc2NTE3MjE= | 9,392 | Model inputs and outputs are ``None`` when converting fine-tuned gpt2 to Tensorflow? | {
"login": "farazk86",
"id": 33456896,
"node_id": "MDQ6VXNlcjMzNDU2ODk2",
"avatar_url": "https://avatars.githubusercontent.com/u/33456896?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/farazk86",
"html_url": "https://github.com/farazk86",
"followers_url": "https://api.github.com/users/far... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"Hey @farazk86,\r\n\r\nthanks for your issue! I've sadly never worked with tflite so not really sure how to best help you here. Maybe @jplu (or @LysandreJik) ? ",
"> Hey @farazk86,\r\n> \r\n> thanks for your issue! I've sadly never worked with tflite so not really sure how to best help you here. Maybe @jplu (or @... | 1,609 | 1,614 | 1,614 | NONE | null | Hi,
I've fine tuned a distilgpt2 model using my own text using ``run_language_modeling.py`` and its working fine after training and ``run_generation.py`` script produces the expected results.
Now I want to convert this to a Tensorflow Lite model and did so by using the following
```python
from transformers i... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9392/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9392/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9391 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9391/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9391/comments | https://api.github.com/repos/huggingface/transformers/issues/9391/events | https://github.com/huggingface/transformers/issues/9391 | 777,579,543 | MDU6SXNzdWU3Nzc1Nzk1NDM= | 9,391 | Similar usage of `past_key_values` in CausalLM and Seq2SeqLM | {
"login": "forest1988",
"id": 2755894,
"node_id": "MDQ6VXNlcjI3NTU4OTQ=",
"avatar_url": "https://avatars.githubusercontent.com/u/2755894?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/forest1988",
"html_url": "https://github.com/forest1988",
"followers_url": "https://api.github.com/users... | [
{
"id": 1990918270,
"node_id": "MDU6TGFiZWwxOTkwOTE4Mjcw",
"url": "https://api.github.com/repos/huggingface/transformers/labels/Good%20First%20Issue",
"name": "Good First Issue",
"color": "bbf794",
"default": false,
"description": ""
}
] | closed | false | null | [] | [
"Hey @forest1988,\r\n\r\nIn order to fully use `BartDecoder` separately from `BartModel` as a `BartForCausalLM` model, we're still waiting on this PR: #9128.\r\nAnd again, you're very correct in your assessment that the behavior between `BartDecoder` and `GPT2` is not fully aligned here. IMO, we should change GPT2'... | 1,609 | 1,611 | 1,611 | CONTRIBUTOR | null | # 🚀 Feature request
It seems GPT-2 and BartDecoder has a different style of `past_key_values`.
In GPT-2, `past_key_values` is explained as below:
(the explanation is from https://huggingface.co/transformers/model_doc/gpt2.html#gpt2model)
```
(parameters)
past_key_values (List[torch.FloatTensor] of length c... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9391/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9391/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9390 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9390/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9390/comments | https://api.github.com/repos/huggingface/transformers/issues/9390/events | https://github.com/huggingface/transformers/pull/9390 | 777,579,109 | MDExOlB1bGxSZXF1ZXN0NTQ3ODM2MzM1 | 9,390 | [trainer] self.model_wrapped + _model_unwrap | {
"login": "stas00",
"id": 10676103,
"node_id": "MDQ6VXNlcjEwNjc2MTAz",
"avatar_url": "https://avatars.githubusercontent.com/u/10676103?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/stas00",
"html_url": "https://github.com/stas00",
"followers_url": "https://api.github.com/users/stas00/fo... | [] | closed | false | null | [] | [
"This should be good now, @sgugger - thanks a lot for all the suggestions!",
"@LysandreJik GitHub Reviewers is down, so tagging you instead."
] | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | This PR adds:
* [x] adds `self.model_wrapped` - to have access to the outmost module regardless of how many times it was wrapped (e.g. under DeepSpeed there is a double wrapping ` DDP(Deepspeed(Transformers Model))`)
* [x] makes sure that `self.model` is always set to the normal model
* [x] fixes a bug where unde... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9390/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9390/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9390",
"html_url": "https://github.com/huggingface/transformers/pull/9390",
"diff_url": "https://github.com/huggingface/transformers/pull/9390.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9390.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9389 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9389/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9389/comments | https://api.github.com/repos/huggingface/transformers/issues/9389/events | https://github.com/huggingface/transformers/pull/9389 | 777,578,223 | MDExOlB1bGxSZXF1ZXN0NTQ3ODM1Njgx | 9,389 | [trainer] self.model_wrapped + _model_unwrap | {
"login": "stas00",
"id": 10676103,
"node_id": "MDQ6VXNlcjEwNjc2MTAz",
"avatar_url": "https://avatars.githubusercontent.com/u/10676103?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/stas00",
"html_url": "https://github.com/stas00",
"followers_url": "https://api.github.com/users/stas00/fo... | [] | closed | false | null | [] | [] | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | This PR adds:
* [x] self.wrapped
https://github.com/huggingface/transformers/pull/9211
in progress | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9389/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9389/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9389",
"html_url": "https://github.com/huggingface/transformers/pull/9389",
"diff_url": "https://github.com/huggingface/transformers/pull/9389.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9389.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/9388 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9388/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9388/comments | https://api.github.com/repos/huggingface/transformers/issues/9388/events | https://github.com/huggingface/transformers/issues/9388 | 777,564,229 | MDU6SXNzdWU3Nzc1NjQyMjk= | 9,388 | Conditional Generation using input_embeds instead of input_ids | {
"login": "frankgandiao",
"id": 31948011,
"node_id": "MDQ6VXNlcjMxOTQ4MDEx",
"avatar_url": "https://avatars.githubusercontent.com/u/31948011?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/frankgandiao",
"html_url": "https://github.com/frankgandiao",
"followers_url": "https://api.github.c... | [] | closed | false | null | [] | [
"Hey @frankgandiao,\r\n\r\nNote that Encoder-Decoder models usually require both `input_ids` and `decoder_input_ids`. Bart is special in a sense that it can automatically create the `decoder_input_ids` from the `input_ids` if you **don't** provide the `decoder_input_ids`. However, the model is not able to automatic... | 1,609 | 1,609 | 1,609 | NONE | null | Hi @patrickvonplaten!
When using input_embeds instead of input_ids as inputs to the BartForConditionalGeneration model, I am not able to generate the result. Could you please take a look? The same code works with GPT2 in place of Bart. Thanks!
Here is the script
```
import torch
from transformers import Bart... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9388/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9388/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9387 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9387/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9387/comments | https://api.github.com/repos/huggingface/transformers/issues/9387/events | https://github.com/huggingface/transformers/issues/9387 | 777,540,791 | MDU6SXNzdWU3Nzc1NDA3OTE= | 9,387 | Where is the impact when output_attentions=True? | {
"login": "celsofranssa",
"id": 11181748,
"node_id": "MDQ6VXNlcjExMTgxNzQ4",
"avatar_url": "https://avatars.githubusercontent.com/u/11181748?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/celsofranssa",
"html_url": "https://github.com/celsofranssa",
"followers_url": "https://api.github.c... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"If `output_attentions=True` memory consumption should increase significantly (for large `sequence_length`) since we now store all attentions of size (`batch_size`, `num_heas`, `sequence_length`, `sequence_length`). This is less significant in training since the stored activations for training consume most RAM anyw... | 1,609 | 1,614 | 1,614 | NONE | null | Is there any impact regarding performance (training/fine-tuning time, GPU memory, batch size, etc.) when `output_attentions=True`?
```python
self.bert_encoder = BertModel.from_pretrained(
hparams.architecture, # "bert-base-uncased"
output_attentions=True)
``` | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9387/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9387/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9386 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9386/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9386/comments | https://api.github.com/repos/huggingface/transformers/issues/9386/events | https://github.com/huggingface/transformers/pull/9386 | 777,522,803 | MDExOlB1bGxSZXF1ZXN0NTQ3Nzk2NjYw | 9,386 | replace apex.normalization.FusedLayerNorm with torch.nn.LayerNorm | {
"login": "stas00",
"id": 10676103,
"node_id": "MDQ6VXNlcjEwNjc2MTAz",
"avatar_url": "https://avatars.githubusercontent.com/u/10676103?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/stas00",
"html_url": "https://github.com/stas00",
"followers_url": "https://api.github.com/users/stas00/fo... | [] | closed | false | null | [] | [
"Merging since it's blocking #9343 ."
] | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | This PR proposes to drop `apex.normalization.FusedLayerNorm` in favor of faster `torch.nn.LayerNorm`.
1. For performance and background details please see the discussions in https://github.com/huggingface/transformers/issues/9377
2. It's also needed for https://github.com/huggingface/transformers/pull/9384 since `... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9386/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9386/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9386",
"html_url": "https://github.com/huggingface/transformers/pull/9386",
"diff_url": "https://github.com/huggingface/transformers/pull/9386.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9386.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9385 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9385/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9385/comments | https://api.github.com/repos/huggingface/transformers/issues/9385/events | https://github.com/huggingface/transformers/pull/9385 | 777,519,130 | MDExOlB1bGxSZXF1ZXN0NTQ3Nzk0MTI1 | 9,385 | [logging] autoflush | {
"login": "stas00",
"id": 10676103,
"node_id": "MDQ6VXNlcjEwNjc2MTAz",
"avatar_url": "https://avatars.githubusercontent.com/u/10676103?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/stas00",
"html_url": "https://github.com/stas00",
"followers_url": "https://api.github.com/users/stas00/fo... | [] | closed | false | null | [] | [] | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | This PR proposes to:
* auto-flush `transformers` logging
When using logging for tracing signals from different parts of the code and which could be mixed with print debug this aids to get all the logging events synchronized.
I don't think this change will introduce any performance impacts.
If it helps som... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9385/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9385/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9385",
"html_url": "https://github.com/huggingface/transformers/pull/9385",
"diff_url": "https://github.com/huggingface/transformers/pull/9385.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9385.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9384 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9384/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9384/comments | https://api.github.com/repos/huggingface/transformers/issues/9384/events | https://github.com/huggingface/transformers/pull/9384 | 777,517,730 | MDExOlB1bGxSZXF1ZXN0NTQ3NzkzMjU0 | 9,384 | [model parallelism] Bart goes parallel | {
"login": "stas00",
"id": 10676103,
"node_id": "MDQ6VXNlcjEwNjc2MTAz",
"avatar_url": "https://avatars.githubusercontent.com/u/10676103?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/stas00",
"html_url": "https://github.com/stas00",
"followers_url": "https://api.github.com/users/stas00/fo... | [
{
"id": 2627272588,
"node_id": "MDU6TGFiZWwyNjI3MjcyNTg4",
"url": "https://api.github.com/repos/huggingface/transformers/labels/Model%20Parallel",
"name": "Model Parallel",
"color": "8B66A5",
"default": false,
"description": "Model Parallelilsm Implementations"
},
{
"id": 2796628... | closed | false | null | [] | [
"That looks great! Model parallelism would be very nice for Bart. We should coordinate here a bit with all the open PRs. I'm also more or less done with the big \"split-bart-into-separate-models\" PR: https://github.com/huggingface/transformers/pull/9343.\r\nThink the merge conflicts can become a bit painful here :... | 1,609 | 1,654 | 1,622 | CONTRIBUTOR | null | This PR implements model parallelism (MP) in Bart.
