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 β | author_association stringclasses 4
values | active_lock_reason stringclasses 2
values | body stringlengths 0 234k β | 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/2612 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2612/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2612/comments | https://api.github.com/repos/huggingface/transformers/issues/2612/events | https://github.com/huggingface/transformers/issues/2612 | 553,675,038 | MDU6SXNzdWU1NTM2NzUwMzg= | 2,612 | Error in fine tuning Roberta for QA | {
"login": "houdaM97",
"id": 43147098,
"node_id": "MDQ6VXNlcjQzMTQ3MDk4",
"avatar_url": "https://avatars.githubusercontent.com/u/43147098?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/houdaM97",
"html_url": "https://github.com/houdaM97",
"followers_url": "https://api.github.com/users/hou... | [
{
"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 | [] | [
"Please [format your code correctly](https://help.github.com/en/github/writing-on-github/creating-and-highlighting-code-blocks). Now it is very unreadable.",
"Thank you for mentioning this, it's done!",
"Something seems to have gone wrong. Can you check? There is a line \"and here is my training script:\" that ... | 1,579 | 1,585 | 1,585 | NONE | null | ## β Questions & Help
<!-- A clear and concise description of the question. -->
Hi,
I tried to fine tune RobertaModel for question answering task, i implemented TFRobertaForQuestionAnswering but when i run the training script i got this error:
tensorflow.python.framework.errors_impl.InvalidArgumentError: indic... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2612/reactions",
"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/2612/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2611 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2611/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2611/comments | https://api.github.com/repos/huggingface/transformers/issues/2611/events | https://github.com/huggingface/transformers/issues/2611 | 553,620,840 | MDU6SXNzdWU1NTM2MjA4NDA= | 2,611 | Finetuning my language model | {
"login": "paulthemagno",
"id": 38130299,
"node_id": "MDQ6VXNlcjM4MTMwMjk5",
"avatar_url": "https://avatars.githubusercontent.com/u/38130299?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/paulthemagno",
"html_url": "https://github.com/paulthemagno",
"followers_url": "https://api.github.c... | [] | closed | false | null | [] | [
"I am having the same problem when finetuning my own language model with run_lm_finetuning.py on camembert Model. I guess this might be related to the fact that it is reading a big file at once (103M, ~500k lines).\r\n Since the code reads whole data at once, it requires so much memory to handle huge corpus.\r\n\r\... | 1,579 | 1,580 | 1,580 | NONE | null | ## β Questions & Help
I have a problem about a finetuning of my own language model ([model](https://mxmdownloads.s3.amazonaws.com/umberto/umberto-commoncrawl-cased-v1.tar.gz) and [sentencepiece](https://mxmdownloads.s3.amazonaws.com/umberto/umberto-commoncrawl-cased-v1-sentencepiece.bpe.model)). I'm trying to use [r... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2611/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/2611/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2610 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2610/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2610/comments | https://api.github.com/repos/huggingface/transformers/issues/2610/events | https://github.com/huggingface/transformers/issues/2610 | 553,573,355 | MDU6SXNzdWU1NTM1NzMzNTU= | 2,610 | run_ner.py huge discrepancy between eval and predict (or "dev" and "test" evaluation modes) | {
"login": "MatejUlcar",
"id": 26550612,
"node_id": "MDQ6VXNlcjI2NTUwNjEy",
"avatar_url": "https://avatars.githubusercontent.com/u/26550612?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/MatejUlcar",
"html_url": "https://github.com/MatejUlcar",
"followers_url": "https://api.github.com/use... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.\n"
] | 1,579 | 1,585 | 1,585 | NONE | null | ## β Questions & Help
I'm comparing two different pretrained Bert models on the NER task. One is a bert-base-multilingual-cased model, which works fine, consistently, as one would expect. The other is our own pretrained multilingual Bert, which is trained on fewer languages and has so far shown better results on tho... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2610/reactions",
"total_count": 2,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 2
} | https://api.github.com/repos/huggingface/transformers/issues/2610/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2609 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2609/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2609/comments | https://api.github.com/repos/huggingface/transformers/issues/2609/events | https://github.com/huggingface/transformers/issues/2609 | 553,449,716 | MDU6SXNzdWU1NTM0NDk3MTY= | 2,609 | Bad Results with Albert | {
"login": "chikubee",
"id": 25073753,
"node_id": "MDQ6VXNlcjI1MDczNzUz",
"avatar_url": "https://avatars.githubusercontent.com/u/25073753?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/chikubee",
"html_url": "https://github.com/chikubee",
"followers_url": "https://api.github.com/users/chi... | [
{
"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. Many people are reporting unstable results or just unexpected results. You can search for issues in this library, and even in other ones (e.g. https://github.com/deepset-ai/FARM/issues/202#issuecomment-577077201). It seems that ALBERT is very sensitive to hyperparameters and even then... For now there seems to... | 1,579 | 1,585 | 1,585 | NONE | null | ## β Questions & Help
Trying to understand why is the cosine similarity between tokens with Albert way bad in comparison to DistilBert.
Any inferences on the same would be helpful.
Thanks in advance.
Embeddings constructed for a token by summing the last 4 encoded layers.
Distance metric: cosine
Results wit... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2609/reactions",
"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/2609/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2608 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2608/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2608/comments | https://api.github.com/repos/huggingface/transformers/issues/2608/events | https://github.com/huggingface/transformers/issues/2608 | 553,445,324 | MDU6SXNzdWU1NTM0NDUzMjQ= | 2,608 | Bug in the command line tool: os.DirEntry not supported in Python 3.5 | {
"login": "netw0rkf10w",
"id": 8569515,
"node_id": "MDQ6VXNlcjg1Njk1MTU=",
"avatar_url": "https://avatars.githubusercontent.com/u/8569515?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/netw0rkf10w",
"html_url": "https://github.com/netw0rkf10w",
"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
},
{
"id": 1802861720,
"node_id": "MDU6TGFiZWwxODAyODYxNzIw",
... | closed | false | null | [] | [
"Though the core of the library is Python 3.5+, the CLI is actually Python3.6+ as for instance the serving subcommand uses FastAPI which is Py36+ only (cc @mfuntowicz)\r\n\r\nDo we have a way to make that clear in the doc @LysandreJik?",
"Personal opinion: supporting 3.6+ only seems realistic and may make mainten... | 1,579 | 1,585 | 1,585 | NONE | null | Hi,
[This line](https://github.com/huggingface/transformers/blob/1a8e87be4e2a1b551175bd6f0f749f3d2289010f/src/transformers/commands/user.py#L162) will cause a syntax error in Python 3.5 as `os.DirEntry` does not exist.
You should either update the code for backward compatibility or update the README replacing 3.5... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2608/reactions",
"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/2608/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2607 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2607/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2607/comments | https://api.github.com/repos/huggingface/transformers/issues/2607/events | https://github.com/huggingface/transformers/pull/2607 | 553,339,959 | MDExOlB1bGxSZXF1ZXN0MzY1Njg5NjA5 | 2,607 | Fix inconsistency between T5WithLMHeadModel's doc and it's behavior | {
"login": "nalourie-ai2",
"id": 23320271,
"node_id": "MDQ6VXNlcjIzMzIwMjcx",
"avatar_url": "https://avatars.githubusercontent.com/u/23320271?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/nalourie-ai2",
"html_url": "https://github.com/nalourie-ai2",
"followers_url": "https://api.github.c... | [] | 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... | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/2607?src=pr&el=h1) Report\n> Merging [#2607](https://codecov.io/gh/huggingface/transformers/pull/2607?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/1a8e87be4e2a1b551175bd6f0f749f3d2289010f?src=pr&el=desc) will **n... | 1,579 | 1,579 | 1,579 | CONTRIBUTOR | null | The doc string for `T5WithLMHeadModel` is currently inconsistent with it's behavior.
The doc string says that the forward method ignores indices of -1 when computing the loss; however, the method instead ignores indices of -100. This pull request changes the method to ignore indices of -1, making the two consistent.... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2607/reactions",
"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/2607/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/2607",
"html_url": "https://github.com/huggingface/transformers/pull/2607",
"diff_url": "https://github.com/huggingface/transformers/pull/2607.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/2607.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/2606 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2606/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2606/comments | https://api.github.com/repos/huggingface/transformers/issues/2606/events | https://github.com/huggingface/transformers/issues/2606 | 553,168,121 | MDU6SXNzdWU1NTMxNjgxMjE= | 2,606 | Upload CLI: on Windows, uniformize paths/urls separators | {
"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
},
{
"id": 1802861720,
"node_id": "MDU6TGFiZWwxODAyODYxNzIw",
... | closed | false | {
"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-... | [
{
"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.gith... | [
"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",
"This was completed but forgot to close the issue."
] | 1,579 | 1,584 | 1,584 | MEMBER | null | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2606/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/2606/timeline | completed | null | null | |
https://api.github.com/repos/huggingface/transformers/issues/2605 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2605/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2605/comments | https://api.github.com/repos/huggingface/transformers/issues/2605/events | https://github.com/huggingface/transformers/issues/2605 | 553,154,537 | MDU6SXNzdWU1NTMxNTQ1Mzc= | 2,605 | glue.py when using mrpc and similar data does not work | {
"login": "nargesam",
"id": 24642904,
"node_id": "MDQ6VXNlcjI0NjQyOTA0",
"avatar_url": "https://avatars.githubusercontent.com/u/24642904?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/nargesam",
"html_url": "https://github.com/nargesam",
"followers_url": "https://api.github.com/users/nar... | [
{
"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! Indeed, you can't access the dictionary by using `example.label`. Are you passing your `tf.data.Dataset` as the `examples` argument to the `glue_convert_examples_to_features`?",
"yes, train_data is the examples. and the spec is: \r\n\r\n{'idx': TensorSpec(shape=(), dtype=tf.string, name=None),\r\n'sentenc... | 1,579 | 1,585 | 1,585 | NONE | null | ## π 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:
* [ ] the official example scripts: (give details):
I am using my dataset with format [idx, sentence1, sentence2, label] in form ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2605/reactions",
"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/2605/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2604 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2604/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2604/comments | https://api.github.com/repos/huggingface/transformers/issues/2604/events | https://github.com/huggingface/transformers/issues/2604 | 553,073,485 | MDU6SXNzdWU1NTMwNzM0ODU= | 2,604 | Can not upload BertTokenizer.from_pretrained() from an AWS S3 bucket | {
"login": "bnicholl",
"id": 26211830,
"node_id": "MDQ6VXNlcjI2MjExODMw",
"avatar_url": "https://avatars.githubusercontent.com/u/26211830?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/bnicholl",
"html_url": "https://github.com/bnicholl",
"followers_url": "https://api.github.com/users/bni... | [
{
"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 Ben,\r\n\r\nUnless I misunderstand what you're trying to do, this is not really what `save_pretrained()` and `from_pretrained()` are made for.\r\n\r\n`save_pretrained()` lets you save a tokenizer or a model _locally_, inside a local folder. (you can then upload those files to your own s3 bucket, or use the `tra... | 1,579 | 1,585 | 1,585 | NONE | null | ## π Migration
the model I am using is BertForSequenceClassification
The problem arises when I serialize my Bert model, and then upload to an AWS S3 bucket. Once my model is inside of S3, I can not import the model via BertTokenizer.from_pretrained()
For example, in order to save my model to S3, my code reads... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2604/reactions",
"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/2604/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2603 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2603/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2603/comments | https://api.github.com/repos/huggingface/transformers/issues/2603/events | https://github.com/huggingface/transformers/issues/2603 | 552,989,439 | MDU6SXNzdWU1NTI5ODk0Mzk= | 2,603 | XLNet: Incorrect segment id for CLS token | {
"login": "ChristophAlt",
"id": 6420705,
"node_id": "MDQ6VXNlcjY0MjA3MDU=",
"avatar_url": "https://avatars.githubusercontent.com/u/6420705?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ChristophAlt",
"html_url": "https://github.com/ChristophAlt",
"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 | [] | [
"Indeed, this is an error ! Thanks for letting us know, it was patched with 088fa7b!",
"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"
] | 1,579 | 1,585 | 1,585 | NONE | null | Thanks for the great work!
It seems that the XLNetTokenizer assigns an incorrect segment id to the CLS token when a single sequence of token ids is provided. If token_ids_1 is None, all segment ids are '0', including the segment id of the CLS token. In my understanding, the segment ids should always differ.
```py... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2603/reactions",
"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/2603/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2602 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2602/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2602/comments | https://api.github.com/repos/huggingface/transformers/issues/2602/events | https://github.com/huggingface/transformers/pull/2602 | 552,957,982 | MDExOlB1bGxSZXF1ZXN0MzY1MzcxNTg5 | 2,602 | Edit a way to get `projected_context_layer` | {
"login": "jinkilee",
"id": 6321520,
"node_id": "MDQ6VXNlcjYzMjE1MjA=",
"avatar_url": "https://avatars.githubusercontent.com/u/6321520?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jinkilee",
"html_url": "https://github.com/jinkilee",
"followers_url": "https://api.github.com/users/jinki... | [] | closed | false | null | [] | [
"Hi, please check the [contribution guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md#start-contributing-pull-requests) for the code quality tests to pass.\r\n\r\nWhy did you edit this, does this solve a bug or add new functionality?",
"Hi i edited this because i read a comment l... | 1,579 | 1,580 | 1,580 | NONE | null | I edited a way to get `projected_context_layer`.
