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/1610 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1610/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1610/comments | https://api.github.com/repos/huggingface/transformers/issues/1610/events | https://github.com/huggingface/transformers/pull/1610 | 511,296,603 | MDExOlB1bGxSZXF1ZXN0MzMxNTIwMzEw | 1,610 | Update setup.py | {
"login": "singhanandVEVO",
"id": 24494214,
"node_id": "MDQ6VXNlcjI0NDk0MjE0",
"avatar_url": "https://avatars.githubusercontent.com/u/24494214?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/singhanandVEVO",
"html_url": "https://github.com/singhanandVEVO",
"followers_url": "https://api.gi... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/1610?src=pr&el=h1) Report\n> Merging [#1610](https://codecov.io/gh/huggingface/transformers/pull/1610?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/ef1b8b2ae5ad1057154a126879f7eb8de685f862?src=pr&el=desc) will **n... | 1,571 | 1,576 | 1,576 | NONE | null | changed update setup file | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1610/reactions",
"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/1610/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/1610",
"html_url": "https://github.com/huggingface/transformers/pull/1610",
"diff_url": "https://github.com/huggingface/transformers/pull/1610.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/1610.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/1609 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1609/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1609/comments | https://api.github.com/repos/huggingface/transformers/issues/1609/events | https://github.com/huggingface/transformers/issues/1609 | 511,294,787 | MDU6SXNzdWU1MTEyOTQ3ODc= | 1,609 | Can the prefix for GPT-2 conditional sampling be very long (longer than context window size)? | {
"login": "leejason",
"id": 4224456,
"node_id": "MDQ6VXNlcjQyMjQ0NTY=",
"avatar_url": "https://avatars.githubusercontent.com/u/4224456?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/leejason",
"html_url": "https://github.com/leejason",
"followers_url": "https://api.github.com/users/leeja... | [
{
"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"
] | 1,571 | 1,577 | 1,577 | NONE | null | Can the prefix for GPT-2 conditional sampling be very long (longer than context window size)? | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1609/reactions",
"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/1609/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1608 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1608/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1608/comments | https://api.github.com/repos/huggingface/transformers/issues/1608/events | https://github.com/huggingface/transformers/pull/1608 | 511,290,520 | MDExOlB1bGxSZXF1ZXN0MzMxNTE1MzEz | 1,608 | Error raised by "tmp_eval_loss += tmp_eval_loss.item()" when using multi-gpu | {
"login": "focox",
"id": 30308731,
"node_id": "MDQ6VXNlcjMwMzA4NzMx",
"avatar_url": "https://avatars.githubusercontent.com/u/30308731?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/focox",
"html_url": "https://github.com/focox",
"followers_url": "https://api.github.com/users/focox/follow... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/1608?src=pr&el=h1) Report\n> Merging [#1608](https://codecov.io/gh/huggingface/transformers/pull/1608?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/ef1b8b2ae5ad1057154a126879f7eb8de685f862?src=pr&el=desc) will **i... | 1,571 | 1,572 | 1,572 | NONE | null | fixed the bug raised by "tmp_eval_loss += tmp_eval_loss.item()" when parallelly using multi-gpu. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1608/reactions",
"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/1608/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/1608",
"html_url": "https://github.com/huggingface/transformers/pull/1608",
"diff_url": "https://github.com/huggingface/transformers/pull/1608.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/1608.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/1607 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1607/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1607/comments | https://api.github.com/repos/huggingface/transformers/issues/1607/events | https://github.com/huggingface/transformers/issues/1607 | 511,207,455 | MDU6SXNzdWU1MTEyMDc0NTU= | 1,607 | failed to download pretrained weights | {
"login": "minapril",
"id": 52790610,
"node_id": "MDQ6VXNlcjUyNzkwNjEw",
"avatar_url": "https://avatars.githubusercontent.com/u/52790610?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/minapril",
"html_url": "https://github.com/minapril",
"followers_url": "https://api.github.com/users/min... | [] | closed | false | null | [] | [
"Hi, this seems to be a network error. Are you sure you have access to the internet on this machine, or is it behind a firewall?",
"I had exactly the same problem Yesterday and s3.amazonaws.com just was not reachable. We also had the same problem with another service as well. After trying for some time it just st... | 1,571 | 1,571 | 1,571 | NONE | null | ## 🐛 Bug
<!-- Important information -->
Model I am using (Bert, XLNet....):BERT
Language I am using the model on (English, Chinese....):English
During downloading pretrained weights with code of modeling_bert.BertForMaskedLM.from_pretrained('bert-base-uncased'),another exception occurred:
![image](https:... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1607/reactions",
"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/1607/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1606 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1606/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1606/comments | https://api.github.com/repos/huggingface/transformers/issues/1606/events | https://github.com/huggingface/transformers/issues/1606 | 511,131,736 | MDU6SXNzdWU1MTExMzE3MzY= | 1,606 | Show pretrained model and config file download address directly in README.md & doc | {
"login": "Sunnycheey",
"id": 32103564,
"node_id": "MDQ6VXNlcjMyMTAzNTY0",
"avatar_url": "https://avatars.githubusercontent.com/u/32103564?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Sunnycheey",
"html_url": "https://github.com/Sunnycheey",
"followers_url": "https://api.github.com/use... | [] | closed | false | null | [] | [
"I feel that this would clutter the README, leading to a bad experience for 99.99% of users. But you can always submit a PR and see what the maintainers think.",
"Just make a PR."
] | 1,571 | 1,572 | 1,572 | NONE | null | ## 🚀 Feature
<!-- A clear and concise description of the feature proposal. Please provide a link to the paper and code in case they exist. -->
Show download address of pretrained model address directly in markdown file.
## Motivation
Since my server cannot directly go to aws server, and it's not configed to prox... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1606/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/1606/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1605 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1605/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1605/comments | https://api.github.com/repos/huggingface/transformers/issues/1605/events | https://github.com/huggingface/transformers/issues/1605 | 511,085,503 | MDU6SXNzdWU1MTEwODU1MDM= | 1,605 | Support for gpt2-medium, gpt2-large and distilgpt2 in pytorch-pretrained-bert 0.6.2 | {
"login": "g-karthik",
"id": 3851993,
"node_id": "MDQ6VXNlcjM4NTE5OTM=",
"avatar_url": "https://avatars.githubusercontent.com/u/3851993?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/g-karthik",
"html_url": "https://github.com/g-karthik",
"followers_url": "https://api.github.com/users/g-... | [
{
"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 | [] | [
"As far as I know pytorch-pretrained-bert development has been discontinued. That makes sense. If you want the new features, you have to upgrade.",
"Well technically what I'm asking for isn't a new feature, it's just backwards-compatibility for the above three model artifacts.\r\n\r\nI can manually add them to th... | 1,571 | 1,577 | 1,577 | NONE | null | ## 🚀 Feature
Request: Inclusion of the below 3 lines in pytorch-pretrained-bert 0.6.2
https://github.com/huggingface/transformers/blob/ef1b8b2ae5ad1057154a126879f7eb8de685f862/transformers/modeling_gpt2.py#L40-L42
## Motivation
Currently, the above 3 lines exist in the latest version of transformers in PyPI,... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1605/reactions",
"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/1605/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1604 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1604/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1604/comments | https://api.github.com/repos/huggingface/transformers/issues/1604/events | https://github.com/huggingface/transformers/pull/1604 | 510,949,096 | MDExOlB1bGxSZXF1ZXN0MzMxMjM2MzEw | 1,604 | Versioning in documentation | {
"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 | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/1604?src=pr&el=h1) Report\n> Merging [#1604](https://codecov.io/gh/huggingface/transformers/pull/1604?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/ef1b8b2ae5ad1057154a126879f7eb8de685f862?src=pr&el=desc) will **d... | 1,571 | 1,572 | 1,572 | MEMBER | null | Several versions of the documentation can now be accessed:
`huggingface.co/transformers` for the master release
`huggingface.co/transformers/v2.1.1` for the 2.1.1 official release and so on. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1604/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/1604/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/1604",
"html_url": "https://github.com/huggingface/transformers/pull/1604",
"diff_url": "https://github.com/huggingface/transformers/pull/1604.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/1604.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/1603 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1603/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1603/comments | https://api.github.com/repos/huggingface/transformers/issues/1603/events | https://github.com/huggingface/transformers/pull/1603 | 510,862,905 | MDExOlB1bGxSZXF1ZXN0MzMxMTYyMjU2 | 1,603 | [scripts] Proposal: add a specific device flag | {
"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-... | [
{
"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/1603?src=pr&el=h1) Report\n> Merging [#1603](https://codecov.io/gh/huggingface/transformers/pull/1603?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/e16d46843a19ab289b82138e4eccec5610a76de7?src=pr&el=desc) will **n... | 1,571 | 1,586 | 1,578 | MEMBER | null | wdyt?
Will do in other scripts if this gets merged.
My use case is I have an instance with multiple GPUs and want to run one generation on `cuda:0`, another one on `cuda:1`, etc. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1603/reactions",
"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/1603/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/1603",
"html_url": "https://github.com/huggingface/transformers/pull/1603",
"diff_url": "https://github.com/huggingface/transformers/pull/1603.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/1603.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/1602 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1602/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1602/comments | https://api.github.com/repos/huggingface/transformers/issues/1602/events | https://github.com/huggingface/transformers/pull/1602 | 510,852,221 | MDExOlB1bGxSZXF1ZXN0MzMxMTUzNTU2 | 1,602 | Fix architectures count | {
"login": "dataista0",
"id": 4383443,
"node_id": "MDQ6VXNlcjQzODM0NDM=",
"avatar_url": "https://avatars.githubusercontent.com/u/4383443?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/dataista0",
"html_url": "https://github.com/dataista0",
"followers_url": "https://api.github.com/users/da... | [] | closed | false | null | [] | [
"Great, thanks!",
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/1602?src=pr&el=h1) Report\n> Merging [#1602](https://codecov.io/gh/huggingface/transformers/pull/1602?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/1cfd9748683db43af2c98da1a19d39f0efc8cc3b?src=... | 1,571 | 1,571 | 1,571 | CONTRIBUTOR | null | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1602/reactions",
"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/1602/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/1602",
"html_url": "https://github.com/huggingface/transformers/pull/1602",
"diff_url": "https://github.com/huggingface/transformers/pull/1602.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/1602.patch",
"merged_at": 1... | |
https://api.github.com/repos/huggingface/transformers/issues/1601 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1601/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1601/comments | https://api.github.com/repos/huggingface/transformers/issues/1601/events | https://github.com/huggingface/transformers/pull/1601 | 510,826,670 | MDExOlB1bGxSZXF1ZXN0MzMxMTMyNTU2 | 1,601 | Clean roberta model & all tokenizers now add special tokens by default (breaking change) | {
"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 | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/1601?src=pr&el=h1) Report\n> Merging [#1601](https://codecov.io/gh/huggingface/transformers/pull/1601?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/079bfb32fba4f2b39d344ca7af88d79a3ff27c7c?src=pr&el=desc) will **d... | 1,571 | 1,572 | 1,572 | MEMBER | null | The RoBERTa model checks that special tokens are in the input sequence as it cannot function as expected if they are not here. This is not the best practice:
- The print method is not handled on TPU, and the check is problematic when tracing the models
- RoBERTa is the only model to print this warning while other m... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1601/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/1601/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/1601",
"html_url": "https://github.com/huggingface/transformers/pull/1601",
"diff_url": "https://github.com/huggingface/transformers/pull/1601.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/1601.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/1600 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1600/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1600/comments | https://api.github.com/repos/huggingface/transformers/issues/1600/events | https://github.com/huggingface/transformers/issues/1600 | 510,777,589 | MDU6SXNzdWU1MTA3Nzc1ODk= | 1,600 | None in openAi-gpt tokenization | {
"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 | [] | [
"Are you using the GPT tokenizer? If not try\r\n\r\n```\r\ntokenizer = transformers.OpenAIGTPTTokenizer()\r\ninput_ids = tokenizer.encode(your_text)\r\n```\r\n\r\n",
"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 yo... | 1,571 | 1,577 | 1,577 | NONE | null | Hi
I want to concat two sentences, and give it to openAI-gpt, I use cl sentence1 sep sentence2 sep
I got none with openai-gpt in the first position, could you tell me what is the expected format? thanks | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1600/reactions",
"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/1600/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1599 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1599/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1599/comments | https://api.github.com/repos/huggingface/transformers/issues/1599/events | https://github.com/huggingface/transformers/issues/1599 | 510,738,059 | MDU6SXNzdWU1MTA3MzgwNTk= | 1,599 | Issue in Cost Function | {
"login": "anandhperumal",
"id": 12907396,
"node_id": "MDQ6VXNlcjEyOTA3Mzk2",
"avatar_url": "https://avatars.githubusercontent.com/u/12907396?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/anandhperumal",
"html_url": "https://github.com/anandhperumal",
"followers_url": "https://api.githu... | [] | closed | false | null | [] | [
"Hi @anandhperumal. Remember that you train GPT-2 by doing next-token prediction, therefore you need to compare the i-th input label--the truth--with what the model predicted: the (i-1)th output. Hence the indices shift.",
"@rlouf oh yeah. Thanks for the input.\r\nif you don't mind can you answer this question as... | 1,571 | 1,571 | 1,571 | NONE | null | ## 🐛 Bug
<!-- Important information -->
Model I am using (Bert, XLNet....): GPT2
Language I am using the model on (English, Chinese....): English
The problem arise when using:
* [X] the official example scripts: (give details)
* [ ] my own modified scripts: (give details)
The tasks I am working on is:... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1599/reactions",
"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/1599/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1598 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1598/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1598/comments | https://api.github.com/repos/huggingface/transformers/issues/1598/events | https://github.com/huggingface/transformers/pull/1598 | 510,639,726 | MDExOlB1bGxSZXF1ZXN0MzMwOTc4NzY5 | 1,598 | changing "out_features" of final linear layer | {
"login": "SKRohit",
"id": 9626333,
"node_id": "MDQ6VXNlcjk2MjYzMzM=",
"avatar_url": "https://avatars.githubusercontent.com/u/9626333?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/SKRohit",
"html_url": "https://github.com/SKRohit",
"followers_url": "https://api.github.com/users/SKRohit/... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/1598?src=pr&el=h1) Report\n> Merging [#1598](https://codecov.io/gh/huggingface/transformers/pull/1598?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/b8c9ea0010a09cca8173e5bdf4af855123aebfc7?src=pr&el=desc) will **d... | 1,571 | 1,573 | 1,573 | CONTRIBUTOR | null | calling `resize_token_embeddings` changes the dimensions of the final linear layer. so changed `out_features` | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1598/reactions",
"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/1598/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/1598",
"html_url": "https://github.com/huggingface/transformers/pull/1598",
"diff_url": "https://github.com/huggingface/transformers/pull/1598.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/1598.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/1597 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1597/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1597/comments | https://api.github.com/repos/huggingface/transformers/issues/1597/events | https://github.com/huggingface/transformers/issues/1597 | 510,620,485 | MDU6SXNzdWU1MTA2MjA0ODU= | 1,597 | _tokenize() got an unexpected keyword argument 'add_prefix_space' in CTRL | {
"login": "BramVanroy",
"id": 2779410,
"node_id": "MDQ6VXNlcjI3Nzk0MTA=",
"avatar_url": "https://avatars.githubusercontent.com/u/2779410?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/BramVanroy",
"html_url": "https://github.com/BramVanroy",
"followers_url": "https://api.github.com/users... | [] | closed | false | null | [] | [
"Hi @BramVanroy, thanks for reporting this. There was an issue in the docstring. It does not use prefix spaces and it does not use a byte-level BPE like GPT-2 does. The docstring should be fixed now."
