url stringlengths 62 66 | repository_url stringclasses 1
value | labels_url stringlengths 76 80 | comments_url stringlengths 71 75 | events_url stringlengths 69 73 | html_url stringlengths 50 56 | id int64 377M 2.15B | node_id stringlengths 18 32 | number int64 1 29.2k | title stringlengths 1 487 | user dict | labels list | state stringclasses 2
values | locked bool 2
classes | assignee dict | assignees list | comments list | created_at int64 1.54k 1.71k | updated_at int64 1.54k 1.71k | closed_at int64 1.54k 1.71k ⌀ | author_association stringclasses 4
values | active_lock_reason stringclasses 2
values | body stringlengths 0 234k ⌀ | reactions dict | timeline_url stringlengths 71 75 | state_reason stringclasses 3
values | draft bool 2
classes | pull_request dict |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
https://api.github.com/repos/huggingface/transformers/issues/809 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/809/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/809/comments | https://api.github.com/repos/huggingface/transformers/issues/809/events | https://github.com/huggingface/transformers/issues/809 | 469,429,682 | MDU6SXNzdWU0Njk0Mjk2ODI= | 809 | Problem loading finetuned XLNet model | {
"login": "igormis",
"id": 6599037,
"node_id": "MDQ6VXNlcjY1OTkwMzc=",
"avatar_url": "https://avatars.githubusercontent.com/u/6599037?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/igormis",
"html_url": "https://github.com/igormis",
"followers_url": "https://api.github.com/users/igormis/... | [] | closed | false | null | [] | [
"What task did you fine-tuned it on?\r\nYou can convert it by running the `convert_xlnet_checkpoint_to_pytorch.py` script with a `--finetuning_task` argument (see [here](https://github.com/huggingface/pytorch-transformers/blob/master/pytorch_transformers/convert_xlnet_checkpoint_to_pytorch.py#L94-L97))\r\n",
"bin... | 1,563 | 1,564 | 1,564 | NONE | null | After fine-tuning an XLNet classification model and obtaining TF checkpoints I converted the checkpoint to pytorch_model.bin and config.json. I need to make prediction on input text, but I have problems loading the models correctly. Any help? | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/809/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/809/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/808 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/808/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/808/comments | https://api.github.com/repos/huggingface/transformers/issues/808/events | https://github.com/huggingface/transformers/issues/808 | 469,380,474 | MDU6SXNzdWU0NjkzODA0NzQ= | 808 | GPT2 model does not have attention mask | {
"login": "Saner3",
"id": 30628796,
"node_id": "MDQ6VXNlcjMwNjI4Nzk2",
"avatar_url": "https://avatars.githubusercontent.com/u/30628796?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Saner3",
"html_url": "https://github.com/Saner3",
"followers_url": "https://api.github.com/users/Saner3/fo... | [] | closed | false | null | [] | [
"Indeed, I will remove this doctring, there is no attention_mask on GPT-2.",
"> Indeed, I will remove this doctring, there is no attention_mask on GPT-2.\r\n\r\nBut what to do if I do want to avoid computing attention on the paddings in the input sequences.",
"@Saner3 @thomwolf I have same question? don't we ne... | 1,563 | 1,566 | 1,566 | NONE | null | Hello, in the doc string of GPT2 model, it says there is an optional input called [attention_mask](https://github.com/huggingface/pytorch-transformers/blob/f289e6cfe46885f260e4f2b3c8a164aa1a567e4c/pytorch_transformers/modeling_gpt2.py#L405) to avoid computing attention on paddings. But actually I cannot find the implem... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/808/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/808/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/807 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/807/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/807/comments | https://api.github.com/repos/huggingface/transformers/issues/807/events | https://github.com/huggingface/transformers/issues/807 | 469,360,451 | MDU6SXNzdWU0NjkzNjA0NTE= | 807 | AttributeError: 'tuple' object has no attribute 'softmax' | {
"login": "Raghavendra15",
"id": 7957331,
"node_id": "MDQ6VXNlcjc5NTczMzE=",
"avatar_url": "https://avatars.githubusercontent.com/u/7957331?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Raghavendra15",
"html_url": "https://github.com/Raghavendra15",
"followers_url": "https://api.github.... | [] | closed | false | null | [] | [
"HI,\r\nI have the same problem!\r\nWhat was the solution here?",
"Me too.",
"Need more information like version of python/pytorch/transformers (all the information requested in the issue templates actually)",
"> Need more information like version of python/pytorch/transformers (all the information requested ... | 1,563 | 1,576 | 1,563 | NONE | null | I get the following error when I use the pytorch transformers, It used to work just fine in the previous pretrained-bert,
Original code: https://github.com/ceshine/pytorch-pretrained-BERT/blob/master/notebooks/Next%20Sentence%20Prediction.ipynb
Code which has error:
model.eval()
res = []
mb = progress_bar(ev... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/807/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/807/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/806 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/806/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/806/comments | https://api.github.com/repos/huggingface/transformers/issues/806/events | https://github.com/huggingface/transformers/pull/806 | 469,311,770 | MDExOlB1bGxSZXF1ZXN0Mjk4NTY5ODA2 | 806 | Fix a path so that a test can run on Windows | {
"login": "wschin",
"id": 3524474,
"node_id": "MDQ6VXNlcjM1MjQ0NzQ=",
"avatar_url": "https://avatars.githubusercontent.com/u/3524474?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/wschin",
"html_url": "https://github.com/wschin",
