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/3812 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3812/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3812/comments | https://api.github.com/repos/huggingface/transformers/issues/3812/events | https://github.com/huggingface/transformers/pull/3812 | 600,591,475 | MDExOlB1bGxSZXF1ZXN0NDAzOTkyMjE4 | 3,812 | Question Answering support for Albert and Roberta in TF | {
"login": "Pierrci",
"id": 5020707,
"node_id": "MDQ6VXNlcjUwMjA3MDc=",
"avatar_url": "https://avatars.githubusercontent.com/u/5020707?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Pierrci",
"html_url": "https://github.com/Pierrci",
"followers_url": "https://api.github.com/users/Pierrci/... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/3812?src=pr&el=h1) Report\n> Merging [#3812](https://codecov.io/gh/huggingface/transformers/pull/3812?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/301bf8d1b43d99efe1fdb5ba15871e975b3cb6cf&el=desc) will **increase... | 1,586 | 1,587 | 1,587 | MEMBER | null | This PR simply adds `TFRobertaForQuestionAnswering` and `TFAlbertForQuestionAnswering` classes (I needed them to do some model conversions!) | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3812/reactions",
"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/3812/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3812",
"html_url": "https://github.com/huggingface/transformers/pull/3812",
"diff_url": "https://github.com/huggingface/transformers/pull/3812.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3812.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/3811 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3811/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3811/comments | https://api.github.com/repos/huggingface/transformers/issues/3811/events | https://github.com/huggingface/transformers/issues/3811 | 600,433,686 | MDU6SXNzdWU2MDA0MzM2ODY= | 3,811 | Pre-trained BART performance on XSum lower than expected | {
"login": "morningmoni",
"id": 8191712,
"node_id": "MDQ6VXNlcjgxOTE3MTI=",
"avatar_url": "https://avatars.githubusercontent.com/u/8191712?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/morningmoni",
"html_url": "https://github.com/morningmoni",
"followers_url": "https://api.github.com/us... | [
{
"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'm having the [exact same issue](https://github.com/pytorch/fairseq/issues/1971), with the official BART code on fairseq.\r\n\r\nThe author is currently looking into it.",
"This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. ... | 1,586 | 1,612 | 1,592 | NONE | null | Greetings,
I am trying to reproduce BART's results on xsum using 'bart-large-xsum' and modified `examples/summarization/bart/evaluate_cnn.py` (max_length=60, min_length=10, beam=6, lenpen=1) but got lower ROUGE scores than reported.
I first obtained comparable results on CNNDM using 'bart-large-cnndm' and the ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3811/reactions",
"total_count": 1,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 1
} | https://api.github.com/repos/huggingface/transformers/issues/3811/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3810 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3810/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3810/comments | https://api.github.com/repos/huggingface/transformers/issues/3810/events | https://github.com/huggingface/transformers/issues/3810 | 600,413,860 | MDU6SXNzdWU2MDA0MTM4NjA= | 3,810 | run_glue.py example doesn't work for distilbert models | {
"login": "ereday",
"id": 13196191,
"node_id": "MDQ6VXNlcjEzMTk2MTkx",
"avatar_url": "https://avatars.githubusercontent.com/u/13196191?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ereday",
"html_url": "https://github.com/ereday",
"followers_url": "https://api.github.com/users/ereday/fo... | [] | closed | false | null | [] | [
"I think I might have broken this on `master` when merging #3688 🤔\r\n\r\nHi @ereday could you please try from the `trainer` branch described in PR #3800?\r\n\r\nOtherwise, hotfixing this in your code should be easy (just remove the `all_token_type_ids` line)"
] | 1,586 | 1,587 | 1,587 | NONE | null | # 🐛 Bug
## Information
Hi all,
I am succesfully able to run the run_glue.py example with BERT, XLNet and other architectures. However, when I try distilbert I got the following error:
```
Traceback (most recent call last):
File "run_glue.py", line 562, in <module>
main()
File "run_glue.py", line... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3810/reactions",
"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/3810/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3809 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3809/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3809/comments | https://api.github.com/repos/huggingface/transformers/issues/3809/events | https://github.com/huggingface/transformers/issues/3809 | 600,399,662 | MDU6SXNzdWU2MDAzOTk2NjI= | 3,809 | Roberta Tokenizer crashes when tokenizing empty string in 2.8.0 | {
"login": "HaokunLiu",
"id": 35565210,
"node_id": "MDQ6VXNlcjM1NTY1MjEw",
"avatar_url": "https://avatars.githubusercontent.com/u/35565210?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/HaokunLiu",
"html_url": "https://github.com/HaokunLiu",
"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
},
{
"id": 1834056635,
"node_id": "MDU6TGFiZWwxODM0MDU2NjM1",
... | closed | false | null | [] | [
"This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.\n",
"Fixed by #4209"
] | 1,586 | 1,592 | 1,592 | NONE | null | ```
import transformers
tokenizer = transformers.AutoTokenizer.from_pretrained("roberta-large")
tokenzer.tokenize("")
```
File "/scratch/wh629/nlu/env/lib/python3.7/site-packages/transformers/tokenization_utils.py", line 1054, in encode_plus
first_ids = get_input_ids(text)
File "/scratch/wh629/nlu/en... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3809/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/3809/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3808 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3808/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3808/comments | https://api.github.com/repos/huggingface/transformers/issues/3808/events | https://github.com/huggingface/transformers/pull/3808 | 600,384,026 | MDExOlB1bGxSZXF1ZXN0NDAzODI5ODE2 | 3,808 | typo: fine-grained token-leven | {
"login": "JonathanSum",
"id": 21982975,
"node_id": "MDQ6VXNlcjIxOTgyOTc1",
"avatar_url": "https://avatars.githubusercontent.com/u/21982975?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/JonathanSum",
"html_url": "https://github.com/JonathanSum",
"followers_url": "https://api.github.com/... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/3808?src=pr&el=h1) Report\n> Merging [#3808](https://codecov.io/gh/huggingface/transformers/pull/3808?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/301bf8d1b43d99efe1fdb5ba15871e975b3cb6cf&el=desc) will **not chan... | 1,586 | 1,587 | 1,587 | CONTRIBUTOR | null | Changing from "fine-grained token-leven" to "fine-grained token-level" | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3808/reactions",
"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/3808/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3808",
"html_url": "https://github.com/huggingface/transformers/pull/3808",
"diff_url": "https://github.com/huggingface/transformers/pull/3808.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3808.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/3807 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3807/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3807/comments | https://api.github.com/repos/huggingface/transformers/issues/3807/events | https://github.com/huggingface/transformers/pull/3807 | 600,321,361 | MDExOlB1bGxSZXF1ZXN0NDAzNzc4NzI3 | 3,807 | isort ignores examples directory | {
"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 | [] | [] | 1,586 | 1,588 | 1,588 | CONTRIBUTOR | null | Temporary solution while we wait for an isort release.
I have my local alias hacked in this way, but I figure new contributors might get confused by the circleci/local isort discrepancy. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3807/reactions",
"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/3807/timeline | null | true | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3807",
"html_url": "https://github.com/huggingface/transformers/pull/3807",
"diff_url": "https://github.com/huggingface/transformers/pull/3807.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3807.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/3806 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3806/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3806/comments | https://api.github.com/repos/huggingface/transformers/issues/3806/events | https://github.com/huggingface/transformers/pull/3806 | 600,232,717 | MDExOlB1bGxSZXF1ZXN0NDAzNzA3NDQ1 | 3,806 | [cleanup] factor out get_head_mask, invert_attn_mask, get_extended_attention_mask | {
"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 | [] | [
"This looks nice",
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/3806?src=pr&el=h1) Report\n> Merging [#3806](https://codecov.io/gh/huggingface/transformers/pull/3806?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/01c37dcdb529bff38aadd51001cb5812e5fe9b21&el=... | 1,586 | 1,587 | 1,587 | CONTRIBUTOR | null | This is three changes applied all over:
1) `get_head_mask` from @LysandreJik is used instead of redundant snippet
2) `get_extended_attention_mask` is used instead of redundant snippet (that also makes causal mask)
3) `invert_attention_mask` is used instead of redundant snippet that doesn't make causal mask.
The... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3806/reactions",
"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/3806/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3806",
"html_url": "https://github.com/huggingface/transformers/pull/3806",
"diff_url": "https://github.com/huggingface/transformers/pull/3806.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3806.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/3805 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3805/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3805/comments | https://api.github.com/repos/huggingface/transformers/issues/3805/events | https://github.com/huggingface/transformers/issues/3805 | 600,084,422 | MDU6SXNzdWU2MDAwODQ0MjI= | 3,805 | Using fill-mask pipeline to get the “score” for a result it didn't suggest | {
"login": "p-christ",
"id": 26346243,
"node_id": "MDQ6VXNlcjI2MzQ2MjQz",
"avatar_url": "https://avatars.githubusercontent.com/u/26346243?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/p-christ",
"html_url": "https://github.com/p-christ",
"followers_url": "https://api.github.com/users/p-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 | [] | [
"This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.\n"
] | 1,586 | 1,592 | 1,592 | NONE | null | Hi lovely huggingface people,
I'm trying to use your fill-mask pipeline in order to get the score for a result it didn't suggest.
For example, if my sentence is `"I ate bacon and <mask> for breakfast"` I can use `pipeline('fill-mask')` to get back predictions and their scores e.g. it might give me back `["eggs",... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3805/reactions",
"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/3805/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3804 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3804/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3804/comments | https://api.github.com/repos/huggingface/transformers/issues/3804/events | https://github.com/huggingface/transformers/issues/3804 | 600,023,544 | MDU6SXNzdWU2MDAwMjM1NDQ= | 3,804 | Calculated offsets are wrong in squad.py | {
"login": "ericperfect",
"id": 52102789,
"node_id": "MDQ6VXNlcjUyMTAyNzg5",
"avatar_url": "https://avatars.githubusercontent.com/u/52102789?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ericperfect",
"html_url": "https://github.com/ericperfect",
"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 because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.\n"
] | 1,586 | 1,592 | 1,592 | NONE | null | - when tokenizer.padding_side == "left" and if tokenizer.pad_token_id in span["input_ids"]:
doc_offset should add last_padding_id_position+1 instead of 0.
start_position = 0
end_position = 0
if is_training and not span_is_impossible:
# For training, if our document chun... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3804/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/3804/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3803 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3803/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3803/comments | https://api.github.com/repos/huggingface/transformers/issues/3803/events | https://github.com/huggingface/transformers/pull/3803 | 600,015,094 | MDExOlB1bGxSZXF1ZXN0NDAzNTM2OTQ2 | 3,803 | Fix bug in max_seq_length for preprocessing in ner example | {
"login": "r-tinn",
"id": 28840171,
"node_id": "MDQ6VXNlcjI4ODQwMTcx",
"avatar_url": "https://avatars.githubusercontent.com/u/28840171?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/r-tinn",
"html_url": "https://github.com/r-tinn",
"followers_url": "https://api.github.com/users/r-tinn/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 | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/3803?src=pr&el=h1) Report\n> Merging [#3803](https://codecov.io/gh/huggingface/transformers/pull/3803?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/01c37dcdb529bff38aadd51001cb5812e5fe9b21&el=desc) will **increase... | 1,586 | 1,594 | 1,594 | NONE | null | **Summary**
If you try to use a value for `max_seq_length` that is less than 128 in the NER example, the maximum sequence length is exceeded when the predictions are made. There is a warning logged for this "Maximum sequence length exceeded: No prediction for.." and predictions cannot be made for these tokens.
**... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3803/reactions",
"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/3803/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3803",
"html_url": "https://github.com/huggingface/transformers/pull/3803",
"diff_url": "https://github.com/huggingface/transformers/pull/3803.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3803.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/3802 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3802/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3802/comments | https://api.github.com/repos/huggingface/transformers/issues/3802/events | https://github.com/huggingface/transformers/pull/3802 | 600,008,109 | MDExOlB1bGxSZXF1ZXN0NDAzNTMxNzMz | 3,802 | Fix examples/translation/t5 to use newstest2014 rather than newstest2013 | {
"login": "tholiao",
"id": 12995527,
"node_id": "MDQ6VXNlcjEyOTk1NTI3",
"avatar_url": "https://avatars.githubusercontent.com/u/12995527?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/tholiao",
"html_url": "https://github.com/tholiao",
"followers_url": "https://api.github.com/users/tholia... | [] | 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... | [
"Hey @tholiao, \r\n\r\nThanks a lot for the PR :-) This looks good so far. Don't worry about running the script against the evaluation set - we can do this once this is merged! \r\n\r\nCan you make sure that the `run_examples_torch` pass? Don't worry too much about the `check_code_quality` test - there have been so... | 1,586 | 1,587 | 1,587 | CONTRIBUTOR | null | Resolves #3759, in addition to minor nits: fixed a bug with argparse arguments + more pythonic file handling + formatted with black and isort.
Please note that I have not yet run the evaluation script against the full newstest2014 test set, as it is rather compute intensive, so the disclaimer at the top of the READ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3802/reactions",
"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/3802/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3802",
"html_url": "https://github.com/huggingface/transformers/pull/3802",
"diff_url": "https://github.com/huggingface/transformers/pull/3802.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3802.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/3801 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3801/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3801/comments | https://api.github.com/repos/huggingface/transformers/issues/3801/events | https://github.com/huggingface/transformers/pull/3801 | 599,999,058 | MDExOlB1bGxSZXF1ZXN0NDAzNTI1MDY1 | 3,801 | Fix bug in GLUE example for models that do not require token_type_ids | {
"login": "douwekiela",
"id": 6024930,
"node_id": "MDQ6VXNlcjYwMjQ5MzA=",
"avatar_url": "https://avatars.githubusercontent.com/u/6024930?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/douwekiela",
"html_url": "https://github.com/douwekiela",
"followers_url": "https://api.github.com/users... | [] | closed | false | null | [] | [
"Hi @douwekiela, thanks for the PR. This should be fixed soon in a more stable way in the soon-to-be-merged #3800 \r\nLet us know if it works.",
"Should be fixed on master by #3800, please open a new issue if you encounter other problems."
] | 1,586 | 1,587 | 1,587 | NONE | null | If you try to run the `run_glue.py` example with e.g. roberta from a fresh install of the library, it errors out with the following error:
```
Traceback (most recent call last):
File "examples/run_glue.py", line 564, in <module>
main()
File "examples/run_glue.py", line 512, in main
train_dataset = l... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3801/reactions",
"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/3801/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3801",
"html_url": "https://github.com/huggingface/transformers/pull/3801",
"diff_url": "https://github.com/huggingface/transformers/pull/3801.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3801.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/3800 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3800/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3800/comments | https://api.github.com/repos/huggingface/transformers/issues/3800/events | https://github.com/huggingface/transformers/pull/3800 | 599,957,708 | MDExOlB1bGxSZXF1ZXN0NDAzNDkzODYw | 3,800 | Trainer | {
"login": "julien-c",
"id": 326577,
"node_id": "MDQ6VXNlcjMyNjU3Nw==",
"avatar_url": "https://avatars.githubusercontent.com/u/326577?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/julien-c",
"html_url": "https://github.com/julien-c",
"followers_url": "https://api.github.com/users/julien-... | [] | closed | true | null | [] | [
"So, thinking about it more, I've re-run a MNLI training with the previous default max_seq_length of 128 (instead of the \"new\" default of tokenizer.max_len – _soon to be renamed_), and training is naturally way faster with the smaller sequence length (light blue line below in relative time):\r\n\r\n<img width=\"1... | 1,586 | 1,587 | 1,587 | MEMBER | resolved | This is a bottom-up refactor of the example scripts (currently `run_glue.py`, `run_language_modeling.py`, `run_ner.py` and `run_multiple_choice.py`) into a Trainer class and associated utilities, as described in [trainer-proposal](https://github.com/julien-c/trainer-proposal).