] | 1,571 | 1,571 | 1,571 | COLLABORATOR | null | ## 🐛 Bug
If you look at [the search results in this repo](https://github.com/huggingface/transformers/search?q=add_prefix_space) for `add_prefix_space`, you'll find gpt2, roberta, and ctrl all document that
> `add_prefix_space`: Requires a space to start the input string => the encoding methods should be called... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1597/reactions",
"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/1597/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1596 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1596/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1596/comments | https://api.github.com/repos/huggingface/transformers/issues/1596/events | https://github.com/huggingface/transformers/issues/1596 | 510,546,258 | MDU6SXNzdWU1MTA1NDYyNTg= | 1,596 | How to use BERT for ENTITY extraction from a Sequence without classification in the NER task ? | {
"login": "ManojPrabhakar",
"id": 5091907,
"node_id": "MDQ6VXNlcjUwOTE5MDc=",
"avatar_url": "https://avatars.githubusercontent.com/u/5091907?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ManojPrabhakar",
"html_url": "https://github.com/ManojPrabhakar",
"followers_url": "https://api.gith... | [] | closed | false | null | [] | [
"I'm a bit confused: you're basically defining the broad case of named entity recognition. Is it not enough to have a binary NER (token-level classification) task for entity vs non-entity?",
"Assuming you have 3-class (PER, ORG, LOC) data with labels:\r\nB-PER, I-PER, B-ORG, I-ORG, B-LOC, I-LOC, as well as O\r\n\... | 1,571 | 1,575 | 1,575 | NONE | null | ## ❓ Questions & Help
<!-- A clear and concise description of the question. -->
My requirement here is given a sentence(sequence), I would like to just extract the entities present in the sequence without classifying them to a type in the NER task. I see that BERT has BertForTokenClassification for NER which does t... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1596/reactions",
"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/1596/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1595 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1595/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1595/comments | https://api.github.com/repos/huggingface/transformers/issues/1595/events | https://github.com/huggingface/transformers/issues/1595 | 510,502,762 | MDU6SXNzdWU1MTA1MDI3NjI= | 1,595 | Using HuggingFace TransfoXLLMHeadModel() with custom Torchtext vocabulary | {
"login": "h56cho",
"id": 52889259,
"node_id": "MDQ6VXNlcjUyODg5MjU5",
"avatar_url": "https://avatars.githubusercontent.com/u/52889259?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/h56cho",
"html_url": "https://github.com/h56cho",
"followers_url": "https://api.github.com/users/h56cho/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",
"I faced the same problem:I solved it by passing the size of your vocabulary (from your custom tokenizer) as a parameter.\r\nI proceeded... | 1,571 | 1,580 | 1,577 | NONE | null | Hello,
I am trying to use the HuggingFace TransfoXLLMHeadModel on WikiText2 dataset under a customized TransfoXLConfig with different vocabulary, and it causing an error. I am not sure how to fix it. Below are my code:
```js
# Import packages
import torch
import torch.nn as nn
import torch.nn.functional as F... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1595/reactions",
"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/1595/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1594 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1594/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1594/comments | https://api.github.com/repos/huggingface/transformers/issues/1594/events | https://github.com/huggingface/transformers/issues/1594 | 510,483,278 | MDU6SXNzdWU1MTA0ODMyNzg= | 1,594 | Make benchmark more flexible (TF or PT) | {
"login": "BramVanroy",
"id": 2779410,
"node_id": "MDQ6VXNlcjI3Nzk0MTA=",
"avatar_url": "https://avatars.githubusercontent.com/u/2779410?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/BramVanroy",
"html_url": "https://github.com/BramVanroy",
"followers_url": "https://api.github.com/users... | [] | closed | false | null | [] | [
"I believe a quick workaround is to just install the pre-built, CPU version of TensorFlow 2.0. If you won't be running the TF benchmarks, it wouldn't affect anything.",
"True, but still not quite flexible. Since the goal of the benchmark script is to, I believe, encourage the community to add there runtimes, it's... | 1,571 | 1,571 | 1,571 | COLLABORATOR | null | I've been trying to run the benchmark, but I gave up after running into a trillion compatibility issues with tensorflow and bazel. To be fair, I just want to contribute and test all there is to test on PyTorch with 4x Tesla V100. It would be great if only the required modules are needed, and not all of them. So only tr... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1594/reactions",
"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/1594/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1593 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1593/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1593/comments | https://api.github.com/repos/huggingface/transformers/issues/1593/events | https://github.com/huggingface/transformers/pull/1593 | 510,480,033 | MDExOlB1bGxSZXF1ZXN0MzMwODQ3MjE4 | 1,593 | Fix AdamW import error for <1.2 | {
"login": "BramVanroy",
"id": 2779410,
"node_id": "MDQ6VXNlcjI3Nzk0MTA=",
"avatar_url": "https://avatars.githubusercontent.com/u/2779410?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/BramVanroy",
"html_url": "https://github.com/BramVanroy",
"followers_url": "https://api.github.com/users... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/1593?src=pr&el=h1) Report\n> Merging [#1593](https://codecov.io/gh/huggingface/transformers/pull/1593?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/702f589848baba97ea4897aa3f0bb937e1ec3bcf?src=pr&el=desc) will **d... | 1,571 | 1,586 | 1,586 | COLLABORATOR | null | closes #1585 | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1593/reactions",
"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/1593/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/1593",
"html_url": "https://github.com/huggingface/transformers/pull/1593",
"diff_url": "https://github.com/huggingface/transformers/pull/1593.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/1593.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/1592 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1592/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1592/comments | https://api.github.com/repos/huggingface/transformers/issues/1592/events | https://github.com/huggingface/transformers/pull/1592 | 510,468,349 | MDExOlB1bGxSZXF1ZXN0MzMwODM3NTk0 | 1,592 | Consider do_lower_case in PreTrainedTokenizer | {
"login": "watkinsm",
"id": 38503580,
"node_id": "MDQ6VXNlcjM4NTAzNTgw",
"avatar_url": "https://avatars.githubusercontent.com/u/38503580?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/watkinsm",
"html_url": "https://github.com/watkinsm",
"followers_url": "https://api.github.com/users/wat... | [] | closed | false | null | [] | [
"this lgtm but let's wait for @thomwolf and @LysandreJik to chime in",
"I'd also like to improve the test cases. I'll try to find some time for that this weekend",
"Nice improvement, it would be even better with tests for DistilBERT and XLNet as both those models make use of the `do_lower_case` argument. Transf... | 1,571 | 1,574 | 1,574 | CONTRIBUTOR | null | As pointed out in #1545, when using an uncased model, and adding a new uncased token, the tokenizer does not correctly identify this in the case that the input text contains the token in a cased format.
For instance, if we load bert-base-uncased into BertTokenizer, and then use .add_tokens() to add "cool-token", we ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1592/reactions",
"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/1592/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/1592",
"html_url": "https://github.com/huggingface/transformers/pull/1592",
"diff_url": "https://github.com/huggingface/transformers/pull/1592.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/1592.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/1591 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1591/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1591/comments | https://api.github.com/repos/huggingface/transformers/issues/1591/events | https://github.com/huggingface/transformers/issues/1591 | 510,402,622 | MDU6SXNzdWU1MTA0MDI2MjI= | 1,591 | Error when trying to reuse hidden states in CTRL | {
"login": "bilal2vec",
"id": 29356759,
"node_id": "MDQ6VXNlcjI5MzU2NzU5",
"avatar_url": "https://avatars.githubusercontent.com/u/29356759?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/bilal2vec",
"html_url": "https://github.com/bilal2vec",
"followers_url": "https://api.github.com/users/... | [] | 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... | [
"The same error occurs with the library installed with `git clone` (_master_ version) + torch v1.3.0 + python v3.6.8.\r\n[Here](https://colab.research.google.com/drive/1nawWX6Lrfh9ZVKyRfTLgFSIG355xkPRy#scrollTo=n93UZjq5EIE_) is a more verbose version of the Colab Notebook posted by @bkkaggle with Google Colab.",
... | 1,571 | 1,573 | 1,573 | CONTRIBUTOR | null | ## 🐛 Bug
<!-- Important information -->
Model I am using (Bert, XLNet....): CTRL
Language I am using the model on (English, Chinese....): English
The problem arise when using:
My own script, the colab link is available [here](https://colab.research.google.com/drive/143T4sBda4r2nDYzmuNwi-ZFTbhJWfeOW)
... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1591/reactions",
"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/1591/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1590 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1590/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1590/comments | https://api.github.com/repos/huggingface/transformers/issues/1590/events | https://github.com/huggingface/transformers/pull/1590 | 510,400,839 | MDExOlB1bGxSZXF1ZXN0MzMwNzgxOTg3 | 1,590 | [WIP] Fixes for TF Roberta (and other models WIP) | {
"login": "tlkh",
"id": 5409617,
"node_id": "MDQ6VXNlcjU0MDk2MTc=",
"avatar_url": "https://avatars.githubusercontent.com/u/5409617?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/tlkh",
"html_url": "https://github.com/tlkh",
"followers_url": "https://api.github.com/users/tlkh/followers",
... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/1590?src=pr&el=h1) Report\n> Merging [#1590](https://codecov.io/gh/huggingface/transformers/pull/1590?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/4d456542e9d381090f9a00b2bcc5a4cb07f6f3f7?src=pr&el=desc) will **n... | 1,571 | 1,573 | 1,573 | CONTRIBUTOR | null | When converting the `run_tf_glue.py` example to the same format at `benchmarks.py` to create a standardized benchmark for training, I ran into errors with **training** the non-BERT models with the normal `model.fit()` method. I am attempting to resolve all the errors I encountered in this PR. In particular, I have fixe... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1590/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/1590/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/1590",
"html_url": "https://github.com/huggingface/transformers/pull/1590",
"diff_url": "https://github.com/huggingface/transformers/pull/1590.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/1590.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/1589 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1589/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1589/comments | https://api.github.com/repos/huggingface/transformers/issues/1589/events | https://github.com/huggingface/transformers/pull/1589 | 510,368,002 | MDExOlB1bGxSZXF1ZXN0MzMwNzU1NzA5 | 1,589 | Fix architectures count | {
"login": "dataista0",
"id": 4383443,
"node_id": "MDQ6VXNlcjQzODM0NDM=",
"avatar_url": "https://avatars.githubusercontent.com/u/4383443?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/dataista0",
"html_url": "https://github.com/dataista0",
"followers_url": "https://api.github.com/users/da... | [] | closed | false | null | [] | [
"Hi! Actually, if we count DistilGPT-2 as a standalone architecture, it should be 10. Do you think you could update it to 10 before we merge? Thanks."
] | 1,571 | 1,576 | 1,576 | CONTRIBUTOR | null | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1589/reactions",
"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/1589/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/1589",
"html_url": "https://github.com/huggingface/transformers/pull/1589",
"diff_url": "https://github.com/huggingface/transformers/pull/1589.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/1589.patch",
"merged_at": n... | |
https://api.github.com/repos/huggingface/transformers/issues/1588 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1588/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1588/comments | https://api.github.com/repos/huggingface/transformers/issues/1588/events | https://github.com/huggingface/transformers/issues/1588 | 510,211,458 | MDU6SXNzdWU1MTAyMTE0NTg= | 1,588 | Using HuggingFace pre-trained transformer to tokenize and generate iterator for a different text than the one it was trained on | {
"login": "h56cho",
"id": 52889259,
"node_id": "MDQ6VXNlcjUyODg5MjU5",
"avatar_url": "https://avatars.githubusercontent.com/u/52889259?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/h56cho",
"html_url": "https://github.com/h56cho",
"followers_url": "https://api.github.com/users/h56cho/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"
] | 1,571 | 1,577 | 1,577 | NONE | null | Hello,
I am trying to do NLP by using HuggingFace transformers, and I have a question. Is it possible to use the pre-trained HuggingFace Transformer-XL and its pre-trained vocabulary to tokenize and generate BPTTIterator for the WikiText2 dataset instead of the WikiText103 that the transformer was originally trained o... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1588/reactions",
"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/1588/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1587 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1587/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1587/comments | https://api.github.com/repos/huggingface/transformers/issues/1587/events | https://github.com/huggingface/transformers/issues/1587 | 509,811,647 | MDU6SXNzdWU1MDk4MTE2NDc= | 1,587 | Sequence to sequence with GPT model | {
"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 | [] | [
"We are currently working on implementing seq2seq for most models in the library (see https://github.com/huggingface/transformers/pull/1455). I won't be ready before a week or two.",
"I'm closing this issue, but feel free to reply in #1506 that we leave open for comments on this implementation."