This is the latest incarnation of generalization of the MP in `transformers`, based on @alexorona's original work. I have done some of it already in https://github.com/huggingface/transformers/pull/9323 and this PR builds upon the other one. It's slightly complex wh... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9384/reactions",
"total_count": 2,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 2,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9384/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9384",
"html_url": "https://github.com/huggingface/transformers/pull/9384",
"diff_url": "https://github.com/huggingface/transformers/pull/9384.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9384.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/9383 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9383/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9383/comments | https://api.github.com/repos/huggingface/transformers/issues/9383/events | https://github.com/huggingface/transformers/issues/9383 | 777,484,807 | MDU6SXNzdWU3Nzc0ODQ4MDc= | 9,383 | [Marian] Doc says `config.add_bias_logits=True`, but config is has `config.add_bias_logits=False` | {
"login": "patrickvonplaten",
"id": 23423619,
"node_id": "MDQ6VXNlcjIzNDIzNjE5",
"avatar_url": "https://avatars.githubusercontent.com/u/23423619?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/patrickvonplaten",
"html_url": "https://github.com/patrickvonplaten",
"followers_url": "https://... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"I think what's going on is that `config.add_final_bias_logits` is unused.\r\n\r\nIn modeling_bart.py line 148 we call\r\n\r\n```python\r\nself.register_buffer(\"final_logits_bias\", torch.zeros((1, self.model.shared.num_embeddings)))\r\n```\r\nregardless of the config, and then if it's in the state dict it will ge... | 1,609 | 1,614 | 1,614 | MEMBER | null | **Question**:
In the docs, it is written that Marian has (contrary to Bart), `config.add_bias_logits=True`: https://huggingface.co/transformers/model_doc/marian.html#implementation-notes. But when looking into the code:
https://github.com/huggingface/transformers/blob/b01f451ca38695c60175b34d245997ef4d18231d/src/t... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9383/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9383/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9382 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9382/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9382/comments | https://api.github.com/repos/huggingface/transformers/issues/9382/events | https://github.com/huggingface/transformers/pull/9382 | 777,475,473 | MDExOlB1bGxSZXF1ZXN0NTQ3NzY0NDY0 | 9,382 | [docs] Fix TF base model examples: outputs.last_hidden_states -> state | {
"login": "ck37",
"id": 50770,
"node_id": "MDQ6VXNlcjUwNzcw",
"avatar_url": "https://avatars.githubusercontent.com/u/50770?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ck37",
"html_url": "https://github.com/ck37",
"followers_url": "https://api.github.com/users/ck37/followers",
"follo... | [] | closed | false | null | [] | [] | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | # What does this PR do?
Fixes a typo in the examples of Tensorflow-based base models, in which the returned last_hidden_state attribute of model output is incorrectly listed as "last_hidden_states".
Fixes #9376
## Before submitting
- [x] This PR fixes a typo or improves the docs (you can dismiss the other ch... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9382/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/huggingface/transformers/issues/9382/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9382",
"html_url": "https://github.com/huggingface/transformers/pull/9382",
"diff_url": "https://github.com/huggingface/transformers/pull/9382.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9382.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9381 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9381/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9381/comments | https://api.github.com/repos/huggingface/transformers/issues/9381/events | https://github.com/huggingface/transformers/pull/9381 | 777,468,806 | MDExOlB1bGxSZXF1ZXN0NTQ3NzU5Nzc0 | 9,381 | [Docs] `past_key_values` return a tuple of tuple as a default | {
"login": "patrickvonplaten",
"id": 23423619,
"node_id": "MDQ6VXNlcjIzNDIzNjE5",
"avatar_url": "https://avatars.githubusercontent.com/u/23423619?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/patrickvonplaten",
"html_url": "https://github.com/patrickvonplaten",
"followers_url": "https://... | [] | closed | false | null | [] | [] | 1,609 | 1,609 | 1,609 | MEMBER | null | # What does this PR do?
<!--
Congratulations! You've made it this far! You're not quite done yet though.
Once merged, your PR is going to appear in the release notes with the title you set, so make sure it's a great title that fully reflects the extent of your awesome contribution.
Then, please replace this w... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9381/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/huggingface/transformers/issues/9381/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9381",
"html_url": "https://github.com/huggingface/transformers/pull/9381",
"diff_url": "https://github.com/huggingface/transformers/pull/9381.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9381.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9380 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9380/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9380/comments | https://api.github.com/repos/huggingface/transformers/issues/9380/events | https://github.com/huggingface/transformers/issues/9380 | 777,455,936 | MDU6SXNzdWU3Nzc0NTU5MzY= | 9,380 | BartModel's `past_key_values` seems to have different explanations in input_doc and output_doc | {
"login": "forest1988",
"id": 2755894,
"node_id": "MDQ6VXNlcjI3NTU4OTQ=",
"avatar_url": "https://avatars.githubusercontent.com/u/2755894?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/forest1988",
"html_url": "https://github.com/forest1988",
"followers_url": "https://api.github.com/users... | [] | closed | false | null | [] | [
"Hey @forest1988,\r\n\r\nThanks for your issue! You're 100% correct. The docs need to be updated here! The output is actually never a list, it should always be a `Tuple(Tuple(torch.FloatTensor))` - I'll make a PR afterward. \r\nAnd in Bart, `past_key_values` always consists of `selt_attn_present_key_value` and `cro... | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | ## Environment info
- `transformers` version: 4.1.1
- Platform: Linux-4.15.0-123-generic-x86_64-with-glibc2.10
- Python version: 3.8.3
- PyTorch version (GPU?): 1.7.0 (True)
- Tensorflow version (GPU?): 2.3.1 (True)
- Using GPU in script?: No
- Using distributed or parallel set-up in script?: No
### W... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9380/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9380/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9379 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9379/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9379/comments | https://api.github.com/repos/huggingface/transformers/issues/9379/events | https://github.com/huggingface/transformers/pull/9379 | 777,454,243 | MDExOlB1bGxSZXF1ZXN0NTQ3NzQ5Nzcx | 9,379 | Improve documentation coverage for Bertweet | {
"login": "Qbiwan",
"id": 69753975,
"node_id": "MDQ6VXNlcjY5NzUzOTc1",
"avatar_url": "https://avatars.githubusercontent.com/u/69753975?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Qbiwan",
"html_url": "https://github.com/Qbiwan",
"followers_url": "https://api.github.com/users/Qbiwan/fo... | [] | closed | false | null | [] | [
"Thanks @Qbiwan!"
] | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | # What does this PR do?
<!--
Congratulations! You've made it this far! You're not quite done yet though.
Once merged, your PR is going to appear in the release notes with the title you set, so make sure it's a great title that fully reflects the extent of your awesome contribution.
Then, please replace this w... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9379/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9379/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9379",
"html_url": "https://github.com/huggingface/transformers/pull/9379",
"diff_url": "https://github.com/huggingface/transformers/pull/9379.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9379.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9378 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9378/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9378/comments | https://api.github.com/repos/huggingface/transformers/issues/9378/events | https://github.com/huggingface/transformers/pull/9378 | 777,454,030 | MDExOlB1bGxSZXF1ZXN0NTQ3NzQ5NjE4 | 9,378 | [Docs] Tokenizer Squad 2.0 example | {
"login": "patrickvonplaten",
"id": 23423619,
"node_id": "MDQ6VXNlcjIzNDIzNjE5",
"avatar_url": "https://avatars.githubusercontent.com/u/23423619?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/patrickvonplaten",
"html_url": "https://github.com/patrickvonplaten",
"followers_url": "https://... | [] | closed | false | null | [] | [
"Thank you for the pr. \r\nPlease also update the documentation here: https://huggingface.co/transformers/custom_datasets.html#qa-squad\r\nLine-> end_positions[-1] = encodings.char_to_token(i, answers[i]['answer_end'] + 1)\r\nto -> end_positions[-1] = tokenizer.model_max_length",
"So what is the state of this i... | 1,609 | 1,633 | 1,609 | MEMBER | null | # What does this PR do?
<!--
Congratulations! You've made it this far! You're not quite done yet though.
Once merged, your PR is going to appear in the release notes with the title you set, so make sure it's a great title that fully reflects the extent of your awesome contribution.