# Regression test and benchmark on `run... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3800/reactions",
"total_count": 8,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 8,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/3800/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3800",
"html_url": "https://github.com/huggingface/transformers/pull/3800",
"diff_url": "https://github.com/huggingface/transformers/pull/3800.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3800.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/3799 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3799/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3799/comments | https://api.github.com/repos/huggingface/transformers/issues/3799/events | https://github.com/huggingface/transformers/issues/3799 | 599,934,554 | MDU6SXNzdWU1OTk5MzQ1NTQ= | 3,799 | Clarification about GPT2LMHeadModel lm_head weights | {
"login": "thesamuel",
"id": 6275391,
"node_id": "MDQ6VXNlcjYyNzUzOTE=",
"avatar_url": "https://avatars.githubusercontent.com/u/6275391?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/thesamuel",
"html_url": "https://github.com/thesamuel",
"followers_url": "https://api.github.com/users/th... | [] | 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... | [
"yes exactly - this should not be a problem :-) "
] | 1,586 | 1,587 | 1,587 | NONE | null | # ❓ Questions & Help
<!-- The GitHub issue tracker is primarly intended for bugs, feature requests,
new models and benchmarks, and migration questions. For all other questions,
we direct you to Stack Overflow (SO) where a whole community of PyTorch and
Tensorflow enthusiast can help you out. Make s... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3799/reactions",
"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/3799/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3798 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3798/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3798/comments | https://api.github.com/repos/huggingface/transformers/issues/3798/events | https://github.com/huggingface/transformers/issues/3798 | 599,894,141 | MDU6SXNzdWU1OTk4OTQxNDE= | 3,798 | Error when using run_generation.py to generate texts with long prompts, specifically for models -XLM and Openai-GPT | {
"login": "AdaUchendu",
"id": 32556160,
"node_id": "MDQ6VXNlcjMyNTU2MTYw",
"avatar_url": "https://avatars.githubusercontent.com/u/32556160?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/AdaUchendu",
"html_url": "https://github.com/AdaUchendu",
"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 because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.\n"
] | 1,586 | 1,592 | 1,592 | NONE | null | # 🐛 Bug
## Information
Model I am using (XLM, OPENAI-GPT):
Language I am using the model on (English):
The problem arises when using:
* [ ] The models to generate texts with long prompts
The tasks I am working on is:
* [ ] Text generation
## To reproduce
Steps to reproduce the behavior:
1. c... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3798/reactions",
"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/3798/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3797 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3797/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3797/comments | https://api.github.com/repos/huggingface/transformers/issues/3797/events | https://github.com/huggingface/transformers/pull/3797 | 599,850,719 | MDExOlB1bGxSZXF1ZXN0NDAzNDA3MzQ5 | 3,797 | [Config, Serialization] more readable config serialization | {
"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 | [] | [
"I would prefer **v2**) because the parameters saved in each of the model configuration files are important for the models behavior and would be nice to see in the config (also to compare to other models' configs)",
"Need to look into it more but on principle this is very nice. (If Configs were `dataclass`-backed... | 1,586 | 1,587 | 1,587 | MEMBER | null | Given the discussion in PR: #3433, we want to make the serialized model conf more readable.
### Problem:
E.g. `bert-base-cased` has the following config on S3:
```
{
"architectures": [
"BertForMaskedLM"
],
"attention_probs_dropout_prob": 0.1,
"hidden_act": "gelu",
"hidden_dropout_prob": 0.1... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3797/reactions",
"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/3797/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3797",
"html_url": "https://github.com/huggingface/transformers/pull/3797",
"diff_url": "https://github.com/huggingface/transformers/pull/3797.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3797.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/3796 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3796/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3796/comments | https://api.github.com/repos/huggingface/transformers/issues/3796/events | https://github.com/huggingface/transformers/issues/3796 | 599,849,691 | MDU6SXNzdWU1OTk4NDk2OTE= | 3,796 | Calculated offsets are wrong | {
"login": "dirkgr",
"id": 920638,
"node_id": "MDQ6VXNlcjkyMDYzOA==",
"avatar_url": "https://avatars.githubusercontent.com/u/920638?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/dirkgr",
"html_url": "https://github.com/dirkgr",
"followers_url": "https://api.github.com/users/dirkgr/follow... | [] | closed | false | null | [] | [
"It's probably normal since Roberta's tokenizers is a byte-level tokenizer which split words at byte-level (ie. possibly smaller than the character unit).\r\n\r\nCc @n1t0 ",
"There is a bug indeed, the offsets shouldn't be shifted after the `<mask>` token. I should be able to fix this.\r\nI'm not sure I'll be abl... | 1,586 | 1,587 | 1,587 | CONTRIBUTOR | null | This is on the latest `master` (from 2020-04-13):
```Python
import transformers
text = 'A, <mask> AllenNLP sentence.'
t = transformers.AutoTokenizer.from_pretrained("roberta-base", use_fast=True, add_special_tokens=True)
x2 = t.encode_plus(text, return_offsets_mapping=True)
print(repr(t.convert_ids_to_tokens(x2['... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3796/reactions",
"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/3796/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3795 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3795/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3795/comments | https://api.github.com/repos/huggingface/transformers/issues/3795/events | https://github.com/huggingface/transformers/pull/3795 | 599,827,938 | MDExOlB1bGxSZXF1ZXN0NDAzMzg5MDQ0 | 3,795 | [Pipelines] Clean pipelines test and remove unnecessary code | {
"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 | [] | [] | 1,586 | 1,587 | 1,587 | MEMBER | null | This PR cleans up pipelines a bit:
1) Fixes non-working pipeline creation test
2) Remove unnecessary code (due to PR #3116) in pipelines as discussed with @thomwolf in PR #3413
Note: Tested on QA pipelines slow tests. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3795/reactions",
"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/3795/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3795",
"html_url": "https://github.com/huggingface/transformers/pull/3795",
"diff_url": "https://github.com/huggingface/transformers/pull/3795.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3795.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/3794 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3794/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3794/comments | https://api.github.com/repos/huggingface/transformers/issues/3794/events | https://github.com/huggingface/transformers/issues/3794 | 599,796,794 | MDU6SXNzdWU1OTk3OTY3OTQ= | 3,794 | Getting large alloc error while evaluating bert-base on NER task | {
"login": "Sumegh-git",
"id": 37850881,
"node_id": "MDQ6VXNlcjM3ODUwODgx",
"avatar_url": "https://avatars.githubusercontent.com/u/37850881?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Sumegh-git",
"html_url": "https://github.com/Sumegh-git",
"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 because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.\n",
"I'm having the same issue. Training is fine but get this error when evaluate."
] | 1,586 | 1,605 | 1,592 | NONE | null | # 🐛 Bug
## Information
Model I am using (Bert-base-multilingual-cased,):
Language I am using the model on (English):
The problem arises when using:
Evaluation
The tasks I am working on is:
My own custom dataset having the same as GermEval task format.
Running on **Colab**.
I believe this is due... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3794/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/3794/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3793 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3793/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3793/comments | https://api.github.com/repos/huggingface/transformers/issues/3793/events | https://github.com/huggingface/transformers/pull/3793 | 599,782,461 | MDExOlB1bGxSZXF1ZXN0NDAzMzUyMzk1 | 3,793 | [Bert] remove hard-coded pad token id | {
"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 | [] | [
"LGTM"
] | 1,586 | 1,587 | 1,587 | MEMBER | null | tiny change to remove hard coded `pad_token_id` in Bert. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3793/reactions",
"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/3793/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3793",
"html_url": "https://github.com/huggingface/transformers/pull/3793",
"diff_url": "https://github.com/huggingface/transformers/pull/3793.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3793.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/3792 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3792/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3792/comments | https://api.github.com/repos/huggingface/transformers/issues/3792/events | https://github.com/huggingface/transformers/issues/3792 | 599,729,141 | MDU6SXNzdWU1OTk3MjkxNDE= | 3,792 | Using run_glue.py on external datasets for fine-tuning a RoBERTa classification model --> Is this possible? | {
"login": "seyonechithrananda",
"id": 46096704,
"node_id": "MDQ6VXNlcjQ2MDk2NzA0",
"avatar_url": "https://avatars.githubusercontent.com/u/46096704?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/seyonechithrananda",
"html_url": "https://github.com/seyonechithrananda",
"followers_url": "ht... | [] | closed | false | null | [] | [
"Hi @seyonechithrananda, do you have a `torch.data.Dataset` for your classification dataset?\r\n\r\nIf you do, this will be pretty easy following #3800 (i.e. the amount of code required should be pretty minimal)",
"Thanks for the response @julien-c! I originally grab a list of SMILES sequences and their correspon... | 1,586 | 1,588 | 1,588 | CONTRIBUTOR | null | # ❓ Questions & Help
<!-- The GitHub issue tracker is primarly intended for bugs, feature requests,
new models and benchmarks, and migration questions. For all other questions,
we direct you to Stack Overflow (SO) where a whole community of PyTorch and
Tensorflow enthusiast can help you out. Make s... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3792/reactions",
"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/3792/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3791 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3791/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3791/comments | https://api.github.com/repos/huggingface/transformers/issues/3791/events | https://github.com/huggingface/transformers/pull/3791 | 599,689,941 | MDExOlB1bGxSZXF1ZXN0NDAzMjc5ODY2 | 3,791 | XLM tokenizer should encode with bos token | {
"login": "LysandreJik",
"id": 30755778,
"node_id": "MDQ6VXNlcjMwNzU1Nzc4",
"avatar_url": "https://avatars.githubusercontent.com/u/30755778?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/LysandreJik",
"html_url": "https://github.com/LysandreJik",
"followers_url": "https://api.github.com/... | [] | closed | false | null | [] | [] | 1,586 | 1,587 | 1,587 | MEMBER | null | XLM tokenizer should behave according to the documentation
closes https://github.com/huggingface/transformers/issues/3788 | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3791/reactions",
"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/3791/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3791",
"html_url": "https://github.com/huggingface/transformers/pull/3791",
"diff_url": "https://github.com/huggingface/transformers/pull/3791.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3791.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/3790 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3790/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3790/comments | https://api.github.com/repos/huggingface/transformers/issues/3790/events | https://github.com/huggingface/transformers/pull/3790 | 599,649,255 | MDExOlB1bGxSZXF1ZXN0NDAzMjQ2Mzgw | 3,790 | Fix token_type_id in BERT question-answering example | {
"login": "siboehm",
"id": 14908678,
"node_id": "MDQ6VXNlcjE0OTA4Njc4",
"avatar_url": "https://avatars.githubusercontent.com/u/14908678?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/siboehm",
"html_url": "https://github.com/siboehm",
"followers_url": "https://api.github.com/users/siboeh... | [] | closed | false | null | [] | [] | 1,586 | 1,587 | 1,587 | CONTRIBUTOR | null | `token_type_id` wasn't being set correctly in the code examples for BERT Question answering. It's being turned into the sequence embedding, hence needs to highlight whether each token belongs to sequence 0 or 1.
For this small example the model returns the correct answer even though the parameter was incorrectly set... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3790/reactions",
"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/3790/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3790",
"html_url": "https://github.com/huggingface/transformers/pull/3790",
"diff_url": "https://github.com/huggingface/transformers/pull/3790.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3790.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/3789 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3789/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3789/comments | https://api.github.com/repos/huggingface/transformers/issues/3789/events | https://github.com/huggingface/transformers/issues/3789 | 599,584,722 | MDU6SXNzdWU1OTk1ODQ3MjI= | 3,789 | Is there a classical transformer model in the project? | {
"login": "980202006",
"id": 24452502,
"node_id": "MDQ6VXNlcjI0NDUyNTAy",
"avatar_url": "https://avatars.githubusercontent.com/u/24452502?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/980202006",
"html_url": "https://github.com/980202006",
"followers_url": "https://api.github.com/users/... | [] | closed | false | null | [] | [
"You might find this useful http://nlp.seas.harvard.edu/2018/04/03/attention.html\r\nI think PyTorch already have it implemented in their library\r\nhttps://pytorch.org/docs/stable/nn.html?highlight=transformer#torch.nn.Transformer",
"thank you，It is a answer for me."
] | 1,586 | 1,586 | 1,586 | NONE | null | Hi,
I am studying in some domain that need the original transformer which is from 《attention is what you all need》. Is there an implementation? | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3789/reactions",
"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/3789/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3788 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3788/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3788/comments | https://api.github.com/repos/huggingface/transformers/issues/3788/events | https://github.com/huggingface/transformers/issues/3788 | 599,548,172 | MDU6SXNzdWU1OTk1NDgxNzI= | 3,788 | Inconsistencies and possible bugs in different tokenizers | {
"login": "alanakbik",
"id": 18665324,
"node_id": "MDQ6VXNlcjE4NjY1MzI0",
"avatar_url": "https://avatars.githubusercontent.com/u/18665324?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/alanakbik",
"html_url": "https://github.com/alanakbik",
"followers_url": "https://api.github.com/users/... | [
{
"id": 1834056635,
"node_id": "MDU6TGFiZWwxODM0MDU2NjM1",
"url": "https://api.github.com/repos/huggingface/transformers/labels/Core:%20Tokenization",
"name": "Core: Tokenization",
"color": "FF4446",
"default": false,
"description": "Internals of the library; Tokenization."
},
{
... | closed | false | {
"login": "LysandreJik",
"id": 30755778,
"node_id": "MDQ6VXNlcjMwNzU1Nzc4",
"avatar_url": "https://avatars.githubusercontent.com/u/30755778?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/LysandreJik",
"html_url": "https://github.com/LysandreJik",
"followers_url": "https://api.github.com/... | [
{
"login": "LysandreJik",
"id": 30755778,
"node_id": "MDQ6VXNlcjMwNzU1Nzc4",
"avatar_url": "https://avatars.githubusercontent.com/u/30755778?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/LysandreJik",
"html_url": "https://github.com/LysandreJik",
"followers_url": "htt... | [
"Hi, and thanks for your report! Indeed some of these seem to be bugs.\r\n\r\n## XLM\r\n\r\nThis seems to be a bug. It doesn't behave as the documentation says, I'm looking into it. Encoding sequences with the bos token instead of the cls token should do the trick.\r\n\r\n## XLNet\r\n\r\nThe way XLNet encode sequen... | 1,586 | 1,587 | 1,587 | NONE | null | # 🐛 Bug
## Information
Over in [Flair](https://github.com/flairNLP/flair/pull/1494) we are integrating your awesome library in our embeddings interfaces. We are using the `AutoTokenizer` class to create one interface for all embeddings. We use the tokenizers to encode strings with special tokens.