] | 1,571 | 1,571 | 1,571 | NONE | null | Hi, I really appreciate if you could tell me if I can build a seq2seq model with gpt2 like this:
I am getting GPT2 run_generation codes, and I want to finetune it in a way, that I give a
sequence as a context, and then generate another sequence with gpt2, and then I minimize
the cross-entropy loss between the gener... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1587/reactions",
"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/1587/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1586 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1586/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1586/comments | https://api.github.com/repos/huggingface/transformers/issues/1586/events | https://github.com/huggingface/transformers/pull/1586 | 509,730,759 | MDExOlB1bGxSZXF1ZXN0MzMwMjE3NTU2 | 1,586 | Add special tokens to documentation for bert examples to resolve issue: #1561 | {
"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 | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/1586?src=pr&el=h1) Report\n> Merging [#1586](https://codecov.io/gh/huggingface/transformers/pull/1586?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/82f6abd98aaa691ca0adfe21e85a17dc6f386497?src=pr&el=desc) will **n... | 1,571 | 1,576 | 1,576 | CONTRIBUTOR | null | **Currently the BERT examples only show the strings encoded without the inclusion of special tokens (e.g. [CLS] and [SEP]) as illustrated below:**
```
tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
sentence = "Hello there, General Kenobi!"
print(tokenizer.encode(sentence))
print(tokenizer.cls_token... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1586/reactions",
"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/1586/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/1586",
"html_url": "https://github.com/huggingface/transformers/pull/1586",
"diff_url": "https://github.com/huggingface/transformers/pull/1586.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/1586.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/1585 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1585/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1585/comments | https://api.github.com/repos/huggingface/transformers/issues/1585/events | https://github.com/huggingface/transformers/issues/1585 | 509,715,862 | MDU6SXNzdWU1MDk3MTU4NjI= | 1,585 | AdamW requires torch>=1.2.0 | {
"login": "carter54",
"id": 26741594,
"node_id": "MDQ6VXNlcjI2NzQxNTk0",
"avatar_url": "https://avatars.githubusercontent.com/u/26741594?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/carter54",
"html_url": "https://github.com/carter54",
"followers_url": "https://api.github.com/users/car... | [
{
"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 don't think that's right. AdamW is implemented in transformers.optimization\r\n\r\nhttps://github.com/huggingface/transformers/blob/82f6abd98aaa691ca0adfe21e85a17dc6f386497/transformers/optimization.py#L107\r\n\r\nAs far as I can see that does not require anything specific to torch 1.2. _However_, if you are try... | 1,571 | 1,586 | 1,586 | NONE | null | ## 🐛 Bug
<!-- Important information -->
AdamW requires torch>=1.2.0, torch < 1.2.0 will cause an importError: cannot import name 'AdamW'
Model I am using (Bert, XLNet....):
Language I am using the model on (English, Chinese....):
The problem arise when using:
* [ ] the official example scripts: (give detai... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1585/reactions",
"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/1585/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1584 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1584/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1584/comments | https://api.github.com/repos/huggingface/transformers/issues/1584/events | https://github.com/huggingface/transformers/pull/1584 | 509,715,658 | MDExOlB1bGxSZXF1ZXN0MzMwMjA1NTEx | 1,584 | Add special tokens to documentation for bert examples to resolve issue: #1561 | {
"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 | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/1584?src=pr&el=h1) Report\n> Merging [#1584](https://codecov.io/gh/huggingface/transformers/pull/1584?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/82f6abd98aaa691ca0adfe21e85a17dc6f386497?src=pr&el=desc) will **n... | 1,571 | 1,571 | 1,571 | CONTRIBUTOR | null | **Currently the BERT examples only show the strings encoded without the inclusion of special tokens (e.g. [CLS] and [SEP]) as illustrated below:**
```
tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
sentence = "Hello there, General Kenobi!"
print(tokenizer.encode(sentence))
print(tokenizer.cls_tok... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1584/reactions",
"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/1584/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/1584",
"html_url": "https://github.com/huggingface/transformers/pull/1584",
"diff_url": "https://github.com/huggingface/transformers/pull/1584.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/1584.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/1583 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1583/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1583/comments | https://api.github.com/repos/huggingface/transformers/issues/1583/events | https://github.com/huggingface/transformers/issues/1583 | 509,696,138 | MDU6SXNzdWU1MDk2OTYxMzg= | 1,583 | Question answering for SQuAD with XLNet | {
"login": "kayoyin",
"id": 44864455,
"node_id": "MDQ6VXNlcjQ0ODY0NDU1",
"avatar_url": "https://avatars.githubusercontent.com/u/44864455?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/kayoyin",
"html_url": "https://github.com/kayoyin",
"followers_url": "https://api.github.com/users/kayoyi... | [
{
"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"
] | 1,571 | 1,577 | 1,577 | NONE | null | ## ❓ Questions & Help
Dear huggingface,
Thank you very much for your great implementation of NLP architectures! I'm currently trying to train an XLNet model for question answering in French.
I studied your code to understand how question answering is done with XLNet, but I am struggling to follow how it works.... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1583/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/1583/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1582 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1582/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1582/comments | https://api.github.com/repos/huggingface/transformers/issues/1582/events | https://github.com/huggingface/transformers/issues/1582 | 509,681,078 | MDU6SXNzdWU1MDk2ODEwNzg= | 1,582 | How does arg --vocab_transform help in extract_distilbert.py? | {
"login": "evehsu",
"id": 17281640,
"node_id": "MDQ6VXNlcjE3MjgxNjQw",
"avatar_url": "https://avatars.githubusercontent.com/u/17281640?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/evehsu",
"html_url": "https://github.com/evehsu",
"followers_url": "https://api.github.com/users/evehsu/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 | [] | [
"Hello @evehsu,\r\n\r\nBERT uses an additional non-linearity before the vocabulary projection (see [here](https://github.com/huggingface/transformers/blob/master/transformers/modeling_bert.py#L381)).\r\nIt's a design choice, as far as I know, XLM doesn't a non-linearity right before the vocab projection (the langua... | 1,571 | 1,577 | 1,577 | NONE | null | ## ❓ Questions & Help
<!-- A clear and concise description of the question. -->
Hi everyone, I'm new to experiment with bert model distillation. When running extract_distilbert.py on my fine tuned bert model, I came across the argusment vocat_transform.
```
if args.vocab_transform:
for w in ['weight', 'b... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1582/reactions",
"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/1582/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1581 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1581/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1581/comments | https://api.github.com/repos/huggingface/transformers/issues/1581/events | https://github.com/huggingface/transformers/issues/1581 | 509,666,574 | MDU6SXNzdWU1MDk2NjY1NzQ= | 1,581 | Is there a computation/speed advantage to batching inputs into `TransformerModel` to reduce its number calls | {
"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"
] | 1,571 | 1,577 | 1,577 | CONTRIBUTOR | null | ## ❓ Questions & Help
<!-- A clear and concise description of the question. -->
For my particular application, I need to have several `output = TransformerModel(inputIDs)` calls per step, from different datasets.
so
```
output1 = TransformerModel(inputIDs_dataset1)
output2 = TransformerModel(inputIDs_dat... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1581/reactions",
"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/1581/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1580 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1580/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1580/comments | https://api.github.com/repos/huggingface/transformers/issues/1580/events | https://github.com/huggingface/transformers/pull/1580 | 509,656,316 | MDExOlB1bGxSZXF1ZXN0MzMwMTYwODcz | 1,580 | Gradient norm clipping should be done right before calling the optimiser | {
"login": "pminervini",
"id": 227357,
"node_id": "MDQ6VXNlcjIyNzM1Nw==",
"avatar_url": "https://avatars.githubusercontent.com/u/227357?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/pminervini",
"html_url": "https://github.com/pminervini",
"followers_url": "https://api.github.com/users/p... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/1580?src=pr&el=h1) Report\n> Merging [#1580](https://codecov.io/gh/huggingface/transformers/pull/1580?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/82f6abd98aaa691ca0adfe21e85a17dc6f386497?src=pr&el=desc) will **n... | 1,571 | 1,571 | 1,571 | CONTRIBUTOR | null | Right now it's done after each step in the gradient accumulation. What do you think? | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1580/reactions",
"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/1580/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/1580",
"html_url": "https://github.com/huggingface/transformers/pull/1580",
"diff_url": "https://github.com/huggingface/transformers/pull/1580.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/1580.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/1579 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1579/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1579/comments | https://api.github.com/repos/huggingface/transformers/issues/1579/events | https://github.com/huggingface/transformers/issues/1579 | 509,626,406 | MDU6SXNzdWU1MDk2MjY0MDY= | 1,579 | seq2seq with gpt2 | {
"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 | [] | [
"Merging with #1506"
] | 1,571 | 1,571 | 1,571 | NONE | null | Hi,
I want to have a seq2seq model from gpt2, if I change the script of "run_lm_finetuning.py" in a way that it gets a sequence, then make it a context ids, and let it generate another sequence, like "run_generation.py" code, then minimize the cross-entropy loss, does it this way, create a seq2seq model? I rgreatly ap... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1579/reactions",
"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/1579/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1578 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1578/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1578/comments | https://api.github.com/repos/huggingface/transformers/issues/1578/events | https://github.com/huggingface/transformers/issues/1578 | 509,611,258 | MDU6SXNzdWU1MDk2MTEyNTg= | 1,578 | distilled gpt2 to be added to run_generation and run_lm_fintuning | {
"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 | [] | [
"Hi, DistilGPT-2 is considered to be a checkpoint of GPT-2 in our library (differently to DistilBERT). You can already use DistilGPT-2 for both of these scripts with the following:\r\n```bash\r\npython run_generation --model_type=gpt2 --model_name_or_path=distilgpt2\r\n```"
] | 1,571 | 1,571 | 1,571 | NONE | null | Hi
I greatly appreciated also adding distilled GPT2 to the codes above, thanks | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1578/reactions",
"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/1578/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1577 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1577/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1577/comments | https://api.github.com/repos/huggingface/transformers/issues/1577/events | https://github.com/huggingface/transformers/pull/1577 | 509,610,055 | MDExOlB1bGxSZXF1ZXN0MzMwMTI3NzY0 | 1,577 | Add feature #1572 which gives support for multiple candidate sequences | {
"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 | null | [] | [
"Please note that main() now returns a list with ```num_samples``` elements inside.\r\n\r\nBecause of this, the test for run_generation.py should be updated to test for ```length``` for each element within the list. This explains why **build_py3_torch** test failed.\r\n\r\nI will update ```ExamplesTests.test_genera... | 1,571 | 1,572 | 1,572 | CONTRIBUTOR | null | **Multiple candidate sequences can be generated by setting ```num_samples > 1``` (still 1 by default).**
EXAMPLE with ```num_samples == 2``` for a GPT2 model:
```
INPUT:
Why did the chicken
OUTPUT:
cross the road <eoq> To go to the other side. <eoa>
eat food <eoq> Because it was hungry <eoa>
```... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1577/reactions",
"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/1577/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/1577",
"html_url": "https://github.com/huggingface/transformers/pull/1577",
"diff_url": "https://github.com/huggingface/transformers/pull/1577.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/1577.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/1576 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1576/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1576/comments | https://api.github.com/repos/huggingface/transformers/issues/1576/events | https://github.com/huggingface/transformers/issues/1576 | 509,596,089 | MDU6SXNzdWU1MDk1OTYwODk= | 1,576 | evaluating on race dataset with checkpoints fine tuned on roberta with fairseq | {
"login": "qshi95",
"id": 23690677,
"node_id": "MDQ6VXNlcjIzNjkwNjc3",
"avatar_url": "https://avatars.githubusercontent.com/u/23690677?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/qshi95",
"html_url": "https://github.com/qshi95",
"followers_url": "https://api.github.com/users/qshi95/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 | [] | [
"Do you get any improvement? \r\nThe ACC of eval and test has a huge gap.\r\n",
" --classification-head when converting the models ",
"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",
... | 1,571 | 1,582 | 1,580 | NONE | null | ## ❓ Questions & Help
<!-- A clear and concise description of the question. -->
I fine tuned a model on race dataset with reberta, following the fairseq instruction, got the result:
| epoch 004 | valid on 'valid' subset: | loss 0.913 | nll_loss 0.003 | ppl 1.00 | num_updates 21849 | best_accuracy 0.846563 | accur... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1576/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/1576/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1575 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1575/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1575/comments | https://api.github.com/repos/huggingface/transformers/issues/1575/events | https://github.com/huggingface/transformers/issues/1575 | 509,586,024 | MDU6SXNzdWU1MDk1ODYwMjQ= | 1,575 | use gpt2 as a seq2seq model | {
"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 | [] | [
"Merging with #1506"
] | 1,571 | 1,571 | 1,571 | NONE | null | Hi
could you assist me please and show me with example on how I can use GPT-2 language model decoding method so train seq2seq model? thanks a lot | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1575/reactions",
"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/1575/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1574 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1574/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1574/comments | https://api.github.com/repos/huggingface/transformers/issues/1574/events | https://github.com/huggingface/transformers/issues/1574 | 509,560,147 | MDU6SXNzdWU1MDk1NjAxNDc= | 1,574 | Why the output is same within a batch use BertForSequenceClassification? | {
"login": "201101050424",
"id": 6504096,
"node_id": "MDQ6VXNlcjY1MDQwOTY=",
"avatar_url": "https://avatars.githubusercontent.com/u/6504096?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/201101050424",
"html_url": "https://github.com/201101050424",
"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 | [] | [
"Hello, could you provide a script so that we may better understand the problem here?",
"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"
] | 1,571 | 1,577 | 1,577 | NONE | null | ## ❓ Questions & Help
I use BertForSequenceClassification for classification task, but the output witin a batch became same after just 2 or 3 batch, the value between different batch is different, really strange.
batch 1 output:
[-0.5966, 0.6081],
[-0.4659, 0.3766],
[-0.3595, 0.1334],
... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1574/reactions",
"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/1574/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1573 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1573/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1573/comments | https://api.github.com/repos/huggingface/transformers/issues/1573/events | https://github.com/huggingface/transformers/issues/1573 | 509,548,923 | MDU6SXNzdWU1MDk1NDg5MjM= | 1,573 | GPT2 attention mask and output masking | {
"login": "anandhperumal",
"id": 12907396,
"node_id": "MDQ6VXNlcjEyOTA3Mzk2",
"avatar_url": "https://avatars.githubusercontent.com/u/12907396?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/anandhperumal",
"html_url": "https://github.com/anandhperumal",
"followers_url": "https://api.githu... | [] | closed | false | null | [] | [
"Closing this because I found my answer",
"Hi, would you mind sharing what the answer you found is? Thank you so much!",
"Sorry for the delay. Gpt2 was trained as a CLM model with a fixed block size of data. So there was no need for attention mask. (That is what I understood). "
] | 1,571 | 1,576 | 1,571 | NONE | null | ## ❓ Questions & Help
<!-- A clear and concise description of the question. -->
I have couple of questions:
1. In the original gpt2 they didn't pad the sequence, so they didn't need a attention mask, but in other cases where we our input sequence is small and we pad the input, don't we need a attention mask?
2... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1573/reactions",
"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/1573/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1572 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1572/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1572/comments | https://api.github.com/repos/huggingface/transformers/issues/1572/events | https://github.com/huggingface/transformers/issues/1572 | 509,545,595 | MDU6SXNzdWU1MDk1NDU1OTU= | 1,572 | Can we generate multiple possible sentences using GPT? | {
"login": "zhaoxy92",
"id": 21225257,
"node_id": "MDQ6VXNlcjIxMjI1MjU3",
"avatar_url": "https://avatars.githubusercontent.com/u/21225257?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/zhaoxy92",
"html_url": "https://github.com/zhaoxy92",
"followers_url": "https://api.github.com/users/zha... | [] | closed | false | null | [] | [
"@zhaoxy92 I happen to have a use case for this as well. I'll add in this feature to the ```run_generation.py```",
"@zhaoxy92 Added this functionality in ```run_generation.py```. You can set the number of candidates generated by setting the argument ```num_samples``` which is set to 1 by default.",
"I think yo... | 1,571 | 1,572 | 1,572 | NONE | null | Hi,
Is there any way to generate multiple candidate text sequences using the pretrained generators? | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1572/reactions",
"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/1572/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1571 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1571/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1571/comments | https://api.github.com/repos/huggingface/transformers/issues/1571/events | https://github.com/huggingface/transformers/issues/1571 | 509,516,014 | MDU6SXNzdWU1MDk1MTYwMTQ= | 1,571 | Pytorch Transformers no longer loads SciBert weights, getting `UnicodeDecodeError`. Worked in pytorch_pretrained_bert | {
"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.... | [] | closed | false | null | [] | [
"`from_pretrained` expects the following files: `vocab.txt`, `config.json` and `pytorch_model.bin`. \r\n\r\nThus, you only need to extract the `weights.tar.gz` archive. \r\n\r\nThen rename `bert_config.json` to `config.json` and pass the path name to the `from_pretrained` method: this should be `/content/scibert_sc... | 1,571 | 1,571 | 1,571 | CONTRIBUTOR | null | ## 🐛 Bug
<!-- Important information -->
When using the old pytorch_pretrained_bert library, I could point the model with `from_pretrained` to the SciBert weights.tar.gz file, and it would load this just. However, if I try this with the Pytorch Transformers, I get this error.