Then, please replace this w... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9378/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9378/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9378",
"html_url": "https://github.com/huggingface/transformers/pull/9378",
"diff_url": "https://github.com/huggingface/transformers/pull/9378.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9378.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9377 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9377/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9377/comments | https://api.github.com/repos/huggingface/transformers/issues/9377/events | https://github.com/huggingface/transformers/issues/9377 | 777,413,099 | MDU6SXNzdWU3Nzc0MTMwOTk= | 9,377 | replacing apex.normalization.FusedLayerNorm with torch.nn.LayerNorm | {
"login": "stas00",
"id": 10676103,
"node_id": "MDQ6VXNlcjEwNjc2MTAz",
"avatar_url": "https://avatars.githubusercontent.com/u/10676103?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/stas00",
"html_url": "https://github.com/stas00",
"followers_url": "https://api.github.com/users/stas00/fo... | [] | closed | false | null | [] | [
"I'm good with changing to `torch.nn.LayerNorm`. At @stas00 - do you know what the advantage of `apex.normalization.FusedLayerNorm` is supposed to be? Why did we add `apex.normalization.FusedLayerNorm` in the first place?",
"Prior to about a year ago, `apex.normalization.FusedLayerNorm` was faster than `torch.nn.... | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | It seems that time has arrived to drop `apex.normalization.FusedLayerNorm` in favor of `torch.nn.LayerNorm`
1. the latter was ported more than a year ago from apex https://github.com/pytorch/pytorch/pull/27634 (around pt-1.4)
2. it's faster than the apex according to my benchmarks https://github.com/pytorch/pytor... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9377/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9377/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9376 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9376/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9376/comments | https://api.github.com/repos/huggingface/transformers/issues/9376/events | https://github.com/huggingface/transformers/issues/9376 | 777,323,655 | MDU6SXNzdWU3NzczMjM2NTU= | 9,376 | [docs] TFRobertaModel example: last_hidden_states -> last_hidden_state | {
"login": "ck37",
"id": 50770,
"node_id": "MDQ6VXNlcjUwNzcw",
"avatar_url": "https://avatars.githubusercontent.com/u/50770?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ck37",
"html_url": "https://github.com/ck37",
"followers_url": "https://api.github.com/users/ck37/followers",
"follo... | [] | closed | false | null | [] | [
"looks like this is what you're looking for: https://github.com/huggingface/transformers/blob/ae333d04b29a25be1a70eaccd6260c294c243c5b/src/transformers/file_utils.py#L842-L855",
"Hey @ck37,\r\n\r\nThanks for your issue! Yes, this typo should be corrected -> it would be great if you could open a PR :-) "
] | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | This is a documentation error on the currently deployed version of https://huggingface.co/transformers/
### Who can help
examples/distillation: @VictorSanh
documentation: @sgugger
## Information
Model I am using: TFRoberta
The problem arises when using:
* [x] the official example scripts
## To rep... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9376/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9376/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9375 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9375/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9375/comments | https://api.github.com/repos/huggingface/transformers/issues/9375/events | https://github.com/huggingface/transformers/pull/9375 | 777,285,914 | MDExOlB1bGxSZXF1ZXN0NTQ3NjI4Njkx | 9,375 | Fix Typo | {
"login": "vanche",
"id": 10228650,
"node_id": "MDQ6VXNlcjEwMjI4NjUw",
"avatar_url": "https://avatars.githubusercontent.com/u/10228650?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/vanche",
"html_url": "https://github.com/vanche",
"followers_url": "https://api.github.com/users/vanche/fo... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"This issue has been automatically marked as stale and been closed because it has not had recent activity. Thank you for your contributions.\n\nIf you think this still needs to be addressed please comment on this thread."
] | 1,609 | 1,614 | 1,614 | CONTRIBUTOR | null | # What does this PR do?
<!--
Congratulations! You've made it this far! You're not quite done yet though.
Once merged, your PR is going to appear in the release notes with the title you set, so make sure it's a great title that fully reflects the extent of your awesome contribution.
Then, please replace this w... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9375/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9375/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9375",
"html_url": "https://github.com/huggingface/transformers/pull/9375",
"diff_url": "https://github.com/huggingface/transformers/pull/9375.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9375.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/9374 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9374/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9374/comments | https://api.github.com/repos/huggingface/transformers/issues/9374/events | https://github.com/huggingface/transformers/issues/9374 | 777,281,096 | MDU6SXNzdWU3NzcyODEwOTY= | 9,374 | How do I handle class imbalance for text data when using pretrained models like BERT? | {
"login": "nikhil6041",
"id": 42090593,
"node_id": "MDQ6VXNlcjQyMDkwNTkz",
"avatar_url": "https://avatars.githubusercontent.com/u/42090593?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/nikhil6041",
"html_url": "https://github.com/nikhil6041",
"followers_url": "https://api.github.com/use... | [] | closed | false | null | [] | [
"Hi!\r\n\r\nYou could try replacing the CrossEntropy loss with [this Dice Loss](https://github.com/fursovia/self-adj-dice), which may help you with the imbalance issues. In the paper linked in the repo they explain the design process. I have tried it with mixed results, although for my (skewed) dataset, weighting t... | 1,609 | 1,698 | 1,615 | NONE | null | I have a skewed dataset consisting of samples of the form:
```
Category 1 10000
Category 2 2000
Category 3 400
Category 4 300
Category 5 100
```
The dataset consists of text with data labeled into one of the five categories. I am trying to use the pretrained models like BERT for the classification t... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9374/reactions",
"total_count": 7,
"+1": 7,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9374/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9373 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9373/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9373/comments | https://api.github.com/repos/huggingface/transformers/issues/9373/events | https://github.com/huggingface/transformers/issues/9373 | 777,275,417 | MDU6SXNzdWU3NzcyNzU0MTc= | 9,373 | how to evaluate models on SUPER_GLUE benchmark? | {
"login": "YoungTimmy",
"id": 39907234,
"node_id": "MDQ6VXNlcjM5OTA3MjM0",
"avatar_url": "https://avatars.githubusercontent.com/u/39907234?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/YoungTimmy",
"html_url": "https://github.com/YoungTimmy",
"followers_url": "https://api.github.com/use... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"This issue has been automatically marked as stale and been closed because it has not had recent activity. Thank you for your contributions.\n\nIf you think this still needs to be addressed please comment on this thread."
] | 1,609 | 1,614 | 1,614 | NONE | null | Hi, I am trying to evaluate models on SUPER_GLUE benchmark.
However, I can load SUPER_GLUE dataset from Transformer but I cant find any metrics of this benchmark.
Is there any script like _**superglue_metrics.py**_ that can evaluate models on superglue?
thanks a lot! :) | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9373/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9373/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9372 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9372/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9372/comments | https://api.github.com/repos/huggingface/transformers/issues/9372/events | https://github.com/huggingface/transformers/issues/9372 | 777,137,329 | MDU6SXNzdWU3NzcxMzczMjk= | 9,372 | Why does datasets get imported when running "from transformers.models.roberta.tokenization_roberta_fast import RobertaTokenizerFast" | {
"login": "vgoklani",
"id": 180487,
"node_id": "MDQ6VXNlcjE4MDQ4Nw==",
"avatar_url": "https://avatars.githubusercontent.com/u/180487?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/vgoklani",
"html_url": "https://github.com/vgoklani",
"followers_url": "https://api.github.com/users/vgoklan... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"hi @vgoklani \r\n\r\n`datasets` is not required for tokenizers, so it's unlikely to get this error when just importing the tokenizer. Are you running any examples scripts? because those require `datasets` lib",
"Hi @patil-suraj Happy New Year!\r\n\r\nHere is the stack trace:\r\n\r\n root@b5d80f9670ea:~/src# i... | 1,609 | 1,614 | 1,614 | NONE | null | When running
from transformers.models.roberta.tokenization_roberta_fast import RobertaTokenizerFast
I get this warning:
ImportWarning: To use `datasets`, the module `pyarrow>=0.16.0` is required, and the current version of `pyarrow` doesn't match this condition. If you are running this in a Google Co... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9372/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9372/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9371 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9371/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9371/comments | https://api.github.com/repos/huggingface/transformers/issues/9371/events | https://github.com/huggingface/transformers/issues/9371 | 777,097,353 | MDU6SXNzdWU3NzcwOTczNTM= | 9,371 | Excessive GPU-GPU communication with GPT2 making multi-GPU training slow? | {
"login": "moyix",
"id": 34380,
"node_id": "MDQ6VXNlcjM0Mzgw",
"avatar_url": "https://avatars.githubusercontent.com/u/34380?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/moyix",
"html_url": "https://github.com/moyix",
"followers_url": "https://api.github.com/users/moyix/followers",
"f... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
},
{
"id": 2604155188,
"node_id": "MDU6TGFiZWwyNjA0MTU1MTg4",
... | closed | false | null | [] | [
"Not an answer to your issue/question, but have you tried running in distributed training (DDP), which is the recommended way of running over multiple GPUs: https://github.com/huggingface/transformers/tree/master/examples#distributed-training-and-mixed-precision\r\n\r\nWould be curious to see the same with/without ... | 1,609 | 1,694 | 1,614 | NONE | null | Summary: on a multi-GPU system, training GPT2 seems to scale poorly unless a very fast GPU-GPU interconnect like NVLink is available. In particular, without NVLink using two GPUs is *slower* than using just one GPU.
## Environment info
- `transformers` version: 4.1.1
- Platform: Linux-5.8.0-rc7-custom-x86_64-wit... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9371/reactions",
"total_count": 5,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 5
} | https://api.github.com/repos/huggingface/transformers/issues/9371/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9370 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9370/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9370/comments | https://api.github.com/repos/huggingface/transformers/issues/9370/events | https://github.com/huggingface/transformers/issues/9370 | 776,942,988 | MDU6SXNzdWU3NzY5NDI5ODg= | 9,370 | Custom train/validation file not supported in run_qa.py | {
"login": "BatMrE",
"id": 48859022,
"node_id": "MDQ6VXNlcjQ4ODU5MDIy",
"avatar_url": "https://avatars.githubusercontent.com/u/48859022?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/BatMrE",
"html_url": "https://github.com/BatMrE",
"followers_url": "https://api.github.com/users/BatMrE/fo... | [] | closed | false | null | [] | [
"Hi @BatMrE \r\n\r\nThe syntax of the command is wrong, there should be no spaces around `=` or you can also just remove the `=`\r\n\r\nSo it should be either like this\r\n```bash\r\npython run_qa.py \\\r\n --model_name_or_path bert-base-uncased \\\r\n --train_file=train-v1.1.json \\\r\n --validation_file=dev-v... | 1,609 | 1,619 | 1,619 | NONE | null | **Environment info**
transformers version: 4.0.1
Platform: Linux-5.4.0-58-generic-x86_64-with-glibc2.10
Python version: 3.8.5
PyTorch version (GPU?): 1.7.1+cu110 (True)
Tensorflow version (GPU?): not installed (NA)
Using GPU in script?: yes
I am trying to pass custom dataset or modified squad dataset (in val... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9370/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9370/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9369 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9369/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9369/comments | https://api.github.com/repos/huggingface/transformers/issues/9369/events | https://github.com/huggingface/transformers/pull/9369 | 776,927,069 | MDExOlB1bGxSZXF1ZXN0NTQ3MzMyMjc3 | 9,369 | TF >= 2.3 cleaning | {
"login": "jplu",
"id": 959590,
"node_id": "MDQ6VXNlcjk1OTU5MA==",
"avatar_url": "https://avatars.githubusercontent.com/u/959590?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jplu",
"html_url": "https://github.com/jplu",
"followers_url": "https://api.github.com/users/jplu/followers",
... | [] | closed | false | null | [] | [] | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | # What does this PR do?
The minimal TF version has recently been fixed to >=2.3, this PR remove all the <2.3 calls, mostly replacing experimental features by their stable ones. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9369/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9369/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9369",
"html_url": "https://github.com/huggingface/transformers/pull/9369",
"diff_url": "https://github.com/huggingface/transformers/pull/9369.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9369.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9368 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9368/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9368/comments | https://api.github.com/repos/huggingface/transformers/issues/9368/events | https://github.com/huggingface/transformers/pull/9368 | 776,917,178 | MDExOlB1bGxSZXF1ZXN0NTQ3MzI0NDc4 | 9,368 | Fix utils on Windows | {
"login": "jplu",
"id": 959590,
"node_id": "MDQ6VXNlcjk1OTU5MA==",
"avatar_url": "https://avatars.githubusercontent.com/u/959590?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jplu",
"html_url": "https://github.com/jplu",
"followers_url": "https://api.github.com/users/jplu/followers",
... | [] | closed | false | null | [] | [] | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | # What does this PR do?