However, we... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3788/reactions",
"total_count": 4,
"+1": 4,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/3788/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3787 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3787/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3787/comments | https://api.github.com/repos/huggingface/transformers/issues/3787/events | https://github.com/huggingface/transformers/issues/3787 | 599,506,289 | MDU6SXNzdWU1OTk1MDYyODk= | 3,787 | In just fouth blocks of the code of the colab notebook "01-training notebook", it just failed. | {
"login": "JonathanSum",
"id": 21982975,
"node_id": "MDQ6VXNlcjIxOTgyOTc1",
"avatar_url": "https://avatars.githubusercontent.com/u/21982975?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/JonathanSum",
"html_url": "https://github.com/JonathanSum",
"followers_url": "https://api.github.com/... | [] | closed | false | null | [] | [
"it seems to be that it can be solved by without installing it from pip",
"> it seems to be that it can be solved by without installing it from pip\r\n\r\ncould you please explain how you solved the issue ? did not understand.\r\nthanks,",
"@uunal, do you have this issue too?",
"@JonathanSum yes, can not find... | 1,586 | 1,587 | 1,586 | CONTRIBUTOR | null | # ❓ Questions & Help
<!-- The GitHub issue tracker is primarly intended for bugs, feature requests,
new models and benchmarks, and migration questions. For all other questions,
we direct you to Stack Overflow (SO) where a whole community of PyTorch and
Tensorflow enthusiast can help you out. Make s... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3787/reactions",
"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/3787/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3786 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3786/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3786/comments | https://api.github.com/repos/huggingface/transformers/issues/3786/events | https://github.com/huggingface/transformers/issues/3786 | 599,344,910 | MDU6SXNzdWU1OTkzNDQ5MTA= | 3,786 | Why force tokens in Bart decoding | {
"login": "chqiwang",
"id": 10123767,
"node_id": "MDQ6VXNlcjEwMTIzNzY3",
"avatar_url": "https://avatars.githubusercontent.com/u/10123767?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/chqiwang",
"html_url": "https://github.com/chqiwang",
"followers_url": "https://api.github.com/users/chq... | [] | 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... | [
"see https://github.com/huggingface/transformers/issues/3668",
"> see #3668\r\n\r\nthanks a lot!"
] | 1,586 | 1,586 | 1,586 | NONE | null | https://github.com/huggingface/transformers/blob/master/src/transformers/modeling_bart.py#L955
What's the meaning of this line? Why decode \<S\> when cur_len = 1? Why not when cur_len = 0? | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3786/reactions",
"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/3786/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3785 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3785/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3785/comments | https://api.github.com/repos/huggingface/transformers/issues/3785/events | https://github.com/huggingface/transformers/issues/3785 | 599,298,672 | MDU6SXNzdWU1OTkyOTg2NzI= | 3,785 | How to fine tune EncoderDecoder model for training a new corpus of data ? | {
"login": "banunitte",
"id": 6847024,
"node_id": "MDQ6VXNlcjY4NDcwMjQ=",
"avatar_url": "https://avatars.githubusercontent.com/u/6847024?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/banunitte",
"html_url": "https://github.com/banunitte",
"followers_url": "https://api.github.com/users/ba... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
},
{
"id": 1843738573,
"node_id": "MDU6TGFiZWwxODQzNzM4NTcz",
... | 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... | [
"We are currently working on implementing the encoder decoder framework. See PR: https://github.com/huggingface/transformers/pull/3383",
"I think in a week it should be ready :-) ",
"This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activit... | 1,586 | 1,596 | 1,596 | NONE | null | is there any documentation available for the same? | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3785/reactions",
"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/3785/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3784 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3784/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3784/comments | https://api.github.com/repos/huggingface/transformers/issues/3784/events | https://github.com/huggingface/transformers/issues/3784 | 599,270,233 | MDU6SXNzdWU1OTkyNzAyMzM= | 3,784 | Convert pytorch-pretrained-bert to new version (transformers) | {
"login": "Oxi84",
"id": 25420033,
"node_id": "MDQ6VXNlcjI1NDIwMDMz",
"avatar_url": "https://avatars.githubusercontent.com/u/25420033?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Oxi84",
"html_url": "https://github.com/Oxi84",
"followers_url": "https://api.github.com/users/Oxi84/follow... | [] | closed | false | null | [] | [
"Have you tried using the exact same methods `tokenize` and `convert_tokens_to_ids`?",
"```py\r\nfrom transformers import AlbertTokenizer, AlbertForMaskedLM\r\ntokenizer = AlbertTokenizer.from_pretrained('albert-base-v2)\r\nmodel = AlbertForMaskedLM.from_pretrained('albert-base-v2) \r\n\r\ntext = \"[CLS] Who wa... | 1,586 | 1,587 | 1,587 | NONE | null | So I have this code for albert:
```py
from transformers import AlbertTokenizer, AlbertForMaskedLM
tokenizer = AlbertTokenizer.from_pretrained('albert-base-v2')
model = AlbertForMaskedLM.from_pretrained('albert-base-v2')
input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3784/reactions",
"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/3784/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3783 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3783/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3783/comments | https://api.github.com/repos/huggingface/transformers/issues/3783/events | https://github.com/huggingface/transformers/issues/3783 | 599,216,990 | MDU6SXNzdWU1OTkyMTY5OTA= | 3,783 | Longformer, a scalable transformer model for long-document NLP tasks | {
"login": "bratao",
"id": 1090152,
"node_id": "MDQ6VXNlcjEwOTAxNTI=",
"avatar_url": "https://avatars.githubusercontent.com/u/1090152?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/bratao",
"html_url": "https://github.com/bratao",
"followers_url": "https://api.github.com/users/bratao/foll... | [] | 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... | [
"Any updates on this? Just curious.",
"Reformer will be added next week and then work will start on Longformer :-) ",
"Look forward to it!",
"Longformer is added now - closing!",
"@patrickvonplaten I have been using `Longformer` self attention with `LongBart` for summarisation recently and have done some si... | 1,586 | 1,594 | 1,591 | NONE | null | # 🌟 New model addition
## Model description
This is an incredible project from the awesome https://github.com/allenai team that solves a big problem in transformers.
From https://twitter.com/i_beltagy/status/1249750021811011591
Excited to share our work on Longformer, a scalable transformer model for long-do... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3783/reactions",
"total_count": 21,
"+1": 10,
"-1": 0,
"laugh": 0,
"hooray": 5,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 6
} | https://api.github.com/repos/huggingface/transformers/issues/3783/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3782 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3782/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3782/comments | https://api.github.com/repos/huggingface/transformers/issues/3782/events | https://github.com/huggingface/transformers/issues/3782 | 599,193,088 | MDU6SXNzdWU1OTkxOTMwODg= | 3,782 | Importing horovod.tensorflow crashes AlbertTokenizer but not BertTokenizer | {
"login": "jarednielsen",
"id": 4564897,
"node_id": "MDQ6VXNlcjQ1NjQ4OTc=",
"avatar_url": "https://avatars.githubusercontent.com/u/4564897?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jarednielsen",
"html_url": "https://github.com/jarednielsen",
"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 | {
"login": "LysandreJik",
"id": 30755778,
"node_id": "MDQ6VXNlcjMwNzU1Nzc4",
"avatar_url": "https://avatars.githubusercontent.com/u/30755778?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/LysandreJik",
"html_url": "https://github.com/LysandreJik",
"followers_url": "https://api.github.com/... | [
{
"login": "LysandreJik",
"id": 30755778,
"node_id": "MDQ6VXNlcjMwNzU1Nzc4",
"avatar_url": "https://avatars.githubusercontent.com/u/30755778?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/LysandreJik",
"html_url": "https://github.com/LysandreJik",
"followers_url": "htt... | [
"I think I remember someone mentioning this before. @LysandreJik does it ring any bell?",
"The error at the time was due to https://github.com/scipy/scipy/issues/11237. I'll look into it and try to reproduce @jarednielsen.",
"The `AlbertTokenizer` is using `SentencePiece` which is based on protobuffs. This seem... | 1,586 | 1,603 | 1,603 | CONTRIBUTOR | null | # 🐛 Bug
## Information
Albert tokenizers began crashing after I reorded my import statements with `isort`. Tracked down the bug to very strange behavior: importing `horovod.tensorflow` before `AlbertTokenizer` causes a crash, while importing `AlbertTokenizer` first does not. This behavior does not occur with `Be... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3782/reactions",
"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/3782/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3781 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3781/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3781/comments | https://api.github.com/repos/huggingface/transformers/issues/3781/events | https://github.com/huggingface/transformers/pull/3781 | 599,185,014 | MDExOlB1bGxSZXF1ZXN0NDAyODgxNDcx | 3,781 | Create model card | {
"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/... | [
{
"id": 1838412367,
"node_id": "MDU6TGFiZWwxODM4NDEyMzY3",
"url": "https://api.github.com/repos/huggingface/transformers/labels/model%20card",
"name": "model card",
"color": "92d5f4",
"default": false,
"description": "Related to pretrained model cards"
}
] | closed | false | null | [] | [] | 1,586 | 1,587 | 1,587 | CONTRIBUTOR | null | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3781/reactions",
"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/3781/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3781",
"html_url": "https://github.com/huggingface/transformers/pull/3781",
"diff_url": "https://github.com/huggingface/transformers/pull/3781.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3781.patch",
"merged_at": 1... | |
https://api.github.com/repos/huggingface/transformers/issues/3780 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3780/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3780/comments | https://api.github.com/repos/huggingface/transformers/issues/3780/events | https://github.com/huggingface/transformers/issues/3780 | 599,179,483 | MDU6SXNzdWU1OTkxNzk0ODM= | 3,780 | language modeling other models | {
"login": "urlocal12",
"id": 61215920,
"node_id": "MDQ6VXNlcjYxMjE1OTIw",
"avatar_url": "https://avatars.githubusercontent.com/u/61215920?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/urlocal12",
"html_url": "https://github.com/urlocal12",
"followers_url": "https://api.github.com/users/... | [] | closed | false | null | [] | [] | 1,586 | 1,586 | 1,586 | NONE | null | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3780/reactions",
"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/3780/timeline | completed | null | null | |
https://api.github.com/repos/huggingface/transformers/issues/3779 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3779/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3779/comments | https://api.github.com/repos/huggingface/transformers/issues/3779/events | https://github.com/huggingface/transformers/issues/3779 | 599,153,038 | MDU6SXNzdWU1OTkxNTMwMzg= | 3,779 | Problem when Converting a Fine-tuned Checkpoint from TF to PyTorch using ALBERTxxlargev1 Model | {
"login": "salrowili",
"id": 56635735,
"node_id": "MDQ6VXNlcjU2NjM1NzM1",
"avatar_url": "https://avatars.githubusercontent.com/u/56635735?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/salrowili",
"html_url": "https://github.com/salrowili",
"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 | {
"login": "LysandreJik",
"id": 30755778,
"node_id": "MDQ6VXNlcjMwNzU1Nzc4",
"avatar_url": "https://avatars.githubusercontent.com/u/30755778?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/LysandreJik",
"html_url": "https://github.com/LysandreJik",
"followers_url": "https://api.github.com/... | [
{
"login": "LysandreJik",
"id": 30755778,
"node_id": "MDQ6VXNlcjMwNzU1Nzc4",
"avatar_url": "https://avatars.githubusercontent.com/u/30755778?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/LysandreJik",
"html_url": "https://github.com/LysandreJik",
"followers_url": "htt... | [
"Same issue here. i still have an error. could you write answer here?",
"our hero LysandreJik assigned this problem to himself. Let's have confidence in him to solve it (:",
"You fine-tuned your TF checkpoint using the original implementation, is that correct?",
"Thanks you, but i find way resolved my problem... | 1,586 | 1,595 | 1,595 | NONE | null | # 🐛 Bug
## Information
Model I am using : ALBERTxxlargeV1
Language I am using the model on : English
The problem arises when using: Converting fine-tuned checkpoint from TF to PyTorch. No Problem with converting pre-trained checkpoints from TF.
* [ ] the official example scripts:
```
!python /content/t... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3779/reactions",
"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/3779/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3778 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3778/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3778/comments | https://api.github.com/repos/huggingface/transformers/issues/3778/events | https://github.com/huggingface/transformers/pull/3778 | 599,134,776 | MDExOlB1bGxSZXF1ZXN0NDAyODQwMDk5 | 3,778 | [Generation, EncoderDecoder] Apply Encoder Decoder 1.5GB memory savings to TF as well | {
"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 | [] | [
"Tested on `RUN SLOW=1 pytest tests/test_modeling_tf_t5.py` and all tests pass."
] | 1,586 | 1,586 | 1,586 | MEMBER | null | As was done by @sshleifer for torch, improved the memory usage for TF Encoder Decoder models.
Straight-forward translation of PR: https://github.com/huggingface/transformers/pull/3370. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3778/reactions",
"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/3778/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3778",
"html_url": "https://github.com/huggingface/transformers/pull/3778",
"diff_url": "https://github.com/huggingface/transformers/pull/3778.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3778.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/3777 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3777/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3777/comments | https://api.github.com/repos/huggingface/transformers/issues/3777/events | https://github.com/huggingface/transformers/pull/3777 | 599,059,182 | MDExOlB1bGxSZXF1ZXN0NDAyNzc4NTcw | 3,777 | [PretrainedTokenizer] Factor out tensor conversion method | {
"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 | [] | [] | 1,586 | 1,587 | 1,587 | CONTRIBUTOR | null | `MBartTokenizer` and `MarianTokenizer` will call the new method. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3777/reactions",
"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/3777/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3777",
"html_url": "https://github.com/huggingface/transformers/pull/3777",
"diff_url": "https://github.com/huggingface/transformers/pull/3777.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3777.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/3776 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3776/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3776/comments | https://api.github.com/repos/huggingface/transformers/issues/3776/events | https://github.com/huggingface/transformers/pull/3776 | 599,040,998 | MDExOlB1bGxSZXF1ZXN0NDAyNzY0ODIy | 3,776 | MBartTokenizer:add language codes | {
"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 | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/3776?src=pr&el=h1) Report\n> Merging [#3776](https://codecov.io/gh/huggingface/transformers/pull/3776?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/6e603cb7892b49a2cbbc10ba859759f92c3fb7a6&el=desc) will **increase... | 1,586 | 1,591 | 1,591 | CONTRIBUTOR | null | The mbart tokenizer is meant to
- not add bos token at the beginning
- end `input_ids` with [eos, src_lang_code]
- end `decoder_input_ids` with [eos, tgt_lang_code]
I have posted a fairseq issue to confirm this, but all that will change is the ordering of special tokens. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3776/reactions",
"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/3776/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3776",
"html_url": "https://github.com/huggingface/transformers/pull/3776",
"diff_url": "https://github.com/huggingface/transformers/pull/3776.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3776.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/3775 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3775/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3775/comments | https://api.github.com/repos/huggingface/transformers/issues/3775/events | https://github.com/huggingface/transformers/issues/3775 | 598,932,000 | MDU6SXNzdWU1OTg5MzIwMDA= | 3,775 | OpusNMT/MarianMT Machine Translation Models | {
"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... | [
{
"id": 1845609017,
"node_id": "MDU6TGFiZWwxODQ1NjA5MDE3",
"url": "https://api.github.com/repos/huggingface/transformers/labels/seq2seq",
"name": "seq2seq",
"color": "fef2c0",
"default": false,
"description": ""
}
] | closed | false | {
"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... | [
{
"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... | [
"So no way to get output logits?",
"The `forward` method returns logits, like other models with language modeling heads. Is that what you meant?"