```
UnicodeDecodeError: 'utf-8' ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1571/reactions",
"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/1571/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1570 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1570/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1570/comments | https://api.github.com/repos/huggingface/transformers/issues/1570/events | https://github.com/huggingface/transformers/pull/1570 | 509,512,914 | MDExOlB1bGxSZXF1ZXN0MzMwMDYwNDk5 | 1,570 | Fix Roberta on TPU | {
"login": "rickysaurav",
"id": 13986039,
"node_id": "MDQ6VXNlcjEzOTg2MDM5",
"avatar_url": "https://avatars.githubusercontent.com/u/13986039?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/rickysaurav",
"html_url": "https://github.com/rickysaurav",
"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 | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/1570?src=pr&el=h1) Report\n> Merging [#1570](https://codecov.io/gh/huggingface/transformers/pull/1570?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/82f6abd98aaa691ca0adfe21e85a17dc6f386497?src=pr&el=desc) will **n... | 1,571 | 1,577 | 1,577 | NONE | null | Fixes #1569
- Revert tf.print() to logger , since tf.print() is an unsupported TPU ops. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1570/reactions",
"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/1570/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/1570",
"html_url": "https://github.com/huggingface/transformers/pull/1570",
"diff_url": "https://github.com/huggingface/transformers/pull/1570.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/1570.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/1569 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1569/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1569/comments | https://api.github.com/repos/huggingface/transformers/issues/1569/events | https://github.com/huggingface/transformers/issues/1569 | 509,512,157 | MDU6SXNzdWU1MDk1MTIxNTc= | 1,569 | TFRobertaForSequenceClassification fails on TPU on Transformers >2.0.0 | {
"login": "rickysaurav",
"id": 13986039,
"node_id": "MDQ6VXNlcjEzOTg2MDM5",
"avatar_url": "https://avatars.githubusercontent.com/u/13986039?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/rickysaurav",
"html_url": "https://github.com/rickysaurav",
"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"
] | 1,571 | 1,577 | 1,577 | NONE | null | ## 🐛 Bug
<!-- Important information -->
Model I am using (TFRobertaForSequenceClassification):
Language I am using the model on (English):
The problem arise when using:
* [ ] the official example scripts: (give details)
* [x] my own modified scripts: (give details)
The tasks I am working on is:
* [ ]... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1569/reactions",
"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/1569/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1568 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1568/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1568/comments | https://api.github.com/repos/huggingface/transformers/issues/1568/events | https://github.com/huggingface/transformers/pull/1568 | 509,504,935 | MDExOlB1bGxSZXF1ZXN0MzMwMDU0NDkw | 1,568 | Fix hanging when loading pretrained models | {
"login": "daemon",
"id": 6188572,
"node_id": "MDQ6VXNlcjYxODg1NzI=",
"avatar_url": "https://avatars.githubusercontent.com/u/6188572?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/daemon",
"html_url": "https://github.com/daemon",
"followers_url": "https://api.github.com/users/daemon/foll... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/1568?src=pr&el=h1) Report\n> Merging [#1568](https://codecov.io/gh/huggingface/transformers/pull/1568?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/82f6abd98aaa691ca0adfe21e85a17dc6f386497?src=pr&el=desc) will **d... | 1,571 | 1,571 | 1,571 | CONTRIBUTOR | null | - Fix hanging when loading pretrained models from the cache without having internet access. This is a widespread issue on supercomputers whose internal compute nodes are firewalled. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1568/reactions",
"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/1568/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/1568",
"html_url": "https://github.com/huggingface/transformers/pull/1568",
"diff_url": "https://github.com/huggingface/transformers/pull/1568.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/1568.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/1567 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1567/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1567/comments | https://api.github.com/repos/huggingface/transformers/issues/1567/events | https://github.com/huggingface/transformers/pull/1567 | 509,420,268 | MDExOlB1bGxSZXF1ZXN0MzMwMDAwMzM2 | 1,567 | Added mixed precision (AMP) to inference benchmark | {
"login": "tlkh",
"id": 5409617,
"node_id": "MDQ6VXNlcjU0MDk2MTc=",
"avatar_url": "https://avatars.githubusercontent.com/u/5409617?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/tlkh",
"html_url": "https://github.com/tlkh",
"followers_url": "https://api.github.com/users/tlkh/followers",
... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/1567?src=pr&el=h1) Report\n> Merging [#1567](https://codecov.io/gh/huggingface/transformers/pull/1567?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/079bfb32fba4f2b39d344ca7af88d79a3ff27c7c?src=pr&el=desc) will **n... | 1,571 | 1,572 | 1,572 | CONTRIBUTOR | null | I added a mixed precision option to the benchmark script and ran it on a DGX Station to get the results. As you can see, we can get between 1.2x to up to 4.5x inference speed depending on model, batch size and sequence length.
**Summary**
| Batch Size | Speedup (XLA only) | Speedup (XLA + AMP) | Min. Seq Len* |
... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1567/reactions",
"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/1567/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/1567",
"html_url": "https://github.com/huggingface/transformers/pull/1567",
"diff_url": "https://github.com/huggingface/transformers/pull/1567.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/1567.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/1566 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1566/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1566/comments | https://api.github.com/repos/huggingface/transformers/issues/1566/events | https://github.com/huggingface/transformers/issues/1566 | 509,419,560 | MDU6SXNzdWU1MDk0MTk1NjA= | 1,566 | error load bert model ：not found model file | {
"login": "iambyd",
"id": 11927058,
"node_id": "MDQ6VXNlcjExOTI3MDU4",
"avatar_url": "https://avatars.githubusercontent.com/u/11927058?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/iambyd",
"html_url": "https://github.com/iambyd",
"followers_url": "https://api.github.com/users/iambyd/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 | [] | [
"I don't understand when you get this error?",
"In order to understand when you've encountered this bug, as suggested by @iedmrc , you've to write down the source code that generates the bug! And please show your environment (Python, Transformers, PyTorch, TensorFlow versions) too! \r\n\r\n> Error content:\r\n> O... | 1,571 | 1,589 | 1,581 | NONE | null | Error content:
OSError: Error no file named ['pytorch_model.bin', 'tf_model.h5', 'model.ckpt.index'] found in directory ./uncased_L-12_H-768_A-12_transformers or `from_tf` set to False
but file exists
| {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1566/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/1566/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1565 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1565/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1565/comments | https://api.github.com/repos/huggingface/transformers/issues/1565/events | https://github.com/huggingface/transformers/issues/1565 | 509,384,799 | MDU6SXNzdWU1MDkzODQ3OTk= | 1,565 | How to add the output word vector of bert to my model | {
"login": "e-tuanzi",
"id": 49483010,
"node_id": "MDQ6VXNlcjQ5NDgzMDEw",
"avatar_url": "https://avatars.githubusercontent.com/u/49483010?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/e-tuanzi",
"html_url": "https://github.com/e-tuanzi",
"followers_url": "https://api.github.com/users/e-t... | [
{
"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 think you can see [here](https://jalammar.github.io/a-visual-guide-to-using-bert-for-the-first-time/). In more details, this tutorial... | 1,571 | 1,582 | 1,582 | NONE | null | ## ❓ Questions & Help
Hello, I am a student who is learning nlp.
Now I want to use the word vector output by bert to apply to my model, but **I can't connect the word vector to the network**. Could you give me an example program or tutorial about this which use textCNN or LSTM. You can sent e-mail to **vvain0208@163... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1565/reactions",
"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/1565/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1564 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1564/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1564/comments | https://api.github.com/repos/huggingface/transformers/issues/1564/events | https://github.com/huggingface/transformers/issues/1564 | 509,380,922 | MDU6SXNzdWU1MDkzODA5MjI= | 1,564 | ALBERT: will it be supported? | {
"login": "xinqipony",
"id": 42603620,
"node_id": "MDQ6VXNlcjQyNjAzNjIw",
"avatar_url": "https://avatars.githubusercontent.com/u/42603620?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/xinqipony",
"html_url": "https://github.com/xinqipony",
"followers_url": "https://api.github.com/users/... | [] | closed | false | null | [] | [
"https://github.com/brightmart/albert_zh",
"Please direct all your questions to the main albert topic. https://github.com/huggingface/transformers/issues/1370 \r\n\r\nPlease close this current topic. It does not add anything.",
"... We should extend the issue template and redirect all ALBERT questions to #1370 ... | 1,571 | 1,571 | 1,571 | NONE | null | will you release an ALBERT model?
it sets the new state of art;
# 🌟New model addition
ALBERT: A LITE BERT FOR SELF-SUPERVISED
LEARNING OF LANGUAGE REPRESENTATIONS
https://arxiv.org/pdf/1909.11942.pdf
## Model description
![image](https://user-images.githubusercontent.com/42603620/67136720-eedec280-f25c-11e... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1564/reactions",
"total_count": 10,
"+1": 6,
"-1": 0,
"laugh": 0,
"hooray": 2,
"confused": 0,
"heart": 0,
"rocket": 2,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/1564/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1563 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1563/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1563/comments | https://api.github.com/repos/huggingface/transformers/issues/1563/events | https://github.com/huggingface/transformers/issues/1563 | 509,361,014 | MDU6SXNzdWU1MDkzNjEwMTQ= | 1,563 | The implementation of grad clipping is not correct when gradient accumulation is enabled | {
"login": "yangyiben",
"id": 6025986,
"node_id": "MDQ6VXNlcjYwMjU5ODY=",
"avatar_url": "https://avatars.githubusercontent.com/u/6025986?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/yangyiben",
"html_url": "https://github.com/yangyiben",
"followers_url": "https://api.github.com/users/ya... | [] | closed | false | null | [] | [
"This should be fixed in https://github.com/huggingface/transformers/pull/1580",
"yeah @yangyiben please let me know if that merge fixes it",
"#1580 is now merged"
] | 1,571 | 1,572 | 1,572 | NONE | null | torch.nn.utils.clip_grad_norm_ should be applied before optimizer.step() not after each backward
## 🐛 Bug
<!-- Important information -->
Model I am using (Bert, XLNet....):
Language I am using the model on (English, Chinese....):
The problem arise when using:
* [ ] the official example scripts: (gi... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1563/reactions",
"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/1563/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1562 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1562/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1562/comments | https://api.github.com/repos/huggingface/transformers/issues/1562/events | https://github.com/huggingface/transformers/issues/1562 | 509,120,090 | MDU6SXNzdWU1MDkxMjAwOTA= | 1,562 | training BERT on coreference resolution | {
"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 | [] | [
"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",
"There's a newer [approach](https://github.com/mandarjoshi90/coref) using BERT but it's using tensorflow 1.14. I wish if we could get th... | 1,571 | 1,597 | 1,577 | NONE | null | Hi
I really appreciate if you could add codes to train BERT on coref resolution dataset of CONLL-2012, thanks | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1562/reactions",
"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/1562/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1561 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1561/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1561/comments | https://api.github.com/repos/huggingface/transformers/issues/1561/events | https://github.com/huggingface/transformers/issues/1561 | 509,107,804 | MDU6SXNzdWU1MDkxMDc4MDQ= | 1,561 | [CLS] & [SEP] tokens missing in documentation | {
"login": "hawkeoni",
"id": 27156990,
"node_id": "MDQ6VXNlcjI3MTU2OTkw",
"avatar_url": "https://avatars.githubusercontent.com/u/27156990?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/hawkeoni",
"html_url": "https://github.com/hawkeoni",
"followers_url": "https://api.github.com/users/haw... | [] | closed | false | null | [] | [
"@hawkeoni [CLS] and [SEP] tokens are added automatically as long as you use the tokenizer, ```BertTokenizer```",
"@enzoampil It doesn't seem to work.\r\nThe following code\r\n```python\r\nfrom transformers import BertTokenizer\r\ntokenizer = BertTokenizer.from_pretrained('bert-base-uncased')\r\nsentence = \"He... | 1,571 | 1,572 | 1,571 | CONTRIBUTOR | null | https://github.com/huggingface/transformers/blob/fd97761c5a977fd22df789d2851cf57c7c9c0930/transformers/modeling_bert.py#L1017-L1023
In this example of bert for token classification input sentence is encoded, but [CLS] & [SEP] tokens are not added. Is this intentional or just a typo?
Do I need to add [CLS] & [SEP] tok... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1561/reactions",
"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/1561/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1560 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1560/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1560/comments | https://api.github.com/repos/huggingface/transformers/issues/1560/events | https://github.com/huggingface/transformers/issues/1560 | 509,033,024 | MDU6SXNzdWU1MDkwMzMwMjQ= | 1,560 | Finetuning OpenAI GPT-2 for another language. | {
"login": "0x01h",
"id": 32897657,
"node_id": "MDQ6VXNlcjMyODk3NjU3",
"avatar_url": "https://avatars.githubusercontent.com/u/32897657?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/0x01h",
"html_url": "https://github.com/0x01h",
"followers_url": "https://api.github.com/users/0x01h/follow... | [
{
"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 | [] | [
"Hello, if you want to try and fine-tune GPT-2 to another language, you can just give the `run_lm_finetuning` script your text in the other language on which you want to fine-tune your model. \r\n\r\nHowever, please be aware that according to the language and its distance to the English language (language on which ... | 1,571 | 1,578 | 1,578 | NONE | null | ## ❓ Questions & Help
Hi,
Is there any option to finetune and use OpenAI GPT-2 for another language except English? | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1560/reactions",
"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/1560/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1559 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1559/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1559/comments | https://api.github.com/repos/huggingface/transformers/issues/1559/events | https://github.com/huggingface/transformers/issues/1559 | 509,011,557 | MDU6SXNzdWU1MDkwMTE1NTc= | 1,559 | Compatibility between DistilBert and Bert models | {
"login": "pvgladkov",
"id": 1284012,
"node_id": "MDQ6VXNlcjEyODQwMTI=",
"avatar_url": "https://avatars.githubusercontent.com/u/1284012?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/pvgladkov",
"html_url": "https://github.com/pvgladkov",
"followers_url": "https://api.github.com/users/pv... | [
{
"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"
] | 1,571 | 1,577 | 1,577 | NONE | null | ## ❓ Questions & Help
I have a regular classification task for sentences in russian language.