This PR fixes `check_repo` for Windows execution. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9368/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9368/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9368",
"html_url": "https://github.com/huggingface/transformers/pull/9368",
"diff_url": "https://github.com/huggingface/transformers/pull/9368.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9368.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9367 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9367/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9367/comments | https://api.github.com/repos/huggingface/transformers/issues/9367/events | https://github.com/huggingface/transformers/pull/9367 | 776,881,276 | MDExOlB1bGxSZXF1ZXN0NTQ3Mjk2ODE2 | 9,367 | Add-support-for-examples-scripts-to-run-on-sagemaker | {
"login": "philschmid",
"id": 32632186,
"node_id": "MDQ6VXNlcjMyNjMyMTg2",
"avatar_url": "https://avatars.githubusercontent.com/u/32632186?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/philschmid",
"html_url": "https://github.com/philschmid",
"followers_url": "https://api.github.com/use... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"@sgugger's 1st point is a good point and we should probably either:\r\n- recommend that users `git clone` the same version of `transformers` as is installed in the DLC image.\r\n- or even find a way to bundle the scripts themselves (or their capabilities) in the image itself, kinda like what I was suggesting befor... | 1,609 | 1,615 | 1,614 | MEMBER | null | Hello Guys,
i am currently working on how we could edit/extend the fine-tuning scripts from `examples/` to work out-of-the-box within sagemaker. For that i adjusted the [`run_glue.py` script](https://github.com/huggingface/transformers/blob/master/examples/text-classification/run_glue.py).
To test it I created a ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9367/reactions",
"total_count": 2,
"+1": 1,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 1
} | https://api.github.com/repos/huggingface/transformers/issues/9367/timeline | null | true | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9367",
"html_url": "https://github.com/huggingface/transformers/pull/9367",
"diff_url": "https://github.com/huggingface/transformers/pull/9367.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9367.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/9366 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9366/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9366/comments | https://api.github.com/repos/huggingface/transformers/issues/9366/events | https://github.com/huggingface/transformers/issues/9366 | 776,810,175 | MDU6SXNzdWU3NzY4MTAxNzU= | 9,366 | How to implement seq2seq attention mask conviniently? | {
"login": "zhizeng8",
"id": 49787234,
"node_id": "MDQ6VXNlcjQ5Nzg3MjM0",
"avatar_url": "https://avatars.githubusercontent.com/u/49787234?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/zhizeng8",
"html_url": "https://github.com/zhizeng8",
"followers_url": "https://api.github.com/users/zhi... | [] | closed | false | null | [] | [
"The image is also on this link.\r\nhttps://img-blog.csdnimg.cn/20191025102941935.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9qYWNrY3VpLmJsb2cuY3Nkbi5uZXQ=,size_16,color_FFFFFF,t_70",
"Hey @zhizeng8, \r\n\r\nIn the future, it would be nice if such questions are posted in t... | 1,609 | 1,619 | 1,619 | NONE | null | BERT's attention mask is square, GPT's attention mask is triangular. How to implement seq2seq attention mask with transformers package conviniently? like the one appears in UniLM, a triangle concatenates a rectangle.
![unilm](https://user-images.githubusercontent.com/49787234/103397155-ff354180-4b71-11eb-8283-1c0f50f5... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9366/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9366/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9365 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9365/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9365/comments | https://api.github.com/repos/huggingface/transformers/issues/9365/events | https://github.com/huggingface/transformers/issues/9365 | 776,702,511 | MDU6SXNzdWU3NzY3MDI1MTE= | 9,365 | Multi turn conversation with Blender Bot | {
"login": "mailong25",
"id": 12481660,
"node_id": "MDQ6VXNlcjEyNDgxNjYw",
"avatar_url": "https://avatars.githubusercontent.com/u/12481660?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/mailong25",
"html_url": "https://github.com/mailong25",
"followers_url": "https://api.github.com/users/... | [] | closed | false | {
"login": "patil-suraj",
"id": 27137566,
"node_id": "MDQ6VXNlcjI3MTM3NTY2",
"avatar_url": "https://avatars.githubusercontent.com/u/27137566?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/patil-suraj",
"html_url": "https://github.com/patil-suraj",
"followers_url": "https://api.github.com/... | [
{
"login": "patil-suraj",
"id": 27137566,
"node_id": "MDQ6VXNlcjI3MTM3NTY2",
"avatar_url": "https://avatars.githubusercontent.com/u/27137566?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/patil-suraj",
"html_url": "https://github.com/patil-suraj",
"followers_url": "htt... | [
"Hi @mailong25 for blenderbot the dialogs are separated by the newline `\\n`.\r\n\r\nSo the text should be \r\n`I am from Vietnam\\nI've never been there, but I've always wanted to go. How do you like it?\\npretty good actually , where you are from ?`\r\n\r\nwhich model are you using, 90M or 3B?\r\n\r\nAlso, could ... | 1,609 | 1,689 | 1,620 | NONE | null | # 🚀 Feature request
Hi there. is there any way to predict a response using multi-turn dialog context for the Blender Bot Model. From your [example](https://huggingface.co/transformers/model_doc/blenderbot.html) I saw that it only use single-turn context.
I tried to use a ` </sep>` token to separate human/bot tur... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9365/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/huggingface/transformers/issues/9365/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9364 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9364/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9364/comments | https://api.github.com/repos/huggingface/transformers/issues/9364/events | https://github.com/huggingface/transformers/issues/9364 | 776,630,873 | MDU6SXNzdWU3NzY2MzA4NzM= | 9,364 | Finetune mbart rouge score difference between training and evaluation part | {
"login": "eymenkagantaspinar",
"id": 58509497,
"node_id": "MDQ6VXNlcjU4NTA5NDk3",
"avatar_url": "https://avatars.githubusercontent.com/u/58509497?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/eymenkagantaspinar",
"html_url": "https://github.com/eymenkagantaspinar",
"followers_url": "ht... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"Hi @Eymen3455 \r\n\r\nIn the `run_eval` command I see that you are setting `n_obs` to 100 while the `finetune_trainer` uses all test examples, could you maybe run eval again with all test examples and see if you get close results?",
"First of all, thank you very much @patil-suraj for your reply. I did as you sai... | 1,609 | 1,626 | 1,614 | NONE | null | ### Environment info
- transformers from source
- google colab
### Information
When I use the `sshleifer/student_cnn_12_6` model for `finetune_trainer.py` and then I run finetuned model in `run_eval.py`, I can get close and high rouge scores.However when I give `facebook/mbart-large-cc25` as model and tokenizer ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9364/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9364/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9363 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9363/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9363/comments | https://api.github.com/repos/huggingface/transformers/issues/9363/events | https://github.com/huggingface/transformers/pull/9363 | 776,608,062 | MDExOlB1bGxSZXF1ZXN0NTQ3MDY0MjE3 | 9,363 | Make sure to use return dict for the encoder call inside RagTokenForGeneration | {
"login": "dblakely",
"id": 20539855,
"node_id": "MDQ6VXNlcjIwNTM5ODU1",
"avatar_url": "https://avatars.githubusercontent.com/u/20539855?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/dblakely",
"html_url": "https://github.com/dblakely",
"followers_url": "https://api.github.com/users/dbl... | [] | closed | false | null | [] | [
"Thanks!"
] | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | ## What does this PR do?
At some point, `return_dict` was set to be `False` by default inside BART. However, this created a type error inside `RagTokenForGeneration`, which was still written with the expectation that `return_dict=True` by default. This PR simply adds `return_dict=True` to the call to the BART encode... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9363/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9363/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9363",
"html_url": "https://github.com/huggingface/transformers/pull/9363",
"diff_url": "https://github.com/huggingface/transformers/pull/9363.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9363.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9362 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9362/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9362/comments | https://api.github.com/repos/huggingface/transformers/issues/9362/events | https://github.com/huggingface/transformers/issues/9362 | 776,524,792 | MDU6SXNzdWU3NzY1MjQ3OTI= | 9,362 | Jupyter Notebook Kernel crashes when tokenizing large dataset | {
"login": "lthiet",
"id": 26815719,
"node_id": "MDQ6VXNlcjI2ODE1NzE5",
"avatar_url": "https://avatars.githubusercontent.com/u/26815719?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lthiet",
"html_url": "https://github.com/lthiet",
"followers_url": "https://api.github.com/users/lthiet/fo... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"This issue has been automatically marked as stale and been closed because it has not had recent activity. Thank you for your contributions.\n\nIf you think this still needs to be addressed please comment on this thread.",
"i still see it's a problem.. running tokenizer",
"We can't really help you debug your sc... | 1,609 | 1,692 | 1,614 | NONE | null | ## Environment info
I am using 2 setups, my personal laptop and a cluster.