Instead of doing
'''
w = (
self.dense.weight.t()
.view(self.num_attention_heads, self.attention_head_size, self.hidden_size)
.to(context_layer.dtype)
)
b = self.dense.bias.to(context_layer.dtype)
projected_context_layer = torch.einsum("bfnd,ndh->bf... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2602/reactions",
"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/2602/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/2602",
"html_url": "https://github.com/huggingface/transformers/pull/2602",
"diff_url": "https://github.com/huggingface/transformers/pull/2602.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/2602.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/2601 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2601/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2601/comments | https://api.github.com/repos/huggingface/transformers/issues/2601/events | https://github.com/huggingface/transformers/issues/2601 | 552,946,884 | MDU6SXNzdWU1NTI5NDY4ODQ= | 2,601 | unexpected keyword argument 'encoder_hidden_states' when using PreTrainedEncoderDecoder | {
"login": "mmsamiei",
"id": 12582703,
"node_id": "MDQ6VXNlcjEyNTgyNzAz",
"avatar_url": "https://avatars.githubusercontent.com/u/12582703?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/mmsamiei",
"html_url": "https://github.com/mmsamiei",
"followers_url": "https://api.github.com/users/mms... | [
{
"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 am facing the same issue. Probably `PreTrainedEncoderDecoder` supports only `Bert` to `Bert` models.",
"I'm getting this error too\... | 1,579 | 1,674 | 1,585 | NONE | null | ## β Questions & Help
After I have defined my seq2seq class using Encoder Decoder Architecture in the following way:
```
from transformers import PreTrainedEncoderDecoder
model = PreTrainedEncoderDecoder.from_pretrained('bert-base-uncased','gpt2')
```
I try to forward tensor through the model in this following ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2601/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/2601/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2600 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2600/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2600/comments | https://api.github.com/repos/huggingface/transformers/issues/2600/events | https://github.com/huggingface/transformers/issues/2600 | 552,777,072 | MDU6SXNzdWU1NTI3NzcwNzI= | 2,600 | Trouble fine tuning multiple choice | {
"login": "KerenzaDoxolodeo",
"id": 7535438,
"node_id": "MDQ6VXNlcjc1MzU0Mzg=",
"avatar_url": "https://avatars.githubusercontent.com/u/7535438?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/KerenzaDoxolodeo",
"html_url": "https://github.com/KerenzaDoxolodeo",
"followers_url": "https://ap... | [
{
"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! I believe this issue could stem from your label being negative as well. Could you check that it doesn't fail when computing the loss with a negative label?",
"I'm not sure what is a negative label in the context of multiple-choice, but here's what I did: it successfully computed loss when the label is 0 (wh... | 1,579 | 1,591 | 1,591 | NONE | null | ## β Questions & Help
Hi! I have issues with fine-tuning the multi-choice BERT and I am stuck on an error and I can use some help. When I tried to fine-tune it with my own dataset, it threw the Error ```RuntimeError: Assertion `cur_target >= 0 && cur_target < n_classes' failed. at /opt/conda/conda-bld/pytorch_1570... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2600/reactions",
"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/2600/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2599 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2599/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2599/comments | https://api.github.com/repos/huggingface/transformers/issues/2599/events | https://github.com/huggingface/transformers/issues/2599 | 552,685,004 | MDU6SXNzdWU1NTI2ODUwMDQ= | 2,599 | Xlnet, Alberta, Roberta are not finetuned for CoLA task | {
"login": "SarikGhazarian",
"id": 31287614,
"node_id": "MDQ6VXNlcjMxMjg3NjE0",
"avatar_url": "https://avatars.githubusercontent.com/u/31287614?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/SarikGhazarian",
"html_url": "https://github.com/SarikGhazarian",
"followers_url": "https://api.gi... | [] | closed | false | null | [] | [
"NM, I was able to solve it by changing some of the hyperparameter values.",
"Hi, glad you could make it work! Do you mind sharing what hyperparameter values you tuned in order to make it work?",
"> Hi, glad you could make it work! Do you mind sharing what hyperparameter values you tuned in order to make it wor... | 1,579 | 1,589 | 1,580 | NONE | null | ## π Bug
I am currently trying to finetune pretrained models on CoLA task by using run_glue.py. Some of the models such as Bert and DistilBert are finetuned correctly as it is expected (the training loss goes down and the evaluation result is as what has been reported). Even though, for other models such as Roberta... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2599/reactions",
"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/2599/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2598 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2598/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2598/comments | https://api.github.com/repos/huggingface/transformers/issues/2598/events | https://github.com/huggingface/transformers/issues/2598 | 552,585,910 | MDU6SXNzdWU1NTI1ODU5MTA= | 2,598 | load tf2 roberta model meet error | {
"login": "bestpredicts",
"id": 12403152,
"node_id": "MDQ6VXNlcjEyNDAzMTUy",
"avatar_url": "https://avatars.githubusercontent.com/u/12403152?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/bestpredicts",
"html_url": "https://github.com/bestpredicts",
"followers_url": "https://api.github.c... | [] | closed | false | null | [] | [
"I am facing the same issue, how did you resolve this? @bestpredicts ",
"Same issue with portuguese bert version\r\n",
"I had the same issue and found that this problem occurs because the default \"RobertaConfig\" is based on \"bert-base-uncased\" config, which is different from \"roberta-base\" config. The rig... | 1,579 | 1,642 | 1,579 | NONE | null | ## β Questions & Help
`
config = RobertaConfig() # print(config) to see settings
config.output_hidden_states = False # Set to True to obtain hidden states
model = TFRobertaModel.from_pretrained('/home/wk/Bert_Pretrained/robert_base/roberta-base-tf_model.h5', config=config) `
errors
`ValueError ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2598/reactions",
"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/2598/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2597 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2597/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2597/comments | https://api.github.com/repos/huggingface/transformers/issues/2597/events | https://github.com/huggingface/transformers/issues/2597 | 552,585,576 | MDU6SXNzdWU1NTI1ODU1NzY= | 2,597 | Transfer Learning on Text Summarization Model | {
"login": "imayachita",
"id": 3615586,
"node_id": "MDQ6VXNlcjM2MTU1ODY=",
"avatar_url": "https://avatars.githubusercontent.com/u/3615586?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/imayachita",
"html_url": "https://github.com/imayachita",
"followers_url": "https://api.github.com/users... | [
{
"id": 1845609017,
"node_id": "MDU6TGFiZWwxODQ1NjA5MDE3",
"url": "https://api.github.com/repos/huggingface/transformers/labels/seq2seq",
"name": "seq2seq",
"color": "fef2c0",
"default": false,
"description": ""
}
] | closed | false | {
"login": "sshleifer",
"id": 6045025,
"node_id": "MDQ6VXNlcjYwNDUwMjU=",
"avatar_url": "https://avatars.githubusercontent.com/u/6045025?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sshleifer",
"html_url": "https://github.com/sshleifer",
"followers_url": "https://api.github.com/users/ss... | [
{
"login": "sshleifer",
"id": 6045025,
"node_id": "MDQ6VXNlcjYwNDUwMjU=",
"avatar_url": "https://avatars.githubusercontent.com/u/6045025?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sshleifer",
"html_url": "https://github.com/sshleifer",
"followers_url": "https://api... | [
"@imayachita If your succeeded to use transfer learning on your own data, please update here.",
"`examples/summarization/bart/run_bart_sum.py` now exists :)"
] | 1,579 | 1,585 | 1,585 | NONE | null | Hi all,
Is there any way to do transfer learning on the Text Summarization model (bertabs-finetuned-cnndm)? I would like to continue training it on my dataset.
The code run_summarization.py only does prediction. Thanks! | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2597/reactions",
"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/2597/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2596 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2596/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2596/comments | https://api.github.com/repos/huggingface/transformers/issues/2596/events | https://github.com/huggingface/transformers/issues/2596 | 552,511,882 | MDU6SXNzdWU1NTI1MTE4ODI= | 2,596 | changing the attention head size in MultiBert | {
"login": "mhajiaghayi",
"id": 28695943,
"node_id": "MDQ6VXNlcjI4Njk1OTQz",
"avatar_url": "https://avatars.githubusercontent.com/u/28695943?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/mhajiaghayi",
"html_url": "https://github.com/mhajiaghayi",
"followers_url": "https://api.github.com/... | [] | closed | false | null | [] | [
"I got it working"
] | 1,579 | 1,584 | 1,584 | NONE | null | ## π Bug
<!-- Important information -->
I'm trying to use a MultiBERT model with not all 12 attention heads but just 8 attention heads. so, in config file, I changed the following keys
config.num_attention_heads = 8
config.hidden_size = 512
config.pooler_fc_size = 512
I assumed similar to layer s... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2596/reactions",
"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/2596/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2595 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2595/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2595/comments | https://api.github.com/repos/huggingface/transformers/issues/2595/events | https://github.com/huggingface/transformers/issues/2595 | 552,449,169 | MDU6SXNzdWU1NTI0NDkxNjk= | 2,595 | RAM leakage when trying to retrieve the hidden states from the GPT-2 model. | {
"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... | [] | closed | false | null | [] | [
"To add more, I fixed my code like below:\r\n\r\n```python\r\n# for loop to calculate TVD\r\ndef TVD_loop(test_i, test_dummy_i, nlayer, best_model):\r\n \r\n TVD_tensor = torch.zeros(test_i.size()[1], (nlayer+1), test_i.size()[0]).float()\r\n \r\n # replace every 0's in TVD_tensor to -2\r\n TVD_tenso... | 1,579 | 1,579 | 1,579 | NONE | null | Hello,
I am trying to retrieve hidden state vectors from my trained GPT-2 model from a loop, and there is a huge RAM leakage associated with the operation. Below are my code:
```python
# for loop to calculate TVD
def TVD_loop(test_i, test_dummy_i, nlayer, best_model):
TVD_tensor = torch.zeros(test_i.... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2595/reactions",
"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/2595/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2594 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2594/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2594/comments | https://api.github.com/repos/huggingface/transformers/issues/2594/events | https://github.com/huggingface/transformers/pull/2594 | 552,386,364 | MDExOlB1bGxSZXF1ZXN0MzY0OTA0NzQz | 2,594 | edited a way to get at AlbertAttention.forward | {
"login": "jinkilee",
"id": 6321520,
"node_id": "MDQ6VXNlcjYzMjE1MjA=",
"avatar_url": "https://avatars.githubusercontent.com/u/6321520?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jinkilee",
"html_url": "https://github.com/jinkilee",
"followers_url": "https://api.github.com/users/jinki... | [] | closed | false | null | [] | [
"will commit with another pull request"
] | 1,579 | 1,579 | 1,579 | NONE | null | '''
# Should find a better way to do this
w = (
self.dense.weight.t()
.view(self.num_attention_heads, self.attention_head_size, self.hidden_size)
.to(context_layer.dtype)
)
b = self.dense.bias.to(context_layer.dtype)
'''
I thought the above code is not necessary.
it can be simply fixed by "mergi... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2594/reactions",
"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/2594/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/2594",
"html_url": "https://github.com/huggingface/transformers/pull/2594",
"diff_url": "https://github.com/huggingface/transformers/pull/2594.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/2594.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/2593 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2593/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2593/comments | https://api.github.com/repos/huggingface/transformers/issues/2593/events | https://github.com/huggingface/transformers/pull/2593 | 552,379,217 | MDExOlB1bGxSZXF1ZXN0MzY0ODk4OTM1 | 2,593 | Added custom model dir to PPLM train | {
"login": "ashirviskas",
"id": 11985242,
"node_id": "MDQ6VXNlcjExOTg1MjQy",
"avatar_url": "https://avatars.githubusercontent.com/u/11985242?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ashirviskas",
"html_url": "https://github.com/ashirviskas",
"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"
] | 1,579 | 1,585 | 1,585 | NONE | null | Just an option to save the model to other than the working directory.
Default functionality hasn't changed. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2593/reactions",
"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/2593/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/2593",
"html_url": "https://github.com/huggingface/transformers/pull/2593",
"diff_url": "https://github.com/huggingface/transformers/pull/2593.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/2593.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/2592 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2592/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2592/comments | https://api.github.com/repos/huggingface/transformers/issues/2592/events | https://github.com/huggingface/transformers/issues/2592 | 552,243,924 | MDU6SXNzdWU1NTIyNDM5MjQ= | 2,592 | RuntimeError: The expanded size of the tensor (449) must match the existing size (2) at non-singleton dimension 2. Target sizes: [4, 2, 449]. Tensor sizes: [1, 2] while using ALBERT | {
"login": "hasnain2808",
"id": 28212972,
"node_id": "MDQ6VXNlcjI4MjEyOTcy",
"avatar_url": "https://avatars.githubusercontent.com/u/28212972?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/hasnain2808",
"html_url": "https://github.com/hasnain2808",
"followers_url": "https://api.github.com/... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | {
"login": "patrickvonplaten",
"id": 23423619,
"node_id": "MDQ6VXNlcjIzNDIzNjE5",
"avatar_url": "https://avatars.githubusercontent.com/u/23423619?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/patrickvonplaten",
"html_url": "https://github.com/patrickvonplaten",
"followers_url": "https://... | [
{
"login": "patrickvonplaten",
"id": 23423619,
"node_id": "MDQ6VXNlcjIzNDIzNjE5",
"avatar_url": "https://avatars.githubusercontent.com/u/23423619?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/patrickvonplaten",
"html_url": "https://github.com/patrickvonplaten",
"follo... | [
"What is the code that you are executing that leads to this error? ",
"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",
"How did you fix the problem?"
] | 1,579 | 1,620 | 1,587 | NONE | null | ## β Questions & Help
<!-- A clear and concise description of the question. -->
I wanted to use ALBERT with a double head
as we have one for openaigpt with the name OpenAIGPTDoubleHeadsModel
I am doing it with taking inspiration from OpenAIGPTDoubleHeadsModel
but I am getting this error
` File "train.py", l... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2592/reactions",
"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/2592/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2591 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2591/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2591/comments | https://api.github.com/repos/huggingface/transformers/issues/2591/events | https://github.com/huggingface/transformers/issues/2591 | 552,161,685 | MDU6SXNzdWU1NTIxNjE2ODU= | 2,591 | What is the f1 score of Squad v2.0 on bert-base? I only got f1 score 74.78. | {
"login": "YJYJLee",
"id": 28900943,
"node_id": "MDQ6VXNlcjI4OTAwOTQz",
"avatar_url": "https://avatars.githubusercontent.com/u/28900943?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/YJYJLee",
"html_url": "https://github.com/YJYJLee",
"followers_url": "https://api.github.com/users/YJYJLe... | [] | closed | false | null | [] | [
"Your result is fine. This [poster](https://web.stanford.edu/class/cs224n/posters/15848021.pdf) says that they achieved **76.70**. Maybe you can get there as well when you train for 2 more epochs.",
"Thank you for your reply! :)",
"Please close the question if the answer suits your needs.",
"Sorry, I closed i... | 1,579 | 1,579 | 1,579 | NONE | null | ## β Questions & Help
<!-- A clear and concise description of the question. -->
Hello, I am doing some experiment of squad v2.0 on bert-base (NOT bert-large).
According to the BERT paper, bert-large achieves f1 score 81.9 with squad v2.0.
Since I couldn't find the official result for bert-base, I am not sure if I... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2591/reactions",
"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/2591/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2590 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2590/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2590/comments | https://api.github.com/repos/huggingface/transformers/issues/2590/events | https://github.com/huggingface/transformers/issues/2590 | 552,092,601 | MDU6SXNzdWU1NTIwOTI2MDE= | 2,590 | run_glue.py, CoLA : MCC goes to 0, in some hyperparameter cases | {
"login": "drcdr",
"id": 18181131,
"node_id": "MDQ6VXNlcjE4MTgxMTMx",
"avatar_url": "https://avatars.githubusercontent.com/u/18181131?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/drcdr",
"html_url": "https://github.com/drcdr",
"followers_url": "https://api.github.com/users/drcdr/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 | [] | [
"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"
] | 1,579 | 1,585 | 1,585 | NONE | null | ## π Bug
<!-- Important information -->
Model I am using: roberta-large
Language I am using the model on: English
The problem arise when using:
* [x] the official example scripts: run_glue.py
The tasks I am working on is:
* [x] an official GLUE/SQUaD task: CoLA
## To Reproduce
Steps to reproduce the... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2590/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/2590/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2589 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2589/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2589/comments | https://api.github.com/repos/huggingface/transformers/issues/2589/events | https://github.com/huggingface/transformers/issues/2589 | 551,969,213 | MDU6SXNzdWU1NTE5NjkyMTM= | 2,589 | run_lm_finetuning.py regenerates examples cache when restored from a checkpoint, is this intended? | {
"login": "mike-athene",
"id": 2640536,
"node_id": "MDQ6VXNlcjI2NDA1MzY=",
"avatar_url": "https://avatars.githubusercontent.com/u/2640536?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/mike-athene",
"html_url": "https://github.com/mike-athene",
"followers_url": "https://api.github.com/us... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"I also use --save_total_limit=10 which triggered another exception in checkpoint deletion code as it tried to delete the cache file like it was a folder\r\n\r\n```\r\nDeleting older checkpoint [./output\\checkpoint-200_cached_lm_512_dataset.txt] due to args.save_total_limit\r\nTraceback (most recent call last):\r... | 1,579 | 1,585 | 1,585 | NONE | null | ## β Questions & Help
Hello,
I am finetuning a gpt2-medium model on a large (800mb+) input via run_lm_finetuning.py on a windown/conda env with recent git checkout of transformers and apex installed (--per_gpu_train_batch_size=1 --fp16 --fp16_opt_level O2 --gradient_accumulation_steps=10 --block_size=512).