I used to train `BertForSequenceClassification` with pretrained Bert from [DeepPavlov](http://docs.deeppavlov.ai/en/master/features/models/bert.html) [RuBERT](http://files.deeppavlov.ai/deeppavlov_data/bert/rubert_cased_L... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1559/reactions",
"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/1559/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1558 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1558/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1558/comments | https://api.github.com/repos/huggingface/transformers/issues/1558/events | https://github.com/huggingface/transformers/issues/1558 | 508,983,477 | MDU6SXNzdWU1MDg5ODM0Nzc= | 1,558 | unable to parse E:/litao/bert/bert-base-cased\config.json as a URL or as a local path | {
"login": "754563116",
"id": 32032029,
"node_id": "MDQ6VXNlcjMyMDMyMDI5",
"avatar_url": "https://avatars.githubusercontent.com/u/32032029?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/754563116",
"html_url": "https://github.com/754563116",
"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 | [] | [
"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",
"Did you find a solution?",
"> Did you find a solution?\r\n\r\nRename \"bert_config.json\" to \"config.json\"."
] | 1,571 | 1,622 | 1,577 | NONE | null | ## ❓ Questions & Help
<!-- A clear and concise description of the question. -->
train_file = 'E:/litao/bert/SQuAD 1.1/train-v1.1.json'
predict_file = 'E:/litao/bert/SQuAD 1.1/dev-v1.1.json'
model_type = 'bert'
model_name_or_path = 'E:/litao/bert/bert-base-cased'
output_dir = 'E:/litao/bert/tran... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1558/reactions",
"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/1558/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1557 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1557/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1557/comments | https://api.github.com/repos/huggingface/transformers/issues/1557/events | https://github.com/huggingface/transformers/issues/1557 | 508,889,984 | MDU6SXNzdWU1MDg4ODk5ODQ= | 1,557 | Tuning BERT on our own data set for multi-class classification problem | {
"login": "rshah1990",
"id": 37735152,
"node_id": "MDQ6VXNlcjM3NzM1MTUy",
"avatar_url": "https://avatars.githubusercontent.com/u/37735152?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/rshah1990",
"html_url": "https://github.com/rshah1990",
"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 | [] | [
"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"
] | 1,571 | 1,577 | 1,577 | NONE | null | I want to tune pre-trained BERT for multi-class classification with **6 million class, 30 million rows & highly imbalance data set.**
Can we tune BERT in batch of classes?
For example, I will take 15 classes (last layer will have only 15 neuron) and train my BERT model & in next batch use that trained model to tra... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1557/reactions",
"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/1557/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1556 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1556/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1556/comments | https://api.github.com/repos/huggingface/transformers/issues/1556/events | https://github.com/huggingface/transformers/issues/1556 | 508,857,672 | MDU6SXNzdWU1MDg4NTc2NzI= | 1,556 | Does the function of 'evaluate()' change the result? | {
"login": "lceustc",
"id": 37608648,
"node_id": "MDQ6VXNlcjM3NjA4NjQ4",
"avatar_url": "https://avatars.githubusercontent.com/u/37608648?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lceustc",
"html_url": "https://github.com/lceustc",
"followers_url": "https://api.github.com/users/lceust... | [] | closed | false | null | [] | [
"Could you specify what script you're running, with which parameters? Did you set a random seed?",
"> Could you specify what script you're running, with which parameters? Did you set a random seed?\r\n\r\nfor SEEDS in 99\r\n\r\ndo\r\nCUDA_VISIBLE_DEVICES=2 python run_glue.py \\\r\n --data_dir '/data/trans... | 1,571 | 1,571 | 1,571 | NONE | null | ## ❓ Questions & Help
<!-- A clear and concise description of the question. -->
when i run RTE task , and logging steps=50: the result is:
gloabl_step 50: 0.8953
global_step 100: 0.8953
gloabl_step 150: 0.8916
global_step 200: 0.8736
but when logging steps =100:
global_step 100: 0.... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1556/reactions",
"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/1556/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1555 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1555/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1555/comments | https://api.github.com/repos/huggingface/transformers/issues/1555/events | https://github.com/huggingface/transformers/pull/1555 | 508,808,738 | MDExOlB1bGxSZXF1ZXN0MzI5NTI5Mjk4 | 1,555 | Sample a constant number of tokens for masking in LM finetuning | {
"login": "rakeshchada",
"id": 2664691,
"node_id": "MDQ6VXNlcjI2NjQ2OTE=",
"avatar_url": "https://avatars.githubusercontent.com/u/2664691?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/rakeshchada",
"html_url": "https://github.com/rakeshchada",
"followers_url": "https://api.github.com/us... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/1555?src=pr&el=h1) Report\n> Merging [#1555](https://codecov.io/gh/huggingface/transformers/pull/1555?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/fd97761c5a977fd22df789d2851cf57c7c9c0930?src=pr&el=desc) will **i... | 1,571 | 1,573 | 1,573 | CONTRIBUTOR | null | For Masked LM fine-tuning, I think both the original BERT and RoBERTa implementations uniformly sample x number of tokens in *each* sequence for masking (where x = mlm_probability * 100 * sequence_length)
However, The current logic in run_lm_finetuning.py does an indepdendent sampling (from bernoulli distribution) for... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1555/reactions",
"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/1555/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/1555",
"html_url": "https://github.com/huggingface/transformers/pull/1555",
"diff_url": "https://github.com/huggingface/transformers/pull/1555.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/1555.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/1554 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1554/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1554/comments | https://api.github.com/repos/huggingface/transformers/issues/1554/events | https://github.com/huggingface/transformers/issues/1554 | 508,786,635 | MDU6SXNzdWU1MDg3ODY2MzU= | 1,554 | GPT2 not in modeltype | {
"login": "tuhinjubcse",
"id": 3104771,
"node_id": "MDQ6VXNlcjMxMDQ3NzE=",
"avatar_url": "https://avatars.githubusercontent.com/u/3104771?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/tuhinjubcse",
"html_url": "https://github.com/tuhinjubcse",
"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 | [] | [
"Hey @tuhinjubcse gpt2 is a text generation model. If you look in the run_glue.py file you will see your options for model selection for using the run_glue.py script.\r\n```\r\nMODEL_CLASSES = {\r\n 'bert': (BertConfig, BertForSequenceClassification, BertTokenizer),\r\n 'xlnet': (XLNetConfig, XLNetForSequence... | 1,571 | 1,581 | 1,581 | NONE | null | ## 🐛 Bug
<!-- Important information -->
Model I am using (Bert, XLNet....): GPT2
Language I am using the model on (English, Chinese....): ENGLISH
The problem arise when using:
* [ ] the official example scripts: (give details) run_glue.py
* [ ] my own modified scripts: (give details)
The tasks I am wo... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1554/reactions",
"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/1554/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1553 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1553/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1553/comments | https://api.github.com/repos/huggingface/transformers/issues/1553/events | https://github.com/huggingface/transformers/pull/1553 | 508,736,364 | MDExOlB1bGxSZXF1ZXN0MzI5NDc1MjUz | 1,553 | Add speed log to examples/run_squad.py | {
"login": "WilliamTambellini",
"id": 109458,
"node_id": "MDQ6VXNlcjEwOTQ1OA==",
"avatar_url": "https://avatars.githubusercontent.com/u/109458?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/WilliamTambellini",
"html_url": "https://github.com/WilliamTambellini",
"followers_url": "https://a... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/1553?src=pr&el=h1) Report\n> Merging [#1553](https://codecov.io/gh/huggingface/transformers/pull/1553?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/fd97761c5a977fd22df789d2851cf57c7c9c0930?src=pr&el=desc) will **i... | 1,571 | 1,572 | 1,572 | CONTRIBUTOR | null | Add a speed estimate log (time per example)
for evaluation to examples/run_squad.py | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1553/reactions",
"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/1553/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/1553",
"html_url": "https://github.com/huggingface/transformers/pull/1553",
"diff_url": "https://github.com/huggingface/transformers/pull/1553.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/1553.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/1552 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1552/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1552/comments | https://api.github.com/repos/huggingface/transformers/issues/1552/events | https://github.com/huggingface/transformers/issues/1552 | 508,694,853 | MDU6SXNzdWU1MDg2OTQ4NTM= | 1,552 | There is not space after generating an 'special token' and the next word using gpt2. | {
"login": "fabrahman",
"id": 22799593,
"node_id": "MDQ6VXNlcjIyNzk5NTkz",
"avatar_url": "https://avatars.githubusercontent.com/u/22799593?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/fabrahman",
"html_url": "https://github.com/fabrahman",
"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 | [] | [
"Can you show exactly how you ran ```run_generation.py```?",
"@enzoampil thanks, this is my command:\r\n```python run_generation.py --model_type=gpt2 --model_name_or_path=gpt2_finetuned/ --top_k 10 --temperature 0.8 --top_p 0.0 --stop_token \"<|endoftext|>\" ```",
"Can you try running ```python run_generation.p... | 1,571 | 1,649 | 1,582 | NONE | null | ## ❓ Questions & Help
Hi,
I have used ```run_lm_finetuning.py``` to finetune gpt2 and then tried to do some generation. I have added a couple of special token to dictionary and the finetuned gpt2 without any problem.
Then when I am doing the generation using ```run_generation.py```, I realized whenever the model g... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1552/reactions",
"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/1552/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1551 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1551/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1551/comments | https://api.github.com/repos/huggingface/transformers/issues/1551/events | https://github.com/huggingface/transformers/pull/1551 | 508,640,194 | MDExOlB1bGxSZXF1ZXN0MzI5Mzk1MzAy | 1,551 | [FIX] fix repetition penalty in `examples/run_generation.py` | {
"login": "leo-du",
"id": 15058481,
"node_id": "MDQ6VXNlcjE1MDU4NDgx",
"avatar_url": "https://avatars.githubusercontent.com/u/15058481?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/leo-du",
"html_url": "https://github.com/leo-du",
"followers_url": "https://api.github.com/users/leo-du/fo... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/1551?src=pr&el=h1) Report\n> Merging [#1551](https://codecov.io/gh/huggingface/transformers/pull/1551?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/c5441946112e68441b46866d114bf8d3c29b0c1d?src=pr&el=desc) will **n... | 1,571 | 1,571 | 1,571 | CONTRIBUTOR | null | The repetition penalty in `examples/run_generation.py` is incorrectly implemented due to the following snippet.
```python
for _ in set(generated):
next_token_logits[_] /= repetition_penalty
```
`generated` is a tensor, and python built-in `set` does not compare tensors correctly, e.... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1551/reactions",
"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/1551/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/1551",
"html_url": "https://github.com/huggingface/transformers/pull/1551",
"diff_url": "https://github.com/huggingface/transformers/pull/1551.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/1551.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/1550 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1550/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1550/comments | https://api.github.com/repos/huggingface/transformers/issues/1550/events | https://github.com/huggingface/transformers/issues/1550 | 508,619,252 | MDU6SXNzdWU1MDg2MTkyNTI= | 1,550 | training BERT from scratch for native language PT-BR? Without init weight | {
"login": "calusbr",
"id": 25322394,
"node_id": "MDQ6VXNlcjI1MzIyMzk0",
"avatar_url": "https://avatars.githubusercontent.com/u/25322394?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/calusbr",
"html_url": "https://github.com/calusbr",
"followers_url": "https://api.github.com/users/calusb... | [
{
"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"
] | 1,571 | 1,577 | 1,577 | NONE | null | I would like to train BERT from scratch for a textual base in PT-BR (8GB data). Is it possible to use the run_lm_finetuning.py code to perform this process without using the multi-language bert model?
I already have a vocab.txt for the PT-BR base and I don't want to load initial weights.
Is there any script or tu... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1550/reactions",
"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/1550/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1549 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1549/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1549/comments | https://api.github.com/repos/huggingface/transformers/issues/1549/events | https://github.com/huggingface/transformers/pull/1549 | 508,546,747 | MDExOlB1bGxSZXF1ZXN0MzI5MzE4OTM4 | 1,549 | Fix token order in xlnet preprocessing for SQuAD | {
"login": "hlums",
"id": 16907204,
"node_id": "MDQ6VXNlcjE2OTA3MjA0",
"avatar_url": "https://avatars.githubusercontent.com/u/16907204?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/hlums",
"html_url": "https://github.com/hlums",
"followers_url": "https://api.github.com/users/hlums/follow... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/1549?src=pr&el=h1) Report\n> Merging [#1549](https://codecov.io/gh/huggingface/transformers/pull/1549?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/8a62835577a2a93642546858b21372e43c1a1ff8?src=pr&el=desc) will **n... | 1,571 | 1,573 | 1,572 | CONTRIBUTOR | null | #947
My current result on SQuAD 1.1
{
"exact": 85.45884578997162,
"f1": 92.5974600601065,
"total": 10570,
"HasAns_exact": 85.45884578997162,
"HasAns_f1": 92.59746006010651,
"HasAns_total": 10570
}
My code validation command
```
python /data/home/hlu/transformers/examples/run_squad.py \
-... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1549/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/1549/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/1549",
"html_url": "https://github.com/huggingface/transformers/pull/1549",
"diff_url": "https://github.com/huggingface/transformers/pull/1549.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/1549.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/1548 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1548/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1548/comments | https://api.github.com/repos/huggingface/transformers/issues/1548/events | https://github.com/huggingface/transformers/pull/1548 | 508,539,560 | MDExOlB1bGxSZXF1ZXN0MzI5MzEzMzU4 | 1,548 | [2.2] - Command-line interface - Pipeline class | {
"login": "thomwolf",
"id": 7353373,
"node_id": "MDQ6VXNlcjczNTMzNzM=",
"avatar_url": "https://avatars.githubusercontent.com/u/7353373?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/thomwolf",
"html_url": "https://github.com/thomwolf",
"followers_url": "https://api.github.com/users/thomw... | [] | 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://... | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/1548?src=pr&el=h1) Report\n> Merging [#1548](https://codecov.io/gh/huggingface/transformers/pull/1548?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/33adab2b91697b3e78af618a21ab9f1176281165?src=pr&el=desc) will **d... | 1,571 | 1,586 | 1,576 | MEMBER | null | Adding a `Pipeline` class that encapsulates a `Tokenizer` and a `Model`.
`Pipelines` take python objects as inputs (lists/dict of string/int/float) and output python objects as well (lists/dict of string/int/float).