My laptop has this environment :
`transformers` version: 4.1.1
- Platform: Darwin-20.2.0-x86_64-i386-64bit
- Python version: 3.7.6
- PyTorch version (GPU?): 1.7.1 (False)
- Tensorflow version (GPU?): 2.2.0 (False)
- Using GPU in sc... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9362/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9362/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9361 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9361/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9361/comments | https://api.github.com/repos/huggingface/transformers/issues/9361/events | https://github.com/huggingface/transformers/issues/9361 | 776,508,406 | MDU6SXNzdWU3NzY1MDg0MDY= | 9,361 | DeBERTa in TF (TFAutoModel): unrecognized configuration class | {
"login": "ck37",
"id": 50770,
"node_id": "MDQ6VXNlcjUwNzcw",
"avatar_url": "https://avatars.githubusercontent.com/u/50770?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ck37",
"html_url": "https://github.com/ck37",
"followers_url": "https://api.github.com/users/ck37/followers",
"follo... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"Hi @ck37 \r\n\r\n`DeBERTa` is currently PyTorch-only so it can't be loaded with `TFAutoModel`. The table on the doc's [homepage](https://huggingface.co/transformers/) shows whether the models have support in PyTorch, TensorFlow, and/or Flax.",
"Gotcha, thanks for the fast response. Do you think the TF side will... | 1,609 | 1,614 | 1,614 | CONTRIBUTOR | null | ## Environment info
- `transformers` version: 4.1.1
- Platform: Linux-3.10.0-957.21.3.el7.x86_64-x86_64-with-centos-7.6.1810-Core
- Python version: 3.6.9
- PyTorch version (GPU?): 1.4.0 (True)
- Tensorflow version (GPU?): 2.2.0 (True)
- Using GPU in script?: No
- Using distributed or parallel set-up in sc... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9361/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9361/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9360 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9360/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9360/comments | https://api.github.com/repos/huggingface/transformers/issues/9360/events | https://github.com/huggingface/transformers/issues/9360 | 776,494,110 | MDU6SXNzdWU3NzY0OTQxMTA= | 9,360 | Loading a set of tokenized files for training | {
"login": "nlpravi",
"id": 1936777,
"node_id": "MDQ6VXNlcjE5MzY3Nzc=",
"avatar_url": "https://avatars.githubusercontent.com/u/1936777?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/nlpravi",
"html_url": "https://github.com/nlpravi",
"followers_url": "https://api.github.com/users/nlpravi/... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"In this case, you could write a custom `dataset` that will read your pickle files and return the examples from `__getitem__` method.\r\n\r\nIf you are looking for an efficient way of pre-tokenizing the dataset, saving/caching it for future use, and loading it for training then I would recommend you to take a look ... | 1,609 | 1,614 | 1,614 | NONE | null | I have a directory of files that are already tokenized using a pretrained tokenizer. Each file is a pickle file containing a list of objects where each object corresponds to a text sequence containing input_ids and attention_masks. The directory has thousands of files. I'm looking for an efficient way to load the data ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9360/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9360/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9359 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9359/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9359/comments | https://api.github.com/repos/huggingface/transformers/issues/9359/events | https://github.com/huggingface/transformers/issues/9359 | 776,434,266 | MDU6SXNzdWU3NzY0MzQyNjY= | 9,359 | Training loss not getting logged | {
"login": "kunalpagarey",
"id": 38290549,
"node_id": "MDQ6VXNlcjM4MjkwNTQ5",
"avatar_url": "https://avatars.githubusercontent.com/u/38290549?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/kunalpagarey",
"html_url": "https://github.com/kunalpagarey",
"followers_url": "https://api.github.c... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"@sgugger is the best suited to answer you",
"This option is not implemented in Trainer.",
"This issue has been automatically marked as stale and been closed because it has not had recent activity. Thank you for your contributions.\n\nIf you think this still needs to be addressed please comment on this thread."... | 1,609 | 1,614 | 1,614 | NONE | null | While training GPT2 using run_clm.py I wanted to track the training loss as well but could not find a way to do that with evaluation strategy = epoch. So I tried to look deeper into the code and found that may be adding `control.should_log = True` after line referred below will start logging training loss after every e... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9359/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9359/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9358 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9358/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9358/comments | https://api.github.com/repos/huggingface/transformers/issues/9358/events | https://github.com/huggingface/transformers/issues/9358 | 776,411,771 | MDU6SXNzdWU3NzY0MTE3NzE= | 9,358 | error while finetuning for Regression task. | {
"login": "SAIVENKATARAJU",
"id": 46083296,
"node_id": "MDQ6VXNlcjQ2MDgzMjk2",
"avatar_url": "https://avatars.githubusercontent.com/u/46083296?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/SAIVENKATARAJU",
"html_url": "https://github.com/SAIVENKATARAJU",
"followers_url": "https://api.gi... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"This issue has been automatically marked as stale and been closed because it has not had recent activity. Thank you for your contributions.\n\nIf you think this still needs to be addressed please comment on this thread.",
"Was this solved somehow?\r\n",
"Ping @Rocketknight1 ",
"Hi, the problem here is that o... | 1,609 | 1,679 | 1,614 | NONE | null | Hi I was trying to perform finetuning regression task . below is my network.
```
from transformers import TFBertForSequenceClassification
# model initialization
base_model = TFBertForSequenceClassification.from_pretrained('bert-base-uncased')
base_model.bert.trainable=False
model=tf.keras.Sequential(base_model)... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9358/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9358/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9357 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9357/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9357/comments | https://api.github.com/repos/huggingface/transformers/issues/9357/events | https://github.com/huggingface/transformers/issues/9357 | 776,375,240 | MDU6SXNzdWU3NzYzNzUyNDA= | 9,357 | Blenderbot-3B config seems to be a little wrong | {
"login": "Narsil",
"id": 204321,
"node_id": "MDQ6VXNlcjIwNDMyMQ==",
"avatar_url": "https://avatars.githubusercontent.com/u/204321?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Narsil",
"html_url": "https://github.com/Narsil",
"followers_url": "https://api.github.com/users/Narsil/follow... | [] | closed | false | {
"login": "patrickvonplaten",
"id": 23423619,
"node_id": "MDQ6VXNlcjIzNDIzNjE5",
"avatar_url": "https://avatars.githubusercontent.com/u/23423619?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/patrickvonplaten",
"html_url": "https://github.com/patrickvonplaten",
"followers_url": "https://... | [
{
"login": "patrickvonplaten",
"id": 23423619,
"node_id": "MDQ6VXNlcjIzNDIzNjE5",
"avatar_url": "https://avatars.githubusercontent.com/u/23423619?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/patrickvonplaten",
"html_url": "https://github.com/patrickvonplaten",
"follo... | [
"This issue has been stale for 1 month.",
"Closing this, blenderbot 90M is very different in Arch as other variants, so it will receive less love (it's not that powerful compared to the others anyway).\r\n\r\nAlso a lot of work was done here : https://github.com/huggingface/transformers/pull/10002"
] | 1,609 | 1,615 | 1,615 | CONTRIBUTOR | null | ## Environment info
<!-- You can run the command `transformers-cli env` and copy-and-paste its output below.
Don't forget to fill out the missing fields in that output! -->
- `transformers` version: 4.1
- Platform: Linux
- Python version: 3.8
- PyTorch version (GPU?):-
- Tensorflow version (GPU?): -... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9357/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9357/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9356 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9356/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9356/comments | https://api.github.com/repos/huggingface/transformers/issues/9356/events | https://github.com/huggingface/transformers/pull/9356 | 776,299,030 | MDExOlB1bGxSZXF1ZXN0NTQ2ODE5ODU2 | 9,356 | [examples/language-modeling] Add dataset download instructions | {
"login": "stas00",
"id": 10676103,
"node_id": "MDQ6VXNlcjEwNjc2MTAz",
"avatar_url": "https://avatars.githubusercontent.com/u/10676103?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/stas00",
"html_url": "https://github.com/stas00",
"followers_url": "https://api.github.com/users/stas00/fo... | [] | closed | false | null | [] | [
"Isn’t this one simply on the [HuggingFace datasets hub](https://huggingface.co/datasets) by the way?\r\n\r\nOp wo 30 dec. 2020 om 07:50 schreef Stas Bekman <notifications@github.com>\r\n\r\n> I had to hunt for instructions to get the dataset used in this set of\r\n> examples, so this PR proposes to add them to REA... | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | I had to hunt for instructions to get the dataset used in this set of examples, so this PR proposes to add them to README.md.
@patrickvonplaten
| {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9356/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9356/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9356",
"html_url": "https://github.com/huggingface/transformers/pull/9356",
"diff_url": "https://github.com/huggingface/transformers/pull/9356.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9356.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/9355 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9355/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9355/comments | https://api.github.com/repos/huggingface/transformers/issues/9355/events | https://github.com/huggingface/transformers/pull/9355 | 776,271,952 | MDExOlB1bGxSZXF1ZXN0NTQ2Nzk3NTEx | 9,355 | Fix typos in README and bugs in RAG example code for end-to-end evaluation and finetuning | {
"login": "yoshitomo-matsubara",
"id": 11156001,
"node_id": "MDQ6VXNlcjExMTU2MDAx",
"avatar_url": "https://avatars.githubusercontent.com/u/11156001?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/yoshitomo-matsubara",
"html_url": "https://github.com/yoshitomo-matsubara",
"followers_url": ... | [] | closed | false | null | [] | [
"Hi @patrickvonplaten,\r\n\r\nThank you for reviewing this PR!\r\nAs commented above, the argument `num_retrieval_workers` in `add_ray_specific_args` is duplicate ([first defined in `add_retriever_specific_args`](https://github.com/huggingface/transformers/blob/8217d4e37fce48490a68af7e8ce902af16318132/examples/rese... | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | # What does this PR do?
<!--
Congratulations! You've made it this far! You're not quite done yet though.
Once merged, your PR is going to appear in the release notes with the title you set, so make sure it's a great title that fully reflects the extent of your awesome contribution.
Then, please replace this w... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9355/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9355/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9355",
"html_url": "https://github.com/huggingface/transformers/pull/9355",
"diff_url": "https://github.com/huggingface/transformers/pull/9355.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9355.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9354 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9354/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9354/comments | https://api.github.com/repos/huggingface/transformers/issues/9354/events | https://github.com/huggingface/transformers/pull/9354 | 776,154,845 | MDExOlB1bGxSZXF1ZXN0NTQ2NjkwOTYz | 9,354 | [test_model_parallelization] multiple fixes | {
"login": "stas00",
"id": 10676103,
"node_id": "MDQ6VXNlcjEwNjc2MTAz",
"avatar_url": "https://avatars.githubusercontent.com/u/10676103?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/stas00",
"html_url": "https://github.com/stas00",
"followers_url": "https://api.github.com/users/stas00/fo... | [
{
"id": 2627272588,
"node_id": "MDU6TGFiZWwyNjI3MjcyNTg4",
"url": "https://api.github.com/repos/huggingface/transformers/labels/Model%20Parallel",
"name": "Model Parallel",
"color": "8B66A5",
"default": false,
"description": "Model Parallelilsm Implementations"
}
] | closed | false | null | [] | [
"Thank you for fixing!"
] | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | There are multiple issues with the current common `test_model_parallelization` test.
The main issue is that it uses `nvidia-smi` to take memory snapshots. This leads to 2 potential problems:
1. these tests must not be run with pytest distributed, as they rely on all the GPUs being unused - and with `-n 2` or higher... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9354/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9354/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9354",
"html_url": "https://github.com/huggingface/transformers/pull/9354",
"diff_url": "https://github.com/huggingface/transformers/pull/9354.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9354.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9353 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9353/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9353/comments | https://api.github.com/repos/huggingface/transformers/issues/9353/events | https://github.com/huggingface/transformers/pull/9353 | 776,141,112 | MDExOlB1bGxSZXF1ZXN0NTQ2NjgwMTM3 | 9,353 | Fixes crash when `compute_metrics` is not passed to `Trainer` in run_mlm example | {
"login": "galtay",
"id": 663051,
"node_id": "MDQ6VXNlcjY2MzA1MQ==",
"avatar_url": "https://avatars.githubusercontent.com/u/663051?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/galtay",
"html_url": "https://github.com/galtay",
"followers_url": "https://api.github.com/users/galtay/follow... | [] | closed | false | null | [] | [
"In master `train` always returns `metrics`.\r\n\r\nIs it possible that you are running the script from master but loading `transformers` that is pre-installed and it is not master? This metrics was added just recently.\r\n\r\nDo you still get the error if you do:\r\n```\r\ngit clone https://github.com/huggingface/... | 1,609 | 1,609 | 1,609 | NONE | null | # What does this PR do?