On in... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2589/reactions",
"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/2589/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2588 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2588/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2588/comments | https://api.github.com/repos/huggingface/transformers/issues/2588/events | https://github.com/huggingface/transformers/issues/2588 | 551,929,414 | MDU6SXNzdWU1NTE5Mjk0MTQ= | 2,588 | how can i download the model manually? | {
"login": "shange1996",
"id": 49185852,
"node_id": "MDQ6VXNlcjQ5MTg1ODUy",
"avatar_url": "https://avatars.githubusercontent.com/u/49185852?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/shange1996",
"html_url": "https://github.com/shange1996",
"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 | [] | [
"- xlnet-base-cased : https://s3.amazonaws.com/models.huggingface.co/bert/xlnet-base-cased-config.json\r\n- xlnet-large-cased : https://s3.amazonaws.com/models.huggingface.co/bert/xlnet-large-cased-config.json\r\n\r\nI got it from [here](https://huggingface.co/transformers/_modules/transformers/configuration_xlnet.... | 1,579 | 1,592 | 1,585 | NONE | null | ## β Questions & Help
<!-- A clear and concise description of the question. -->
I want to download the model manually because of my network. But now I can only find the download address of bert. Where is the address of all models? Such as XLNETγ | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2588/reactions",
"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/2588/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2587 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2587/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2587/comments | https://api.github.com/repos/huggingface/transformers/issues/2587/events | https://github.com/huggingface/transformers/issues/2587 | 551,915,182 | MDU6SXNzdWU1NTE5MTUxODI= | 2,587 | The accuracy of XLNet | {
"login": "chlorane",
"id": 39242468,
"node_id": "MDQ6VXNlcjM5MjQyNDY4",
"avatar_url": "https://avatars.githubusercontent.com/u/39242468?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/chlorane",
"html_url": "https://github.com/chlorane",
"followers_url": "https://api.github.com/users/chl... | [
{
"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 sure that you are using the right models and not just `BertModel`? You also have to change the tokenizer completely.\r\n\r\nInstead of something like\r\n\r\n```python\r\nmodel = BertModel.from_pretrained('bert-base-uncased')\r\ntokenizer = BertTokenizer.from_pretrained('bert-base-uncased')\r\n```\r\n\r\nYo... | 1,579 | 1,589 | 1,585 | NONE | null | ## π Migration
<!-- Important information -->
Model I am using (Bert, XLNet....):
XLNet
Language I am using the model on (English, Chinese....):
English
The problem arise when using:
[ ] the official example scripts: (give details)
[ * ] my own modified scripts: (give details)
I use my own scripts under... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2587/reactions",
"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/2587/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2586 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2586/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2586/comments | https://api.github.com/repos/huggingface/transformers/issues/2586/events | https://github.com/huggingface/transformers/issues/2586 | 551,896,193 | MDU6SXNzdWU1NTE4OTYxOTM= | 2,586 | PyTorch 1.2 has released API 'torch.nn.Transformer'οΌso it's better to modify the source code with the official python API | {
"login": "daydayfun",
"id": 39835967,
"node_id": "MDQ6VXNlcjM5ODM1OTY3",
"avatar_url": "https://avatars.githubusercontent.com/u/39835967?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/daydayfun",
"html_url": "https://github.com/daydayfun",
"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 has been suggested a while back when this was first introduced (we're at 1.4 now). This is possibly impractical to do since it is likely that many people are still on 1.0<=x<1.2. ",
"This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further act... | 1,579 | 1,585 | 1,585 | NONE | null | ## π Feature
<!-- A clear and concise description of the feature proposal. Please provide a link to the paper and code in case they exist. -->
## Motivation
It's better to Modify modeling_bert.py with official API 'torch.nn.Transformer' of PyTorch 1.2
## Additional context
https://pytorch.org/docs/stabl... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2586/reactions",
"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/2586/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2585 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2585/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2585/comments | https://api.github.com/repos/huggingface/transformers/issues/2585/events | https://github.com/huggingface/transformers/issues/2585 | 551,869,598 | MDU6SXNzdWU1NTE4Njk1OTg= | 2,585 | Attibute ErrorοΌβNoneTypeβ object has no attribute 'seek' and OSError | {
"login": "MrChenJianhui",
"id": 45336867,
"node_id": "MDQ6VXNlcjQ1MzM2ODY3",
"avatar_url": "https://avatars.githubusercontent.com/u/45336867?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/MrChenJianhui",
"html_url": "https://github.com/MrChenJianhui",
"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 | [] | [
"python version is 3.7.3",
"Hi, you would need to provide more information than that for us to help you. What code made you run into this error? ",
"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 contr... | 1,579 | 1,585 | 1,585 | NONE | null | My pytorch version is 1.4.0+cpu and tensorflow version is 2.0.0-dev20191002.
/torch/serialization.py,line 289,in_check_seekable
βNoneTypeβ object has no attribute 'seek'
You can only torch.load from a file that is seekable.Please pre_load the data into a buffer like io.BytesIO and try to load from it instead.
But h... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2585/reactions",
"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/2585/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2584 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2584/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2584/comments | https://api.github.com/repos/huggingface/transformers/issues/2584/events | https://github.com/huggingface/transformers/issues/2584 | 551,860,131 | MDU6SXNzdWU1NTE4NjAxMzE= | 2,584 | what's the structure of the model saved after fine-tuning ? | {
"login": "JiangYanting",
"id": 44471391,
"node_id": "MDQ6VXNlcjQ0NDcxMzkx",
"avatar_url": "https://avatars.githubusercontent.com/u/44471391?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/JiangYanting",
"html_url": "https://github.com/JiangYanting",
"followers_url": "https://api.github.c... | [] | closed | false | null | [] | [
"I'm afraid I don't understand your question. The pretrained model is an architecture whose weights have already been trained on some task (typically (M)LM and NSP/SOP). When you finetune the model, the architecture stays exactly the same but the weights are finetuned to best fit your task.",
"@BramVanroy I tried... | 1,579 | 1,579 | 1,579 | NONE | null | ## β Questions & Help
<!-- A clear and concise description of the question. -->
Hello ! I'm wondering what is the structure of the model saved after fine-tuning. For example, after the sequence classification fine-tuning , how to show the layers information of newly-formed model ? Is new model's sentence vector di... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2584/reactions",
"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/2584/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2583 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2583/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2583/comments | https://api.github.com/repos/huggingface/transformers/issues/2583/events | https://github.com/huggingface/transformers/issues/2583 | 551,849,821 | MDU6SXNzdWU1NTE4NDk4MjE= | 2,583 | How to start a server and client to get feature vectors | {
"login": "duan348733684",
"id": 26431015,
"node_id": "MDQ6VXNlcjI2NDMxMDE1",
"avatar_url": "https://avatars.githubusercontent.com/u/26431015?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/duan348733684",
"html_url": "https://github.com/duan348733684",
"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"
] | 1,579 | 1,585 | 1,585 | NONE | null | ## β Questions & Help
<!-- A clear and concise description of the question. -->
How to start a server and client to get feature vectorsοΌor Which part of the code should I study in https://github.com/huggingface/transformers.git.
| {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2583/reactions",
"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/2583/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2582 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2582/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2582/comments | https://api.github.com/repos/huggingface/transformers/issues/2582/events | https://github.com/huggingface/transformers/issues/2582 | 551,832,180 | MDU6SXNzdWU1NTE4MzIxODA= | 2,582 | XLM-Roberta checkpoint redundant weight | {
"login": "NProkoptsev",
"id": 13507236,
"node_id": "MDQ6VXNlcjEzNTA3MjM2",
"avatar_url": "https://avatars.githubusercontent.com/u/13507236?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/NProkoptsev",
"html_url": "https://github.com/NProkoptsev",
"followers_url": "https://api.github.com/... | [] | closed | false | null | [] | [] | 1,579 | 1,579 | 1,579 | NONE | null | deleted | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2582/reactions",
"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/2582/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2581 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2581/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2581/comments | https://api.github.com/repos/huggingface/transformers/issues/2581/events | https://github.com/huggingface/transformers/issues/2581 | 551,823,128 | MDU6SXNzdWU1NTE4MjMxMjg= | 2,581 | Invalid argument: assertion failed: [Condition x == y did not hold element-wise:] [x (loss/output_1_loss/SparseSoftmaxCrossEntropyWithLogits/Shape_1:0) = ] [32 1] [y (loss/output_1_loss/SparseSoftmaxCrossEntropyWithLogits/strided_slice:0) = ] [32 128] | {
"login": "noambaron1989",
"id": 40001659,
"node_id": "MDQ6VXNlcjQwMDAxNjU5",
"avatar_url": "https://avatars.githubusercontent.com/u/40001659?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/noambaron1989",
"html_url": "https://github.com/noambaron1989",
"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 | [] | [
"Hi, GLUE is a sequence classification task, not a token classification task. The model you're using classifies tokens instead of entires sequences, and therefore has a different output than what is expected by the GLUE task.\r\n\r\nChange this line:\r\n\r\n```py\r\nmodel = TFBertForTokenClassification.from_pretrai... | 1,579 | 1,585 | 1,585 | NONE | null | I'm trying to run TFBertForTokenClassification with tensorflow_datasets.load('glue/sst2'):
```py
import tensorflow as tf
import tensorflow_datasets
from transformers import *
tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
model = TFBertForTokenClassification.from_pretrained('bert-base-uncased'... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2581/reactions",
"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/2581/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2580 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2580/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2580/comments | https://api.github.com/repos/huggingface/transformers/issues/2580/events | https://github.com/huggingface/transformers/issues/2580 | 551,802,588 | MDU6SXNzdWU1NTE4MDI1ODg= | 2,580 | glue_convert_examples_to_features in glue.py runs to errors | {
"login": "nargesam",
"id": 24642904,
"node_id": "MDQ6VXNlcjI0NjQyOTA0",
"avatar_url": "https://avatars.githubusercontent.com/u/24642904?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/nargesam",
"html_url": "https://github.com/nargesam",
"followers_url": "https://api.github.com/users/nar... | [] | closed | false | null | [] | [
"# Load dataset, tokenizer, model from pretrained model/vocabulary\r\ntokenizer = BertTokenizer.from_pretrained('bert-base-cased')\r\nmodel = TFBertForSequenceClassification.from_pretrained('bert-base-cased', force_download=True)\r\ndata = tensorflow_datasets.load('glue/mrpc')\r\n\r\nprint(\"checkpoint on data\")\r... | 1,579 | 1,579 | 1,579 | NONE | null | ## π Bug
<!-- Important information -->
Model I am using: Bert
Language I am using the model on (English)
The problem arise when using:
* [x] the official example scripts: (give details):
I have a venv running with TF2.0 and transformers, and I am running mrpc dataset with BERT. Here's the code:
trai... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2580/reactions",
"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/2580/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2579 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2579/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2579/comments | https://api.github.com/repos/huggingface/transformers/issues/2579/events | https://github.com/huggingface/transformers/pull/2579 | 551,768,277 | MDExOlB1bGxSZXF1ZXN0MzY0NDI1Nzg4 | 2,579 | Added saving to custom dir in PPLM train | {
"login": "ashirviskas",
"id": 11985242,
"node_id": "MDQ6VXNlcjExOTg1MjQy",
"avatar_url": "https://avatars.githubusercontent.com/u/11985242?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ashirviskas",
"html_url": "https://github.com/ashirviskas",
"followers_url": "https://api.github.com/... | [] | closed | false | null | [] | [
"The CI was a dependency glitch that was fixed on master since, you can rebase on master if you want it to go away.",
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/2579?src=pr&el=h1) Report\n> :exclamation: No coverage uploaded for pull request base (`master@cb13c8a`). [Click here to learn what... | 1,579 | 1,579 | 1,579 | NONE | null | Just an option to save the model to other than the working directory.
Default functionality hasn't changed. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2579/reactions",
"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/2579/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/2579",
"html_url": "https://github.com/huggingface/transformers/pull/2579",
"diff_url": "https://github.com/huggingface/transformers/pull/2579.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/2579.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/2578 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2578/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2578/comments | https://api.github.com/repos/huggingface/transformers/issues/2578/events | https://github.com/huggingface/transformers/issues/2578 | 551,763,691 | MDU6SXNzdWU1NTE3NjM2OTE= | 2,578 | GPT2TokenizerFast object has no attribute 'with_pre_tokenizer' | {
"login": "armheb",
"id": 18460769,
"node_id": "MDQ6VXNlcjE4NDYwNzY5",
"avatar_url": "https://avatars.githubusercontent.com/u/18460769?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/armheb",
"html_url": "https://github.com/armheb",
"followers_url": "https://api.github.com/users/armheb/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 | [] | [
"Hi, could you specify your versions of `tokenizers` and `transformers`? I believe you're running on an older version of `transformers`, could you install from source: `pip install git+https://github.com/huggingface/transformers` and let me know if it fixes this issue? Thank you.",
"This issue has been automatica... | 1,579 | 1,590 | 1,590 | NONE | null | ## π 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:
tokenizer = GPT2TokenizerFast.from_pretrained('gpt2')
The tasks I am working on is:
* [ ] my own task or dataset:
## To R... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2578/reactions",
"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/2578/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2577 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2577/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2577/comments | https://api.github.com/repos/huggingface/transformers/issues/2577/events | https://github.com/huggingface/transformers/issues/2577 | 551,738,014 | MDU6SXNzdWU1NTE3MzgwMTQ= | 2,577 | always occur error:AssertionError | {
"login": "jiangjiaqi6",
"id": 33390819,
"node_id": "MDQ6VXNlcjMzMzkwODE5",
"avatar_url": "https://avatars.githubusercontent.com/u/33390819?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jiangjiaqi6",
"html_url": "https://github.com/jiangjiaqi6",
"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 | [] | [
"It seems like the paths to your data files are incorrect.\r\n\r\nAre you sure they're not at `./dataset/wiki.train.raw ` (notice the leading `.`)",
"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 contri... | 1,579 | 1,585 | 1,585 | NONE | null | ## β Questions & Help
<!-- A clear and concise description of the question. -->
I put the wiki.train.raw and the wiki.test.raw in /dataset,
then run the command:
python run_lm_finetuning.py --output_dir=output --model_type=roberta --model_name_or_path=roberta-base --do_train --train_data_file=... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2577/reactions",
"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/2577/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2576 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2576/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2576/comments | https://api.github.com/repos/huggingface/transformers/issues/2576/events | https://github.com/huggingface/transformers/pull/2576 | 551,724,704 | MDExOlB1bGxSZXF1ZXN0MzY0Mzk0NDUx | 2,576 | fill_mask helper | {
"login": "julien-c",
"id": 326577,
"node_id": "MDQ6VXNlcjMyNjU3Nw==",
"avatar_url": "https://avatars.githubusercontent.com/u/326577?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/julien-c",
"html_url": "https://github.com/julien-c",
"followers_url": "https://api.github.com/users/julien-... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/2576?src=pr&el=h1) Report\n> Merging [#2576](https://codecov.io/gh/huggingface/transformers/pull/2576?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/9d87eafd118739a4c121d69d7cff425264f01e1c?src=pr&el=desc) will **d... | 1,579 | 1,580 | 1,580 | MEMBER | null | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2576/reactions",
"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/2576/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/2576",
"html_url": "https://github.com/huggingface/transformers/pull/2576",
"diff_url": "https://github.com/huggingface/transformers/pull/2576.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/2576.patch",
"merged_at": 1... | |
https://api.github.com/repos/huggingface/transformers/issues/2575 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2575/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2575/comments | https://api.github.com/repos/huggingface/transformers/issues/2575/events | https://github.com/huggingface/transformers/pull/2575 | 551,721,001 | MDExOlB1bGxSZXF1ZXN0MzY0MzkxNjMx | 2,575 | Fix examples/run_tf_ner.py label encoding error #2559 | {
"login": "HuiyingLi",
"id": 1331543,
"node_id": "MDQ6VXNlcjEzMzE1NDM=",
"avatar_url": "https://avatars.githubusercontent.com/u/1331543?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/HuiyingLi",
"html_url": "https://github.com/HuiyingLi",
"followers_url": "https://api.github.com/users/Hu... | [
{
"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/2575?src=pr&el=h1) Report\n> Merging [#2575](https://codecov.io/gh/huggingface/transformers/pull/2575?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/1a8e87be4e2a1b551175bd6f0f749f3d2289010f?src=pr&el=desc) will **n... | 1,579 | 1,585 | 1,585 | NONE | null | This is an explanation and a proposed fix for #2559
The code set `pad_token_label_id = 0`, and increase the total number of labels `num_labels = len(labels) + 1`, but made no change to the label list. Thus the first label in label list has the same index as pad_token_label_id.