`Pipelines` can be used to query and train models and should be framework agnostic (default to TF ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1548/reactions",
"total_count": 6,
"+1": 4,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 2
} | https://api.github.com/repos/huggingface/transformers/issues/1548/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/1548",
"html_url": "https://github.com/huggingface/transformers/pull/1548",
"diff_url": "https://github.com/huggingface/transformers/pull/1548.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/1548.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/1547 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1547/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1547/comments | https://api.github.com/repos/huggingface/transformers/issues/1547/events | https://github.com/huggingface/transformers/issues/1547 | 508,527,704 | MDU6SXNzdWU1MDg1Mjc3MDQ= | 1,547 | Is it possible/is there a plan to enable continued pretraining? | {
"login": "oligiles0",
"id": 20796307,
"node_id": "MDQ6VXNlcjIwNzk2MzA3",
"avatar_url": "https://avatars.githubusercontent.com/u/20796307?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/oligiles0",
"html_url": "https://github.com/oligiles0",
"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 @oligiles0, you can actually use ```run_lm_finetuning.py``` for this. You can find more details in the **RoBERTa/BERT and masked language modeling** section in the README",
"> Hi @oligiles0, you can actually use `run_lm_finetuning.py` for this. You can find more details in the **RoBERTa/BERT and masked langua... | 1,571 | 1,581 | 1,581 | NONE | null | ## 🚀 Feature
Standardised interface to pretrain various Transformers with standardised expectations with regards to formatting training data.
## Motivation
To achieve state of the art within a given domain it is not sufficient to take models pretrained on nonspecific literature (wikipedia/books/etc). The ide... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1547/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/1547/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1546 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1546/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1546/comments | https://api.github.com/repos/huggingface/transformers/issues/1546/events | https://github.com/huggingface/transformers/issues/1546 | 508,470,322 | MDU6SXNzdWU1MDg0NzAzMjI= | 1,546 | Q / Note: BERT Masked-LM fails to predict last token in sequence if it is not punctuation | {
"login": "IngoMarquart",
"id": 44617909,
"node_id": "MDQ6VXNlcjQ0NjE3OTA5",
"avatar_url": "https://avatars.githubusercontent.com/u/44617909?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/IngoMarquart",
"html_url": "https://github.com/IngoMarquart",
"followers_url": "https://api.github.c... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"If you play with the script a bit, you can see that the loss for BERT with the MLM head is actually quite high, as someone suggested elsewhere, this may be due to pre-training on different tasks than just MLM",
"This issue has been automatically marked as stale because it has not had recent activity. It will be ... | 1,571 | 1,577 | 1,577 | NONE | null | ## ❓ Questions & Help
I am playing with BERT to see what the distributions of the prediction for a MASK token are. I wrote a quick script that successively masks all words in an input sequence.
This is based on the implementation in the examples (e.g. the lm finetuning script and the examples in the documentation)... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1546/reactions",
"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/1546/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1545 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1545/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1545/comments | https://api.github.com/repos/huggingface/transformers/issues/1545/events | https://github.com/huggingface/transformers/issues/1545 | 508,457,835 | MDU6SXNzdWU1MDg0NTc4MzU= | 1,545 | Adding new tokens to uncased tokenizers - case insensitivity is lost | {
"login": "mbednarski",
"id": 13330503,
"node_id": "MDQ6VXNlcjEzMzMwNTAz",
"avatar_url": "https://avatars.githubusercontent.com/u/13330503?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/mbednarski",
"html_url": "https://github.com/mbednarski",
"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 | [] | [
"Hmm looks like BertTokenizer's super class handles `.add_tokens()` and the first steps of `.tokenize()`, and doesn't really seem to consider whether the tokens should be made lowercase. I'm not sure whether it's intentional, but I'll make a PR and find out :smile: \r\n\r\nIn the meantime, it might be a good idea t... | 1,571 | 1,577 | 1,577 | NONE | null | ## ❓ Questions & Help
Hello!
I'm trying to add new tokens to bert-base-uncased. Let's say my token is '**cool-token**' and it was not present in the original vocab
```
from transformers import BertTokenizer
tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
print(tokenizer.tokenize('Sentence wit... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1545/reactions",
"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/1545/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1544 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1544/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1544/comments | https://api.github.com/repos/huggingface/transformers/issues/1544/events | https://github.com/huggingface/transformers/issues/1544 | 508,414,669 | MDU6SXNzdWU1MDg0MTQ2Njk= | 1,544 | the num_labels in run_squad | {
"login": "zhujun5164",
"id": 49580602,
"node_id": "MDQ6VXNlcjQ5NTgwNjAy",
"avatar_url": "https://avatars.githubusercontent.com/u/49580602?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/zhujun5164",
"html_url": "https://github.com/zhujun5164",
"followers_url": "https://api.github.com/use... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"Hey @zhujun5164, 2 is the right setting of num_labels for the task. If you look at the model they use (say Bert is BertForQuestionAnswering), you'll see that they get two outputs for each position which is from the num_labels = 2. The two outputs correspond to the start_logits position and the end_logits position.... | 1,571 | 1,577 | 1,577 | NONE | null | ## ❓ Questions & Help
In the run_squad, I have not find any code to define the num_labels. In the modeling_utils, the num_labels have been default as 2, but in the Question & Answer task it may predict the start_position and end_position in the inputs. Is there the code have missing consider the reset of num_labels, o... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1544/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/1544/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1543 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1543/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1543/comments | https://api.github.com/repos/huggingface/transformers/issues/1543/events | https://github.com/huggingface/transformers/issues/1543 | 508,297,036 | MDU6SXNzdWU1MDgyOTcwMzY= | 1,543 | Where is pytorch-pretrained-BERT? | {
"login": "ShallTearchen",
"id": 25740395,
"node_id": "MDQ6VXNlcjI1NzQwMzk1",
"avatar_url": "https://avatars.githubusercontent.com/u/25740395?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ShallTearchen",
"html_url": "https://github.com/ShallTearchen",
"followers_url": "https://api.githu... | [] | closed | false | null | [] | [
"`pytorch-pretrained-BERT` is this library, but four or five months ago. It evolved into `pytorch-transformers` as more models were added to the library, before becoming `transformers` as we now have a front-end for both pytorch and tensorflow.",
"is this still an issue?",
"I don't think so. In my opinion, @Sha... | 1,571 | 1,575 | 1,575 | NONE | null | ## ❓ Questions & Help
<!-- A clear and concise description of the question. -->
As the title shows, where is pytorch-pretrained-BERT? Please tell me the path, THX. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1543/reactions",
"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/1543/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1542 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1542/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1542/comments | https://api.github.com/repos/huggingface/transformers/issues/1542/events | https://github.com/huggingface/transformers/issues/1542 | 508,281,913 | MDU6SXNzdWU1MDgyODE5MTM= | 1,542 | Running CTRL Model On Google Colab Environment | {
"login": "yusufani",
"id": 35346311,
"node_id": "MDQ6VXNlcjM1MzQ2MzEx",
"avatar_url": "https://avatars.githubusercontent.com/u/35346311?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/yusufani",
"html_url": "https://github.com/yusufani",
"followers_url": "https://api.github.com/users/yus... | [] | closed | false | null | [] | [
"[In the official repo,](https://github.com/salesforce/ctrl) you can find a 'lower_memory' branch. You can take a look there. As always, you can try to make the batch size and max sequence length smaller, too.",
"Thank you for your help",
"Please close this topic if you have no further questions. "
] | 1,571 | 1,571 | 1,571 | NONE | null | ## ❓ Questions & Help
As you know Google Colab environment has ****12 GB** ram** limit . When i want to run [run_generation.py](https://github.com/huggingface/transformers/blob/master/examples/run_generation.py) file , Colab automatically stops the process. How much ram does the CTRL Model need? Where can I learn th... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1542/reactions",
"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/1542/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1541 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1541/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1541/comments | https://api.github.com/repos/huggingface/transformers/issues/1541/events | https://github.com/huggingface/transformers/issues/1541 | 508,229,991 | MDU6SXNzdWU1MDgyMjk5OTE= | 1,541 | Type of model for each GLUE task | {
"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 | [] | [
"All models on GLUE should use BertForSequenceClassification (MNLI is 3 class, STS-B is 1 class).",
"As specified in the documentation, for `XxxForSequenceClassification` models:\r\n\r\n```\r\nIf ``config.num_labels == 1`` a regression loss is computed (Mean-Square loss),\r\nIf ``config.num_labels > 1`` a classif... | 1,571 | 1,573 | 1,573 | NONE | null | ## ❓ Questions & Help
There are nine GLUE tasks, and I wanted to verify which BERT model type is best suited for each task. Can anyone confirm these matchings? I am not sure what to do for STS-B especially, and am unsure if BertForMultipleChoice is perhaps the correct option for MNLI.
| {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1541/reactions",
"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/1541/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1540 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1540/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1540/comments | https://api.github.com/repos/huggingface/transformers/issues/1540/events | https://github.com/huggingface/transformers/issues/1540 | 508,168,936 | MDU6SXNzdWU1MDgxNjg5MzY= | 1,540 | Should the option to run on TPU in run_glue.py use some sort of xla data parallelizer ? | {
"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 | [] | [
"Indeed, it would be great to improve the current TPU script to include better optimization, such as using the TPU DataParallel from Pytorch. We haven't gotten to it yet and we'll probably do so soon.\r\n\r\nWe'd be very happy to welcome a PR too! :)",
"Sounds good, I'm trying to figure it out. The part I'm stuck... | 1,571 | 1,577 | 1,577 | CONTRIBUTOR | null | ## ❓ Questions & Help
<!-- A clear and concise description of the question. -->
The xla API ( https://github.com/pytorch/xla/blob/master/API_GUIDE.md ) and the TPU colab examples ( https://github.com/pytorch/xla/tree/master/contrib/colab ) each parallelize their data, either using a `torch_xla.distributed.paralle... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1540/reactions",
"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/1540/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1539 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1539/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1539/comments | https://api.github.com/repos/huggingface/transformers/issues/1539/events | https://github.com/huggingface/transformers/issues/1539 | 508,166,767 | MDU6SXNzdWU1MDgxNjY3Njc= | 1,539 | A couple of noob-to-transformers questions | {
"login": "GrahamboJangles",
"id": 36944031,
"node_id": "MDQ6VXNlcjM2OTQ0MDMx",
"avatar_url": "https://avatars.githubusercontent.com/u/36944031?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/GrahamboJangles",
"html_url": "https://github.com/GrahamboJangles",
"followers_url": "https://api... | [
{
"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",
"@GrahamboJangles\r\n\r\n1) If you want to provide _context_ into any model offered by Transformers, you can **fine-tune** the model you... | 1,571 | 1,582 | 1,582 | NONE | null | ## ❓ Questions & Help
#### If you don't want to or don't know the answer to all of these, just answer some that you know!
1. How is it that you can provide context to these models? Say, if you want to summarize or pull data from a text. Do you have to train it on that text or just put it somehow in the prompt?
2. Ca... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1539/reactions",
"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/1539/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1538 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1538/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1538/comments | https://api.github.com/repos/huggingface/transformers/issues/1538/events | https://github.com/huggingface/transformers/issues/1538 | 508,140,503 | MDU6SXNzdWU1MDgxNDA1MDM= | 1,538 | Fine-tune RoBERTa on WikiText-2 | {
"login": "estoica111",
"id": 56658382,
"node_id": "MDQ6VXNlcjU2NjU4Mzgy",
"avatar_url": "https://avatars.githubusercontent.com/u/56658382?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/estoica111",
"html_url": "https://github.com/estoica111",
"followers_url": "https://api.github.com/use... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"Hi, could you give us a bit more information? For example, you seem to be running this on a GPU, are you running on a distributed setting? Could you list your software versions (python, torch, transformers)?",
"Thank you for your response. I am running on a single machine with one gpu,\nPython 3.6.8, pytorch_tra... | 1,571 | 1,586 | 1,586 | NONE | null | ## ❓ Questions & Help
I am trying to train Roberta using the run_lm_finetuning.py script and TRAIN_FILE=wiki.train.raw, TEST_FILE=wiki.test.raw, basically, I use the demo data (wikiText-2) as described at https://huggingface.co/transformers/examples.html
CUDA_LAUNCH_BLOCKING=1 python run_lm_finetuning.py \
-... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1538/reactions",
"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/1538/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1537 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1537/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1537/comments | https://api.github.com/repos/huggingface/transformers/issues/1537/events | https://github.com/huggingface/transformers/issues/1537 | 508,006,416 | MDU6SXNzdWU1MDgwMDY0MTY= | 1,537 | Behavior of Masked-LM BERT, dependence on masked token | {
"login": "IngoMarquart",
"id": 44617909,
"node_id": "MDQ6VXNlcjQ0NjE3OTA5",
"avatar_url": "https://avatars.githubusercontent.com/u/44617909?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/IngoMarquart",
"html_url": "https://github.com/IngoMarquart",
"followers_url": "https://api.github.c... | [] | closed | false | null | [] | [
"Yes, this is nice illustration of the discrepancy between Bert's training (in which masked tokens are provided) and Bert's testing (in which no masked token is provided).",
"I've noticed that this also frequently occurs when the last token in the sentence is masked. \r\n\r\nFor example,\r\n`['[CLS]', 'donald', '... | 1,571 | 1,586 | 1,571 | NONE | null | I am experimenting with the masked-LM for BERT to understand how the masking affects predictions of the other tokens.