If I'm understanding the example run_mlm code correctly, the `metrics` attribute will not be present on the `train_result` object if `compute_metrics` is not passed to `Trainer`. This edit prevents the script from attempting to write the metrics to file if they don't exist.
## Who can rev... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9353/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9353/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9353",
"html_url": "https://github.com/huggingface/transformers/pull/9353",
"diff_url": "https://github.com/huggingface/transformers/pull/9353.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9353.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/9352 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9352/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9352/comments | https://api.github.com/repos/huggingface/transformers/issues/9352/events | https://github.com/huggingface/transformers/pull/9352 | 776,123,334 | MDExOlB1bGxSZXF1ZXN0NTQ2NjY1OTQy | 9,352 | [trainer] parametrize default output_dir | {
"login": "stas00",
"id": 10676103,
"node_id": "MDQ6VXNlcjEwNjc2MTAz",
"avatar_url": "https://avatars.githubusercontent.com/u/10676103?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/stas00",
"html_url": "https://github.com/stas00",
"followers_url": "https://api.github.com/users/stas00/fo... | [] | closed | false | null | [] | [] | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | This PR:
* fixes trainer to have the logger agree with the actual default `output_dir`, by setting it in one place and passing it as an argument to both places. The current logger falsely informs the user that `output_dir` is the current path, while using `tmp_trainer` as the path.
@patrickvonplaten, @sgugger
| {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9352/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9352/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9352",
"html_url": "https://github.com/huggingface/transformers/pull/9352",
"diff_url": "https://github.com/huggingface/transformers/pull/9352.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9352.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9351 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9351/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9351/comments | https://api.github.com/repos/huggingface/transformers/issues/9351/events | https://github.com/huggingface/transformers/issues/9351 | 776,113,709 | MDU6SXNzdWU3NzYxMTM3MDk= | 9,351 | XLNet evaluation on SQuAD | {
"login": "slvcsl",
"id": 25265140,
"node_id": "MDQ6VXNlcjI1MjY1MTQw",
"avatar_url": "https://avatars.githubusercontent.com/u/25265140?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/slvcsl",
"html_url": "https://github.com/slvcsl",
"followers_url": "https://api.github.com/users/slvcsl/fo... | [
{
"id": 2392046359,
"node_id": "MDU6TGFiZWwyMzkyMDQ2MzU5",
"url": "https://api.github.com/repos/huggingface/transformers/labels/Good%20Second%20Issue",
"name": "Good Second Issue",
"color": "dd935a",
"default": false,
"description": "Issues that are more difficult to do than \"Good First... | closed | false | null | [] | [
"Pinging @sgugger here. Think he has more knowledge about the training script than I do.",
"This is linked to [this issue](https://github.com/huggingface/tokenizers/issues/552) in the tokenizers repo. Until this is solved, the script `run_qa` does not work properly with XLNet (the offset mappings computed are inc... | 1,609 | 1,633 | 1,614 | NONE | null | ## Environment info
- `transformers` version: 4.2.0dev0
- Platform: Linux-5.3.0-64-generic-x86_64-with-debian-buster-sid
- Python version: 3.7.4
- PyTorch version (GPU?): 1.7.1+cu101 (True)
- Tensorflow version (GPU?): not installed (NA)
- Using GPU in script?: yes
- Using distributed or parallel set-up in scrip... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9351/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9351/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9350 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9350/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9350/comments | https://api.github.com/repos/huggingface/transformers/issues/9350/events | https://github.com/huggingface/transformers/pull/9350 | 776,101,060 | MDExOlB1bGxSZXF1ZXN0NTQ2NjQ3Nzg1 | 9,350 | [apex.normalizations.FusedLayerNorm] torch.cuda.is_available() is redundant as apex handles that internally | {
"login": "stas00",
"id": 10676103,
"node_id": "MDQ6VXNlcjEwNjc2MTAz",
"avatar_url": "https://avatars.githubusercontent.com/u/10676103?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/stas00",
"html_url": "https://github.com/stas00",
"followers_url": "https://api.github.com/users/stas00/fo... | [] | closed | false | null | [] | [
"Thanks a lot for digging into this @stas00 "
] | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | This PR is a follow up to https://github.com/huggingface/transformers/issues/9338
According to https://github.com/huggingface/transformers/issues/9338#issuecomment-752242098 we can just remove the `torch.cuda.is_available()` check before importing `apex.normalizations.FusedLayerNorm` and the multiprocess problem wil... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9350/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9350/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9350",
"html_url": "https://github.com/huggingface/transformers/pull/9350",
"diff_url": "https://github.com/huggingface/transformers/pull/9350.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9350.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9349 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9349/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9349/comments | https://api.github.com/repos/huggingface/transformers/issues/9349/events | https://github.com/huggingface/transformers/pull/9349 | 776,060,307 | MDExOlB1bGxSZXF1ZXN0NTQ2NjE0MDY4 | 9,349 | [prophetnet] wrong import | {
"login": "stas00",
"id": 10676103,
"node_id": "MDQ6VXNlcjEwNjc2MTAz",
"avatar_url": "https://avatars.githubusercontent.com/u/10676103?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/stas00",
"html_url": "https://github.com/stas00",
"followers_url": "https://api.github.com/users/stas00/fo... | [] | closed | false | null | [] | [] | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | ```
python -c "from apex.normalization import FusedProphetNetLayerNorm"
Traceback (most recent call last):
File "<string>", line 1, in <module>
ImportError: cannot import name 'FusedProphetNetLayerNorm' from 'apex.normalization' (/home/stas/anaconda3/envs/main-38/lib/python3.8/site-packages/apex/normalization/__i... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9349/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9349/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9349",
"html_url": "https://github.com/huggingface/transformers/pull/9349",
"diff_url": "https://github.com/huggingface/transformers/pull/9349.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9349.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9348 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9348/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9348/comments | https://api.github.com/repos/huggingface/transformers/issues/9348/events | https://github.com/huggingface/transformers/pull/9348 | 776,056,672 | MDExOlB1bGxSZXF1ZXN0NTQ2NjExMDQ2 | 9,348 | Fix TF Longformer | {
"login": "jplu",
"id": 959590,
"node_id": "MDQ6VXNlcjk1OTU5MA==",
"avatar_url": "https://avatars.githubusercontent.com/u/959590?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jplu",
"html_url": "https://github.com/jplu",
"followers_url": "https://api.github.com/users/jplu/followers",
... | [] | closed | false | null | [] | [
"I have already ran the slow tests as well and they all pass!"
] | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | # What does this PR do?
This PR aims to fix the TF Longformer version in order to make it graph compliant. As seen offline with @patrickvonplaten `all_global_attentions` now is added in the output when `output_attentions=True`. The global attentions are filled with zeros in case `is_global_attn` is False (see line 8... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9348/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9348/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9348",
"html_url": "https://github.com/huggingface/transformers/pull/9348",
"diff_url": "https://github.com/huggingface/transformers/pull/9348.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9348.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9347 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9347/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9347/comments | https://api.github.com/repos/huggingface/transformers/issues/9347/events | https://github.com/huggingface/transformers/pull/9347 | 776,045,660 | MDExOlB1bGxSZXF1ZXN0NTQ2NjAyMjE3 | 9,347 | [trainer] --model_parallel hasn't been implemented for most models | {
"login": "stas00",
"id": 10676103,
"node_id": "MDQ6VXNlcjEwNjc2MTAz",
"avatar_url": "https://avatars.githubusercontent.com/u/10676103?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/stas00",
"html_url": "https://github.com/stas00",
"followers_url": "https://api.github.com/users/stas00/fo... | [
{
"id": 2627272588,
"node_id": "MDU6TGFiZWwyNjI3MjcyNTg4",
"url": "https://api.github.com/repos/huggingface/transformers/labels/Model%20Parallel",
"name": "Model Parallel",
"color": "8B66A5",
"default": false,
"description": "Model Parallelilsm Implementations"
}
] | closed | false | null | [] | [
"@alexorona proposed to have the `model_parallel` method in `PreTrainedModel`, https://github.com/huggingface/transformers/pull/9323#issuecomment-752352280 which then would break this code as it'd be then present in all models.\r\n\r\nI see this PR as a quick band-aid since we released the new cl arg w/o checking t... | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | Apparently we unleashed `--model_parallel` in trainer w/o checking if the model supports MP (most don't). This PR:
* [x] checks whether the model supports MP and asserts otherwise
* [x] fixes the cl arg help to note that the flag will only work if the model supports MP
As we are gradually starting to build MP-s... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9347/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9347/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9347",
"html_url": "https://github.com/huggingface/transformers/pull/9347",
"diff_url": "https://github.com/huggingface/transformers/pull/9347.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9347.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9346 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9346/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9346/comments | https://api.github.com/repos/huggingface/transformers/issues/9346/events | https://github.com/huggingface/transformers/pull/9346 | 776,027,621 | MDExOlB1bGxSZXF1ZXN0NTQ2NTg4MTI3 | 9,346 | [Seq2Seq Templates] Add forgotten imports to templates | {
"login": "patrickvonplaten",
"id": 23423619,
"node_id": "MDQ6VXNlcjIzNDIzNjE5",
"avatar_url": "https://avatars.githubusercontent.com/u/23423619?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/patrickvonplaten",
"html_url": "https://github.com/patrickvonplaten",
"followers_url": "https://... | [] | closed | false | null | [] | [] | 1,609 | 1,609 | 1,609 | MEMBER | null | # What does this PR do?
<!--
Congratulations! You've made it this far! You're not quite done yet though.
Once merged, your PR is going to appear in the release notes with the title you set, so make sure it's a great title that fully reflects the extent of your awesome contribution.