] | 1,586 | 1,589 | 1,589 | CONTRIBUTOR | null | ### Model description
1,026 Language Pair Models, downloadable [here](http://opus.nlpl.eu/Opus-MT/)
Trained with Marian C++ [library](https://github.com/marian-nmt/marian)
### Open source status
* [ x] the model implementation is available: (give details)
* [ x] the model weights are available: (give details)
... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3775/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/3775/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3774 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3774/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3774/comments | https://api.github.com/repos/huggingface/transformers/issues/3774/events | https://github.com/huggingface/transformers/issues/3774 | 598,880,066 | MDU6SXNzdWU1OTg4ODAwNjY= | 3,774 | Making Simple whitespace tokenizer and then using that tokenizer to make a language model from scratch? | {
"login": "ishaansharma",
"id": 8963395,
"node_id": "MDQ6VXNlcjg5NjMzOTU=",
"avatar_url": "https://avatars.githubusercontent.com/u/8963395?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ishaansharma",
"html_url": "https://github.com/ishaansharma",
"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 | [] | [
"I am also looking for a solution to this problem.",
"I want the tokenizer to do something like this one\r\nhttps://github.com/huggingface/transformers/issues/1036#issuecomment-522201118",
"This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further ... | 1,586 | 1,683 | 1,592 | NONE | null | # ❓ How can I make a whitespace tokenizer and use it to build a language model from scratch using transformers.
<!-- The GitHub issue tracker is primarly intended for bugs, feature requests,
new models and benchmarks, and migration questions. For all other questions,
we direct you to Stack Overflow (SO) ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3774/reactions",
"total_count": 15,
"+1": 12,
"-1": 0,
"laugh": 0,
"hooray": 1,
"confused": 0,
"heart": 1,
"rocket": 1,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/3774/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3773 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3773/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3773/comments | https://api.github.com/repos/huggingface/transformers/issues/3773/events | https://github.com/huggingface/transformers/issues/3773 | 598,830,629 | MDU6SXNzdWU1OTg4MzA2Mjk= | 3,773 | Why the first item of the config.json of bert is "architectures": ["BertForMaskedLM"] | {
"login": "janyChan",
"id": 20128716,
"node_id": "MDQ6VXNlcjIwMTI4NzE2",
"avatar_url": "https://avatars.githubusercontent.com/u/20128716?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/janyChan",
"html_url": "https://github.com/janyChan",
"followers_url": "https://api.github.com/users/jan... | [] | closed | false | null | [] | [
"This is a property we added to be able to know what kind of final layers the model has. It is used for instance to enable tagging and filtering on our model hub: https://huggingface.co/models"
] | 1,586 | 1,586 | 1,586 | NONE | null | {
"architectures": [
"BertForMaskedLM"
],
"attention_probs_dropout_prob": 0.1,
"directionality": "bidi",
"hidden_act": "gelu",
"hidden_dropout_prob": 0.1,
"hidden_size": 768,
"initializer_range": 0.02,
"intermediate_size": 3072,
"max_position_embeddings": 512,
"num_attentio... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3773/reactions",
"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/3773/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3772 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3772/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3772/comments | https://api.github.com/repos/huggingface/transformers/issues/3772/events | https://github.com/huggingface/transformers/pull/3772 | 598,778,303 | MDExOlB1bGxSZXF1ZXN0NDAyNTU2Nzg1 | 3,772 | [TFT5, Cache] Add cache to TFT5 | {
"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 | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/3772?src=pr&el=h1) Report\n> Merging [#3772](https://codecov.io/gh/huggingface/transformers/pull/3772?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/7972a4019f4bc9f85fd358f42249b90f9cd27c68&el=desc) will **increase... | 1,586 | 1,587 | 1,587 | MEMBER | null | This PR adds caching for TF T5.
This PR is a straight-forward translation from the caching mechanism introduced in PR: #3682. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3772/reactions",
"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/3772/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3772",
"html_url": "https://github.com/huggingface/transformers/pull/3772",
"diff_url": "https://github.com/huggingface/transformers/pull/3772.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3772.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/3771 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3771/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3771/comments | https://api.github.com/repos/huggingface/transformers/issues/3771/events | https://github.com/huggingface/transformers/issues/3771 | 598,775,862 | MDU6SXNzdWU1OTg3NzU4NjI= | 3,771 | Cannot find the script | {
"login": "Mahmedturk",
"id": 48975334,
"node_id": "MDQ6VXNlcjQ4OTc1MzM0",
"avatar_url": "https://avatars.githubusercontent.com/u/48975334?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Mahmedturk",
"html_url": "https://github.com/Mahmedturk",
"followers_url": "https://api.github.com/use... | [] | closed | false | null | [] | [
"[it's in `./src/transformers`.](https://github.com/huggingface/transformers/tree/master/src/transformers)"
] | 1,586 | 1,586 | 1,586 | NONE | null | hi,
where can i find the script "convert_tf_checkpoint_to_pytorch.py" as i have to use BioBERT model for GLUE tasks. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3771/reactions",
"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/3771/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3770 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3770/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3770/comments | https://api.github.com/repos/huggingface/transformers/issues/3770/events | https://github.com/huggingface/transformers/issues/3770 | 598,746,692 | MDU6SXNzdWU1OTg3NDY2OTI= | 3,770 | Getting error AttributeError: 'BertOnlyMLMHead' object has no attribute 'bias' when giving TF path | {
"login": "kennysmith12",
"id": 61227472,
"node_id": "MDQ6VXNlcjYxMjI3NDcy",
"avatar_url": "https://avatars.githubusercontent.com/u/61227472?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/kennysmith12",
"html_url": "https://github.com/kennysmith12",
"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 | [] | [
"To load TF models you should use the TF class `TFBertForMaskedLM` and not `BertForMaskedLM` which is the PyTorch class.",
"Thanks for the response\r\nI found that BertForPreTraining also allows to loads when the TF model",
"This issue has been automatically marked as stale because it has not had recent activi... | 1,586 | 1,592 | 1,592 | NONE | null | # 🐛 Bug
## Information
Model I am using BERT:
Language I am using the model on English:
The problem arises when using:
the official example scripts:
The tasks I am working on is:
my own task or dataset: (give details below)
## To reproduce
Steps to reproduce the behavior:
I have this l... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3770/reactions",
"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/3770/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3769 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3769/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3769/comments | https://api.github.com/repos/huggingface/transformers/issues/3769/events | https://github.com/huggingface/transformers/issues/3769 | 598,655,107 | MDU6SXNzdWU1OTg2NTUxMDc= | 3,769 | Text generation with Transformer-XL stops at <eos> token. | {
"login": "urlocal12",
"id": 61215920,
"node_id": "MDQ6VXNlcjYxMjE1OTIw",
"avatar_url": "https://avatars.githubusercontent.com/u/61215920?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/urlocal12",
"html_url": "https://github.com/urlocal12",
"followers_url": "https://api.github.com/users/... | [
{
"id": 1834059054,
"node_id": "MDU6TGFiZWwxODM0MDU5MDU0",
"url": "https://api.github.com/repos/huggingface/transformers/labels/Ex:%20Generation",
"name": "Ex: Generation",
"color": "06EFF8",
"default": false,
"description": "Natural Language Generation"
}
] | 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... | [
"The `EOS` token stand for `End Of Sentence`, and is used as a STOP token.\r\n\r\nI.E. when the model generate this token, it literally means the generation is done and should be stop. You can control this behavior with the `min_length` option, which force the model to not produce `EOS` token before the minimum len... | 1,586 | 1,591 | 1,591 | NONE | null | Hi,
I was running the [text generation notebook](https://colab.research.google.com/github/huggingface/blog/blob/master/notebooks/02_how_to_generate.ipynb), but replaced the GPT-2 model with Transformer-XL, and when I tried to generate text it would always stop at the <<e>eos> token no matter what the max length was.... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3769/reactions",
"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/3769/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3768 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3768/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3768/comments | https://api.github.com/repos/huggingface/transformers/issues/3768/events | https://github.com/huggingface/transformers/pull/3768 | 598,561,567 | MDExOlB1bGxSZXF1ZXN0NDAyMzkxODcz | 3,768 | [PL examples]: fix progress bar bug | {
"login": "hugoabonizio",
"id": 1206395,
"node_id": "MDQ6VXNlcjEyMDYzOTU=",
"avatar_url": "https://avatars.githubusercontent.com/u/1206395?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/hugoabonizio",
"html_url": "https://github.com/hugoabonizio",
"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 because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.\n"
] | 1,586 | 1,594 | 1,594 | CONTRIBUTOR | null | As shown by @prabalbansal in #3576, the fine tuning script for seq2seq models is failing with the following error:
This PR includes the fix suggested by @sshleifer in [this comment](https://github.com/hugging... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3768/reactions",
"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/3768/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3768",
"html_url": "https://github.com/huggingface/transformers/pull/3768",
"diff_url": "https://github.com/huggingface/transformers/pull/3768.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3768.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/3767 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3767/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3767/comments | https://api.github.com/repos/huggingface/transformers/issues/3767/events | https://github.com/huggingface/transformers/issues/3767 | 598,546,948 | MDU6SXNzdWU1OTg1NDY5NDg= | 3,767 | Issues in Training GPT-2 Model from Scratch (Text Generation-Identifying Epoch Value-Perplexity Calculation) | {
"login": "mhd-git-test",
"id": 63552654,
"node_id": "MDQ6VXNlcjYzNTUyNjU0",
"avatar_url": "https://avatars.githubusercontent.com/u/63552654?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/mhd-git-test",
"html_url": "https://github.com/mhd-git-test",
"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 | [] | [
"Maybe your dataset is too small to make much sense even if you get a smaller perplexity.",
"Thanks for your response.\r\n\r\nHere are the sizes of my corpus:\r\n\r\n1. Training set ~ 77 MB\r\n2. Validation set ~ 10 MB\r\n3. Testing set ~ 10 MB\r\n\r\nI build this corpus as under:\r\n\r\n1. Distribute all files i... | 1,586 | 1,593 | 1,592 | NONE | null | Dear all,
I have trained a GPT-2 model from scratch by following a tutorial mentioned at this [link](https://huggingface.co/blog/how-to-train).
I am mentioning following important code snippets:
```
from pathlib import Path
from tokenizers import ByteLevelBPETokenizer
paths = [str(x) for x in Path(".").glob... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3767/reactions",
"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/3767/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3766 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3766/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3766/comments | https://api.github.com/repos/huggingface/transformers/issues/3766/events | https://github.com/huggingface/transformers/pull/3766 | 598,534,237 | MDExOlB1bGxSZXF1ZXN0NDAyMzcyODUx | 3,766 | Fix shuffling issue for distributed training (#3721) | {
"login": "elk-cloner",
"id": 5828101,
"node_id": "MDQ6VXNlcjU4MjgxMDE=",
"avatar_url": "https://avatars.githubusercontent.com/u/5828101?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/elk-cloner",
"html_url": "https://github.com/elk-cloner",
"followers_url": "https://api.github.com/users... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/3766?src=pr&el=h1) Report\n> Merging [#3766](https://codecov.io/gh/huggingface/transformers/pull/3766?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/7972a4019f4bc9f85fd358f42249b90f9cd27c68&el=desc) will **decrease... | 1,586 | 1,586 | 1,586 | CONTRIBUTOR | null | possible solution for issue [(#3721)](https://github.com/huggingface/transformers/issues/3721) | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3766/reactions",
"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/3766/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3766",
"html_url": "https://github.com/huggingface/transformers/pull/3766",
"diff_url": "https://github.com/huggingface/transformers/pull/3766.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3766.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/3765 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3765/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3765/comments | https://api.github.com/repos/huggingface/transformers/issues/3765/events | https://github.com/huggingface/transformers/issues/3765 | 598,512,981 | MDU6SXNzdWU1OTg1MTI5ODE= | 3,765 | Input format for a BertTokenClassification task | {
"login": "Kc2fresh",
"id": 20489184,
"node_id": "MDQ6VXNlcjIwNDg5MTg0",
"avatar_url": "https://avatars.githubusercontent.com/u/20489184?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Kc2fresh",
"html_url": "https://github.com/Kc2fresh",
"followers_url": "https://api.github.com/users/Kc2... | [
{
"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 because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.\n"
] | 1,586 | 1,592 | 1,592 | NONE | null | # ❓ Questions & Help
https://stackoverflow.com/questions/61168882/processing-and-handling-input-ids-when-using-bert-for-token-classification
So i want to use BERT for semantic entity extraction. This is not quite the same as NER or POS tagging.
for example eg- Given a sentence:
```
A=The leak could have been s... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3765/reactions",
"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/3765/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3764 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3764/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3764/comments | https://api.github.com/repos/huggingface/transformers/issues/3764/events | https://github.com/huggingface/transformers/issues/3764 | 598,497,988 | MDU6SXNzdWU1OTg0OTc5ODg= | 3,764 | long text classification | {
"login": "gogokre",
"id": 44871498,
"node_id": "MDQ6VXNlcjQ0ODcxNDk4",
"avatar_url": "https://avatars.githubusercontent.com/u/44871498?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/gogokre",
"html_url": "https://github.com/gogokre",
"followers_url": "https://api.github.com/users/gogokr... | [
{
"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 | [] | [
"Following!",
"> Following!\r\n\r\nI don't know what it means.\r\n",
"Use transformer-XL",
"Longformer is exactly designed for your use case (https://github.com/allenai/longformer)",
"This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further ac... | 1,586 | 1,623 | 1,593 | NONE | null | I want to binary classify documents with max length greater than 512 with bert.