Following instructions in README take... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2575/reactions",
"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/2575/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/2575",
"html_url": "https://github.com/huggingface/transformers/pull/2575",
"diff_url": "https://github.com/huggingface/transformers/pull/2575.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/2575.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/2574 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2574/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2574/comments | https://api.github.com/repos/huggingface/transformers/issues/2574/events | https://github.com/huggingface/transformers/issues/2574 | 551,715,399 | MDU6SXNzdWU1NTE3MTUzOTk= | 2,574 | is RoBERTa-base.json in s3 wrong? | {
"login": "skyf0cker",
"id": 32331521,
"node_id": "MDQ6VXNlcjMyMzMxNTIx",
"avatar_url": "https://avatars.githubusercontent.com/u/32331521?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/skyf0cker",
"html_url": "https://github.com/skyf0cker",
"followers_url": "https://api.github.com/users/... | [] | closed | false | null | [] | [
"Yes this file is correct."
] | 1,579 | 1,579 | 1,579 | NONE | null | Q:
when i open the json file downloaded from this url:
`https://s3.amazonaws.com/models.huggingface.co/bert/roberta-large-mnli-vocab.json`,
i found there is so many wrong strange code in it,like this:
> {"<s>": 0, "<pad>": 1, "</s>": 2, "<unk>": 3, ".": 4, "Δ the": 5, ",": 6, "Δ to": 7, "Δ and": 8, "Δ of": 9, "Δ a": 10,... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2574/reactions",
"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/2574/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2573 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2573/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2573/comments | https://api.github.com/repos/huggingface/transformers/issues/2573/events | https://github.com/huggingface/transformers/issues/2573 | 551,709,489 | MDU6SXNzdWU1NTE3MDk0ODk= | 2,573 | Is RoBERTa's pair of sequences tokenizer correct with double </s> | {
"login": "keldLundgaard",
"id": 1884495,
"node_id": "MDQ6VXNlcjE4ODQ0OTU=",
"avatar_url": "https://avatars.githubusercontent.com/u/1884495?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/keldLundgaard",
"html_url": "https://github.com/keldLundgaard",
"followers_url": "https://api.github.... | [] | closed | false | null | [] | [
"Yes, this is how RoBERTa was trained. "
] | 1,579 | 1,579 | 1,579 | NONE | null | In RoBERTa's build_input_with_special_tokens, the comment says
```
A RoBERTa sequence has the following format:
single sequence: <s> X </s>
pair of sequences: <s> A </s></s> B </s>
```
I find the double `</s></s>` very peculiar. Can you please verify that it should not be `</s><s>` (as a normal XML tag)... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2573/reactions",
"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/2573/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2572 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2572/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2572/comments | https://api.github.com/repos/huggingface/transformers/issues/2572/events | https://github.com/huggingface/transformers/issues/2572 | 551,705,738 | MDU6SXNzdWU1NTE3MDU3Mzg= | 2,572 | Bert TPU fine-tuning works on Colab but not in GCP | {
"login": "jswift24",
"id": 1891204,
"node_id": "MDQ6VXNlcjE4OTEyMDQ=",
"avatar_url": "https://avatars.githubusercontent.com/u/1891204?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jswift24",
"html_url": "https://github.com/jswift24",
"followers_url": "https://api.github.com/users/jswif... | [
{
"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, your error states: \r\n\r\n```\r\nOne possible root cause is the client and server binaries are not built with the same version. Please make sure the operation or function is registered in the binary running in this process.\r\n```\r\n\r\nDo you have the same TensorFlow versions for your TPU and your VM?",
"... | 1,579 | 1,601 | 1,585 | NONE | null | ## π 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:
* [ ] 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/2572/reactions",
"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/2572/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2571 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2571/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2571/comments | https://api.github.com/repos/huggingface/transformers/issues/2571/events | https://github.com/huggingface/transformers/issues/2571 | 551,681,286 | MDU6SXNzdWU1NTE2ODEyODY= | 2,571 | Why isn't BERT doing wordpiece tokenization? | {
"login": "jianwolf",
"id": 24360583,
"node_id": "MDQ6VXNlcjI0MzYwNTgz",
"avatar_url": "https://avatars.githubusercontent.com/u/24360583?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jianwolf",
"html_url": "https://github.com/jianwolf",
"followers_url": "https://api.github.com/users/jia... | [] | closed | false | null | [] | [
"Even if I do `add_special_tokens=True` when encoding, I get\r\n\r\n```\r\n[CLS] why isn ' t my card working [SEP]\r\n```\r\n\r\nwhich is still not wordpiece tokenization.",
"When using `encode` and `decode` you're performing the full tokenization steps each time:\r\n\r\nencode: tokenizing -> convert tokens to id... | 1,579 | 1,579 | 1,579 | NONE | null | My code is
```
from transformers import BertTokenizer
tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
text = '''why isn't my card working'''
encoded = tokenizer.encode(text, add_special_tokens=False)
text_tokenized = tokenizer.decode(encoded, clean_up_tokenization_spaces=False)
print(text_tokeni... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2571/reactions",
"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/2571/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2570 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2570/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2570/comments | https://api.github.com/repos/huggingface/transformers/issues/2570/events | https://github.com/huggingface/transformers/pull/2570 | 551,672,846 | MDExOlB1bGxSZXF1ZXN0MzY0MzUzMTY1 | 2,570 | [run_lm_finetuning] Train from scratch | {
"login": "julien-c",
"id": 326577,
"node_id": "MDQ6VXNlcjMyNjU3Nw==",
"avatar_url": "https://avatars.githubusercontent.com/u/326577?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/julien-c",
"html_url": "https://github.com/julien-c",
"followers_url": "https://api.github.com/users/julien-... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/2570?src=pr&el=h1) Report\n> Merging [#2570](https://codecov.io/gh/huggingface/transformers/pull/2570?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/65a89a89768f5922e51cdc7d49990d731e3f2c03?src=pr&el=desc) will **n... | 1,579 | 1,579 | 1,579 | MEMBER | null | Ability to train a model from scratch, rather than finetune a pretrained one. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2570/reactions",
"total_count": 2,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 2,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/2570/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/2570",
"html_url": "https://github.com/huggingface/transformers/pull/2570",
"diff_url": "https://github.com/huggingface/transformers/pull/2570.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/2570.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/2569 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2569/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2569/comments | https://api.github.com/repos/huggingface/transformers/issues/2569/events | https://github.com/huggingface/transformers/pull/2569 | 551,651,488 | MDExOlB1bGxSZXF1ZXN0MzY0MzM1NTg2 | 2,569 | Add lower bound to tqdm for tqdm.auto | {
"login": "brendan-ai2",
"id": 16342367,
"node_id": "MDQ6VXNlcjE2MzQyMzY3",
"avatar_url": "https://avatars.githubusercontent.com/u/16342367?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/brendan-ai2",
"html_url": "https://github.com/brendan-ai2",
"followers_url": "https://api.github.com/... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/2569?src=pr&el=h1) Report\n> Merging [#2569](https://codecov.io/gh/huggingface/transformers/pull/2569?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/65a89a89768f5922e51cdc7d49990d731e3f2c03?src=pr&el=desc) will **n... | 1,579 | 1,579 | 1,579 | CONTRIBUTOR | null | - It appears that `tqdm` only introduced `tqdm.auto` in 4.27.
- See https://github.com/tqdm/tqdm/releases/tag/v4.27.0.
- Without a lower bound I received an error when importing `transformers` in an environment where I already had `tqdm` installed.
- `transformers` version:
```
$ pip list | grep transformers
tran... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2569/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/2569/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/2569",
"html_url": "https://github.com/huggingface/transformers/pull/2569",
"diff_url": "https://github.com/huggingface/transformers/pull/2569.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/2569.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/2568 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2568/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2568/comments | https://api.github.com/repos/huggingface/transformers/issues/2568/events | https://github.com/huggingface/transformers/issues/2568 | 551,633,973 | MDU6SXNzdWU1NTE2MzM5NzM= | 2,568 | Finetuning ALBERT using examples/run_lm_finetuning.py | {
"login": "jianwolf",
"id": 24360583,
"node_id": "MDQ6VXNlcjI0MzYwNTgz",
"avatar_url": "https://avatars.githubusercontent.com/u/24360583?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jianwolf",
"html_url": "https://github.com/jianwolf",
"followers_url": "https://api.github.com/users/jia... | [] | closed | false | null | [] | [
"You're right, ALBERT should work out of the box with the fine tuning script as addressed at #2008 by @thomwolf. It's not too tough to fine-tune ALBERT with the script as reference, and there should also be a PR to add ALBERT and some other language models sometime in the near future",
"Thank you!"
] | 1,579 | 1,579 | 1,579 | NONE | null | ## π Feature
The current run_lm_finetuning.py script seems to not have ALBERT added. We should be able to finetune ALBERT in the same way we do to other models in your library.
| {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2568/reactions",
"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/2568/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2567 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2567/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2567/comments | https://api.github.com/repos/huggingface/transformers/issues/2567/events | https://github.com/huggingface/transformers/issues/2567 | 551,629,855 | MDU6SXNzdWU1NTE2Mjk4NTU= | 2,567 | Bert perform way worse than simple LSTM+Glove | {
"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 | [] | [
"Well, how are you actually using it? Are you actually fine-tuning the model? What's your train loop?",
"I actually just solved the issue. It seems that the code I posted was\ncorrect, but it has to do with where I placed the scheduler.step().\n\nOn Mon, Jan 20, 2020 at 6:57 AM Bram Vanroy <notifications@github.c... | 1,579 | 1,580 | 1,580 | NONE | null | Hi, I am doing a very straightforward entity classification task, but Bert is not giving a good result. I am wondering if there is something wrong with my code.
My task is give a sentence and an entity boundary in that sentence, I predict entity type.
Here is my code to prepare input data. The basic idea is I ha... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2567/reactions",
"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/2567/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2566 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2566/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2566/comments | https://api.github.com/repos/huggingface/transformers/issues/2566/events | https://github.com/huggingface/transformers/issues/2566 | 551,571,037 | MDU6SXNzdWU1NTE1NzEwMzc= | 2,566 | question about tokenizer changes original sequence length | {
"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 | [] | [] | 1,579 | 1,579 | 1,579 | NONE | null | Hi, I am working on an entity classification task where I know the entity boundary and the context.
when I use tokenizer to encode the entire sequence, some token got split up to word pieces that will change the original length of the sequence. I want to extract the states only for the entity, but since the seq lengt... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2566/reactions",
"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/2566/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2565 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2565/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2565/comments | https://api.github.com/repos/huggingface/transformers/issues/2565/events | https://github.com/huggingface/transformers/issues/2565 | 551,570,174 | MDU6SXNzdWU1NTE1NzAxNzQ= | 2,565 | Optionally convert output of FeatureExtraction pipeline to list | {
"login": "lambdaofgod",
"id": 3647577,
"node_id": "MDQ6VXNlcjM2NDc1Nzc=",
"avatar_url": "https://avatars.githubusercontent.com/u/3647577?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lambdaofgod",
"html_url": "https://github.com/lambdaofgod",
"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
},
{
"id": 1771187924,
"node_id": "MDU6TGFiZWwxNzcxMTg3OTI0",
... | closed | false | null | [] | [
"Hi @lambdaofgod,\r\n\r\n.tolist() call is there as Python's lists are more compatible with CSV/JSON serialisation than numpy array.\r\n\r\nDid you have a chance to get number of how actually slower it is ? If the difference is non negligible then we might have a look to optimise .tolist() only when serialising thr... | 1,579 | 1,587 | 1,587 | NONE | null | What is the purpose of .tolist in FeatureExtraction pipeline?
Why is it called? Is this because of some kind of compatibility issue?
If someone needs to use __call__ a lot, it only slows it down. I've tried subclassing FeatureExtractionPipeline, but it's very ugly since then I can't just use it with pipelines.pipel... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2565/reactions",
"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/2565/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2564 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2564/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2564/comments | https://api.github.com/repos/huggingface/transformers/issues/2564/events | https://github.com/huggingface/transformers/pull/2564 | 551,554,439 | MDExOlB1bGxSZXF1ZXN0MzY0MjU1NDMw | 2,564 | Fix glue processor failing on tf datasets | {
"login": "neonbjb",
"id": 833082,
"node_id": "MDQ6VXNlcjgzMzA4Mg==",
"avatar_url": "https://avatars.githubusercontent.com/u/833082?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/neonbjb",
"html_url": "https://github.com/neonbjb",
"followers_url": "https://api.github.com/users/neonbjb/fo... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/2564?src=pr&el=h1) Report\n> Merging [#2564](https://codecov.io/gh/huggingface/transformers/pull/2564?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/6d5049a24d5906ece3fd9b68fb3abe1a0b6bb049?src=pr&el=desc) will **d... | 1,579 | 1,579 | 1,579 | CONTRIBUTOR | null | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2564/reactions",
"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/2564/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/2564",
"html_url": "https://github.com/huggingface/transformers/pull/2564",
"diff_url": "https://github.com/huggingface/transformers/pull/2564.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/2564.patch",
"merged_at": 1... | |
https://api.github.com/repos/huggingface/transformers/issues/2563 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2563/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2563/comments | https://api.github.com/repos/huggingface/transformers/issues/2563/events | https://github.com/huggingface/transformers/pull/2563 | 551,369,367 | MDExOlB1bGxSZXF1ZXN0MzY0MTAzMTM1 | 2,563 | Fix typo in examples/run_squad.py | {
"login": "whitedelay",
"id": 38174055,
"node_id": "MDQ6VXNlcjM4MTc0MDU1",
"avatar_url": "https://avatars.githubusercontent.com/u/38174055?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/whitedelay",
"html_url": "https://github.com/whitedelay",
"followers_url": "https://api.github.com/use... | [] | closed | false | null | [] | [
"Great, thanks!"
] | 1,579 | 1,579 | 1,579 | CONTRIBUTOR | null | Rul -> Run | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2563/reactions",
"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/2563/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/2563",
"html_url": "https://github.com/huggingface/transformers/pull/2563",
"diff_url": "https://github.com/huggingface/transformers/pull/2563.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/2563.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/2562 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2562/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2562/comments | https://api.github.com/repos/huggingface/transformers/issues/2562/events | https://github.com/huggingface/transformers/issues/2562 | 551,323,073 | MDU6SXNzdWU1NTEzMjMwNzM= | 2,562 | Architectures for Dialogue | {
"login": "lukasfrank",
"id": 9821158,
"node_id": "MDQ6VXNlcjk4MjExNTg=",
"avatar_url": "https://avatars.githubusercontent.com/u/9821158?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lukasfrank",
"html_url": "https://github.com/lukasfrank",
"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 check out [DialoGPT](https://huggingface.co/microsoft/DialoGPT-large) by @dreasysnail?",
"This issue has been automatically m... | 1,579 | 1,590 | 1,590 | NONE | null | ## β Questions & Help
Hi π
I'm trying to build a dialogue system which should reply based on a history, memory (which is represented as a string) and a confidence if the memory content is correct and should be used.