Of course using no [MASK] is not the intended usage, nor is it to predict each token in the sentence. But my understanding is that the LM head is a separate softmax classifier, taking the final embed... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1537/reactions",
"total_count": 6,
"+1": 6,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/1537/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1536 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1536/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1536/comments | https://api.github.com/repos/huggingface/transformers/issues/1536/events | https://github.com/huggingface/transformers/issues/1536 | 507,824,313 | MDU6SXNzdWU1MDc4MjQzMTM= | 1,536 | Penalize high confident false negative classifications? | {
"login": "callenilsson",
"id": 16915094,
"node_id": "MDQ6VXNlcjE2OTE1MDk0",
"avatar_url": "https://avatars.githubusercontent.com/u/16915094?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/callenilsson",
"html_url": "https://github.com/callenilsson",
"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 | [] | [
"The softmax function specifically uses exponentiation to exacerbate the differences in scores (to get the soft 'max'). You can normalize scores by other means than a softmax.\r\n\r\nRelated to your title: using a log loss will penalize wrong predictions with high confidence more (e.g. BCE).",
"Related read is th... | 1,571 | 1,577 | 1,577 | NONE | null | ## ❓ Questions & Help
I added the line `logits = torch.nn.functional.softmax(logits)` to convert binary classifications to a confidence score between 0.0 - 1.0. However, the predictions are very harsh being really close to either 0.0 or 1.0 and not somewhere in between. Is there a way to penalize the model from being ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1536/reactions",
"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/1536/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1535 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1535/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1535/comments | https://api.github.com/repos/huggingface/transformers/issues/1535/events | https://github.com/huggingface/transformers/issues/1535 | 507,820,870 | MDU6SXNzdWU1MDc4MjA4NzA= | 1,535 | Why the output of DistilBertModel is inconsistent with BertModel?! | {
"login": "amirj",
"id": 1645137,
"node_id": "MDQ6VXNlcjE2NDUxMzc=",
"avatar_url": "https://avatars.githubusercontent.com/u/1645137?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/amirj",
"html_url": "https://github.com/amirj",
"followers_url": "https://api.github.com/users/amirj/follower... | [] | closed | false | null | [] | [
"Hello @amirj,\r\n\r\nThe \"pooled_output\" is the hidden state of the `[CLS]`. It is this hidden state that is used for classification tasks for instance (see DistilBertForSequenceClassification). So you could retrieve it by filtering out `hidden_states `.\r\n\r\nThe reason why there is no linear transformation in... | 1,571 | 1,644 | 1,571 | NONE | null | [The output of DistilBertModel](https://github.com/huggingface/transformers/blob/be916cb3fb4579e278ceeaec11a6524662797d7f/transformers/modeling_distilbert.py#L468) does not contain [pooled_output as available in BERT model](https://github.com/huggingface/transformers/blob/be916cb3fb4579e278ceeaec11a6524662797d7f/transf... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1535/reactions",
"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/1535/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1534 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1534/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1534/comments | https://api.github.com/repos/huggingface/transformers/issues/1534/events | https://github.com/huggingface/transformers/issues/1534 | 507,798,363 | MDU6SXNzdWU1MDc3OTgzNjM= | 1,534 | run_ner.py file with Distill Bert | {
"login": "amankedia",
"id": 8494998,
"node_id": "MDQ6VXNlcjg0OTQ5OTg=",
"avatar_url": "https://avatars.githubusercontent.com/u/8494998?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/amankedia",
"html_url": "https://github.com/amankedia",
"followers_url": "https://api.github.com/users/am... | [] | closed | false | null | [] | [
"I have the same issue. Did you end up using Distilbert?",
"Not sure how well it will perform. Casing is an important feature used in many NER tasks. So I would say it _could_ work, but ymmv. For reference: https://stackoverflow.com/questions/56384231/case-sensitive-entity-recognition",
"RoBERTa is cased so you... | 1,571 | 1,572 | 1,572 | NONE | null | ## ❓ Questions & Help
<!-- A clear and concise description of the question. -->
I wish to use the Distill Bert model for NER. I am not sure if it will work with it directly. Any suggestions on that end would be great.
Also, what values should the parameters **--model_type** and **--model_name_or_path** take for Di... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1534/reactions",
"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/1534/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1533 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1533/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1533/comments | https://api.github.com/repos/huggingface/transformers/issues/1533/events | https://github.com/huggingface/transformers/issues/1533 | 507,793,041 | MDU6SXNzdWU1MDc3OTMwNDE= | 1,533 | Add vocabulary gives sequence length warning | {
"login": "callenilsson",
"id": 16915094,
"node_id": "MDQ6VXNlcjE2OTE1MDk0",
"avatar_url": "https://avatars.githubusercontent.com/u/16915094?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/callenilsson",
"html_url": "https://github.com/callenilsson",
"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 | [] | [
"Hi, this warning means that the sequence you have encoded is longer than the maximum sequence length the model can handle. It isn't related to the tokens you have added.\r\n\r\nRoBERTa can only handle sequences of a maximum of 512 tokens, so you should make sure you only pass sequences of a max length of 512 or el... | 1,571 | 1,581 | 1,581 | NONE | null | ## ❓ Questions & Help
I'm trying to add extra vocabulary to RoBERTa using the `tokenizer.add_tokens()` function. However, when training I get the following warning message:
`WARNING - transformers.tokenization_utils - Token indices sequence length is longer than the specified maximum sequence length for this mode... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1533/reactions",
"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/1533/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1532 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1532/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1532/comments | https://api.github.com/repos/huggingface/transformers/issues/1532/events | https://github.com/huggingface/transformers/issues/1532 | 507,709,752 | MDU6SXNzdWU1MDc3MDk3NTI= | 1,532 | 'BertForSequenceClassification' is not defined 'DUMMY_INPUTS' is not defined | {
"login": "roccqqck",
"id": 34628766,
"node_id": "MDQ6VXNlcjM0NjI4NzY2",
"avatar_url": "https://avatars.githubusercontent.com/u/34628766?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/roccqqck",
"html_url": "https://github.com/roccqqck",
"followers_url": "https://api.github.com/users/roc... | [] | closed | false | null | [] | [
"next time i \r\n```\r\nimport torch\r\n```\r\nit showed\r\n```\r\n---------------------------------------------------------------------------\r\nNameError Traceback (most recent call last)\r\n<ipython-input-14-71a5c3f94250> in <module>\r\n----> 1 pytorch_model = BertForSequenceClass... | 1,571 | 1,614 | 1,571 | NONE | null | transformers-2.1.1
https://github.com/huggingface/transformers#quick-tour-tf-20-training-and-pytorch-interoperability
i just copied and paste and run the code.
it showed
```
NameError: name 'BertForSequenceClassification' is not defined
```
i can't even
```
from transformers import BertForSequenceClassi... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1532/reactions",
"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/1532/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1531 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1531/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1531/comments | https://api.github.com/repos/huggingface/transformers/issues/1531/events | https://github.com/huggingface/transformers/issues/1531 | 507,635,191 | MDU6SXNzdWU1MDc2MzUxOTE= | 1,531 | why xlnet requires a long prompt for short inputs while Bert does not ? | {
"login": "muiPomeranian",
"id": 29085131,
"node_id": "MDQ6VXNlcjI5MDg1MTMx",
"avatar_url": "https://avatars.githubusercontent.com/u/29085131?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/muiPomeranian",
"html_url": "https://github.com/muiPomeranian",
"followers_url": "https://api.githu... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.\n"
] | 1,571 | 1,576 | 1,576 | NONE | null | hey guys,
Q1)
can someone give some more insight what @thomwolf explaining about?
'''
#846
The main reason you get bad performance is that XLNet is not good on short inputs (comes from the way it is pretrained, always having a long memory and only guessing a few words in the sequence).
The run_generation exa... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1531/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/1531/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1530 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1530/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1530/comments | https://api.github.com/repos/huggingface/transformers/issues/1530/events | https://github.com/huggingface/transformers/issues/1530 | 507,582,508 | MDU6SXNzdWU1MDc1ODI1MDg= | 1,530 | Plan to support UniLM ? | {
"login": "astariul",
"id": 43774355,
"node_id": "MDQ6VXNlcjQzNzc0MzU1",
"avatar_url": "https://avatars.githubusercontent.com/u/43774355?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/astariul",
"html_url": "https://github.com/astariul",
"followers_url": "https://api.github.com/users/ast... | [] | closed | false | null | [] | [
"This is on our mid-term roadmap.\r\nWe have a project adding Seq2seq models and UniLM will be part of this project."
] | 1,571 | 1,571 | 1,571 | CONTRIBUTOR | null | # 🌟New model addition
## Model description
**UniLM** : Pre-trained transformer for sequence to sequence generation.
Paper : https://arxiv.org/pdf/1905.03197.pdf
## Open Source status
* [x] the model implementation is available: **[official Pytorch](https://github.com/microsoft/unilm)**
* [x] the model ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1530/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/1530/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1529 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1529/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1529/comments | https://api.github.com/repos/huggingface/transformers/issues/1529/events | https://github.com/huggingface/transformers/issues/1529 | 507,567,684 | MDU6SXNzdWU1MDc1Njc2ODQ= | 1,529 | Hight CPU and low GPU on XLNet | {
"login": "laiviet",
"id": 29224591,
"node_id": "MDQ6VXNlcjI5MjI0NTkx",
"avatar_url": "https://avatars.githubusercontent.com/u/29224591?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/laiviet",
"html_url": "https://github.com/laiviet",
"followers_url": "https://api.github.com/users/laivie... | [
{
"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 also meet the above problem with XLNet. The GPU usage is very low and unstable, but the CPU usage is very high. The running speed is very low.\r\n\r\nAre there any ops running on CPU rather than GPU in your XLNet implementation? How to improve the GPU usage and speed up the running speed ? Thanks!\r\n\r\nEnviron... | 1,571 | 1,592 | 1,585 | NONE | null | ## 🐛 Bug
I am running Bert, GPT, GPT2, XLNET. I got very high CPU usage (e.g. 16 cores) with XLNet while the others (Bert, GPT, GPT2) dont.
For BERT, GPT, GPT2: CPU 1 cores, 100%GPU
For XLNet: CPU 16 cores, 50 to 60% GPU
Is there any hidden implementation which requires CPU?
<!-- Important informatio... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1529/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/1529/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1528 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1528/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1528/comments | https://api.github.com/repos/huggingface/transformers/issues/1528/events | https://github.com/huggingface/transformers/issues/1528 | 507,503,848 | MDU6SXNzdWU1MDc1MDM4NDg= | 1,528 | Question about hidden states in GPT2 | {
"login": "weiguowilliam",
"id": 31396452,
"node_id": "MDQ6VXNlcjMxMzk2NDUy",
"avatar_url": "https://avatars.githubusercontent.com/u/31396452?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/weiguowilliam",
"html_url": "https://github.com/weiguowilliam",
"followers_url": "https://api.githu... | [] | closed | false | null | [] | [
"Hi! The vector of the `hidden_states` is indeed of shape `(13, seq_len, 768)`. The first value (`hidden_states[0]`), of shape `(seq_len, 768)` corresponds to the sum of the word + positional embeddings. The subsequent values are added every time the model goes through an attention layer.\r\n\r\nWithout taking into... | 1,571 | 1,572 | 1,572 | NONE | null | ## ❓ Questions & Help
<!-- A clear and concise description of the question. -->
>tokenizer = GPT2Tokenizer.from_pretrained('gpt2')
>model = GPT2LMHeadModel.from_pretrained('gpt2',output_hidden_states=True)
>model.eval()
>input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute")).unsqueeze(0) # Batch ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1528/reactions",
"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/1528/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1527 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1527/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1527/comments | https://api.github.com/repos/huggingface/transformers/issues/1527/events | https://github.com/huggingface/transformers/issues/1527 | 507,407,932 | MDU6SXNzdWU1MDc0MDc5MzI= | 1,527 | Training GPT or GPT-2 from scratch | {
"login": "LeonCrashCode",
"id": 5652525,
"node_id": "MDQ6VXNlcjU2NTI1MjU=",
"avatar_url": "https://avatars.githubusercontent.com/u/5652525?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/LeonCrashCode",
"html_url": "https://github.com/LeonCrashCode",
"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 | [] | [
"I think, just create an instance of the model (without loading from pretrained one), switch it to train mode and run. That's all.",
"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",
"... | 1,571 | 1,585 | 1,581 | NONE | null | ## ❓ Questions & Help
<!-- A clear and concise description of the question. -->
I am trying to retrain GPT or GPT-2 from scratch, is there any implementation for this? | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1527/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/1527/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1526 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1526/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1526/comments | https://api.github.com/repos/huggingface/transformers/issues/1526/events | https://github.com/huggingface/transformers/issues/1526 | 507,341,324 | MDU6SXNzdWU1MDczNDEzMjQ= | 1,526 | Alignment of tokens - 'extract_features_aligned_to_words' from fairseq roberta? | {
"login": "sidsvash26",
"id": 10222970,
"node_id": "MDQ6VXNlcjEwMjIyOTcw",
"avatar_url": "https://avatars.githubusercontent.com/u/10222970?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sidsvash26",
"html_url": "https://github.com/sidsvash26",
"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",
"It seems this is still unsupported by Huggingface?"
] | 1,571 | 1,605 | 1,576 | NONE | null | ## ❓ Questions & Help
I'm using RoBERTa pretrained model to get embeddings for a dataset. But I want to get the embeddings as per the tokenization which is already present in my dataset. So basically I would want to average the embeddings of a token's BPE if that token in my dataset is getting split into different B... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1526/reactions",
"total_count": 5,
"+1": 5,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/1526/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1525 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1525/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1525/comments | https://api.github.com/repos/huggingface/transformers/issues/1525/events | https://github.com/huggingface/transformers/issues/1525 | 507,282,099 | MDU6SXNzdWU1MDcyODIwOTk= | 1,525 | Understanding run_glue in distributed mode | {
"login": "BramVanroy",
"id": 2779410,
"node_id": "MDQ6VXNlcjI3Nzk0MTA=",
"avatar_url": "https://avatars.githubusercontent.com/u/2779410?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/BramVanroy",
"html_url": "https://github.com/BramVanroy",
"followers_url": "https://api.github.com/users... | [] | closed | false | null | [] | [
"To understand how DDP synchronize across processes, you can read:\r\n- the official doc for DDP: https://pytorch.org/docs/stable/nn.html?highlight=distributed%20data%20parallel#torch.nn.parallel.DistributedDataParallel\r\n- this detailed blog post I did a few months ago: https://medium.com/huggingface/training-lar... | 1,571 | 1,571 | 1,571 | COLLABORATOR | null | ## ❓ Questions & Help
In my own project I am building on top of `transformers` and I'd like to take advantage of DDP. For inspiration I've been looking at how different libraries implement that, as well as how `transformers` handles it. In particular, I've been looking at [`run_glue`](https://github.com/huggingface/... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1525/reactions",
"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/1525/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1524 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1524/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1524/comments | https://api.github.com/repos/huggingface/transformers/issues/1524/events | https://github.com/huggingface/transformers/issues/1524 | 507,270,125 | MDU6SXNzdWU1MDcyNzAxMjU= | 1,524 | Question on AllenNLP vocabulary and huggingface BERT out of sync | {
"login": "pruksmhc",
"id": 10094008,
"node_id": "MDQ6VXNlcjEwMDk0MDA4",
"avatar_url": "https://avatars.githubusercontent.com/u/10094008?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/pruksmhc",
"html_url": "https://github.com/pruksmhc",
"followers_url": "https://api.github.com/users/pru... | [
{
"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"
] | 1,571 | 1,576 | 1,576 | NONE | null | Perhaps this should also be posted in the allennlp repo.