Is there any reference or code for how to do this? | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3764/reactions",
"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/3764/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3763 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3763/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3763/comments | https://api.github.com/repos/huggingface/transformers/issues/3763/events | https://github.com/huggingface/transformers/pull/3763 | 598,493,015 | MDExOlB1bGxSZXF1ZXN0NDAyMzQ0MDA1 | 3,763 | [CI] Add CircleCI workflow to build docs for preview | {
"login": "harupy",
"id": 17039389,
"node_id": "MDQ6VXNlcjE3MDM5Mzg5",
"avatar_url": "https://avatars.githubusercontent.com/u/17039389?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/harupy",
"html_url": "https://github.com/harupy",
"followers_url": "https://api.github.com/users/harupy/fo... | [] | closed | false | null | [] | [
"The built documentation:\r\n\r\nhttps://30100-155220641-gh.circle-artifacts.com/0/docs/_build/html/index.html",
"How to view the built documentation:\r\n\r\n",
"@LysandreJik this might interest you!",
... | 1,586 | 1,587 | 1,587 | CONTRIBUTOR | null | This PR adds a CircleCI workflow to build the documentation and store it as an artifact so that we can preview it and verify it's rendered properly. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3763/reactions",
"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/3763/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3763",
"html_url": "https://github.com/huggingface/transformers/pull/3763",
"diff_url": "https://github.com/huggingface/transformers/pull/3763.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3763.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/3762 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3762/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3762/comments | https://api.github.com/repos/huggingface/transformers/issues/3762/events | https://github.com/huggingface/transformers/issues/3762 | 598,473,795 | MDU6SXNzdWU1OTg0NzM3OTU= | 3,762 | PPLM Write With Transformer demo not working | {
"login": "songproducer",
"id": 597346,
"node_id": "MDQ6VXNlcjU5NzM0Ng==",
"avatar_url": "https://avatars.githubusercontent.com/u/597346?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/songproducer",
"html_url": "https://github.com/songproducer",
"followers_url": "https://api.github.com/u... | [] | closed | false | null | [] | [
"cc @julien-c ",
"See https://github.com/huggingface/transformers/issues/4661#issuecomment-636911923"
] | 1,586 | 1,591 | 1,591 | NONE | null | # 🐛 Bug
## Information
Model I am using: PPLM
Language I am using the model on (English, Chinese ...): English
The problem arises when using:
* [ ] the official example scripts: (give details below)
ios Safari and Firefox
The tasks I am working on is:
Trying out the demo
## To reproduce
Steps t... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3762/reactions",
"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/3762/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3761 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3761/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3761/comments | https://api.github.com/repos/huggingface/transformers/issues/3761/events | https://github.com/huggingface/transformers/issues/3761 | 598,462,759 | MDU6SXNzdWU1OTg0NjI3NTk= | 3,761 | Summarization pipeline fails to initialize | {
"login": "singulart",
"id": 7863785,
"node_id": "MDQ6VXNlcjc4NjM3ODU=",
"avatar_url": "https://avatars.githubusercontent.com/u/7863785?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/singulart",
"html_url": "https://github.com/singulart",
"followers_url": "https://api.github.com/users/si... | [] | 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... | [
"You're using `transformers` version **2.6.0**, but T5 model was released in version **2.7.0**.\r\n\r\nUpdate your library with :\r\n\r\n`pip install --upgrade transformers`",
"Thanks, error is not reproduced in 2.8.0"
] | 1,586 | 1,586 | 1,586 | NONE | null | # 🐛 Bug
## Information
Model I am using (T5):
Language I am using the model on (Englis):
The problem arises when using:
* [+] the official example scripts: (give details below)
https://huggingface.co/transformers/main_classes/pipelines.html#transformers.SummarizationPipeline
* [ ] my own modified script... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3761/reactions",
"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/3761/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3760 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3760/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3760/comments | https://api.github.com/repos/huggingface/transformers/issues/3760/events | https://github.com/huggingface/transformers/issues/3760 | 598,438,038 | MDU6SXNzdWU1OTg0MzgwMzg= | 3,760 | Quick question difference output of Bert models compared to Electra | {
"login": "Stuffooh",
"id": 50005268,
"node_id": "MDQ6VXNlcjUwMDA1MjY4",
"avatar_url": "https://avatars.githubusercontent.com/u/50005268?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Stuffooh",
"html_url": "https://github.com/Stuffooh",
"followers_url": "https://api.github.com/users/Stu... | [] | closed | false | null | [] | [
"For questions like this it's often best to dive into the source code. It helps understanding it a lot.\r\n\r\nBut your example seems wrong, you may have made some copy-pasting errors. If `output[1][0].shape = [768]` then also, because tensors across an axis must have same dimensions (like matrices) `output[2][0].s... | 1,586 | 1,586 | 1,586 | NONE | null | Hi everyone,
I am a little confused at the moment.
When I run `outputs = model.roberta(input_ids, attention_mask)` or `model.albert(input_ids, attention_mask)` the length of output is 3 and looks like this:
`output[0].shape = [4, 249, 768]`
`output[1][0].shape = [768]`
`output[2][0].shape = [4,249,768]`
W... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3760/reactions",
"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/3760/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3759 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3759/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3759/comments | https://api.github.com/repos/huggingface/transformers/issues/3759/events | https://github.com/huggingface/transformers/issues/3759 | 598,429,201 | MDU6SXNzdWU1OTg0MjkyMDE= | 3,759 | Why does `examples/translation/t5` test on newstest2013 rather than newstest2014? | {
"login": "tholiao",
"id": 12995527,
"node_id": "MDQ6VXNlcjEyOTk1NTI3",
"avatar_url": "https://avatars.githubusercontent.com/u/12995527?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/tholiao",
"html_url": "https://github.com/tholiao",
"followers_url": "https://api.github.com/users/tholia... | [] | 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... | [
"Hey @tholiao,\r\n\r\nThanks for the catch! You're right it should be the newstest2014! Do you want to open a PR to change it? Or I can do it as well",
"No worries, I'll submit a PR.",
"@patrickvonplaten, would you mind running evaluate_wmt.py on this branch and computing sacreBLEU via command line? (`cat newst... | 1,586 | 1,587 | 1,587 | CONTRIBUTOR | null | # Details
<!-- Description of your issue -->
The example in examples/translation/t5 uses `newstest2013`, but authors report against `newstest2014` (presumably newstest2014.full):
> Since this is our final set of experiments, we report results on the test set rather than
the validation set. For CNN/Daily Mail, we us... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3759/reactions",
"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/3759/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3758 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3758/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3758/comments | https://api.github.com/repos/huggingface/transformers/issues/3758/events | https://github.com/huggingface/transformers/pull/3758 | 598,424,469 | MDExOlB1bGxSZXF1ZXN0NDAyMjk3MzA2 | 3,758 | Pipeline for Text Generation: GenerationPipeline | {
"login": "enzoampil",
"id": 39557688,
"node_id": "MDQ6VXNlcjM5NTU3Njg4",
"avatar_url": "https://avatars.githubusercontent.com/u/39557688?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/enzoampil",
"html_url": "https://github.com/enzoampil",
"followers_url": "https://api.github.com/users/... | [] | 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... | [
"Hi @enzoampil, \r\n\r\nThanks again for the PR - I reviewed it. I think we can start by deleting a lot of the code and keeping it simple. It can be quite hard to get used to all the \"under-the-hood\" behavior that happens in pipelines. I think we should stick to the format that was used for the `summarization` pi... | 1,586 | 1,587 | 1,587 | CONTRIBUTOR | null | ### This PR implements a text generation pipeline, `GenerationPipeline`, which works on any `ModelWithLMHead` head, and resolves issue #3728
This pipeline predicts the words that will follow a specified text prompt for autoregressive language models. I've registered it to the pipeline function using `gpt2` as the de... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3758/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/3758/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3758",
"html_url": "https://github.com/huggingface/transformers/pull/3758",
"diff_url": "https://github.com/huggingface/transformers/pull/3758.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3758.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/3757 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3757/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3757/comments | https://api.github.com/repos/huggingface/transformers/issues/3757/events | https://github.com/huggingface/transformers/issues/3757 | 598,345,015 | MDU6SXNzdWU1OTgzNDUwMTU= | 3,757 | Dealing with class imbalance | {
"login": "al-yakubovich",
"id": 12928778,
"node_id": "MDQ6VXNlcjEyOTI4Nzc4",
"avatar_url": "https://avatars.githubusercontent.com/u/12928778?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/al-yakubovich",
"html_url": "https://github.com/al-yakubovich",
"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 | [] | [
"Did you find anything on this? Been digging around for the same, seems like `from_pretrained` used to allow `weight` as an arg?",
"This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.\n"
] | 1,586 | 1,595 | 1,595 | NONE | null | Are there any built-in methods for dealing with class imbalance in BERT? | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3757/reactions",
"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/3757/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3756 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3756/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3756/comments | https://api.github.com/repos/huggingface/transformers/issues/3756/events | https://github.com/huggingface/transformers/pull/3756 | 598,319,008 | MDExOlB1bGxSZXF1ZXN0NDAyMjI1Mzky | 3,756 | Trace log probs on generation | {
"login": "aced125",
"id": 44452903,
"node_id": "MDQ6VXNlcjQ0NDUyOTAz",
"avatar_url": "https://avatars.githubusercontent.com/u/44452903?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/aced125",
"html_url": "https://github.com/aced125",
"followers_url": "https://api.github.com/users/aced12... | [
{
"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 | {
"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... | [
"Thanks for the PR @aced125 - could you run the slow tests as well and see whether they pass? \r\nI think checking these three tests should be good enough:\r\n\r\n`RUN_SLOW=1 pytest tests/test_modeling_gpt2.py`\r\n`RUN_SLOW=1 pytest tests/test_modeling_t5.py`\r\n`RUN_SLOW=1 pytest tests/test_modeling_bart.py`",
"... | 1,586 | 1,659 | 1,594 | NONE | null | This PR makes a **few code-line changes** to accomplish the following:
- We want to trace the log probabilities of tokens generated during generation, so that we can do policy gradient methods (e.g we want to improve ROUGE scores for summarization by using RL).
- This requires keeping track of the computation gra... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3756/reactions",
"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/3756/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3756",
"html_url": "https://github.com/huggingface/transformers/pull/3756",
"diff_url": "https://github.com/huggingface/transformers/pull/3756.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3756.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/3755 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3755/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3755/comments | https://api.github.com/repos/huggingface/transformers/issues/3755/events | https://github.com/huggingface/transformers/pull/3755 | 598,301,731 | MDExOlB1bGxSZXF1ZXN0NDAyMjEzNjE2 | 3,755 | [Docs] Add DialoGPT | {
"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 | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/3755?src=pr&el=h1) Report\n> Merging [#3755](https://codecov.io/gh/huggingface/transformers/pull/3755?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/7972a4019f4bc9f85fd358f42249b90f9cd27c68&el=desc) will **increase... | 1,586 | 1,587 | 1,587 | MEMBER | null | This PR adds DialoGPT to the model page and links the models on the model page https://github.com/huggingface/transformers#model-architectures to the docs. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3755/reactions",
"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/3755/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3755",
"html_url": "https://github.com/huggingface/transformers/pull/3755",
"diff_url": "https://github.com/huggingface/transformers/pull/3755.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3755.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/3754 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3754/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3754/comments | https://api.github.com/repos/huggingface/transformers/issues/3754/events | https://github.com/huggingface/transformers/issues/3754 | 598,288,060 | MDU6SXNzdWU1OTgyODgwNjA= | 3,754 | Deprecation warning due to invalid escape sequences in Python 3.7 | {
"login": "tirkarthi",
"id": 3972343,
"node_id": "MDQ6VXNlcjM5NzIzNDM=",
"avatar_url": "https://avatars.githubusercontent.com/u/3972343?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/tirkarthi",
"html_url": "https://github.com/tirkarthi",
"followers_url": "https://api.github.com/users/ti... | [] | closed | false | null | [] | [
"This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.\n",
"The bug is still valid and I have raised https://github.com/huggingface/transformers/pull/4924"
] | 1,586 | 1,592 | 1,592 | CONTRIBUTOR | null | # 🐛 Bug
## To reproduce
Deprecation warnings are raised due to invalid escape sequences. This can be fixed by using raw strings or escaping the literals.
Steps to reproduce the behavior:
```
find . -iname '*.py' | grep -v example | xargs -P 4 -I{} python3.8 -Wall -m py_compile {}
./src/transformers/tok... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3754/reactions",
"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/3754/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3753 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3753/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3753/comments | https://api.github.com/repos/huggingface/transformers/issues/3753/events | https://github.com/huggingface/transformers/issues/3753 | 598,273,944 | MDU6SXNzdWU1OTgyNzM5NDQ= | 3,753 | How to speed up the transformer inference? | {
"login": "hahadashi",
"id": 7497649,
"node_id": "MDQ6VXNlcjc0OTc2NDk=",
"avatar_url": "https://avatars.githubusercontent.com/u/7497649?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/hahadashi",
"html_url": "https://github.com/hahadashi",
"followers_url": "https://api.github.com/users/ha... | [] | 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... | [
"Hi @hahadashi, \r\n\r\nCan you add a code snippet so that we know which model you are using and so that we can reproduce the behavior? ",
"> Hi @hahadashi,\r\n> \r\n> Can you add a code snippet so that we know which model you are using and so that we can reproduce the behavior?\r\n\r\nthx your response, i use th... | 1,586 | 1,586 | 1,586 | NONE | null | i meet a problem, if i direct use the model to inference, it is very slow. and i try to split the model (encoder, decoder), i freeze enc ckpt model and dec ckpt model to enc pb model and dec pb model(use tf C++ to run). only compute enc pb model one time and use dec pb model to complete prediction. but its speed is als... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3753/reactions",
"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/3753/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3752 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3752/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3752/comments | https://api.github.com/repos/huggingface/transformers/issues/3752/events | https://github.com/huggingface/transformers/issues/3752 | 598,272,095 | MDU6SXNzdWU1OTgyNzIwOTU= | 3,752 | uss | {
"login": "hahadashi",
"id": 7497649,
"node_id": "MDQ6VXNlcjc0OTc2NDk=",
"avatar_url": "https://avatars.githubusercontent.com/u/7497649?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/hahadashi",
"html_url": "https://github.com/hahadashi",
"followers_url": "https://api.github.com/users/ha... | [] | 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... | [
"Was this opened by mistake? Can we close it? "
] | 1,586 | 1,586 | 1,586 | NONE | null | # ❓ Questions & Help
<!-- The GitHub issue tracker is primarly intended for bugs, feature requests,
new models and benchmarks, and migration questions. For all other questions,
we direct you to Stack Overflow (SO) where a whole community of PyTorch and
Tensorflow enthusiast can help you out. Make s... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3752/reactions",
"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/3752/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3751 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3751/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3751/comments | https://api.github.com/repos/huggingface/transformers/issues/3751/events | https://github.com/huggingface/transformers/issues/3751 | 598,259,638 | MDU6SXNzdWU1OTgyNTk2Mzg= | 3,751 | Etract all last hidden states of the input dequences for Question and answering Bert | {
"login": "LincLabUCCS",
"id": 30666434,
"node_id": "MDQ6VXNlcjMwNjY2NDM0",
"avatar_url": "https://avatars.githubusercontent.com/u/30666434?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/LincLabUCCS",
"html_url": "https://github.com/LincLabUCCS",
"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 | [] | [
"Modify the script :)",
"This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.\n"
] | 1,586 | 1,592 | 1,592 | NONE | null | Hello everyone,
I am using run_squad.py to finetune Bert for question answering. I would like to save and later extract all last hidden states for all sequences for further use. For example, if I have a dataset of 100 sequences, 64 length, and 768 features. I will eventually have a tensor of (100,64,768).
Is there... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3751/reactions",
"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/3751/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3750 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3750/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3750/comments | https://api.github.com/repos/huggingface/transformers/issues/3750/events | https://github.com/huggingface/transformers/issues/3750 | 598,248,109 | MDU6SXNzdWU1OTgyNDgxMDk= | 3,750 | ImportError: cannot import name 'HfArgumentParser' from 'transformers' | {
"login": "songproducer",
"id": 597346,
"node_id": "MDQ6VXNlcjU5NzM0Ng==",
"avatar_url": "https://avatars.githubusercontent.com/u/597346?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/songproducer",
"html_url": "https://github.com/songproducer",
"followers_url": "https://api.github.com/u... | [] | closed | false | null | [] | [
"You need to install from source, as explained [here](https://github.com/huggingface/transformers#run-the-examples).",
"Thanks!"