Here two examples:
- history: _Hi_
memory: _name: Max_
confidence: _0.2_
=> expected out... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2562/reactions",
"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/2562/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2561 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2561/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2561/comments | https://api.github.com/repos/huggingface/transformers/issues/2561/events | https://github.com/huggingface/transformers/issues/2561 | 551,305,554 | MDU6SXNzdWU1NTEzMDU1NTQ= | 2,561 | Model upload and sharing - delete, update, rename.... | {
"login": "miki537",
"id": 59996473,
"node_id": "MDQ6VXNlcjU5OTk2NDcz",
"avatar_url": "https://avatars.githubusercontent.com/u/59996473?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/miki537",
"html_url": "https://github.com/miki537",
"followers_url": "https://api.github.com/users/miki53... | [] | closed | false | null | [] | [
"Hi @miki537, we already have `transformers-cli s3 rm ____` but it is not super well documented.\r\n\r\nI'll improve the documentation on that point. Also `transformers-cli upload` will overwrite existing files with the same name so you can already update files.\r\n\r\nS3 doesn't not support moving/renaming files s... | 1,579 | 1,604 | 1,580 | NONE | null | ## π Feature
It would be great to have the option of deleting, renaming and adding description to the community models. I saw that there are already some errors in the model names which can probably not be fixed because of this missing functionality.
We should have something like:
transformers-cli delete
tran... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2561/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/2561/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2560 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2560/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2560/comments | https://api.github.com/repos/huggingface/transformers/issues/2560/events | https://github.com/huggingface/transformers/issues/2560 | 551,231,487 | MDU6SXNzdWU1NTEyMzE0ODc= | 2,560 | why this implementation didn't apply residual and layer norm? | {
"login": "reniew",
"id": 32028135,
"node_id": "MDQ6VXNlcjMyMDI4MTM1",
"avatar_url": "https://avatars.githubusercontent.com/u/32028135?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/reniew",
"html_url": "https://github.com/reniew",
"followers_url": "https://api.github.com/users/reniew/fo... | [] | closed | false | null | [] | [
"The layer normalization happening after the attention is visible [here](https://github.com/huggingface/transformers/blob/master/src/transformers/modeling_albert.py#L255).\r\n\r\nThe `inner_group_num` is used to better understand how many layers are in a specific group. It is set to 1 in all the configurations tha... | 1,579 | 1,582 | 1,582 | NONE | null | ## β Questions & Help
In ALBERT implementation code `modeling_albert.py`, i can't find applying skip-connection and layer normalizing after multi-head attention layer. I didn't read about this tequnique.
Is there a special reason about it?
One more, i saw the argument `inner_group_num` in `AlbertLayerGroup` cla... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2560/reactions",
"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/2560/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2559 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2559/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2559/comments | https://api.github.com/repos/huggingface/transformers/issues/2559/events | https://github.com/huggingface/transformers/issues/2559 | 551,164,146 | MDU6SXNzdWU1NTExNjQxNDY= | 2,559 | Prediction on NER Tensorflow 2 | {
"login": "imayachita",
"id": 3615586,
"node_id": "MDQ6VXNlcjM2MTU1ODY=",
"avatar_url": "https://avatars.githubusercontent.com/u/3615586?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/imayachita",
"html_url": "https://github.com/imayachita",
"followers_url": "https://api.github.com/users... | [] | closed | false | null | [] | [
"Yes, the output is wrong. \r\nI think the run_tf_ner.py script has a bug where the labels are off by 1. \r\nAnd the off-by-1 prediction result is sent for evaluation `metrics.classification_report(y_true, y_pred, digits=4)` therefore the evaluation result is wrong too. ",
"Thanks @HuiyingLi! The workaround works... | 1,579 | 1,579 | 1,579 | NONE | null | Hi,
I tried running the implementation of NER on Tensorflow 2. I have a problem doing the prediction. Seems like the label to index are off. Here is some examples:
```
SOCCER B-ORG
- B-ORG
JAPAN B-MISC
GET B-ORG
LUCKY B-MISC
WIN B-ORG
, B-ORG
CHINA B-MISC
IN B-ORG
SURPRISE O
DEFEAT B-ORG
. B-ORG
Nadi... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2559/reactions",
"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/2559/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2558 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2558/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2558/comments | https://api.github.com/repos/huggingface/transformers/issues/2558/events | https://github.com/huggingface/transformers/pull/2558 | 551,160,888 | MDExOlB1bGxSZXF1ZXN0MzYzOTMzMTQ0 | 2,558 | solve the exception: [AttributeError: 'bool' object has no attribute 'mean'] | {
"login": "7will10",
"id": 31179081,
"node_id": "MDQ6VXNlcjMxMTc5MDgx",
"avatar_url": "https://avatars.githubusercontent.com/u/31179081?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/7will10",
"html_url": "https://github.com/7will10",
"followers_url": "https://api.github.com/users/7will1... | [] | closed | false | null | [] | [
"Hi! I'd like to replicate the error you had with the `AttributeError`. Could you let me know in which situation you faced this error?"
] | 1,579 | 1,581 | 1,581 | NONE | null | modified method simple_accuracy(),
before:
it's (preds == labels).mean()
This will cause an exception[AttributeError: 'bool' object has no attribute 'mean'],
then after update:
change to accuracy_score(labels,preds),
use this method accuracy_score() in package sklearn.metrics | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2558/reactions",
"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/2558/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/2558",
"html_url": "https://github.com/huggingface/transformers/pull/2558",
"diff_url": "https://github.com/huggingface/transformers/pull/2558.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/2558.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/2557 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2557/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2557/comments | https://api.github.com/repos/huggingface/transformers/issues/2557/events | https://github.com/huggingface/transformers/pull/2557 | 551,110,027 | MDExOlB1bGxSZXF1ZXN0MzYzODkyMDM3 | 2,557 | Fix BasicTokenizer to respect `never_split` parameters | {
"login": "DeNeutoy",
"id": 16001974,
"node_id": "MDQ6VXNlcjE2MDAxOTc0",
"avatar_url": "https://avatars.githubusercontent.com/u/16001974?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/DeNeutoy",
"html_url": "https://github.com/DeNeutoy",
"followers_url": "https://api.github.com/users/DeN... | [] | closed | false | null | [] | [
"Not sure how to fix that last CI build, seems unrelated?",
"Unrelated Heisenbug, relaunched the CI",
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/2557?src=pr&el=h1) Report\n> Merging [#2557](https://codecov.io/gh/huggingface/transformers/pull/2557?src=pr&el=desc) into [master](https://codec... | 1,579 | 1,579 | 1,579 | CONTRIBUTOR | null | `never_split` was not being passed to `_split_on_punc`, causing special tokens to be split apart. Failing test (in first commit) demonstrates the problem. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2557/reactions",
"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/2557/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/2557",
"html_url": "https://github.com/huggingface/transformers/pull/2557",
"diff_url": "https://github.com/huggingface/transformers/pull/2557.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/2557.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/2556 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2556/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2556/comments | https://api.github.com/repos/huggingface/transformers/issues/2556/events | https://github.com/huggingface/transformers/issues/2556 | 551,020,346 | MDU6SXNzdWU1NTEwMjAzNDY= | 2,556 | Quantized model not preserved when imported using from_pretrained() | {
"login": "trevorpfiz",
"id": 24904780,
"node_id": "MDQ6VXNlcjI0OTA0Nzgw",
"avatar_url": "https://avatars.githubusercontent.com/u/24904780?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/trevorpfiz",
"html_url": "https://github.com/trevorpfiz",
"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 | [] | [
"@LysandreJik any ideas on this? I am itching to use a quantized BERT model in production, but it does not work when loaded in :(",
"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",
"... | 1,579 | 1,594 | 1,585 | NONE | null | ## π 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:
When I import the saved quantized model using `from_pretrained()`, the model's size is inflated to the pre-quantized version. The... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2556/reactions",
"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/2556/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2555 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2555/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2555/comments | https://api.github.com/repos/huggingface/transformers/issues/2555/events | https://github.com/huggingface/transformers/pull/2555 | 551,016,011 | MDExOlB1bGxSZXF1ZXN0MzYzODE0MDg1 | 2,555 | Fix output name | {
"login": "glicerico",
"id": 23503930,
"node_id": "MDQ6VXNlcjIzNTAzOTMw",
"avatar_url": "https://avatars.githubusercontent.com/u/23503930?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/glicerico",
"html_url": "https://github.com/glicerico",
"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 | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/2555?src=pr&el=h1) Report\n> Merging [#2555](https://codecov.io/gh/huggingface/transformers/pull/2555?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/6e2c28a14a3d171e8c4d3838429abb1d69456df5?src=pr&el=desc) will **n... | 1,579 | 1,584 | 1,584 | NONE | null | Output variable name `all_hidden_states` found in README is inconsistent with documentation's `hidden_states`: https://huggingface.co/transformers/model_doc/bert.html#bertmodel | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2555/reactions",
"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/2555/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/2555",
"html_url": "https://github.com/huggingface/transformers/pull/2555",
"diff_url": "https://github.com/huggingface/transformers/pull/2555.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/2555.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/2554 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2554/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2554/comments | https://api.github.com/repos/huggingface/transformers/issues/2554/events | https://github.com/huggingface/transformers/issues/2554 | 551,013,629 | MDU6SXNzdWU1NTEwMTM2Mjk= | 2,554 | CTRL tokenizer has no special tokens to indicate EOS | {
"login": "erickrf",
"id": 294483,
"node_id": "MDQ6VXNlcjI5NDQ4Mw==",
"avatar_url": "https://avatars.githubusercontent.com/u/294483?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/erickrf",
"html_url": "https://github.com/erickrf",
"followers_url": "https://api.github.com/users/erickrf/fo... | [] | closed | false | null | [] | [] | 1,579 | 1,582 | 1,582 | CONTRIBUTOR | null | ## π Bug
The `generate` method from `PreTrainedModel` by default uses index 0 as EOS. This is a problem with CTRL, because its tokenizer has the word `the` mapped to this id.
Actually the CTRL has no special tokens besides UNK:
```
tokenizer = CTRLTokenizer.from_pretrained('ctrl')
tokenizer.special_tokens_m... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2554/reactions",
"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/2554/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2553 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2553/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2553/comments | https://api.github.com/repos/huggingface/transformers/issues/2553/events | https://github.com/huggingface/transformers/issues/2553 | 551,003,765 | MDU6SXNzdWU1NTEwMDM3NjU= | 2,553 | Model not learning when using albert-base-v2 -- ALBERT | {
"login": "trevorpfiz",
"id": 24904780,
"node_id": "MDQ6VXNlcjI0OTA0Nzgw",
"avatar_url": "https://avatars.githubusercontent.com/u/24904780?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/trevorpfiz",
"html_url": "https://github.com/trevorpfiz",
"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 | [] | [
"The interesting things is that why do u use another modelβs tokenizer to\nprocess data .\n\nDo u know each model tokenizer is a map function which map ID to token ?\nSo for each model , the same word for them is mapping to different iD . So\nit will not learning . Do u read paper ?\n\nOn Fri, Jan 17, 2020 at 03:19... | 1,579 | 1,592 | 1,585 | NONE | null | ## π Bug
<!-- Important information -->
Model I am using (Bert, XLNet....): AlbertForSequenceClassification
Language I am using the model on (English, Chinese....): English
The problem arise when using:
When I use `albert-base-v2` instead of `albert-base-v1` for the model and tokenizer, the model does not... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2553/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/2553/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2552 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2552/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2552/comments | https://api.github.com/repos/huggingface/transformers/issues/2552/events | https://github.com/huggingface/transformers/pull/2552 | 550,975,266 | MDExOlB1bGxSZXF1ZXN0MzYzNzgxOTEy | 2,552 | fix #2549 | {
"login": "Guitaricet",
"id": 2821124,
"node_id": "MDQ6VXNlcjI4MjExMjQ=",
"avatar_url": "https://avatars.githubusercontent.com/u/2821124?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Guitaricet",
"html_url": "https://github.com/Guitaricet",
"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",
"=(",
"This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activ... | 1,579 | 1,594 | 1,594 | NONE | null | closes #2549
proposed solution for unsupported operand type error in tokenizer.batch_encode_plus | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2552/reactions",
"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/2552/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/2552",
"html_url": "https://github.com/huggingface/transformers/pull/2552",
"diff_url": "https://github.com/huggingface/transformers/pull/2552.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/2552.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/2551 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2551/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2551/comments | https://api.github.com/repos/huggingface/transformers/issues/2551/events | https://github.com/huggingface/transformers/issues/2551 | 550,971,592 | MDU6SXNzdWU1NTA5NzE1OTI= | 2,551 | EnvironmentError OSError: Couldn't reach server | {
"login": "baggyg",
"id": 4569081,
"node_id": "MDQ6VXNlcjQ1NjkwODE=",
"avatar_url": "https://avatars.githubusercontent.com/u/4569081?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/baggyg",
"html_url": "https://github.com/baggyg",
"followers_url": "https://api.github.com/users/baggyg/foll... | [] | closed | false | null | [] | [
"Can you try again? It seems that the server is reachable now. Of course you must be connected to the Internet.",
"The server was reachable. I try the same URL in my browser at the time of doing it and it loaded fine. Its just via python / transformers that he problem occurs (I've tried everyday for 3 days now). ... | 1,579 | 1,584 | 1,579 | NONE | null | ## π 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: (give details)
* [ ] my own modified scripts: (give details)
The tasks I am working on is:... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2551/reactions",
"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/2551/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2550 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2550/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2550/comments | https://api.github.com/repos/huggingface/transformers/issues/2550/events | https://github.com/huggingface/transformers/issues/2550 | 550,957,418 | MDU6SXNzdWU1NTA5NTc0MTg= | 2,550 | fast gpt2 inference | {
"login": "rajarsheem",
"id": 6441313,
"node_id": "MDQ6VXNlcjY0NDEzMTM=",
"avatar_url": "https://avatars.githubusercontent.com/u/6441313?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/rajarsheem",
"html_url": "https://github.com/rajarsheem",
"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 | [] | [
"Have given some thought to using TorchScript but since my input sequence length changes each time, the easier tracing approach won't work.\r\n\r\nONNX also faced the same problem.",
"Hi rajarsheem,\r\nCan you please share your code that doing the batch inference with variable-length sequences?\r\n\r\nThanks",
... | 1,579 | 1,589 | 1,589 | NONE | null | I have a fine-tuned ```GPT2LMHeadModel``` (gpt2-medium) which I am using to run inference on large data (>60M sequences) offline. At each iteration, my input is a batch of 30 variable-length sequences which gets padded according to the max length of the batch. My current speed is around 8 secs/iter and input sequences ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2550/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/2550/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2549 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2549/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2549/comments | https://api.github.com/repos/huggingface/transformers/issues/2549/events | https://github.com/huggingface/transformers/issues/2549 | 550,954,140 | MDU6SXNzdWU1NTA5NTQxNDA= | 2,549 | unsupported operand type error in tokenizer.batch_encode_plus | {
"login": "Guitaricet",
"id": 2821124,
"node_id": "MDQ6VXNlcjI4MjExMjQ=",
"avatar_url": "https://avatars.githubusercontent.com/u/2821124?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Guitaricet",
"html_url": "https://github.com/Guitaricet",
"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 | [] | [
"I am fixing this issue and just understood that `return_attention_masks` is supposed to work only if `return_tensors is not None`. But it is not mentioned in the docstring nor raises an error. Also, in the case of `is_tf_available` and `return_tensors == 'pt'` current code would return tensorflow maks.\r\nI'd sug... | 1,579 | 1,585 | 1,585 | NONE | null | ## π Bug
<!-- Important information -->
Model I am using (Bert, XLNet....): any model
Language I am using the model on (English, Chinese....): any language
The problem arise when using: tokenizer object
The tasks I am working on is: my own tasks
## To Reproduce
Steps to reproduce the behavior:
us... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2549/reactions",
"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/2549/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2548 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2548/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2548/comments | https://api.github.com/repos/huggingface/transformers/issues/2548/events | https://github.com/huggingface/transformers/issues/2548 | 550,920,835 | MDU6SXNzdWU1NTA5MjA4MzU= | 2,548 | SQuAD convert_examples_to_features skipping doc tokens when they exceed max_seq_length | {