Then, please replace this w... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9346/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9346/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9346",
"html_url": "https://github.com/huggingface/transformers/pull/9346",
"diff_url": "https://github.com/huggingface/transformers/pull/9346.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9346.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9345 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9345/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9345/comments | https://api.github.com/repos/huggingface/transformers/issues/9345/events | https://github.com/huggingface/transformers/issues/9345 | 776,012,959 | MDU6SXNzdWU3NzYwMTI5NTk= | 9,345 | Training of BART slow on TPU - aten ops investigation | {
"login": "phtephanx",
"id": 24647404,
"node_id": "MDQ6VXNlcjI0NjQ3NDA0",
"avatar_url": "https://avatars.githubusercontent.com/u/24647404?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/phtephanx",
"html_url": "https://github.com/phtephanx",
"followers_url": "https://api.github.com/users/... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"Hi @phtephanx \r\n\r\nThank you for troubleshooting the problem. I haven't really dived into this problem yet.\r\n\r\nMaybe @LysandreJik (our TPU expert ) will be able to help here once he's back from vacation ;)\r\n\r\nAlso cc @patrickvonplaten ",
"Hey @phtephanx,\r\n\r\nThanks for opening this issue! I'm actua... | 1,609 | 1,614 | 1,614 | NONE | null | Referencing: https://github.com/huggingface/transformers/issues/8339
Versions:
* `transformers==4.0.1`
* `pytorch==1.7.0`
* `pytorch_xla==1.7.0`
### problem
I've been trying to find out why training of `BartForConditionalGeneration` with `Trainer` is so **slow on TPU**. With slow I mean >30 min/batch (8000... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9345/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/huggingface/transformers/issues/9345/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9344 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9344/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9344/comments | https://api.github.com/repos/huggingface/transformers/issues/9344/events | https://github.com/huggingface/transformers/issues/9344 | 775,980,263 | MDU6SXNzdWU3NzU5ODAyNjM= | 9,344 | MBart prepare_seq2seq_batch | {
"login": "Chiyu-Song",
"id": 66252554,
"node_id": "MDQ6VXNlcjY2MjUyNTU0",
"avatar_url": "https://avatars.githubusercontent.com/u/66252554?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Chiyu-Song",
"html_url": "https://github.com/Chiyu-Song",
"followers_url": "https://api.github.com/use... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"Hi @Chiyu-Song \r\n\r\nNot sure what exactly is the issue here. What do you mean by \r\n> encoded target sequence from \"prepare_seq2seq_batch\" is inconsistent with its description\r\n\r\nAlso would be nice if the code snippet is formatted, bit hard to read :)",
"Thank you for your reply @patrickvonplaten, let... | 1,609 | 1,614 | 1,614 | NONE | null | - `transformers` version: 4.1.1
- mBART: @patrickvonplaten
## Information
Model I am using (Bert, XLNet ...): mBART
The problem arises when using:
* [x] the official example scripts: (give details below)
`example_english_phrase = "UN Chief Says There Is No Military Solution in Syria"
expected_translation_roman... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9344/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9344/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9343 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9343/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9343/comments | https://api.github.com/repos/huggingface/transformers/issues/9343/events | https://github.com/huggingface/transformers/pull/9343 | 775,963,195 | MDExOlB1bGxSZXF1ZXN0NTQ2NTQwNTI1 | 9,343 | [PyTorch Bart] Split Bart into different models | {
"login": "patrickvonplaten",
"id": 23423619,
"node_id": "MDQ6VXNlcjIzNDIzNjE5",
"avatar_url": "https://avatars.githubusercontent.com/u/23423619?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/patrickvonplaten",
"html_url": "https://github.com/patrickvonplaten",
"followers_url": "https://... | [] | closed | false | null | [] | [
"One important comment I forgot to add to my review: I don't think we should adapt the `research_project` to the new structure as it has been pinned to an earlier version of transformers (3.5.1). So apart from the duplicate file deleted, the other changes should be reverted IMO."
] | 1,609 | 1,609 | 1,609 | MEMBER | null | # What does this PR do?
This PR splits all Bart-like models into their own respective classes for PyTorch models only. This is more in line with the general philosophy of the library to have self-contained model files.
As discussed with @jplu, the TF models will be separated in a future PR as there are still some... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9343/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/huggingface/transformers/issues/9343/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9343",
"html_url": "https://github.com/huggingface/transformers/pull/9343",
"diff_url": "https://github.com/huggingface/transformers/pull/9343.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9343.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9342 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9342/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9342/comments | https://api.github.com/repos/huggingface/transformers/issues/9342/events | https://github.com/huggingface/transformers/pull/9342 | 775,958,486 | MDExOlB1bGxSZXF1ZXN0NTQ2NTM3MDY1 | 9,342 | [Seq2Seq Templates] Add embedding scale to templates | {
"login": "patrickvonplaten",
"id": 23423619,
"node_id": "MDQ6VXNlcjIzNDIzNjE5",
"avatar_url": "https://avatars.githubusercontent.com/u/23423619?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/patrickvonplaten",
"html_url": "https://github.com/patrickvonplaten",
"followers_url": "https://... | [] | closed | false | null | [] | [] | 1,609 | 1,609 | 1,609 | MEMBER | null | # What does this PR do?
<!--
Congratulations! You've made it this far! You're not quite done yet though.
Once merged, your PR is going to appear in the release notes with the title you set, so make sure it's a great title that fully reflects the extent of your awesome contribution.
Then, please replace this w... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9342/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9342/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9342",
"html_url": "https://github.com/huggingface/transformers/pull/9342",
"diff_url": "https://github.com/huggingface/transformers/pull/9342.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9342.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9341 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9341/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9341/comments | https://api.github.com/repos/huggingface/transformers/issues/9341/events | https://github.com/huggingface/transformers/pull/9341 | 775,954,409 | MDExOlB1bGxSZXF1ZXN0NTQ2NTM0MDA0 | 9,341 | [PyTorch Bart] Split Bart | {
"login": "patrickvonplaten",
"id": 23423619,
"node_id": "MDQ6VXNlcjIzNDIzNjE5",
"avatar_url": "https://avatars.githubusercontent.com/u/23423619?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/patrickvonplaten",
"html_url": "https://github.com/patrickvonplaten",
"followers_url": "https://... | [] | closed | false | null | [] | [] | 1,609 | 1,609 | 1,609 | MEMBER | null | # What does this PR do?
<!--
Congratulations! You've made it this far! You're not quite done yet though.
Once merged, your PR is going to appear in the release notes with the title you set, so make sure it's a great title that fully reflects the extent of your awesome contribution.
Then, please replace this w... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9341/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9341/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9341",
"html_url": "https://github.com/huggingface/transformers/pull/9341",
"diff_url": "https://github.com/huggingface/transformers/pull/9341.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9341.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/9340 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9340/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9340/comments | https://api.github.com/repos/huggingface/transformers/issues/9340/events | https://github.com/huggingface/transformers/issues/9340 | 775,912,595 | MDU6SXNzdWU3NzU5MTI1OTU= | 9,340 | Possible bug in `train_batch_size` | {
"login": "EmilyAlsentzer",
"id": 7334040,
"node_id": "MDQ6VXNlcjczMzQwNDA=",
"avatar_url": "https://avatars.githubusercontent.com/u/7334040?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/EmilyAlsentzer",
"html_url": "https://github.com/EmilyAlsentzer",
"followers_url": "https://api.gith... | [] | closed | false | {
"login": "sgugger",
"id": 35901082,
"node_id": "MDQ6VXNlcjM1OTAxMDgy",
"avatar_url": "https://avatars.githubusercontent.com/u/35901082?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sgugger",
"html_url": "https://github.com/sgugger",
"followers_url": "https://api.github.com/users/sgugge... | [
{
"login": "sgugger",
"id": 35901082,
"node_id": "MDQ6VXNlcjM1OTAxMDgy",
"avatar_url": "https://avatars.githubusercontent.com/u/35901082?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sgugger",
"html_url": "https://github.com/sgugger",
"followers_url": "https://api.git... | [
"Think @sgugger can best answer here when he's back from holiday :-) ",
"You misunderstand the flag `model_parallel`, it's not there to enable the use of several GPUs as this is done automatically by the `Trainer` (you have to set `CUDA_VISIBLE_DEVICES` to just one GPU if you don't want the Trainer to use them al... | 1,609 | 1,609 | 1,609 | CONTRIBUTOR | null | ## Environment info
- `transformers` version: 4.1.1
- Platform: Linux-4.4.0-62-generic-x86_64-with-glibc2.10
- Python version: 3.8.5
- PyTorch version (GPU?): 1.7.1 (True)
- Tensorflow version (GPU?): not installed (NA)
- Using GPU in script?: yes
- Using distributed or parallel set-up in script?: no
### Who ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9340/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9340/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9339 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9339/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9339/comments | https://api.github.com/repos/huggingface/transformers/issues/9339/events | https://github.com/huggingface/transformers/issues/9339 | 775,887,376 | MDU6SXNzdWU3NzU4ODczNzY= | 9,339 | Arrow file is too large when saving vector data | {
"login": "weiwangorg",
"id": 22360336,
"node_id": "MDQ6VXNlcjIyMzYwMzM2",
"avatar_url": "https://avatars.githubusercontent.com/u/22360336?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/weiwangorg",
"html_url": "https://github.com/weiwangorg",
"followers_url": "https://api.github.com/use... | [] | closed | false | null | [] | [
"I think it's expected because with `bert-base` each token will be embedded as a 768-dimensional vector. So if an example has n tokens then the size of embedding will be `n*768` and these are all 32-bits floating-point numbers.",
"Yes. I use datasets and I think this is a question about datasets, how to save vect... | 1,609 | 1,610 | 1,610 | NONE | null | I computed the sentence embedding of each sentence of bookcorpus data using bert base and saved them to disk. I used 20M sentences and the obtained arrow file is about 59GB while the original text file is only about 1.3GB. Are there any ways to reduce the size of the arrow file? | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9339/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9339/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9338 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9338/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9338/comments | https://api.github.com/repos/huggingface/transformers/issues/9338/events | https://github.com/huggingface/transformers/issues/9338 | 775,839,513 | MDU6SXNzdWU3NzU4Mzk1MTM= | 9,338 | Multiprocessing CUDA issues when importing transformers | {
"login": "maxjeblick",
"id": 24281881,
"node_id": "MDQ6VXNlcjI0MjgxODgx",
"avatar_url": "https://avatars.githubusercontent.com/u/24281881?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/maxjeblick",
"html_url": "https://github.com/maxjeblick",
"followers_url": "https://api.github.com/use... | [] | closed | false | null | [] | [
"Hey @maxjeblick,\r\n\r\nWe do need to be able to call `torch.cuda` in our model code. We cannot \"forbid\" calls to `torch.cuda` in order to allow quite specific use cases where `.cuda()` is mixed with multiprocessing. IMO, using multiprocessing in combination with `.cuda()` is quite an edge case. Are you trying t... | 1,609 | 1,610 | 1,610 | NONE | null | ## Environment info
- `transformers` version: 4.2.0dev0
- Platform: Linux-4.15.0-128-generic-x86_64-with-Ubuntu-18.04-bionic
- Python version: 3.6.9
- PyTorch version (GPU?): 1.6.0 (True)
- Tensorflow version (GPU?): not installed (NA)
- Using GPU in script?: Yes
- Using distributed or parallel set-up in script?... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9338/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9338/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9337 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9337/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9337/comments | https://api.github.com/repos/huggingface/transformers/issues/9337/events | https://github.com/huggingface/transformers/pull/9337 | 775,814,190 | MDExOlB1bGxSZXF1ZXN0NTQ2NDIxODEx | 9,337 | [WIP][Research projects] Add folder for Music AI / Music Transformers | {
"login": "asigalov61",
"id": 56325539,
"node_id": "MDQ6VXNlcjU2MzI1NTM5",
"avatar_url": "https://avatars.githubusercontent.com/u/56325539?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/asigalov61",
"html_url": "https://github.com/asigalov61",
"followers_url": "https://api.github.com/use... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"Hey @asigalov61, \r\n\r\nThe folder should be under `examples/research_projects` => so at `examples/research_projects/music_transformers`. There you are very free to add the files you want. If you want people to use your code, we suggest to add nice & readable code with a well-thought API, and a nice README.md tha... | 1,609 | 1,614 | 1,614 | NONE | null | A new dir specifically for Music AI/Music Transformers.