] | 1,586 | 1,586 | 1,586 | NONE | null | Hi! When running:
```
python ./examples/run_glue.py \
--model_type bert \
--model_name_or_path bert-base-uncased \
--task_name $TASK_NAME \
--do_train \
--do_eval \
--data_dir $GLUE_DIR/$TASK_NAME \
--max_seq_length 128 \
--per_gpu_eval_batch_size=8 \
--per_gpu_train_b... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3750/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/3750/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3749 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3749/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3749/comments | https://api.github.com/repos/huggingface/transformers/issues/3749/events | https://github.com/huggingface/transformers/issues/3749 | 598,216,515 | MDU6SXNzdWU1OTgyMTY1MTU= | 3,749 | Question about whitespace filtering in squad data processor | {
"login": "Santosh-Gupta",
"id": 5524261,
"node_id": "MDQ6VXNlcjU1MjQyNjE=",
"avatar_url": "https://avatars.githubusercontent.com/u/5524261?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Santosh-Gupta",
"html_url": "https://github.com/Santosh-Gupta",
"followers_url": "https://api.github.... | [
{
"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 because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.\n"
] | 1,586 | 1,592 | 1,592 | CONTRIBUTOR | null | Usually to detect for whitespaces in python, the `isspace()` built-in is used. But I noticed for the squad data processor, this is used instead
```
def _is_whitespace(c):
if c == " " or c == "\t" or c == "\r" or c == "\n" or ord(c) == 0x202F:
return True
return False
```
https://github.com/... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3749/reactions",
"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/3749/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3748 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3748/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3748/comments | https://api.github.com/repos/huggingface/transformers/issues/3748/events | https://github.com/huggingface/transformers/issues/3748 | 598,212,227 | MDU6SXNzdWU1OTgyMTIyMjc= | 3,748 | Slow training time on BERT pretraining on multiple gpu compare to single gpu | {
"login": "ntubertchen",
"id": 7036778,
"node_id": "MDQ6VXNlcjcwMzY3Nzg=",
"avatar_url": "https://avatars.githubusercontent.com/u/7036778?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ntubertchen",
"html_url": "https://github.com/ntubertchen",
"followers_url": "https://api.github.com/us... | [] | closed | false | null | [] | [
"I solve this problem with distributed training.\r\n\r\nThe DataParallel is really slow with a lot of defects, just don't use it.\r\n\r\n",
"Hi @ntubertchen, how did you make distributed training to work with the `run_language_modeling.py` script? Or did you do everything from scratch?",
"run_language_modeling.... | 1,586 | 1,595 | 1,586 | NONE | null | # ❓ Questions & Help
Hello, I'm pretraining BERT on two RTX 2080Ti with 15 batch size on each gpu. They update every 40 steps.
The training time to 28000 steps of 2 2080Ti is around 9 days and it only takes 1 2080Ti 10 days to achieve the same training time. I was wondering whether is training time is accurate or... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3748/reactions",
"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/3748/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3747 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3747/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3747/comments | https://api.github.com/repos/huggingface/transformers/issues/3747/events | https://github.com/huggingface/transformers/issues/3747 | 598,190,302 | MDU6SXNzdWU1OTgxOTAzMDI= | 3,747 | text generation like lorem ipsum but human readable | {
"login": "ghost",
"id": 10137,
"node_id": "MDQ6VXNlcjEwMTM3",
"avatar_url": "https://avatars.githubusercontent.com/u/10137?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ghost",
"html_url": "https://github.com/ghost",
"followers_url": "https://api.github.com/users/ghost/followers",
"f... | [
{
"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 | [] | [
"For conditional output generation, you might want to take a look at `CTRL` or you would have to fine-tuned gpt2. To just produce \"any\" fake text, you could just use `GPT2` out of the box as in this demo: \r\nhttps://transformer.huggingface.co/doc/gpt2-large",
"Thanks for your reply :-)\r\n\r\nWhat is cTRL ? d... | 1,586 | 1,598 | 1,598 | NONE | null | Hi guys,
Hope you are all well !
We would like to add a fake text generator based on transformers/gpt-2 for the wordpress module called https://github.com/bordoni/fakerpress. For now, It uses only an old and not comprehensible lorem ipsum generator.
Is it possible to generate fake text (paragraphs, headings, t... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3747/reactions",
"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/3747/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3746 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3746/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3746/comments | https://api.github.com/repos/huggingface/transformers/issues/3746/events | https://github.com/huggingface/transformers/pull/3746 | 598,179,848 | MDExOlB1bGxSZXF1ZXN0NDAyMTMyNzE5 | 3,746 | Added README huseinzol05/albert-tiny-bahasa-cased | {
"login": "huseinzol05",
"id": 19810909,
"node_id": "MDQ6VXNlcjE5ODEwOTA5",
"avatar_url": "https://avatars.githubusercontent.com/u/19810909?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/huseinzol05",
"html_url": "https://github.com/huseinzol05",
"followers_url": "https://api.github.com/... | [
{
"id": 1838412367,
"node_id": "MDU6TGFiZWwxODM4NDEyMzY3",
"url": "https://api.github.com/repos/huggingface/transformers/labels/model%20card",
"name": "model card",
"color": "92d5f4",
"default": false,
"description": "Related to pretrained model cards"
}
] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/3746?src=pr&el=h1) Report\n> Merging [#3746](https://codecov.io/gh/huggingface/transformers/pull/3746?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/700ccf6e35616fcbee59de81edd60cec9e14fb6b&el=desc) will **increase... | 1,586 | 1,586 | 1,586 | CONTRIBUTOR | null | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3746/reactions",
"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/3746/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3746",
"html_url": "https://github.com/huggingface/transformers/pull/3746",
"diff_url": "https://github.com/huggingface/transformers/pull/3746.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3746.patch",
"merged_at": 1... | |
https://api.github.com/repos/huggingface/transformers/issues/3745 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3745/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3745/comments | https://api.github.com/repos/huggingface/transformers/issues/3745/events | https://github.com/huggingface/transformers/pull/3745 | 598,166,246 | MDExOlB1bGxSZXF1ZXN0NDAyMTIzNDY2 | 3,745 | Add `qas_id` to SquadResult and SquadExample | {
"login": "jarednielsen",
"id": 4564897,
"node_id": "MDQ6VXNlcjQ1NjQ4OTc=",
"avatar_url": "https://avatars.githubusercontent.com/u/4564897?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jarednielsen",
"html_url": "https://github.com/jarednielsen",
"followers_url": "https://api.github.com... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/3745?src=pr&el=h1) Report\n> Merging [#3745](https://codecov.io/gh/huggingface/transformers/pull/3745?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/700ccf6e35616fcbee59de81edd60cec9e14fb6b&el=desc) will **increase... | 1,586 | 1,588 | 1,587 | CONTRIBUTOR | null | I'm in the process of adding a `run_tf_squad.py` script, per https://github.com/huggingface/transformers/issues/3685.
This PR:
- Fixes a buggy variable name: `all_example_indices` actually refers to feature indices, so I've renamed it to `all_feature_indices`. This can be verified by adding a breakpoint at https:... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3745/reactions",
"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/3745/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3745",
"html_url": "https://github.com/huggingface/transformers/pull/3745",
"diff_url": "https://github.com/huggingface/transformers/pull/3745.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3745.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/3744 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3744/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3744/comments | https://api.github.com/repos/huggingface/transformers/issues/3744/events | https://github.com/huggingface/transformers/issues/3744 | 598,074,403 | MDU6SXNzdWU1OTgwNzQ0MDM= | 3,744 | Turning off Verbosity on QA model using Pipeline | {
"login": "WeiyangSun",
"id": 34964824,
"node_id": "MDQ6VXNlcjM0OTY0ODI0",
"avatar_url": "https://avatars.githubusercontent.com/u/34964824?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/WeiyangSun",
"html_url": "https://github.com/WeiyangSun",
"followers_url": "https://api.github.com/use... | [] | closed | false | {
"login": "mfuntowicz",
"id": 2241520,
"node_id": "MDQ6VXNlcjIyNDE1MjA=",
"avatar_url": "https://avatars.githubusercontent.com/u/2241520?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/mfuntowicz",
"html_url": "https://github.com/mfuntowicz",
"followers_url": "https://api.github.com/users... | [
{
"login": "mfuntowicz",
"id": 2241520,
"node_id": "MDQ6VXNlcjIyNDE1MjA=",
"avatar_url": "https://avatars.githubusercontent.com/u/2241520?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/mfuntowicz",
"html_url": "https://github.com/mfuntowicz",
"followers_url": "https://... | [
"I was just wondering the same thing."
] | 1,586 | 1,587 | 1,587 | NONE | null | Hi,
I am looking for a way to turn off the logs warnings. Please refer to the screenshot.
This is currently what I am doing.
![image](https://user-images.githubusercontent.com/34964824/79020549-ba2... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3744/reactions",
"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/3744/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3743 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3743/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3743/comments | https://api.github.com/repos/huggingface/transformers/issues/3743/events | https://github.com/huggingface/transformers/pull/3743 | 598,066,734 | MDExOlB1bGxSZXF1ZXN0NDAyMDQ3MDAy | 3,743 | JIT not compatible with PyTorch/XLA | {
"login": "LysandreJik",
"id": 30755778,
"node_id": "MDQ6VXNlcjMwNzU1Nzc4",
"avatar_url": "https://avatars.githubusercontent.com/u/30755778?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/LysandreJik",
"html_url": "https://github.com/LysandreJik",
"followers_url": "https://api.github.com/... | [] | closed | false | null | [] | [
"cc @jysohn23 "
] | 1,586 | 1,587 | 1,587 | MEMBER | null | Tracing with JIT is not supported by TPUs. If `torch_xla` is detected in the environment, the `gelu_new` method won't be traced.
If tracing is done, the line:
```py
model = xm.send_cpu_data_to_device(model, xm.xla_device())
```
in `modeling_utils.py`
will raise:
```py
TypeError: can't pickle torch._C.Sc... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3743/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/3743/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3743",
"html_url": "https://github.com/huggingface/transformers/pull/3743",
"diff_url": "https://github.com/huggingface/transformers/pull/3743.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3743.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/3742 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3742/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3742/comments | https://api.github.com/repos/huggingface/transformers/issues/3742/events | https://github.com/huggingface/transformers/pull/3742 | 598,065,848 | MDExOlB1bGxSZXF1ZXN0NDAyMDQ2MzA1 | 3,742 | Fix `glue_convert_examples_to_features` API breakage | {
"login": "jysohn23",
"id": 19496130,
"node_id": "MDQ6VXNlcjE5NDk2MTMw",
"avatar_url": "https://avatars.githubusercontent.com/u/19496130?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jysohn23",
"html_url": "https://github.com/jysohn23",
"followers_url": "https://api.github.com/users/jys... | [] | closed | false | null | [] | [
"👍 "
] | 1,586 | 1,586 | 1,586 | COLLABORATOR | null | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3742/reactions",
"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/3742/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3742",
"html_url": "https://github.com/huggingface/transformers/pull/3742",
"diff_url": "https://github.com/huggingface/transformers/pull/3742.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3742.patch",
"merged_at": 1... | |
https://api.github.com/repos/huggingface/transformers/issues/3741 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3741/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3741/comments | https://api.github.com/repos/huggingface/transformers/issues/3741/events | https://github.com/huggingface/transformers/issues/3741 | 598,058,940 | MDU6SXNzdWU1OTgwNTg5NDA= | 3,741 | Tokenizer Encode More than 2 inputs | {
"login": "pertschuk",
"id": 6379823,
"node_id": "MDQ6VXNlcjYzNzk4MjM=",
"avatar_url": "https://avatars.githubusercontent.com/u/6379823?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/pertschuk",
"html_url": "https://github.com/pertschuk",
"followers_url": "https://api.github.com/users/pe... | [
{
"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 because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.\n",
"Had this problem been solved?",
"@Zhylkaaa It seems that the problem is still unsolved. Are there any easier ways to encode more tha... | 1,586 | 1,627 | 1,592 | NONE | null | # 🚀 Feature request
Increasingly I'm seeing more than 2 inputs in some cases to BERT model, separated by [SEP] tokens. Often this helps by including context or for pairwise search ranking.
## Motivation
Right now I have to manually add the [SEP] token and concat the output to two tokenizer.encode_plus calls.... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3741/reactions",
"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/3741/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3740 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3740/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3740/comments | https://api.github.com/repos/huggingface/transformers/issues/3740/events | https://github.com/huggingface/transformers/pull/3740 | 597,992,046 | MDExOlB1bGxSZXF1ZXN0NDAxOTg4ODg3 | 3,740 | [WIP] EncoderDecoder model that works | {
"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 | [] | [] | 1,586 | 1,586 | 1,586 | CONTRIBUTOR | null | Continuing #3383 from @patrickvonplaten to facilitate MarianMT project.
### Targeted API:
```python
model = EncoderDecoderModel.from_model_names('bert-base-uncased', 'bert-based-uncased')
model.save_pretrained('bert2bert')
model.from_pretrained('bert2bert') # way 2
```
### TODO
- support test_common
- tes... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3740/reactions",
"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/3740/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3740",
"html_url": "https://github.com/huggingface/transformers/pull/3740",
"diff_url": "https://github.com/huggingface/transformers/pull/3740.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3740.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/3739 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3739/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3739/comments | https://api.github.com/repos/huggingface/transformers/issues/3739/events | https://github.com/huggingface/transformers/pull/3739 | 597,988,814 | MDExOlB1bGxSZXF1ZXN0NDAxOTg2NDk2 | 3,739 | Seq2seq generation with prefix | {
"login": "yjernite",
"id": 10469459,
"node_id": "MDQ6VXNlcjEwNDY5NDU5",
"avatar_url": "https://avatars.githubusercontent.com/u/10469459?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/yjernite",
"html_url": "https://github.com/yjernite",
"followers_url": "https://api.github.com/users/yje... | [
{
"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 | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/3739?src=pr&el=h1) Report\n> Merging [#3739](https://codecov.io/gh/huggingface/transformers/pull/3739?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/7a7fdf71f80452fcae064bd016f06e9a0f0f19ed&el=desc) will **decrease... | 1,586 | 1,594 | 1,594 | MEMBER | null | This PR introduces two small changes in the way model.generate() works.