"login": "ofirzaf",
"id": 18296312,
"node_id": "MDQ6VXNlcjE4Mjk2MzEy",
"avatar_url": "https://avatars.githubusercontent.com/u/18296312?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ofirzaf",
"html_url": "https://github.com/ofirzaf",
"followers_url": "https://api.github.com/users/ofirza... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | {
"login": "LysandreJik",
"id": 30755778,
"node_id": "MDQ6VXNlcjMwNzU1Nzc4",
"avatar_url": "https://avatars.githubusercontent.com/u/30755778?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/LysandreJik",
"html_url": "https://github.com/LysandreJik",
"followers_url": "https://api.github.com/... | [
{
"login": "LysandreJik",
"id": 30755778,
"node_id": "MDQ6VXNlcjMwNzU1Nzc4",
"avatar_url": "https://avatars.githubusercontent.com/u/30755778?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/LysandreJik",
"html_url": "https://github.com/LysandreJik",
"followers_url": "htt... | [
"Indeed, this looks like a bug, thank you for opening an issue. I'll take a look at it.",
"This issue stems from the two arguments: `max_seq_length=128` and `doc_stride=128`.\r\n\r\nWould you mind telling me the expected behavior when putting a doc stride as big as the maximum sequence length? Since the sequence ... | 1,579 | 1,584 | 1,584 | CONTRIBUTOR | null | ## π Bug
<!-- Important information -->
Model I am using (Bert, XLNet....): BERT
Language I am using the model on (English, Chinese....): English
Using Transformers v2.3.0 installed from pypi
The problem arise when using:
transformers/data/processors/squad.py + BertTokenizer
The tasks I am working o... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2548/reactions",
"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/2548/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2547 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2547/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2547/comments | https://api.github.com/repos/huggingface/transformers/issues/2547/events | https://github.com/huggingface/transformers/issues/2547 | 550,809,089 | MDU6SXNzdWU1NTA4MDkwODk= | 2,547 | AlbertDoublehHeadsModel | {
"login": "hasnain2808",
"id": 28212972,
"node_id": "MDQ6VXNlcjI4MjEyOTcy",
"avatar_url": "https://avatars.githubusercontent.com/u/28212972?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/hasnain2808",
"html_url": "https://github.com/hasnain2808",
"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"
] | 1,579 | 1,584 | 1,584 | NONE | null | # πNew model addition
## Model description
Like we have OpenAIGPTDoubleHeadsModel. I actually want to know if someone is already working on similar model for Albert
If not that with some help I would want to contribute towards it
<!-- Important information -->
## Open Source status
* [ ] the model implemen... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2547/reactions",
"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/2547/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2546 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2546/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2546/comments | https://api.github.com/repos/huggingface/transformers/issues/2546/events | https://github.com/huggingface/transformers/issues/2546 | 550,750,245 | MDU6SXNzdWU1NTA3NTAyNDU= | 2,546 | Unable to generate ALBERT embeddings of size 128 | {
"login": "MaheshChandrra",
"id": 13826929,
"node_id": "MDQ6VXNlcjEzODI2OTI5",
"avatar_url": "https://avatars.githubusercontent.com/u/13826929?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/MaheshChandrra",
"html_url": "https://github.com/MaheshChandrra",
"followers_url": "https://api.gi... | [] | closed | false | null | [] | [
"Please format your post with [code blocks](https://help.github.com/en/github/writing-on-github/creating-and-highlighting-code-blocks).\r\n\r\nThis seems like a very general question, where you want to change the size of a dimension of an output tensor. There are different approaches to this. If you want to do this... | 1,579 | 1,579 | 1,579 | NONE | null | ## β Questions & Help
<!-- A clear and concise description of the question. -->
Hi Team Hugging face,
Due to memory issues I wanted to migrate from BERT to ALBERT,tried the model present in the transformers,but I'm unable to generate the embedding's of size 128,all I get in outputs is 768 dimension embedding's... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2546/reactions",
"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/2546/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2545 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2545/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2545/comments | https://api.github.com/repos/huggingface/transformers/issues/2545/events | https://github.com/huggingface/transformers/pull/2545 | 550,707,167 | MDExOlB1bGxSZXF1ZXN0MzYzNTYzNDk0 | 2,545 | modified method simple_accuracy(), solve the exception | {
"login": "7will10",
"id": 31179081,
"node_id": "MDQ6VXNlcjMxMTc5MDgx",
"avatar_url": "https://avatars.githubusercontent.com/u/31179081?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/7will10",
"html_url": "https://github.com/7will10",
"followers_url": "https://api.github.com/users/7will1... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/2545?src=pr&el=h1) Report\n> Merging [#2545](https://codecov.io/gh/huggingface/transformers/pull/2545?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/7833dfccac0d7d74e12d2b2be1f6caa6e895ca73?src=pr&el=desc) will **i... | 1,579 | 1,579 | 1,579 | NONE | null | modified method simple_accuracy(),
before:
it's (preds == labels).mean()
This will cause an exception[AttributeError: 'bool' object has no attribute 'mean'],
then after update:
change to accuracy_score(labels,preds),
use this method accuracy_score() in package sklearn.metrics | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2545/reactions",
"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/2545/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/2545",
"html_url": "https://github.com/huggingface/transformers/pull/2545",
"diff_url": "https://github.com/huggingface/transformers/pull/2545.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/2545.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/2544 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2544/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2544/comments | https://api.github.com/repos/huggingface/transformers/issues/2544/events | https://github.com/huggingface/transformers/pull/2544 | 550,699,946 | MDExOlB1bGxSZXF1ZXN0MzYzNTU3NjY2 | 2,544 | modified method simple_accuracy() | {
"login": "7will10",
"id": 31179081,
"node_id": "MDQ6VXNlcjMxMTc5MDgx",
"avatar_url": "https://avatars.githubusercontent.com/u/31179081?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/7will10",
"html_url": "https://github.com/7will10",
"followers_url": "https://api.github.com/users/7will1... | [] | closed | false | null | [] | [] | 1,579 | 1,579 | 1,579 | NONE | null | modified method simple_accuracy(),
before:
it's (preds == labels).mean()
This will cause an exception[AttributeError: 'bool' object has no attribute 'mean'],
then after update:
change to accuracy_score(labels,preds),
use this method accuracy_score() in package sklearn.metrics | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2544/reactions",
"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/2544/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/2544",
"html_url": "https://github.com/huggingface/transformers/pull/2544",
"diff_url": "https://github.com/huggingface/transformers/pull/2544.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/2544.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/2543 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2543/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2543/comments | https://api.github.com/repos/huggingface/transformers/issues/2543/events | https://github.com/huggingface/transformers/pull/2543 | 550,694,323 | MDExOlB1bGxSZXF1ZXN0MzYzNTUzMDc0 | 2,543 | modified method simple_accuracy() | {
"login": "7will10",
"id": 31179081,
"node_id": "MDQ6VXNlcjMxMTc5MDgx",
"avatar_url": "https://avatars.githubusercontent.com/u/31179081?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/7will10",
"html_url": "https://github.com/7will10",
"followers_url": "https://api.github.com/users/7will1... | [] | closed | false | null | [] | [] | 1,579 | 1,579 | 1,579 | NONE | null | modified method simple_accuracy(),
before:
it's (preds == labels).mean()
This will cause an exception[AttributeError: 'bool' object has no attribute 'mean'],
then after update:
change to accuracy_score(labels,preds),
use this method accuracy_score() in package sklearn.metrics | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2543/reactions",
"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/2543/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/2543",
"html_url": "https://github.com/huggingface/transformers/pull/2543",
"diff_url": "https://github.com/huggingface/transformers/pull/2543.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/2543.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/2542 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2542/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2542/comments | https://api.github.com/repos/huggingface/transformers/issues/2542/events | https://github.com/huggingface/transformers/issues/2542 | 550,579,413 | MDU6SXNzdWU1NTA1Nzk0MTM= | 2,542 | Dynamic Quantization on ALBERT (pytorch) | {
"login": "Rachnas",
"id": 11646550,
"node_id": "MDQ6VXNlcjExNjQ2NTUw",
"avatar_url": "https://avatars.githubusercontent.com/u/11646550?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Rachnas",
"html_url": "https://github.com/Rachnas",
"followers_url": "https://api.github.com/users/Rachna... | [
{
"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 know this doesn't directly answer the question, but I have been playing around with quantization of BERT and everything is good until I want to load the model into my notebook. The size of the model is inflated back to over 400 MB from under 200 MB, and the accuracy takes a huge hit. I noticed this when I trie... | 1,579 | 1,614 | 1,585 | NONE | null | ## β Questions & Help
Hi,
Thank you for providing great documentation on quantization:
https://pytorch.org/tutorials/intermediate/dynamic_quantization_bert_tutorial.html
I am trying similar steps on Albert Pytorch model, converted "albert-base-v1" to quantized one by applying dynamic quantization on linear layers... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2542/reactions",
"total_count": 2,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 2
} | https://api.github.com/repos/huggingface/transformers/issues/2542/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2541 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2541/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2541/comments | https://api.github.com/repos/huggingface/transformers/issues/2541/events | https://github.com/huggingface/transformers/issues/2541 | 550,570,009 | MDU6SXNzdWU1NTA1NzAwMDk= | 2,541 | squad convert example to features potential bug | {
"login": "phuongpm241",
"id": 29219768,
"node_id": "MDQ6VXNlcjI5MjE5NzY4",
"avatar_url": "https://avatars.githubusercontent.com/u/29219768?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/phuongpm241",
"html_url": "https://github.com/phuongpm241",
"followers_url": "https://api.github.com/... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | {
"login": "LysandreJik",
"id": 30755778,
"node_id": "MDQ6VXNlcjMwNzU1Nzc4",
"avatar_url": "https://avatars.githubusercontent.com/u/30755778?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/LysandreJik",
"html_url": "https://github.com/LysandreJik",
"followers_url": "https://api.github.com/... | [
{
"login": "LysandreJik",
"id": 30755778,
"node_id": "MDQ6VXNlcjMwNzU1Nzc4",
"avatar_url": "https://avatars.githubusercontent.com/u/30755778?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/LysandreJik",
"html_url": "https://github.com/LysandreJik",
"followers_url": "htt... | [
"The new and old versions of SQuAD should behave exactly the same when building features. Do you think you could provide an example script that replicates this issue, so that I may take a look at it?",
"Possibly related to #2548 ",
"I was trying to run `run_squad.py` using this script\r\n```\r\nCUDA_VISIBLE_DEV... | 1,579 | 1,584 | 1,584 | NONE | null | ## β Questions & Help
<!-- A clear and concise description of the question. -->
I used an older version of run_squad.py (and everything else in the example). My dataset contains very long documents (1000-2000 tokens). In the past, convert_example_to_features returns about 913 features per 12 examples. However, afte... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2541/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/2541/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2540 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2540/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2540/comments | https://api.github.com/repos/huggingface/transformers/issues/2540/events | https://github.com/huggingface/transformers/pull/2540 | 550,519,215 | MDExOlB1bGxSZXF1ZXN0MzYzNDEyMDQ4 | 2,540 | [PyTorch 1.4] Fix failing torchscript test for xlnet | {
"login": "julien-c",
"id": 326577,
"node_id": "MDQ6VXNlcjMyNjU3Nw==",
"avatar_url": "https://avatars.githubusercontent.com/u/326577?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/julien-c",
"html_url": "https://github.com/julien-c",
"followers_url": "https://api.github.com/users/julien-... | [] | closed | false | null | [] | [] | 1,579 | 1,579 | 1,579 | MEMBER | null | model.parameters() order is apparently not stable (only for xlnet, for some reason) | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2540/reactions",
"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/2540/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/2540",
"html_url": "https://github.com/huggingface/transformers/pull/2540",
"diff_url": "https://github.com/huggingface/transformers/pull/2540.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/2540.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/2539 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2539/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2539/comments | https://api.github.com/repos/huggingface/transformers/issues/2539/events | https://github.com/huggingface/transformers/issues/2539 | 550,507,196 | MDU6SXNzdWU1NTA1MDcxOTY= | 2,539 | Finetuning TFDistilBertForQuestionAnswering on SQuAD | {
"login": "melwazir",
"id": 37004311,
"node_id": "MDQ6VXNlcjM3MDA0MzEx",
"avatar_url": "https://avatars.githubusercontent.com/u/37004311?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/melwazir",
"html_url": "https://github.com/melwazir",
"followers_url": "https://api.github.com/users/mel... | [
{
"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, \r\n\r\n**Regarding the lack of tf examples**:\r\nI am looking for a similar example ( a squad tf one) and found this [issue](https://github.com/huggingface/transformers/issues/2387), where @LysandreJik mentioned that he is currently working on exactly that. \r\n\r\n**Regarding your specific error**: \r\nsquad... | 1,579 | 1,585 | 1,585 | NONE | null | ## β Questions & Help
<!-- A clear and concise description of the question. -->
Hi. I'm trying to finetune a TFDistilBertForQuestionAnswering model on the SQuAD 1.1 dataset, but I'm getting the following error at the "fit" statement:
ValueError: Failed to find data adapter that can handle input: (<class 'list'... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2539/reactions",
"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/2539/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2538 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2538/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2538/comments | https://api.github.com/repos/huggingface/transformers/issues/2538/events | https://github.com/huggingface/transformers/pull/2538 | 550,488,329 | MDExOlB1bGxSZXF1ZXN0MzYzMzg2NTg5 | 2,538 | :lipstick: super | {
"login": "julien-c",
"id": 326577,
"node_id": "MDQ6VXNlcjMyNjU3Nw==",
"avatar_url": "https://avatars.githubusercontent.com/u/326577?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/julien-c",
"html_url": "https://github.com/julien-c",
"followers_url": "https://api.github.com/users/julien-... | [] | closed | false | null | [] | [
"The CI errors are unrelated to this PR (got the same ones on a commit to master), so I'll try to fix them on another branch @thomwolf @LysandreJik ",
"CI error fixed in #2540 "
] | 1,579 | 1,579 | 1,579 | MEMBER | null | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2538/reactions",
"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/2538/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/2538",
"html_url": "https://github.com/huggingface/transformers/pull/2538",
"diff_url": "https://github.com/huggingface/transformers/pull/2538.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/2538.patch",
"merged_at": 1... | |
https://api.github.com/repos/huggingface/transformers/issues/2537 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2537/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2537/comments | https://api.github.com/repos/huggingface/transformers/issues/2537/events | https://github.com/huggingface/transformers/issues/2537 | 550,467,541 | MDU6SXNzdWU1NTA0Njc1NDE= | 2,537 | [Question] Help needed to understand how torch.distributed.barrier() works | {
"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 | [] | [
"I've provided an answer on Stack Overflow. Please close the issue here on Github. Thanks.",
"Thanks @BramVanroy for the detailed answer! "
] | 1,579 | 1,579 | 1,579 | CONTRIBUTOR | null | ## β Questions & Help
<!-- A clear and concise description of the question. -->
I have been trying to understand how torch.distributed.barrier() is being used in the examples in this repo. I posted [this stackoverflow question](https://stackoverflow.com/questions/59760328/how-does-torch-distributed-barrier-work). ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2537/reactions",
"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/2537/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2536 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2536/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2536/comments | https://api.github.com/repos/huggingface/transformers/issues/2536/events | https://github.com/huggingface/transformers/issues/2536 | 550,460,269 | MDU6SXNzdWU1NTA0NjAyNjk= | 2,536 | Universal Sentence Encoder | {
"login": "rdisipio",
"id": 7974270,
"node_id": "MDQ6VXNlcjc5NzQyNzA=",
"avatar_url": "https://avatars.githubusercontent.com/u/7974270?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/rdisipio",
"html_url": "https://github.com/rdisipio",
"followers_url": "https://api.github.com/users/rdisi... | [
{
"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 | [] | [
"+1 !!\r\nAt reply.ai we have been using USE a lot for Semantic Retrieval. What most impressed us was the Q&A dual encoder model. Works better than anything else I know in case you need semantic similarity between a query and contexts.\r\nIt's true that Tensorflow Hub makes it super easy to work with. But we use yo... | 1,579 | 1,677 | 1,604 | NONE | null | # πNew model addition
## Model description
Encoder of greater-than-word length text trained on a variety of data.