I'm currently trying to use a pretrained model (clinicalBERT) with a different set of vocabulary with huggingface's BertModel. Even though the code runs, I'm not 100% convinced that the vocabulary index to weight mappings are synced between the allennlp vocabul... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1524/reactions",
"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/1524/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1523 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1523/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1523/comments | https://api.github.com/repos/huggingface/transformers/issues/1523/events | https://github.com/huggingface/transformers/issues/1523 | 507,183,517 | MDU6SXNzdWU1MDcxODM1MTc= | 1,523 | Why the codes of training BERT from scratch are deprecated | {
"login": "anhnt170489",
"id": 24732444,
"node_id": "MDQ6VXNlcjI0NzMyNDQ0",
"avatar_url": "https://avatars.githubusercontent.com/u/24732444?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/anhnt170489",
"html_url": "https://github.com/anhnt170489",
"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 | [] | [
"They were community provided and the core team didn't have the bandwidth to maintain them.\r\n\r\nAlso we want to limit the number of single-model examples now and favor examples that work for a range of models.\r\n\r\nIf you want to update them to the current version of the repo and add the various models (for in... | 1,571 | 1,576 | 1,576 | NONE | null | ## ❓ Questions & Help
I'm wondering why the team removed the codes of training BERT from scratch, including pregenerate_training_data.py and finetune_on_pregenerated.py. They're very helpful and I still continue developing them to train the BERT as well as Roberta from scratch. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1523/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/1523/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1522 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1522/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1522/comments | https://api.github.com/repos/huggingface/transformers/issues/1522/events | https://github.com/huggingface/transformers/issues/1522 | 507,116,851 | MDU6SXNzdWU1MDcxMTY4NTE= | 1,522 | When to support Albert? | {
"login": "c0derm4n",
"id": 18226382,
"node_id": "MDQ6VXNlcjE4MjI2Mzgy",
"avatar_url": "https://avatars.githubusercontent.com/u/18226382?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/c0derm4n",
"html_url": "https://github.com/c0derm4n",
"followers_url": "https://api.github.com/users/c0d... | [] | closed | false | null | [] | [
"Please use the search function. There's an open issue about albert here: https://github.com/huggingface/transformers/issues/1370"
] | 1,571 | 1,571 | 1,571 | NONE | null | ## ❓ Questions & Help
<!-- A clear and concise description of the question. -->
Do you have any plan to support Google's new model-ALBERT? | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1522/reactions",
"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/1522/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1521 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1521/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1521/comments | https://api.github.com/repos/huggingface/transformers/issues/1521/events | https://github.com/huggingface/transformers/issues/1521 | 507,080,976 | MDU6SXNzdWU1MDcwODA5NzY= | 1,521 | Downloading model in distributed mode | {
"login": "BramVanroy",
"id": 2779410,
"node_id": "MDQ6VXNlcjI3Nzk0MTA=",
"avatar_url": "https://avatars.githubusercontent.com/u/2779410?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/BramVanroy",
"html_url": "https://github.com/BramVanroy",
"followers_url": "https://api.github.com/users... | [] | closed | false | null | [] | [
"This should be fixed in most of the examples through the use of `torch.distributed.barrier`.\r\nE.g. here: https://github.com/huggingface/transformers/blob/master/examples/run_glue.py#L473\r\n\r\nDon't hesitate to submit a PR if some examples don't make use of this technique yet.",
"Thanks for the quick reply! S... | 1,571 | 1,571 | 1,571 | COLLABORATOR | null | ## 🐛 Bug
When running in distributed mode with `n` processes, a new model will be download `n` times. I don't think that's what you want. I found [this related issue](https://github.com/huggingface/transformers/issues/44) but that only fixed the race condition. Downloads still happen in parallel. Is there a way to ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1521/reactions",
"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/1521/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1520 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1520/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1520/comments | https://api.github.com/repos/huggingface/transformers/issues/1520/events | https://github.com/huggingface/transformers/issues/1520 | 507,068,447 | MDU6SXNzdWU1MDcwNjg0NDc= | 1,520 | Changelog | {
"login": "BramVanroy",
"id": 2779410,
"node_id": "MDQ6VXNlcjI3Nzk0MTA=",
"avatar_url": "https://avatars.githubusercontent.com/u/2779410?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/BramVanroy",
"html_url": "https://github.com/BramVanroy",
"followers_url": "https://api.github.com/users... | [] | closed | false | null | [] | [
"Hi @BramVanroy, we detail the changes in the [\"Releases\" section](https://github.com/huggingface/transformers/releases). Are you thinking of something different?\r\n\r\nHaving a documentation per-version is on our roadmap, it should help tremendously regarding version changes.",
"Ah, I was looking inside diff... | 1,571 | 1,571 | 1,571 | COLLABORATOR | null | ## 🚀 Add changelog between versions
New versions are pushed to PyPi at a steady pace, but it's not evident to find the changes that new versions bring. Is there a changelog anywhere? Something similar to a HISTORY file would be nice. I think it would definitely contribute to better documentation!
| {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1520/reactions",
"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/1520/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1519 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1519/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1519/comments | https://api.github.com/repos/huggingface/transformers/issues/1519/events | https://github.com/huggingface/transformers/issues/1519 | 507,029,636 | MDU6SXNzdWU1MDcwMjk2MzY= | 1,519 | Accuracy drop in finetuning roBERTa | {
"login": "Arjunsankarlal",
"id": 28828445,
"node_id": "MDQ6VXNlcjI4ODI4NDQ1",
"avatar_url": "https://avatars.githubusercontent.com/u/28828445?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Arjunsankarlal",
"html_url": "https://github.com/Arjunsankarlal",
"followers_url": "https://api.gi... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.\n"
] | 1,571 | 1,576 | 1,576 | NONE | null | ## ❓ How to achieve GLUE leaderboard acc for QQP task trained with roBERTa?
I am trying to finetune roberta-base model for Quora Question Pair task. In the [GLUE Leaderboard](url) the accuracy claimed is F1 / Accuracy is 74.3/90.2
I am training with the following command to fine tune the roberta-base model,
> ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1519/reactions",
"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/1519/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1518 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1518/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1518/comments | https://api.github.com/repos/huggingface/transformers/issues/1518/events | https://github.com/huggingface/transformers/issues/1518 | 506,944,288 | MDU6SXNzdWU1MDY5NDQyODg= | 1,518 | Predefined token classification | {
"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 | [] | [
"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"
] | 1,571 | 1,576 | 1,576 | NONE | null | Hello,
I am just wondering if "BertForTokenClassification" can be modified to classify predefined tokens (just targeted tokens). E.g. in NER it identifies the entity and then classify it, but in my case I want to classify the targeted tokens only in a sentence with a predefined labels. I thought of adding the target... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1518/reactions",
"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/1518/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1517 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1517/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1517/comments | https://api.github.com/repos/huggingface/transformers/issues/1517/events | https://github.com/huggingface/transformers/issues/1517 | 506,913,371 | MDU6SXNzdWU1MDY5MTMzNzE= | 1,517 | Unable to import TF models | {
"login": "tylerjthomas9",
"id": 36181311,
"node_id": "MDQ6VXNlcjM2MTgxMzEx",
"avatar_url": "https://avatars.githubusercontent.com/u/36181311?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/tylerjthomas9",
"html_url": "https://github.com/tylerjthomas9",
"followers_url": "https://api.githu... | [] | closed | false | null | [] | [
"Can you run the following and report back? It might be that you have some namespace conflict.\r\n\r\n```python\r\n! pip list | grep \"tensorflow\" # Check tensorflow==2.0.0, tensorflow-gpu==2.0.0\r\n! pip list | grep \"transformers\" # Check transformers>=2.0.0\r\n```\r\n",
"Cleaning the environment fixed the ... | 1,571 | 1,572 | 1,571 | NONE | null | ## 🐛 Bug
<!-- Important information -->
Model I am using (Bert, XLNet....): Bert
Language I am using the model on (English, Chinese....): English
The problem arise when using:
* [x] the official example scripts: Quick tour TF 2.0 training and PyTorch interoperability from github homepage
## To Reproduc... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1517/reactions",
"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/1517/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1516 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1516/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1516/comments | https://api.github.com/repos/huggingface/transformers/issues/1516/events | https://github.com/huggingface/transformers/pull/1516 | 506,859,955 | MDExOlB1bGxSZXF1ZXN0MzI3OTYxMzk4 | 1,516 | Fused optimizer and gradient clipper using apex | {
"login": "slayton58",
"id": 4992598,
"node_id": "MDQ6VXNlcjQ5OTI1OTg=",
"avatar_url": "https://avatars.githubusercontent.com/u/4992598?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/slayton58",
"html_url": "https://github.com/slayton58",
"followers_url": "https://api.github.com/users/sl... | [
{
"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/1516?src=pr&el=h1) Report\n> Merging [#1516](https://codecov.io/gh/huggingface/transformers/pull/1516?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/1d4d07025635c998acf8c7abab426b013e87206c?src=pr&el=desc) will **d... | 1,571 | 1,583 | 1,583 | CONTRIBUTOR | null | Significant (40ms / iter for XLNet squad finetuning) performance increase.
Also adds fused grad clipping, gives further ~30ms / iter saving in the same XLNet squad case.
Redefines the `AdamW` implementation such that the existing code will be used if apex's multi_tensor_apply code isn't available, and should dro... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1516/reactions",
"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/1516/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/1516",
"html_url": "https://github.com/huggingface/transformers/pull/1516",
"diff_url": "https://github.com/huggingface/transformers/pull/1516.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/1516.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/1515 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1515/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1515/comments | https://api.github.com/repos/huggingface/transformers/issues/1515/events | https://github.com/huggingface/transformers/issues/1515 | 506,754,502 | MDU6SXNzdWU1MDY3NTQ1MDI= | 1,515 | Main and train for CTRL model | {
"login": "roholazandie",
"id": 7584674,
"node_id": "MDQ6VXNlcjc1ODQ2NzQ=",
"avatar_url": "https://avatars.githubusercontent.com/u/7584674?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/roholazandie",
"html_url": "https://github.com/roholazandie",
"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 | [] | [
"The CTRL model has been added to the [run_generation](https://github.com/huggingface/transformers/blob/master/examples/run_generation.py) script as of now. We will implement it in other scripts as time goes on, but as it has the same API as the other models hosted on our repo it the training script would be very s... | 1,571 | 1,576 | 1,576 | NONE | null | ## 🚀 Feature
I have seen the CTRL model has been added to the repo but I don't see any script to run or train it. Is this going to be added soon?
| {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1515/reactions",
"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/1515/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1514 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1514/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1514/comments | https://api.github.com/repos/huggingface/transformers/issues/1514/events | https://github.com/huggingface/transformers/issues/1514 | 506,736,406 | MDU6SXNzdWU1MDY3MzY0MDY= | 1,514 | /pytorch/aten/src/THC/THCTensorScatterGather.cu:100: void THCudaTensor_gatherKernel(TensorInfo<Real, IndexType>, TensorInfo<Real, IndexType>, TensorInfo<long, IndexType>, int, IndexType) [with IndexType = unsigned int, Real = float, Dims = 3]: block: [4,0,0], thread: [319,0,0] Assertion `indexValue >= 0 && indexValue <... | {
"login": "DeepFool",
"id": 48155157,
"node_id": "MDQ6VXNlcjQ4MTU1MTU3",
"avatar_url": "https://avatars.githubusercontent.com/u/48155157?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/DeepFool",
"html_url": "https://github.com/DeepFool",
"followers_url": "https://api.github.com/users/Dee... | [
{
"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 | [] | [
"Could you please provide more information? Where does this error occur? Are you using one of our example scripts? I believe there are templates you can use so we may help you more efficiently.",
"> Could you please provide more information? Where does this error occur? Are you using one of our example scripts? I... | 1,571 | 1,582 | 1,582 | NONE | null | ## ❓ Questions & Help
<!-- A clear and concise description of the question. -->
I have no idea about this error。 | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1514/reactions",
"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/1514/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/1513 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1513/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1513/comments | https://api.github.com/repos/huggingface/transformers/issues/1513/events | https://github.com/huggingface/transformers/pull/1513 | 506,717,326 | MDExOlB1bGxSZXF1ZXN0MzI3ODUwOTMy | 1,513 | Force einsum to run in fp16 | {
"login": "slayton58",
"id": 4992598,
"node_id": "MDQ6VXNlcjQ5OTI1OTg=",
"avatar_url": "https://avatars.githubusercontent.com/u/4992598?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/slayton58",
"html_url": "https://github.com/slayton58",
"followers_url": "https://api.github.com/users/sl... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/1513?src=pr&el=h1) Report\n> Merging [#1513](https://codecov.io/gh/huggingface/transformers/pull/1513?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/f62f992cf7aa7f1e4eb0d1ef912bd06d26c4dd8c?src=pr&el=desc) will **n... | 1,571 | 1,626 | 1,571 | CONTRIBUTOR | null | As noted in the comments, this will force `torch.einsum` to run in fp16 for the squad finetuning task (it should be valid for other tasks, but I haven't verified that) when run with `--fp16_opt_level="O1"` which is the default.
Otherwise, `torch.einsum` is treated as a "promote" operation by `apex.amp`, and if any a... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1513/reactions",
"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/1513/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/1513",
"html_url": "https://github.com/huggingface/transformers/pull/1513",
"diff_url": "https://github.com/huggingface/transformers/pull/1513.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/1513.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/1512 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1512/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1512/comments | https://api.github.com/repos/huggingface/transformers/issues/1512/events | https://github.com/huggingface/transformers/pull/1512 | 506,517,337 | MDExOlB1bGxSZXF1ZXN0MzI3Njk2MjY1 | 1,512 | Fix import error in script to convert faisreq roberta checkpoints | {
"login": "louismartin",
"id": 12654189,
"node_id": "MDQ6VXNlcjEyNjU0MTg5",
"avatar_url": "https://avatars.githubusercontent.com/u/12654189?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/louismartin",
"html_url": "https://github.com/louismartin",
"followers_url": "https://api.github.com/... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/1512?src=pr&el=h1) Report\n> Merging [#1512](https://codecov.io/gh/huggingface/transformers/pull/1512?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/a701c9b32126f1e6974d9fcb3a5c3700527d8559?src=pr&el=desc) will **n... | 1,571 | 1,571 | 1,571 | CONTRIBUTOR | null | Fix ImportError in `convert_roberta_original_pytorch_checkpoint_to_pytorch.py`, see #1459.
| {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1512/reactions",
"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/1512/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/1512",
"html_url": "https://github.com/huggingface/transformers/pull/1512",
"diff_url": "https://github.com/huggingface/transformers/pull/1512.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/1512.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/1511 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/1511/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/1511/comments | https://api.github.com/repos/huggingface/transformers/issues/1511/events | https://github.com/huggingface/transformers/pull/1511 | 506,470,872 | MDExOlB1bGxSZXF1ZXN0MzI3NjYwNzUy | 1,511 | Run squad with all model lq | {
"login": "qianliu0708",
"id": 20182089,
"node_id": "MDQ6VXNlcjIwMTgyMDg5",
"avatar_url": "https://avatars.githubusercontent.com/u/20182089?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/qianliu0708",
"html_url": "https://github.com/qianliu0708",
"followers_url": "https://api.github.com/... | [] | closed | false | null | [] | [
"This does not seem to be related to our repo. Closing. Please don't reopen unless you want to submit a real PR."
] | 1,571 | 1,571 | 1,571 | NONE | null | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/1511/reactions",
"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/1511/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/1511",
"html_url": "https://github.com/huggingface/transformers/pull/1511",
"diff_url": "https://github.com/huggingface/transformers/pull/1511.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/1511.patch",
"merged_at": n... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.