Previously, the function took in an input_ids argument which had different behaviors in the seq2seq and language model settings: in language modeling, input_ids could be used to provide a prefix for the generation, while in seq2seq, input_ids re... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3739/reactions",
"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/3739/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3739",
"html_url": "https://github.com/huggingface/transformers/pull/3739",
"diff_url": "https://github.com/huggingface/transformers/pull/3739.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3739.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/3738 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3738/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3738/comments | https://api.github.com/repos/huggingface/transformers/issues/3738/events | https://github.com/huggingface/transformers/pull/3738 | 597,977,636 | MDExOlB1bGxSZXF1ZXN0NDAxOTc4MDcw | 3,738 | [docs] The use of `do_lower_case` in scripts is on its way to depreca… | {
"login": "julien-c",
"id": 326577,
"node_id": "MDQ6VXNlcjMyNjU3Nw==",
"avatar_url": "https://avatars.githubusercontent.com/u/326577?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/julien-c",
"html_url": "https://github.com/julien-c",
"followers_url": "https://api.github.com/users/julien-... | [] | closed | false | null | [] | [] | 1,586 | 1,586 | 1,586 | MEMBER | null | …tion
Will close #3633
Will close #3584
Will close #3491 | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3738/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/3738/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3738",
"html_url": "https://github.com/huggingface/transformers/pull/3738",
"diff_url": "https://github.com/huggingface/transformers/pull/3738.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3738.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/3737 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3737/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3737/comments | https://api.github.com/repos/huggingface/transformers/issues/3737/events | https://github.com/huggingface/transformers/issues/3737 | 597,939,058 | MDU6SXNzdWU1OTc5MzkwNTg= | 3,737 | Seq2Seq: decoder hidden_states shape not tested | {
"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 | {
"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... | [
{
"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... | [
"same for `T5`. I thought a bit about how to correctly sort the `encoder_hidden_states`, `decoder_hidden_states` output (if the user wants to output it). I think it's not easy at all to implement this cleanly...namedtuples would make this so much easier, so maybe just wait until we add those? ",
"Makes sense. \r\... | 1,586 | 1,651 | 1,591 | CONTRIBUTOR | null | [this line] points at encoder hidden states for Bart.
https://github.com/huggingface/transformers/blob/2ee410560e45ae3c619dc1e0b0fc4d257c48e18a/tests/test_modeling_common.py#L464 | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3737/reactions",
"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/3737/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3736 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3736/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3736/comments | https://api.github.com/repos/huggingface/transformers/issues/3736/events | https://github.com/huggingface/transformers/pull/3736 | 597,932,963 | MDExOlB1bGxSZXF1ZXN0NDAxOTQyOTgz | 3,736 | updated dutch squad model card | {
"login": "borhenryk",
"id": 35457598,
"node_id": "MDQ6VXNlcjM1NDU3NTk4",
"avatar_url": "https://avatars.githubusercontent.com/u/35457598?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/borhenryk",
"html_url": "https://github.com/borhenryk",
"followers_url": "https://api.github.com/users/... | [] | closed | false | null | [] | [] | 1,586 | 1,586 | 1,586 | CONTRIBUTOR | null | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3736/reactions",
"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/3736/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3736",
"html_url": "https://github.com/huggingface/transformers/pull/3736",
"diff_url": "https://github.com/huggingface/transformers/pull/3736.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3736.patch",
"merged_at": 1... | |
https://api.github.com/repos/huggingface/transformers/issues/3735 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3735/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3735/comments | https://api.github.com/repos/huggingface/transformers/issues/3735/events | https://github.com/huggingface/transformers/issues/3735 | 597,930,505 | MDU6SXNzdWU1OTc5MzA1MDU= | 3,735 | Pipeline for text generation | {
"login": "r0levrai",
"id": 22660388,
"node_id": "MDQ6VXNlcjIyNjYwMzg4",
"avatar_url": "https://avatars.githubusercontent.com/u/22660388?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/r0levrai",
"html_url": "https://github.com/r0levrai",
"followers_url": "https://api.github.com/users/r0l... | [] | closed | false | null | [] | [
"There is currently no text generation pipeline, but the `generate` method on both PyTorch/TensorFlow models is here for that purpose!\r\n\r\n[Here's](https://huggingface.co/transformers/usage.html#causal-language-modeling) an example using that method for generating from left context :).",
"You are right, I thou... | 1,586 | 1,605 | 1,586 | NONE | null | Hello,
* pipelines concise syntax and features are really nice, but there is none for text generation from left context
* `examples/run_generation.py` concise syntax (and some model-specific preprocessing) are really nice, but it is made for use via CLI and not from code
Any chance we see a text generation pipelin... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3735/reactions",
"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/3735/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3734 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3734/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3734/comments | https://api.github.com/repos/huggingface/transformers/issues/3734/events | https://github.com/huggingface/transformers/pull/3734 | 597,928,533 | MDExOlB1bGxSZXF1ZXN0NDAxOTM5NjE5 | 3,734 | [Config, Caching] Remove `output_past` everywhere and replace by `use_cache` argument | {
"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 | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/3734?src=pr&el=h1) Report\n> Merging [#3734](https://codecov.io/gh/huggingface/transformers/pull/3734?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/7972a4019f4bc9f85fd358f42249b90f9cd27c68&el=desc) will **increase... | 1,586 | 1,586 | 1,586 | MEMBER | null | The `config.output_past` variable is removed and replaced by a function argument `use_cache`.
The reasons for this were explained in PR: https://github.com/huggingface/transformers/pull/3682
Affected models are:
T5, Bart, GPT2, XLNet, CTRL, TFGPT2, TFCTRL, TFXLNET
It is made sure that the change **does not brea... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3734/reactions",
"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/3734/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3734",
"html_url": "https://github.com/huggingface/transformers/pull/3734",
"diff_url": "https://github.com/huggingface/transformers/pull/3734.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3734.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/3733 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3733/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3733/comments | https://api.github.com/repos/huggingface/transformers/issues/3733/events | https://github.com/huggingface/transformers/issues/3733 | 597,881,207 | MDU6SXNzdWU1OTc4ODEyMDc= | 3,733 | How i take an OpenAIGPTDoubleHeadsModel from run_language_modeling.py script? | {
"login": "nikkon3",
"id": 41228217,
"node_id": "MDQ6VXNlcjQxMjI4MjE3",
"avatar_url": "https://avatars.githubusercontent.com/u/41228217?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/nikkon3",
"html_url": "https://github.com/nikkon3",
"followers_url": "https://api.github.com/users/nikkon... | [
{
"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 | [] | [
"What are you trying to do exactly? If using our scripts, you would usually pre-train the transformer model (without the heads, so it doesn't make sense to use a double heads model), and you would then fine-tune the model with the heads on a specific dataset. \r\n\r\nIf you can explain a bit more what you're trying... | 1,586 | 1,591 | 1,591 | NONE | null | I train a gpt2 type model from scratch with run_language_modeling.py .
But i want to take an OpenAIGPTDoubleHeadsModel as my model.
My config.json is below. What i should change?
config = {
"architectures": [
"gpt2"
],
"model_name_or_path": None ,
"model_type": "gpt2",
"vocab_size":5000, ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3733/reactions",
"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/3733/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3732 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3732/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3732/comments | https://api.github.com/repos/huggingface/transformers/issues/3732/events | https://github.com/huggingface/transformers/issues/3732 | 597,808,226 | MDU6SXNzdWU1OTc4MDgyMjY= | 3,732 | Fine tuning XLMRoberta for Question Answering | {
"login": "wasiahmad",
"id": 17520413,
"node_id": "MDQ6VXNlcjE3NTIwNDEz",
"avatar_url": "https://avatars.githubusercontent.com/u/17520413?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/wasiahmad",
"html_url": "https://github.com/wasiahmad",
"followers_url": "https://api.github.com/users/... | [] | closed | false | null | [] | [] | 1,586 | 1,586 | 1,586 | NONE | null | # ❓ Questions & Help
<!-- The GitHub issue tracker is primarly intended for bugs, feature requests,
new models and benchmarks, and migration questions. For all other questions,
we direct you to Stack Overflow (SO) where a whole community of PyTorch and
Tensorflow enthusiast can help you out. Make s... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3732/reactions",
"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/3732/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3731 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3731/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3731/comments | https://api.github.com/repos/huggingface/transformers/issues/3731/events | https://github.com/huggingface/transformers/issues/3731 | 597,777,813 | MDU6SXNzdWU1OTc3Nzc4MTM= | 3,731 | Loading pipeline("summarization") failed | {
"login": "asymmetric-supernova",
"id": 17072697,
"node_id": "MDQ6VXNlcjE3MDcyNjk3",
"avatar_url": "https://avatars.githubusercontent.com/u/17072697?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/asymmetric-supernova",
"html_url": "https://github.com/asymmetric-supernova",
"followers_url... | [] | 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... | [
"The default summarization pipeline doesn't have support for TF unfortunately, but we should probably add an explicit error message @sshleifer ",
"+1",
"You can use `T5's` TF summarization though, like:\r\n`pipeline(\"summarization\", model=\"t5-base\", framework=\"tf\")`",
"I think the error has nothing to d... | 1,586 | 1,591 | 1,591 | NONE | null | Hi guys, when I try to load pipeline("summarization") I get the following error:
`Traceback (most recent call last):
File "<frozen importlib._bootstrap>", line 983, in _find_and_load
File "<frozen importlib._bootstrap>", line 967, in _find_and_load_unlocked
File "<frozen importlib._bootstrap>", line 677, in... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3731/reactions",
"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/3731/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3730 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3730/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3730/comments | https://api.github.com/repos/huggingface/transformers/issues/3730/events | https://github.com/huggingface/transformers/issues/3730 | 597,775,828 | MDU6SXNzdWU1OTc3NzU4Mjg= | 3,730 | OOM error when resuming training from a checkpoint | {
"login": "timsoraro",
"id": 61194445,
"node_id": "MDQ6VXNlcjYxMTk0NDQ1",
"avatar_url": "https://avatars.githubusercontent.com/u/61194445?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/timsoraro",
"html_url": "https://github.com/timsoraro",
"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, unfortunately, we have no way of helping without having more information. What scripts are you using? What model? What Python version? What transformers version? \r\n\r\nIt would be wonderful if you could use the issue template and describe exactly the issue so that we may help.",
"Hi, I'm sorry. I'm using [... | 1,586 | 1,595 | 1,595 | NONE | null | # 🐛 Bug
Some previous issues #2954 described a memory leak when resuming training from a checkpoint. I still get an OOM error when resuming training from a checkpoint. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3730/reactions",
"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/3730/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3729 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3729/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3729/comments | https://api.github.com/repos/huggingface/transformers/issues/3729/events | https://github.com/huggingface/transformers/pull/3729 | 597,648,707 | MDExOlB1bGxSZXF1ZXN0NDAxNzIxNDkz | 3,729 | exbert links for my albert model cards | {
"login": "elgeish",
"id": 6879673,
"node_id": "MDQ6VXNlcjY4Nzk2NzM=",
"avatar_url": "https://avatars.githubusercontent.com/u/6879673?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/elgeish",
"html_url": "https://github.com/elgeish",
"followers_url": "https://api.github.com/users/elgeish/... | [
{
"id": 1838412367,
"node_id": "MDU6TGFiZWwxODM4NDEyMzY3",
"url": "https://api.github.com/repos/huggingface/transformers/labels/model%20card",
"name": "model card",
"color": "92d5f4",
"default": false,
"description": "Related to pretrained model cards"
}
] | closed | false | null | [] | [
"Hi @elgeish , you also need to add a \r\n\r\n```\r\ntags:\r\n- exbert\r\n```\r\n\r\nto the metadata block.\r\n\r\n",
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/3729?src=pr&el=h1) Report\n> Merging [#3729](https://codecov.io/gh/huggingface/transformers/pull/3729?src=pr&el=desc) into [master]... | 1,586 | 1,587 | 1,587 | CONTRIBUTOR | null | Adding links for exbert visualization. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3729/reactions",
"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/3729/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3729",
"html_url": "https://github.com/huggingface/transformers/pull/3729",
"diff_url": "https://github.com/huggingface/transformers/pull/3729.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3729.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/3728 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3728/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3728/comments | https://api.github.com/repos/huggingface/transformers/issues/3728/events | https://github.com/huggingface/transformers/issues/3728 | 597,576,947 | MDU6SXNzdWU1OTc1NzY5NDc= | 3,728 | Checking that the LM actually trained | {
"login": "nikkon3",
"id": 41228217,
"node_id": "MDQ6VXNlcjQxMjI4MjE3",
"avatar_url": "https://avatars.githubusercontent.com/u/41228217?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/nikkon3",
"html_url": "https://github.com/nikkon3",
"followers_url": "https://api.github.com/users/nikkon... | [
{
"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 | [] | [
"Yes: simply `model.generate()` (not even a need for a Pipeline in that case)\r\n\r\ncc @patrickvonplaten ",
"I'd check if 'GPT2' works by sampling from a simple prompt. E.g.:\r\n\r\n```\r\noutput = model.generate(tokenizer.encode('The president', return_tensors='pt'), do_sample=True)\r\ntokenizer.decode(output[0... | 1,586 | 1,591 | 1,591 | NONE | null | I have trained a gpt2 from scratch with the way that is decribed in that post https://huggingface.co/blog/how-to-train .
Just in the step 4, where he checks if the trained model actually works, he uses from pipeline the
"fill-mask" but that works only for models with masked language modeling objective.
Exists s... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3728/reactions",
"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/3728/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3727 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3727/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3727/comments | https://api.github.com/repos/huggingface/transformers/issues/3727/events | https://github.com/huggingface/transformers/issues/3727 | 597,478,910 | MDU6SXNzdWU1OTc0Nzg5MTA= | 3,727 | ValueError: Cannot reshape a tensor - TFBertForSequenceClassification | {
"login": "DMells",
"id": 23553543,
"node_id": "MDQ6VXNlcjIzNTUzNTQz",
"avatar_url": "https://avatars.githubusercontent.com/u/23553543?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/DMells",
"html_url": "https://github.com/DMells",
"followers_url": "https://api.github.com/users/DMells/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 because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.\n"
] | 1,586 | 1,592 | 1,592 | NONE | null | I'm building a multiclass text classification model using Keras and BERT.
To convert my inputs to the required bert format, I'm using the `encode_plus` method found in the BertTokenizer class [found here][1]
The data is a paragraph of sentences per feature, and has a single label (of 45 labels in total)
**The ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3727/reactions",
"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/3727/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3726 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3726/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3726/comments | https://api.github.com/repos/huggingface/transformers/issues/3726/events | https://github.com/huggingface/transformers/pull/3726 | 597,472,860 | MDExOlB1bGxSZXF1ZXN0NDAxNTc5Mjgz | 3,726 | Separate input_ids and decoder_input_ids in model.generate() | {
"login": "yjernite",
"id": 10469459,
"node_id": "MDQ6VXNlcjEwNDY5NDU5",
"avatar_url": "https://avatars.githubusercontent.com/u/10469459?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/yjernite",
"html_url": "https://github.com/yjernite",
"followers_url": "https://api.github.com/users/yje... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/3726?src=pr&el=h1) Report\n> Merging [#3726](https://codecov.io/gh/huggingface/transformers/pull/3726?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/bc65afc4dfac3badf3de3be395d4023b44c61bdd&el=desc) will **increase... | 1,586 | 1,586 | 1,586 | MEMBER | null | This makes the generation behavior more similar for sequence-to-sequence and language models, and allows us to initialize decoding with a prefix for the encoder-decoder setting. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3726/reactions",
"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/3726/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3726",
"html_url": "https://github.com/huggingface/transformers/pull/3726",
"diff_url": "https://github.com/huggingface/transformers/pull/3726.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3726.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/3725 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3725/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3725/comments | https://api.github.com/repos/huggingface/transformers/issues/3725/events | https://github.com/huggingface/transformers/issues/3725 | 597,462,081 | MDU6SXNzdWU1OTc0NjIwODE= | 3,725 | Fine-tuning for paraphrasing tasks | {
"login": "anmoljagetia",
"id": 6505326,
"node_id": "MDQ6VXNlcjY1MDUzMjY=",
"avatar_url": "https://avatars.githubusercontent.com/u/6505326?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/anmoljagetia",
"html_url": "https://github.com/anmoljagetia",
"followers_url": "https://api.github.com... | [
{
"id": 1260952223,
"node_id": "MDU6TGFiZWwxMjYwOTUyMjIz",
"url": "https://api.github.com/repos/huggingface/transformers/labels/Discussion",
"name": "Discussion",
"color": "22870e",
"default": false,
"description": "Discussion on a topic (keep it focused or open a new issue though)"
},... | closed | false | null | [] | [
"This might help: https://huggingface.co/transformers/usage.html#sequence-classification",
"^^ These are inference examples. Do you know how can I _retrain_ ?",
"I would stress that this topic is quite interesting and useful. A good generative model for paraphrasing may help with text classification with small ... | 1,586 | 1,595 | 1,595 | NONE | null | # ❓ Questions & Help
I asked on SO and was downvoted since it is considered "off-site" and is against their Terms of Service. My question is somewhat simple. How do I fine-tune a GPT-2 model for the task of paraphrasing like the paper: https://www.aclweb.org/anthology/D19-5623.pdf
A link to my SO question : https:/... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3725/reactions",
"total_count": 7,
"+1": 6,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 1
} | https://api.github.com/repos/huggingface/transformers/issues/3725/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3724 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3724/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3724/comments | https://api.github.com/repos/huggingface/transformers/issues/3724/events | https://github.com/huggingface/transformers/issues/3724 | 597,419,418 | MDU6SXNzdWU1OTc0MTk0MTg= | 3,724 | Has anyone used run_language_modelling.py to train a gpt 2 from scratch? | {
"login": "nikkon3",
"id": 41228217,
"node_id": "MDQ6VXNlcjQxMjI4MjE3",
"avatar_url": "https://avatars.githubusercontent.com/u/41228217?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/nikkon3",
"html_url": "https://github.com/nikkon3",
"followers_url": "https://api.github.com/users/nikkon... | [
{
"id": 1260952223,
"node_id": "MDU6TGFiZWwxMjYwOTUyMjIz",
"url": "https://api.github.com/repos/huggingface/transformers/labels/Discussion",
"name": "Discussion",
"color": "22870e",
"default": false,
"description": "Discussion on a topic (keep it focused or open a new issue though)"
},... | closed | false | null | [] | [
"Hi @nikkon3, the special tokens for `gpt2` are automatically set when you import `GPT2Tokenizer`.\r\n\r\nThe code below shows that `'<|endoftext|>'` is the special token used for BOS (beginning of sequence), EOS (end of sequence), and UNK (unknown - out of vocabulary).\r\n\r\n```\r\nfrom transformers import GPT2To... | 1,586 | 1,702 | 1,592 | NONE | null | I have read this post https://huggingface.co/blog/how-to-train and i would like to do the train from scratch, for a gpt2 type model.