## Open Source status
* [ ] the model implementation is available: see paper https://arxiv.org/abs/1803.11175
* [ ] the model weights are available: available from tfhub: https://tfhub.dev/g... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2536/reactions",
"total_count": 11,
"+1": 11,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/2536/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2535 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2535/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2535/comments | https://api.github.com/repos/huggingface/transformers/issues/2535/events | https://github.com/huggingface/transformers/pull/2535 | 550,431,924 | MDExOlB1bGxSZXF1ZXN0MzYzMzM5Njcz | 2,535 | Tokenizer.from_pretrained: fetch all possible files remotely | {
"login": "julien-c",
"id": 326577,
"node_id": "MDQ6VXNlcjMyNjU3Nw==",
"avatar_url": "https://avatars.githubusercontent.com/u/326577?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/julien-c",
"html_url": "https://github.com/julien-c",
"followers_url": "https://api.github.com/users/julien-... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/2535?src=pr&el=h1) Report\n> Merging [#2535](https://codecov.io/gh/huggingface/transformers/pull/2535?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/eb59e9f70513b538d2174d4ea1efea7ba8554b58?src=pr&el=desc) will **d... | 1,579 | 1,579 | 1,579 | MEMBER | null | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2535/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/2535/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/2535",
"html_url": "https://github.com/huggingface/transformers/pull/2535",
"diff_url": "https://github.com/huggingface/transformers/pull/2535.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/2535.patch",
"merged_at": 1... | |
https://api.github.com/repos/huggingface/transformers/issues/2534 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2534/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2534/comments | https://api.github.com/repos/huggingface/transformers/issues/2534/events | https://github.com/huggingface/transformers/issues/2534 | 550,241,386 | MDU6SXNzdWU1NTAyNDEzODY= | 2,534 | DistilBERT accuracies on the glue test set. | {
"login": "smr97",
"id": 18290261,
"node_id": "MDQ6VXNlcjE4MjkwMjYx",
"avatar_url": "https://avatars.githubusercontent.com/u/18290261?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/smr97",
"html_url": "https://github.com/smr97",
"followers_url": "https://api.github.com/users/smr97/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 | [] | [
"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 want the numbers too.",
"You can check the model card for the evaluation results: https://huggingface.co/distilbert-base-uncased-fi... | 1,579 | 1,657 | 1,584 | NONE | null | ## β Questions & Help
<!-- A clear and concise description of the question. -->
I need to compare my research against distilBERT as a baseline for a paper in progress. I went through your publication and found that you don't report accuracies on the glue test set and instead on the dev set. TINY BERT publication by... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2534/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/2534/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2533 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2533/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2533/comments | https://api.github.com/repos/huggingface/transformers/issues/2533/events | https://github.com/huggingface/transformers/issues/2533 | 550,236,707 | MDU6SXNzdWU1NTAyMzY3MDc= | 2,533 | Gradient accumulation | {
"login": "okanlv",
"id": 29547397,
"node_id": "MDQ6VXNlcjI5NTQ3Mzk3",
"avatar_url": "https://avatars.githubusercontent.com/u/29547397?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/okanlv",
"html_url": "https://github.com/okanlv",
"followers_url": "https://api.github.com/users/okanlv/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"
] | 1,579 | 1,584 | 1,584 | NONE | null | Shouldn't we include `len(train_dataloader)` along with `step` here considering `len(train_dataloader)` might be a odd number? In that case, we could accumulate the gradients more times than `gradient_accumulation_steps`.
https://github.com/huggingface/transformers/blob/0412f3d9298cdb8ba7f69570753ec6a07d240c87/exam... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2533/reactions",
"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/2533/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2532 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2532/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2532/comments | https://api.github.com/repos/huggingface/transformers/issues/2532/events | https://github.com/huggingface/transformers/pull/2532 | 550,221,008 | MDExOlB1bGxSZXF1ZXN0MzYzMTY1ODEx | 2,532 | Automatic testing of examples 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/2532?src=pr&el=h1) Report\n> Merging [#2532](https://codecov.io/gh/huggingface/transformers/pull/2532?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/cefd51c50cc08be8146c1151544495968ce8f2ad?src=pr&el=desc) will **i... | 1,579 | 1,579 | 1,579 | MEMBER | null | Adds a test that tests the examples in the documentation.
Addsa "Glossary" page for recurring arguments.
Updates the documentation of pytorch & tensorflow models.
Models done:
- [x] ALBERT
- [x] BERT
- [x] GPT-2
- [x] GPT
- [x] Transformer XL
- [x] XLNet
- [x] XLM
- [x] CamemBERT
- [x] RoBERTa
- [x] Dist... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2532/reactions",
"total_count": 1,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 1,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/2532/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/2532",
"html_url": "https://github.com/huggingface/transformers/pull/2532",
"diff_url": "https://github.com/huggingface/transformers/pull/2532.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/2532.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/2531 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2531/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2531/comments | https://api.github.com/repos/huggingface/transformers/issues/2531/events | https://github.com/huggingface/transformers/pull/2531 | 550,120,560 | MDExOlB1bGxSZXF1ZXN0MzYzMDgyOTgx | 2,531 | Serving improvements | {
"login": "mfuntowicz",
"id": 2241520,
"node_id": "MDQ6VXNlcjIyNDE1MjA=",
"avatar_url": "https://avatars.githubusercontent.com/u/2241520?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/mfuntowicz",
"html_url": "https://github.com/mfuntowicz",
"followers_url": "https://api.github.com/users... | [] | closed | false | null | [] | [
"> This PR brings some improvements over the CLI serving command.\r\n> \r\n> Changes:\r\n> \r\n> * Expose the possibility to change the number of underlying FastAPI workers.\r\n> * Make forward() async so it doesn't timeout in the middle a requests.\r\n> * Fixed USE_TF, USE_TORCH env vars fighting each other.\r\n\r... | 1,579 | 1,651 | 1,579 | MEMBER | null | This PR brings some improvements over the CLI serving command.
Changes:
- Expose the possibility to change the number of underlying FastAPI workers.
- Make forward() async so it doesn't timeout in the middle a requests.
- Fixed USE_TF, USE_TORCH env vars fighting each other. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2531/reactions",
"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/2531/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/2531",
"html_url": "https://github.com/huggingface/transformers/pull/2531",
"diff_url": "https://github.com/huggingface/transformers/pull/2531.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/2531.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/2530 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2530/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2530/comments | https://api.github.com/repos/huggingface/transformers/issues/2530/events | https://github.com/huggingface/transformers/issues/2530 | 550,054,948 | MDU6SXNzdWU1NTAwNTQ5NDg= | 2,530 | SentencePiece Error with AlbertTokenizer using google pretrained chinese model | {
"login": "ubuntu733",
"id": 5701884,
"node_id": "MDQ6VXNlcjU3MDE4ODQ=",
"avatar_url": "https://avatars.githubusercontent.com/u/5701884?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ubuntu733",
"html_url": "https://github.com/ubuntu733",
"followers_url": "https://api.github.com/users/ub... | [
{
"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, this implementation of ALBERT only supports SentencePiece as its tokenizer.",
"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"
] | 1,579 | 1,584 | 1,584 | NONE | null | ## π Bug
<!-- Important information -->
Model I am using albert:
Language I am using the model on , Chinese:
The problem arise when using:
* [ ] the official example scripts: (give details)
`AlbertTokenizer.from_pretrained(vocab)`
It shows:
> Traceback (most recent call last):
File "/home/shench... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2530/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/2530/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2529 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2529/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2529/comments | https://api.github.com/repos/huggingface/transformers/issues/2529/events | https://github.com/huggingface/transformers/issues/2529 | 550,043,911 | MDU6SXNzdWU1NTAwNDM5MTE= | 2,529 | Updating the issue template, directing general question to SO | {
"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... | [
{
"id": 1260952223,
"node_id": "MDU6TGFiZWwxMjYwOTUyMjIz",
"url": "https://api.github.com/repos/huggingface/transformers/labels/Discussion",
"name": "Discussion",
"color": "22870e",
"default": false,
"description": "Discussion on a topic (keep it focused or open a new issue though)"
},... | closed | false | null | [] | [
"Hi Bram, first of all we want to reiterate our appreciation for what you've been doing β the community is very lucky to have you. \r\n\r\nYou raise some good points. Would you like to update the issue templates, updating what needs to be updated + linking to Stack Overflow for support requests?\r\n\r\nIn the longe... | 1,579 | 1,606 | 1,585 | COLLABORATOR | null | ## π Feature
In the last couple of months, `transformers` has seen an exponential increase in interest; you have exceeded 20k stars, congrats! @thomwolf wrote a blog post on how to open-source your code for a larger audience, but as expected, a side-effect is that you'll get more issues and more pull requests that ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2529/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/2529/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2528 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2528/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2528/comments | https://api.github.com/repos/huggingface/transformers/issues/2528/events | https://github.com/huggingface/transformers/issues/2528 | 550,040,692 | MDU6SXNzdWU1NTAwNDA2OTI= | 2,528 | [Question] Add extra sublayer for each layer of Transformer | {
"login": "robinsongh381",
"id": 42966248,
"node_id": "MDQ6VXNlcjQyOTY2MjQ4",
"avatar_url": "https://avatars.githubusercontent.com/u/42966248?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/robinsongh381",
"html_url": "https://github.com/robinsongh381",
"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 | [] | [
"Anything is possible, if you want to! But it's not straightforward, I think. You can have a look at `BertLayer` where I would assume that you make your changes.\r\n\r\nhttps://github.com/huggingface/transformers/blob/dfe012ad9d6b6f0c9d30bc508b9f1e4c42280c07/src/transformers/modeling_bert.py#L365-L373",
"This iss... | 1,579 | 1,584 | 1,584 | NONE | null | ## β Questions & Help
Hello !
BERT base has 12 layers and each layer includes follwing sublayer ( and of course, add norm)
` {self-attention -> feed-foward} `
I was wondering if there is a way of adding extra unit to this sublayer , for example,
`{self-attention -> feed-foward -> **LSTM**}`
Thanks
| {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2528/reactions",
"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/2528/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2527 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2527/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2527/comments | https://api.github.com/repos/huggingface/transformers/issues/2527/events | https://github.com/huggingface/transformers/issues/2527 | 550,020,997 | MDU6SXNzdWU1NTAwMjA5OTc= | 2,527 | How to get the output in other layers from Bert? | {
"login": "zy4bvb",
"id": 50108711,
"node_id": "MDQ6VXNlcjUwMTA4NzEx",
"avatar_url": "https://avatars.githubusercontent.com/u/50108711?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/zy4bvb",
"html_url": "https://github.com/zy4bvb",
"followers_url": "https://api.github.com/users/zy4bvb/fo... | [] | closed | false | null | [] | [
"Have a look at the [documentation](https://huggingface.co/transformers/model_doc/bert.html#bertmodel), particularly at the point about 'outputs'. You'll see that when you use `output_hidden_states=True`, you'll get _all_ outputs back, like so:\r\n\r\n```python\r\nmodel = BertModel.from_pretrained('bert-base-uncas... | 1,579 | 1,579 | 1,579 | NONE | null | ## β Questions & Help
I want to analyze the information that every bert layers contains. But i found the BertModel only output the sentence embbeding and the CLS embbeding.