Created as suggested by Patrick von Platen.
I am still figuring out PRs so please correct this PR if I've done something wrong.
Thank you for your help/guidance and for the welcome to the Huggingface community :)
Looking forward to contributing what I... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9337/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9337/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9337",
"html_url": "https://github.com/huggingface/transformers/pull/9337",
"diff_url": "https://github.com/huggingface/transformers/pull/9337.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9337.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/9336 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9336/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9336/comments | https://api.github.com/repos/huggingface/transformers/issues/9336/events | https://github.com/huggingface/transformers/issues/9336 | 775,799,871 | MDU6SXNzdWU3NzU3OTk4NzE= | 9,336 | "RuntimeError: Input, output and indices must be on the current device" when trying to finetune MBart | {
"login": "mespla",
"id": 6533192,
"node_id": "MDQ6VXNlcjY1MzMxOTI=",
"avatar_url": "https://avatars.githubusercontent.com/u/6533192?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/mespla",
"html_url": "https://github.com/mespla",
"followers_url": "https://api.github.com/users/mespla/foll... | [] | closed | false | null | [] | [
"Hey @mespla,\r\n\r\nThanks for your issue! I'm afraid at the moment, we're really unsure whether we want to keep supporting all the bash scripts in `examples/seq2seq`. In a couple of weeks, we plan on having a single concise training script for seq2seq models.\r\n\r\ncc @sgugger \r\n\r\nAlso tagging @stas00, @pati... | 1,609 | 1,609 | 1,609 | NONE | null | ### Environment info
- Platform: Linux-4.15.0-123-generic-x86_64-with-glibc2.10
- Tried `transformers` versions 4.1.1 (installed with pip) and 4.2.2 (installed from master branch of the repository)
- Python version: 3.7
- PyTorch version: 1.7
- Tensorflow version: 2.4
- Number of available GPU: 2 (GeForce RTX 208... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9336/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9336/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9335 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9335/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9335/comments | https://api.github.com/repos/huggingface/transformers/issues/9335/events | https://github.com/huggingface/transformers/issues/9335 | 775,799,207 | MDU6SXNzdWU3NzU3OTkyMDc= | 9,335 | Data Loading as a Service | {
"login": "mingruimingrui",
"id": 18568364,
"node_id": "MDQ6VXNlcjE4NTY4MzY0",
"avatar_url": "https://avatars.githubusercontent.com/u/18568364?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/mingruimingrui",
"html_url": "https://github.com/mingruimingrui",
"followers_url": "https://api.gi... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"@lhoestq - this might be interesting for you! Any good tips from your side?",
"Interesting ! Cool features could be reading from s3, gcp etc.\r\nAlso maybe memory mapping can help speed up things a bit.\r\n\r\nStreaming datasets this way is something we'd like to add in the `datasets` library at one point since ... | 1,609 | 1,614 | 1,614 | CONTRIBUTOR | null | # 🚀 Not a Feature request, what am I here for then?
Well, mainly I'd just like to get some feedback from fellow software engineers. Some of the frustrations I've experienced might not have been big issues at all or there could have been easy ways to get around them which I've failed to notice. Getting roasted can b... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9335/reactions",
"total_count": 3,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 3
} | https://api.github.com/repos/huggingface/transformers/issues/9335/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9334 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9334/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9334/comments | https://api.github.com/repos/huggingface/transformers/issues/9334/events | https://github.com/huggingface/transformers/pull/9334 | 775,448,381 | MDExOlB1bGxSZXF1ZXN0NTQ2MTMwNjY5 | 9,334 | [Seq2Seq Templates] Correct some TF-serving errors and add gradient checkpointing to PT by default. | {
"login": "patrickvonplaten",
"id": 23423619,
"node_id": "MDQ6VXNlcjIzNDIzNjE5",
"avatar_url": "https://avatars.githubusercontent.com/u/23423619?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/patrickvonplaten",
"html_url": "https://github.com/patrickvonplaten",
"followers_url": "https://... | [] | closed | false | null | [] | [] | 1,609 | 1,609 | 1,609 | MEMBER | null | # What does this PR do?
This PR improves the Seq2Seq model templates.
Notably:
- a too model-specific test is removed from PyTorch
- gradient checkpointing is added to PyTorch
- some tf-serving incompatible statements are removed
| {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9334/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/huggingface/transformers/issues/9334/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9334",
"html_url": "https://github.com/huggingface/transformers/pull/9334",
"diff_url": "https://github.com/huggingface/transformers/pull/9334.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9334.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/9333 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9333/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9333/comments | https://api.github.com/repos/huggingface/transformers/issues/9333/events | https://github.com/huggingface/transformers/issues/9333 | 775,419,307 | MDU6SXNzdWU3NzU0MTkzMDc= | 9,333 | TF Longformer has some graph compilation/execution issue | {
"login": "jplu",
"id": 959590,
"node_id": "MDQ6VXNlcjk1OTU5MA==",
"avatar_url": "https://avatars.githubusercontent.com/u/959590?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jplu",
"html_url": "https://github.com/jplu",
"followers_url": "https://api.github.com/users/jplu/followers",
... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"Can we define `is_global_attn` in the config?",
"Or can we assume that `is_global_attn == output_attentions` ? The main issue here is that we cannot build a returned value that depends of a tensor that is created during the execution (same issue we had with `output_attentions` and `output_hidden_states` before ... | 1,609 | 1,614 | 1,614 | CONTRIBUTOR | null | TF longformer has the following issues to make it 100% graph compilation/execution compliant. I succeed to fix most of the issues but two still remains:
1. The first issue starts at line [1762](https://github.com/huggingface/transformers/blob/master/src/transformers/models/longformer/modeling_tf_longformer.py#L1762... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9333/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9333/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9332 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9332/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9332/comments | https://api.github.com/repos/huggingface/transformers/issues/9332/events | https://github.com/huggingface/transformers/issues/9332 | 775,394,791 | MDU6SXNzdWU3NzUzOTQ3OTE= | 9,332 | block sparse bert | {
"login": "pingpiang2019",
"id": 55866494,
"node_id": "MDQ6VXNlcjU1ODY2NDk0",
"avatar_url": "https://avatars.githubusercontent.com/u/55866494?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/pingpiang2019",
"html_url": "https://github.com/pingpiang2019",
"followers_url": "https://api.githu... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"I can produce your error. @madlag running the following code:\r\n\r\n```python\r\nfrom transformers import pipeline\r\n\r\nqa_pipeline = pipeline(\r\n \"question-answering\",\r\n model=\"madlag/bert-base-uncased-squad1.1-block-sparse-0.09-ampere-v1\",\r\n tokenizer=\"madlag/bert-base-uncased-squad1.1-bloc... | 1,609 | 1,614 | 1,614 | NONE | null | I got following error while running example usage from https://huggingface.co/madlag/bert-base-uncased-squad1.1-block-sparse-0.09-ampere-v1
Do I need specific torch or transformers setup? Thanks in advance!
Some weights of BertModel were not initialized from the model checkpoint at madlag/bert-base-uncased-squa... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9332/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9332/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/9331 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9331/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9331/comments | https://api.github.com/repos/huggingface/transformers/issues/9331/events | https://github.com/huggingface/transformers/pull/9331 | 775,390,487 | MDExOlB1bGxSZXF1ZXN0NTQ2MDg1NzUx | 9,331 | [WIP] Temp work on pipelines. | {
"login": "Narsil",
"id": 204321,
"node_id": "MDQ6VXNlcjIwNDMyMQ==",
"avatar_url": "https://avatars.githubusercontent.com/u/204321?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Narsil",
"html_url": "https://github.com/Narsil",
"followers_url": "https://api.github.com/users/Narsil/follow... | [] | closed | false | null | [] | [] | 1,609 | 1,614 | 1,614 | CONTRIBUTOR | null | # What does this PR do?
<!--
Congratulations! You've made it this far! You're not quite done yet though.
Once merged, your PR is going to appear in the release notes with the title you set, so make sure it's a great title that fully reflects the extent of your awesome contribution.
Then, please replace this with a d... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9331/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9331/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/9331",
"html_url": "https://github.com/huggingface/transformers/pull/9331",
"diff_url": "https://github.com/huggingface/transformers/pull/9331.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/9331.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/9330 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/9330/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/9330/comments | https://api.github.com/repos/huggingface/transformers/issues/9330/events | https://github.com/huggingface/transformers/issues/9330 | 775,374,598 | MDU6SXNzdWU3NzUzNzQ1OTg= | 9,330 | Fail to reload tokenizer from save_pretrained method | {
"login": "jc-hou",
"id": 30210529,
"node_id": "MDQ6VXNlcjMwMjEwNTI5",
"avatar_url": "https://avatars.githubusercontent.com/u/30210529?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jc-hou",
"html_url": "https://github.com/jc-hou",
"followers_url": "https://api.github.com/users/jc-hou/fo... | [] | closed | false | null | [] | [
"Hi @jc-hou,\r\n\r\nThe `Auto*` classes require the `config.json` (which is saved when you save the model) file to find the correct model/tokenizer class for loading the model/tokenizer. To directly load the tokenizer without the model use the specific tokenizer class, in this case, `BertTokenizer`.",
"Hi, thanks... | 1,609 | 1,609 | 1,609 | NONE | null | Hi,
to reproduce:
```
from transformers import AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained('bert-base-uncased')
tokenizer.save_pretrained(".")
tokenizer = AutoTokenizer.from_pretrained(".")
```
with error msg:
```
file ./config.json not found
Traceback (most recent call last):
File "/data/star... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/9330/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/9330/timeline | completed | null | null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.