One question also, that i have, is the special tokens that he uses in that post for tokenizer will be the same and for a tokenizer that i will use for a gpt2 model? | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3724/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/3724/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3723 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3723/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3723/comments | https://api.github.com/repos/huggingface/transformers/issues/3723/events | https://github.com/huggingface/transformers/issues/3723 | 597,393,483 | MDU6SXNzdWU1OTczOTM0ODM= | 3,723 | How to get multiple answers from the context using BertForQuestionAnswering | {
"login": "MaheshChandrra",
"id": 13826929,
"node_id": "MDQ6VXNlcjEzODI2OTI5",
"avatar_url": "https://avatars.githubusercontent.com/u/13826929?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/MaheshChandrra",
"html_url": "https://github.com/MaheshChandrra",
"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 might help https://github.com/google-research/bert/issues/657",
"Thanks @chutaklee for the quick response, but may I please know whether we can do the same with the existing pretrained BERT model,by changing any parameters anywhere,as I currently do have question-answer pairs and I'm not not training the m... | 1,586 | 1,598 | 1,592 | NONE | null | How do I get multiple answers from the text using **BertForQuestionAnswering**, just like for the below question there are two possible answers:
1. a nice puppet
2. a software engineer
**Below is the code snippet for the same:**
```
from transformers import BertTokenizer, BertForQuestionAnswering
import tor... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3723/reactions",
"total_count": 3,
"+1": 3,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/3723/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3722 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3722/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3722/comments | https://api.github.com/repos/huggingface/transformers/issues/3722/events | https://github.com/huggingface/transformers/pull/3722 | 597,362,060 | MDExOlB1bGxSZXF1ZXN0NDAxNDg5NDk2 | 3,722 | Integrate Bert-like model on Flax runtime. | {
"login": "mfuntowicz",
"id": 2241520,
"node_id": "MDQ6VXNlcjIyNDE1MjA=",
"avatar_url": "https://avatars.githubusercontent.com/u/2241520?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/mfuntowicz",
"html_url": "https://github.com/mfuntowicz",
"followers_url": "https://api.github.com/users... | [] | closed | false | null | [] | [
"As you said, Jax is a library that interact with numpy to provide additional features: autodiff, auto-vectorization [(**vmap**)](https://jax.readthedocs.io/en/latest/notebooks/quickstart.html#Auto-vectorization-with-vmap) and auto-parallelization [(**pmap**)](https://jax.readthedocs.io/en/latest/jax.html#paralleli... | 1,586 | 1,603 | 1,603 | MEMBER | null | This Pull Request attempts to bring support for [Flax](https://github.com/google/flax) framework as part of transformers.
Main focus as been put on providing BERT-like models, principally by making it possible to load PyTorch checkpoints and doing the necessary conversions (few) directly on the fly. Supports also p... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3722/reactions",
"total_count": 14,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 5,
"confused": 0,
"heart": 7,
"rocket": 2,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/3722/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3722",
"html_url": "https://github.com/huggingface/transformers/pull/3722",
"diff_url": "https://github.com/huggingface/transformers/pull/3722.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3722.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/3721 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3721/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3721/comments | https://api.github.com/repos/huggingface/transformers/issues/3721/events | https://github.com/huggingface/transformers/issues/3721 | 597,230,864 | MDU6SXNzdWU1OTcyMzA4NjQ= | 3,721 | DistributedSampler can't shuffle the dataset | {
"login": "elk-cloner",
"id": 5828101,
"node_id": "MDQ6VXNlcjU4MjgxMDE=",
"avatar_url": "https://avatars.githubusercontent.com/u/5828101?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/elk-cloner",
"html_url": "https://github.com/elk-cloner",
"followers_url": "https://api.github.com/users... | [] | closed | false | null | [] | [
"I think you are right",
"Isn't there the same issue in other places?\r\nE.g. in trainer.py: https://github.com/huggingface/transformers/blob/97a375484c618496691982f62518130f294bb9a8/src/transformers/trainer.py#L305-L307",
"I forgot to re-add this in Trainer when merging #3800 \r\n\r\nIt's on my todo-list, but ... | 1,586 | 1,588 | 1,586 | CONTRIBUTOR | null | # 🐛 Bug
## Information
I'm trying to fine-tune BERT model using ```run_language_modeling.py```.
Language I am using the model on is Persian:
The problem arises when using:
* [x] the official example scripts: (give details below)
* [ ] my own modified scripts: (give details below)
The tasks I am work... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3721/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/3721/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3720 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3720/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3720/comments | https://api.github.com/repos/huggingface/transformers/issues/3720/events | https://github.com/huggingface/transformers/issues/3720 | 597,215,669 | MDU6SXNzdWU1OTcyMTU2Njk= | 3,720 | Disable @torch.no_grad() for model.generate() ? | {
"login": "Laksh1997",
"id": 59830552,
"node_id": "MDQ6VXNlcjU5ODMwNTUy",
"avatar_url": "https://avatars.githubusercontent.com/u/59830552?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Laksh1997",
"html_url": "https://github.com/Laksh1997",
"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
},
{
"id": 1834059054,
"node_id": "MDU6TGFiZWwxODM0MDU5MDU0",
... | 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... | [
"At the moment the only solution seems to be copying and pasting the entire generation code, as well as making a few changes that comes along with it, to avoid this issue.",
"One solution I propose is to add an argument `with_grad` which defaults to False.\r\nThen, add this as the first line in the generate code:... | 1,586 | 1,702 | 1,596 | NONE | null | # ❓ Questions & Help
Is there any way to do this?
| {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3720/reactions",
"total_count": 3,
"+1": 3,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/3720/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3719 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3719/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3719/comments | https://api.github.com/repos/huggingface/transformers/issues/3719/events | https://github.com/huggingface/transformers/issues/3719 | 597,209,891 | MDU6SXNzdWU1OTcyMDk4OTE= | 3,719 | Unable to load german BERT model | {
"login": "dakshvar22",
"id": 8708249,
"node_id": "MDQ6VXNlcjg3MDgyNDk=",
"avatar_url": "https://avatars.githubusercontent.com/u/8708249?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/dakshvar22",
"html_url": "https://github.com/dakshvar22",
"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 | [] | [
"That particular model doesn't have a TF version: https://huggingface.co/bert-base-german-dbmdz-cased#list-files\r\n\r\nHowever, you should be able to convert the PyTorch version to TF, using the `from_pt=True` flag.",
"Thanks for clarifying that @julien-c . Is it possible to add this information(regarding availa... | 1,586 | 1,592 | 1,592 | NONE | null | # 🐛 Bug
## Information
Model I am using (Bert, XLNet ...): BERT
Language I am using the model on (English, Chinese ...): German
The problem arises when using:
* [x] the official example scripts: (give details below)
* [ ] my own modified scripts: (give details below)
The tasks I am working on is:
* [... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3719/reactions",
"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/3719/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3718 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3718/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3718/comments | https://api.github.com/repos/huggingface/transformers/issues/3718/events | https://github.com/huggingface/transformers/issues/3718 | 597,188,002 | MDU6SXNzdWU1OTcxODgwMDI= | 3,718 | loading from tf_ckp and this showed up: AttributeError: 'BertCrf' object has no attribute 'bias' . | {
"login": "ghost",
"id": 10137,
"node_id": "MDQ6VXNlcjEwMTM3",
"avatar_url": "https://avatars.githubusercontent.com/u/10137?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ghost",
"html_url": "https://github.com/ghost",
"followers_url": "https://api.github.com/users/ghost/followers",
"f... | [] | closed | false | null | [] | [] | 1,586 | 1,586 | 1,586 | NONE | null | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3718/reactions",
"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/3718/timeline | completed | null | null | |
https://api.github.com/repos/huggingface/transformers/issues/3717 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3717/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3717/comments | https://api.github.com/repos/huggingface/transformers/issues/3717/events | https://github.com/huggingface/transformers/issues/3717 | 597,182,829 | MDU6SXNzdWU1OTcxODI4Mjk= | 3,717 | how to use transformers with gpu | {
"login": "xiongma",
"id": 30991932,
"node_id": "MDQ6VXNlcjMwOTkxOTMy",
"avatar_url": "https://avatars.githubusercontent.com/u/30991932?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/xiongma",
"html_url": "https://github.com/xiongma",
"followers_url": "https://api.github.com/users/xiongm... | [] | closed | false | null | [] | [
"This is a question more suitable to Stack Overflow or another forum"
] | 1,586 | 1,586 | 1,586 | NONE | null | I want to load a model with gpu in transformers, but it seem like the model always load in cpu
my os deepin 15.11 python 3.7.5 pytorch-gpu 1.4 transformers 2.8 | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3717/reactions",
"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/3717/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3716 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3716/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3716/comments | https://api.github.com/repos/huggingface/transformers/issues/3716/events | https://github.com/huggingface/transformers/pull/3716 | 597,176,958 | MDExOlB1bGxSZXF1ZXN0NDAxMzM2Mjky | 3,716 | Shift labels internally within TransfoXLLMHeadModel when called with labels | {
"login": "TevenLeScao",
"id": 26709476,
"node_id": "MDQ6VXNlcjI2NzA5NDc2",
"avatar_url": "https://avatars.githubusercontent.com/u/26709476?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/TevenLeScao",
"html_url": "https://github.com/TevenLeScao",
"followers_url": "https://api.github.com/... | [] | closed | false | null | [] | [
"I would also like to return a (1,)-sized tensor for the loss when called with labels as that's easier for the user, what the models do, and what the old documentation said TransfoXLLMHeadModel did."
] | 1,586 | 1,586 | 1,586 | CONTRIBUTOR | null | Fixes #3711 . | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3716/reactions",
"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/3716/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/3716",
"html_url": "https://github.com/huggingface/transformers/pull/3716",
"diff_url": "https://github.com/huggingface/transformers/pull/3716.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/3716.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/3715 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3715/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3715/comments | https://api.github.com/repos/huggingface/transformers/issues/3715/events | https://github.com/huggingface/transformers/issues/3715 | 597,152,006 | MDU6SXNzdWU1OTcxNTIwMDY= | 3,715 | How can i conditional fine-tuning with GPT2? | {
"login": "toriving",
"id": 42434734,
"node_id": "MDQ6VXNlcjQyNDM0NzM0",
"avatar_url": "https://avatars.githubusercontent.com/u/42434734?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/toriving",
"html_url": "https://github.com/toriving",
"followers_url": "https://api.github.com/users/tor... | [] | 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... | [
"To me, this sound more like a case where encoder-decoder models like `T5` or `Bart` should be fine-tuned. The encoder would encode the \"context\" and the decoder would be teacher-forced on the sentence.",
"> To me, this sound more like a case where encoder-decoder models like `T5` or `Bart` should be fine-tune... | 1,586 | 1,601 | 1,591 | NONE | null | I can use run_generation.py to create a statement by adding context.
But is there a way to do fine-tuning based on condition (context)?
For example, when data of "context [SEP] sentence" is input, the "context" is used to obtain the hidden state without learning.
In addition, the "sentence" is learned with the lan... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3715/reactions",
"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/3715/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3714 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3714/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3714/comments | https://api.github.com/repos/huggingface/transformers/issues/3714/events | https://github.com/huggingface/transformers/issues/3714 | 597,150,867 | MDU6SXNzdWU1OTcxNTA4Njc= | 3,714 | Zero shot multilingual BERT | {
"login": "paulthemagno",
"id": 38130299,
"node_id": "MDQ6VXNlcjM4MTMwMjk5",
"avatar_url": "https://avatars.githubusercontent.com/u/38130299?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/paulthemagno",
"html_url": "https://github.com/paulthemagno",
"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 | [] | [
"I think what you did there is probably best described as transfer learning.\r\n\r\n> Or maybe the 0 shot would be in the case in which I finetuned with an Italian dataset and then I evaluated on a corpus of another language?\r\n\r\nYeah, this is closer to the way the term \"zero-shot\" is being used in the field r... | 1,586 | 1,592 | 1,592 | NONE | null | ❓ Questions & Help
I have a doubt about the usage of multilingual BERT.
I did a domain adaptation on the language model **[`BERT-Base, Multilingual Cased`](https://storage.googleapis.com/bert_models/2018_11_23/multi_cased_L-12_H-768_A-12.zip)** with a dataset of a slightly different kind of text. This dataset is ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3714/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/3714/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/3713 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/3713/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/3713/comments | https://api.github.com/repos/huggingface/transformers/issues/3713/events | https://github.com/huggingface/transformers/issues/3713 | 597,120,321 | MDU6SXNzdWU1OTcxMjAzMjE= | 3,713 | cannot determine what will be the cardinality of the output after applying glue_convert_examples_to_features [TF 2.2.0rcx] | {
"login": "tarrade",
"id": 12021701,
"node_id": "MDQ6VXNlcjEyMDIxNzAx",
"avatar_url": "https://avatars.githubusercontent.com/u/12021701?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/tarrade",
"html_url": "https://github.com/tarrade",
"followers_url": "https://api.github.com/users/tarrad... | [] | closed | false | null | [] | [
"This issue might be of interest to @jplu ",
"Hey @tarrade \r\n\r\nUsually it is not advised to use the cardinality function for several reasons, the biggest two are: 1) it is still experimental, 2) cardinality works only with TF datasets created with `from_tensors` or `from_tensor_slices` which is not the case i... | 1,586 | 1,586 | 1,586 | NONE | null | # 🐛 Bug
## Information
With Tensorflow 2.2.0 (2.2.0rc2) we should be able to see the number of entries in the data without looking over them and using tf.data.experimental.cardinality.
One issue that I found is that after applying `glue_convert_examples_to_features` tf.data.experimental.cardinality is not abl... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/3713/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/3713/timeline | completed | null | null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.