<!-- A clear and concise description of the question. -->
| {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2527/reactions",
"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/2527/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2526 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2526/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2526/comments | https://api.github.com/repos/huggingface/transformers/issues/2526/events | https://github.com/huggingface/transformers/pull/2526 | 549,946,278 | MDExOlB1bGxSZXF1ZXN0MzYyOTQwOTcx | 2,526 | modified method simple_accuracy(), before:(preds == labels).mean() This⦠| {
"login": "7will10",
"id": 31179081,
"node_id": "MDQ6VXNlcjMxMTc5MDgx",
"avatar_url": "https://avatars.githubusercontent.com/u/31179081?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/7will10",
"html_url": "https://github.com/7will10",
"followers_url": "https://api.github.com/users/7will1... | [] | closed | false | null | [] | [] | 1,579 | 1,579 | 1,579 | NONE | null | modified method simple_accuracy(),
before:
it's (preds == labels).mean()
This will cause an exception[AttributeError: 'bool' object has no attribute 'mean'],
then after update:
change to accuracy_score(labels,preds),
use this method accuracy_score() in package sklearn.metrics | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2526/reactions",
"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/2526/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/2526",
"html_url": "https://github.com/huggingface/transformers/pull/2526",
"diff_url": "https://github.com/huggingface/transformers/pull/2526.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/2526.patch",
"merged_at": n... |
https://api.github.com/repos/huggingface/transformers/issues/2525 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2525/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2525/comments | https://api.github.com/repos/huggingface/transformers/issues/2525/events | https://github.com/huggingface/transformers/issues/2525 | 549,909,207 | MDU6SXNzdWU1NDk5MDkyMDc= | 2,525 | Error when running demo script in T5Model | {
"login": "g-jing",
"id": 44223191,
"node_id": "MDQ6VXNlcjQ0MjIzMTkx",
"avatar_url": "https://avatars.githubusercontent.com/u/44223191?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/g-jing",
"html_url": "https://github.com/g-jing",
"followers_url": "https://api.github.com/users/g-jing/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 found the error is because the `lm_labels` is not poped out. One possible solution: change https://github.com/huggingface/transformers/blob/dfe012ad9d6b6f0c9d30bc508b9f1e4c42280c07/src/transformers/modeling_t5.py#L864 to \r\n```\r\n lm_labels = kwargs.pop('decoder_lm_labels', None)\r\n if not lm_la... | 1,579 | 1,584 | 1,584 | NONE | null | ## π Bug
<!-- Important information -->
Model I am using (Bert, XLNet....): T5
Language I am using the model on (English, Chinese....): English
The problem arise when using:
* [ ] the official example scripts: When I use the example script in https://github.com/huggingface/transformers/blob/dfe012ad9d6b6f... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2525/reactions",
"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/2525/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2524 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2524/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2524/comments | https://api.github.com/repos/huggingface/transformers/issues/2524/events | https://github.com/huggingface/transformers/issues/2524 | 549,901,641 | MDU6SXNzdWU1NDk5MDE2NDE= | 2,524 | Will you release the pre-train script for T5? | {
"login": "g-jing",
"id": 44223191,
"node_id": "MDQ6VXNlcjQ0MjIzMTkx",
"avatar_url": "https://avatars.githubusercontent.com/u/44223191?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/g-jing",
"html_url": "https://github.com/g-jing",
"followers_url": "https://api.github.com/users/g-jing/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"
] | 1,579 | 1,584 | 1,584 | NONE | null | ## β Questions & Help
<!-- A clear and concise description of the question. -->
There is a fine-tune script for BERT/GPT https://github.com/huggingface/transformers/blob/master/examples/run_lm_finetuning.py. Will you include T5 into this script? Thanks | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2524/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/2524/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2523 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2523/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2523/comments | https://api.github.com/repos/huggingface/transformers/issues/2523/events | https://github.com/huggingface/transformers/issues/2523 | 549,857,822 | MDU6SXNzdWU1NDk4NTc4MjI= | 2,523 | Tokenizer encoding functions don't support 'left' and 'right' values for `pad_to_max_length` | {
"login": "ranamihir",
"id": 8270471,
"node_id": "MDQ6VXNlcjgyNzA0NzE=",
"avatar_url": "https://avatars.githubusercontent.com/u/8270471?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ranamihir",
"html_url": "https://github.com/ranamihir",
"followers_url": "https://api.github.com/users/ra... | [] | closed | false | null | [] | [
"Hi, the documentation could definitely be improved in that regard but what the docstring means is that it will follow the class attribute `padding_side`:\r\n\r\n```py\r\n>>> from transformers import BertTokenizer\r\n>>> tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')\r\n>>> text = 'Eiffel Tower'\r\n... | 1,579 | 1,579 | 1,579 | NONE | null | ## π Bug
In the tokenizer encoding functions (`encode`, `encode_plus`, etc.), it seems `pad_to_max_length` only supports boolean values. In the [documentation](https://github.com/huggingface/transformers/blob/master/src/transformers/tokenization_utils.py#L801-L805), it's mentioned it can also be set to `left` or `r... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2523/reactions",
"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/2523/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2522 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2522/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2522/comments | https://api.github.com/repos/huggingface/transformers/issues/2522/events | https://github.com/huggingface/transformers/issues/2522 | 549,674,540 | MDU6SXNzdWU1NDk2NzQ1NDA= | 2,522 | https://s3.amazonaws.com/models.huggingface.co/xxx/pytorch_model.bin failed or can not open at xxx/.cache/xxxxxxxxxx | {
"login": "578123043",
"id": 16147509,
"node_id": "MDQ6VXNlcjE2MTQ3NTA5",
"avatar_url": "https://avatars.githubusercontent.com/u/16147509?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/578123043",
"html_url": "https://github.com/578123043",
"followers_url": "https://api.github.com/users/... | [] | closed | false | null | [] | [
"The correct URL is `https://s3.amazonaws.com/models.huggingface.co/bert/roberta-base-pytorch_model.bin`\r\n\r\nDid you see the url above somewhere?",
"\r\nTo resoleved this , adding `config_class pretrai... | 1,579 | 1,579 | 1,579 | NONE | null | https://s3.amazonaws.com/models.huggingface.co/bert/roberta-base/pytorch_model.bin
Could you access on https://s3.amazonaws.com/models.huggingface.co/bert/roberta-base/pytorch_model.bin?
| {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2522/reactions",
"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/2522/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2521 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2521/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2521/comments | https://api.github.com/repos/huggingface/transformers/issues/2521/events | https://github.com/huggingface/transformers/pull/2521 | 549,632,026 | MDExOlB1bGxSZXF1ZXN0MzYyNjg3NTEw | 2,521 | Bias should be resized with the weights | {
"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/2521?src=pr&el=h1) Report\n> Merging [#2521](https://codecov.io/gh/huggingface/transformers/pull/2521?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/6c32d8bb95aa81de6a047cca5ae732b93b9db020?src=pr&el=desc) will **i... | 1,579 | 1,579 | 1,579 | MEMBER | null | Created a link between the linear layer bias and the model attribute bias. This does not change anything for the user nor for the conversion scripts, but allows the `resize_token_embeddings` method to resize the bias as well as the weights of the decoder.
Added a test. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2521/reactions",
"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/2521/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/2521",
"html_url": "https://github.com/huggingface/transformers/pull/2521",
"diff_url": "https://github.com/huggingface/transformers/pull/2521.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/2521.patch",
"merged_at": 1... |
https://api.github.com/repos/huggingface/transformers/issues/2520 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2520/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2520/comments | https://api.github.com/repos/huggingface/transformers/issues/2520/events | https://github.com/huggingface/transformers/issues/2520 | 549,598,736 | MDU6SXNzdWU1NDk1OTg3MzY= | 2,520 | Descriptions of shared models and interaction with contributors | {
"login": "chinisan",
"id": 59874749,
"node_id": "MDQ6VXNlcjU5ODc0NzQ5",
"avatar_url": "https://avatars.githubusercontent.com/u/59874749?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/chinisan",
"html_url": "https://github.com/chinisan",
"followers_url": "https://api.github.com/users/chi... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"In a sense, this is related to the discussion that we had over at https://github.com/huggingface/transformers/pull/2281#issuecomment-570574944. The [answer](https://github.com/huggingface/transformers/pull/2281#issuecomment-571418343) by @julien-c was that they are aware of the difficulties and sensitivities that ... | 1,579 | 1,585 | 1,585 | NONE | null | ## π Feature
It would be nice to have more room for descriptions from the contributors of shared models since at the moment one can only guess from the title as to what the model does and what was improved from existing models.
Additionally, ways of interactions with the contributors such as comments and upvote... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2520/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/2520/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2519 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2519/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2519/comments | https://api.github.com/repos/huggingface/transformers/issues/2519/events | https://github.com/huggingface/transformers/issues/2519 | 549,585,017 | MDU6SXNzdWU1NDk1ODUwMTc= | 2,519 | Does calling fit() method on TFBertForSequenceClassification change the weights of internal pre-trained bert? | {
"login": "hamidreza-ghader",
"id": 8751979,
"node_id": "MDQ6VXNlcjg3NTE5Nzk=",
"avatar_url": "https://avatars.githubusercontent.com/u/8751979?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/hamidreza-ghader",
"html_url": "https://github.com/hamidreza-ghader",
"followers_url": "https://ap... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"I can't tell for sure about the TF version, but I would assume it's the same as the one in PyTorch, in which case yes: all weights are changed. You can freeze layers, though. Here (for PyTorch nn.Module) only freezing the embeddings:\r\n\r\n```python\r\nbert = BertModel.from_pretrained('bert-base-uncased')\r\nfor ... | 1,579 | 1,584 | 1,584 | NONE | null | Hi all,
Let's say, I have a TFBertForSequenceClassification object and I call fit method on it. Does it change the weights of the internal TFBertMainLayer too or it only trains the weights of the Dropout and the classifier Dense layers?
Best | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2519/reactions",
"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/2519/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2518 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2518/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2518/comments | https://api.github.com/repos/huggingface/transformers/issues/2518/events | https://github.com/huggingface/transformers/issues/2518 | 549,493,220 | MDU6SXNzdWU1NDk0OTMyMjA= | 2,518 | Type of Training file needed for finetuning | {
"login": "ankush20m",
"id": 45195876,
"node_id": "MDQ6VXNlcjQ1MTk1ODc2",
"avatar_url": "https://avatars.githubusercontent.com/u/45195876?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ankush20m",
"html_url": "https://github.com/ankush20m",
"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, currently the `run_lm_finetuning` script does not take into account the line returns to split the data. It splits the data according to the maximum length the model will allow (which is 512 tokens for BERT), as it is generally used to fine-tune a model on a lengthy text corpus.\r\n\r\nIf you want to do a line ... | 1,578 | 1,585 | 1,585 | NONE | null | ## β Questions & Help
<!-- A clear and concise description of the question. -->
What kind of **training text** file needed for `run_lm_finetuning.py` script. I have created the **text file** in which I have put **line by line sentences**.
Is this format correct for finetuning?
Because I want to finetune the **BER... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2518/reactions",
"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/2518/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2517 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2517/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2517/comments | https://api.github.com/repos/huggingface/transformers/issues/2517/events | https://github.com/huggingface/transformers/issues/2517 | 549,460,271 | MDU6SXNzdWU1NDk0NjAyNzE= | 2,517 | Save only Bert Model after training a Sequence Classification Task/ LM finetuning Task. | {
"login": "rahulbaburaj",
"id": 8917417,
"node_id": "MDQ6VXNlcjg5MTc0MTc=",
"avatar_url": "https://avatars.githubusercontent.com/u/8917417?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/rahulbaburaj",
"html_url": "https://github.com/rahulbaburaj",
"followers_url": "https://api.github.com... | [] | closed | false | null | [] | [
"By saving only BERT, do you mean saving only the transformer and not the classification layer as well?",
"@rahulbaburaj you can use the code snippet below, change the 'bert-base-uncased' to your fine-tuned model directory.\r\n\r\n```python\r\n# load config\r\nconf = BertConfig.from_pretrained('bert-base-uncased... | 1,578 | 1,579 | 1,579 | NONE | null | ## β Questions & Help
<!-- A clear and concise description of the question. -->
1) How do I save only the BERT Model after finetuning on a Sequence Classification Task/ LM finetuning Task
2) How to load only BERT Model from a saved model trained on Sequence Classification Task/ LM finetuning Task | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2517/reactions",
"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/2517/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2516 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2516/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2516/comments | https://api.github.com/repos/huggingface/transformers/issues/2516/events | https://github.com/huggingface/transformers/pull/2516 | 549,304,736 | MDExOlB1bGxSZXF1ZXN0MzYyNDIxOTQ5 | 2,516 | update | {
"login": "hjc3613",
"id": 37894838,
"node_id": "MDQ6VXNlcjM3ODk0ODM4",
"avatar_url": "https://avatars.githubusercontent.com/u/37894838?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/hjc3613",
"html_url": "https://github.com/hjc3613",
"followers_url": "https://api.github.com/users/hjc361... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/transformers/pull/2516?src=pr&el=h1) Report\n> Merging [#2516](https://codecov.io/gh/huggingface/transformers/pull/2516?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/transformers/commit/51d2683fdcffc03f79dcbdc373628d449d1a0385?src=pr&el=desc) will **d... | 1,578 | 1,579 | 1,579 | NONE | null | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2516/reactions",
"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/2516/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/2516",
"html_url": "https://github.com/huggingface/transformers/pull/2516",
"diff_url": "https://github.com/huggingface/transformers/pull/2516.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/2516.patch",
"merged_at": n... | |
https://api.github.com/repos/huggingface/transformers/issues/2515 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2515/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2515/comments | https://api.github.com/repos/huggingface/transformers/issues/2515/events | https://github.com/huggingface/transformers/issues/2515 | 549,300,414 | MDU6SXNzdWU1NDkzMDA0MTQ= | 2,515 | How to use transformers to convert batch sentences into word vectors??? | {
"login": "duan348733684",
"id": 26431015,
"node_id": "MDQ6VXNlcjI2NDMxMDE1",
"avatar_url": "https://avatars.githubusercontent.com/u/26431015?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/duan348733684",
"html_url": "https://github.com/duan348733684",
"followers_url": "https://api.githu... | [] | closed | false | null | [] | [
"I'm sorry but this is not how you should ask questions to begin with, second it is very general. There are tons of tutorials about this kind of stuff. You can have a look at a notebook that I made. It shows you how to get a feature vector for your input sentence. https://github.com/BramVanroy/bert-for-inference/bl... | 1,578 | 1,579 | 1,579 | NONE | null | ## β Questions & Help
<!-- A clear and concise description of the question. -->
How to use transformers to convert batch sentences into word vectors???
| {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2515/reactions",
"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/2515/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2514 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2514/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2514/comments | https://api.github.com/repos/huggingface/transformers/issues/2514/events | https://github.com/huggingface/transformers/issues/2514 | 549,127,868 | MDU6SXNzdWU1NDkxMjc4Njg= | 2,514 | T5 Masked LM -- pre-trained model import? | {
"login": "moscow25",
"id": 1473764,
"node_id": "MDQ6VXNlcjE0NzM3NjQ=",
"avatar_url": "https://avatars.githubusercontent.com/u/1473764?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/moscow25",
"html_url": "https://github.com/moscow25",
"followers_url": "https://api.github.com/users/mosco... | [
{
"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 found a solution https://github.com/huggingface/transformers/issues/3985#issue-606998741"
] | 1,578 | 1,665 | 1,584 | CONTRIBUTOR | null | ## β Questions & Help
Hi, thanks for merging the T5 model! However it is not clear to me how to use the pretrained model for masked language modeling. It appears that the model example only returns a hidden state, or `T5WithLMHeadModel` which is not clear what this is doing -- it tends to return the same token for m... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2514/reactions",
"total_count": 8,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 8
} | https://api.github.com/repos/huggingface/transformers/issues/2514/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/2513 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/2513/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/2513/comments | https://api.github.com/repos/huggingface/transformers/issues/2513/events | https://github.com/huggingface/transformers/issues/2513 | 549,061,548 | MDU6SXNzdWU1NDkwNjE1NDg= | 2,513 | Error in AlbertForMaskedLM with add_tokens and model.resize_token_embeddings | {
"login": "mckunkel",
"id": 9967035,
"node_id": "MDQ6VXNlcjk5NjcwMzU=",
"avatar_url": "https://avatars.githubusercontent.com/u/9967035?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/mckunkel",
"html_url": "https://github.com/mckunkel",
"followers_url": "https://api.github.com/users/mckun... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | {
"login": "LysandreJik",
"id": 30755778,
"node_id": "MDQ6VXNlcjMwNzU1Nzc4",
"avatar_url": "https://avatars.githubusercontent.com/u/30755778?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/LysandreJik",
"html_url": "https://github.com/LysandreJik",
"followers_url": "https://api.github.com/... | [
{
"login": "LysandreJik",
"id": 30755778,
"node_id": "MDQ6VXNlcjMwNzU1Nzc4",
"avatar_url": "https://avatars.githubusercontent.com/u/30755778?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/LysandreJik",
"html_url": "https://github.com/LysandreJik",
"followers_url": "htt... | [
"Hi, I've pushed a fix that was just merged in `master`. Could you please try and install from source:\r\n```py\r\npip install git+https://github.com/huggingface/transformers\r\n```\r\nand tell me if you face the same error?",
"Greetings, \r\nThanks for the reply. \r\nI do not get the same error anymore, I get a ... | 1,578 | 1,585 | 1,585 | NONE | null | ## π Bug
Model I am using: Albert & Bert
Language I am using the model on English
The problem arise when using:
* [X] the official example scripts: run_lm_finetuning.py
The tasks I am working on is:
* [X] an official GLUE/SQUaD task: mlm
## To Reproduce
Steps to reproduce the behavior:
1. add ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/2513/reactions",
"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/2513/timeline | completed | null | null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.