"followers_url": "https://api.github.com/users/wschin/foll... | [] | closed | false | null | [] | [
"Ok for that, thanks @wschin"
] | 1,563 | 1,566 | 1,566 | CONTRIBUTOR | null | The path for a temporal file is hard coded, so the test fails on Windows. This PR changes that line to a more platform-natural path. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/806/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/806/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/806",
"html_url": "https://github.com/huggingface/transformers/pull/806",
"diff_url": "https://github.com/huggingface/transformers/pull/806.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/806.patch",
"merged_at": 15663... |
https://api.github.com/repos/huggingface/transformers/issues/805 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/805/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/805/comments | https://api.github.com/repos/huggingface/transformers/issues/805/events | https://github.com/huggingface/transformers/issues/805 | 469,270,852 | MDU6SXNzdWU0NjkyNzA4NTI= | 805 | Where is "run_bert_classifier.py"? | {
"login": "amirj",
"id": 1645137,
"node_id": "MDQ6VXNlcjE2NDUxMzc=",
"avatar_url": "https://avatars.githubusercontent.com/u/1645137?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/amirj",
"html_url": "https://github.com/amirj",
"followers_url": "https://api.github.com/users/amirj/follower... | [] | closed | false | null | [] | [
"It's now `run_glue.py`",
"Hi @thomwolf \r\nDoc needs changes from run_bert_classifier to run_glue\r\nhttps://huggingface.co/pytorch-transformers/examples.html",
"Hey, just another headsup @thomwolf \r\nThis Doc also needs changing for the run_bert_classifier.py:\r\nhttps://huggingface.co/transformers/v1.1.0/ex... | 1,563 | 1,606 | 1,563 | NONE | null | Thanks for this great repo.
Is there any equivalent to [the previous run_bert_classifier.py](https://github.com/huggingface/pytorch-pretrained-BERT/tree/master/examples/run_bert_classifier.py)?
| {
"url": "https://api.github.com/repos/huggingface/transformers/issues/805/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/805/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/804 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/804/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/804/comments | https://api.github.com/repos/huggingface/transformers/issues/804/events | https://github.com/huggingface/transformers/issues/804 | 469,195,603 | MDU6SXNzdWU0NjkxOTU2MDM= | 804 | Answers to Bullet/List Items by bert | {
"login": "ghost",
"id": 10137,
"node_id": "MDQ6VXNlcjEwMTM3",
"avatar_url": "https://avatars.githubusercontent.com/u/10137?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ghost",
"html_url": "https://github.com/ghost",
"followers_url": "https://api.github.com/users/ghost/followers",
"f... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.\n"
] | 1,563 | 1,569 | 1,569 | NONE | null | Hi,
There are lists in document (Bullet items), and I am running BERT (non trained as well as squad trained). But seems BERT does not understands Bullets/lines starting with number or star.
Will any text preprocessing help? | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/804/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/804/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/803 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/803/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/803/comments | https://api.github.com/repos/huggingface/transformers/issues/803/events | https://github.com/huggingface/transformers/issues/803 | 469,135,790 | MDU6SXNzdWU0NjkxMzU3OTA= | 803 | AssertionError in BERT-Quickstart example | {
"login": "marcalt94",
"id": 44497700,
"node_id": "MDQ6VXNlcjQ0NDk3NzAw",
"avatar_url": "https://avatars.githubusercontent.com/u/44497700?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/marcalt94",
"html_url": "https://github.com/marcalt94",
"followers_url": "https://api.github.com/users/... | [] | closed | false | null | [] | [
"What is the full line of the assertion you are testing?\r\nAnd what is your input text?",
"I think the mistake was with me, sorry"
] | 1,563 | 1,563 | 1,563 | NONE | null | Hey
I tried running the Quickstart example with my own little text. Everything works fine until I get to the ```assert tokenized_text ==... ``` part. When I try to enter my text instead of the Jim Henson text, I get the following error message: ```Traceback (most recent call last):
File "<stdin>", line 1, in <mod... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/803/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/803/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/802 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/802/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/802/comments | https://api.github.com/repos/huggingface/transformers/issues/802/events | https://github.com/huggingface/transformers/issues/802 | 469,086,281 | MDU6SXNzdWU0NjkwODYyODE= | 802 | fp16+xlnet did not gain any speed increase | {
"login": "fyubang",
"id": 25549892,
"node_id": "MDQ6VXNlcjI1NTQ5ODky",
"avatar_url": "https://avatars.githubusercontent.com/u/25549892?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/fyubang",
"html_url": "https://github.com/fyubang",
"followers_url": "https://api.github.com/users/fyuban... | [
{
"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 makes heavy use of `torch.einsum()` but I'm not sure this method is fp16 compatible.\r\nIt's also quite slow currently so maybe in the mid/long-term it would be good to change these einsum to standard matmul. I won't have time to do that very soon though.",
"As as a suggestion, you can add ```apex.amp.regi... | 1,563 | 1,656 | 1,574 | NONE | null | Hi,
I tried fp16 + xlnet, it did not work.
when I set opt_level='O2', the memory was half, but it was much slower than fp32.
when I set opt_level='O1', the memory was original, and it has similar speed with fp32.
Environment: v100, cuda, 10.1, torch 1.1
The environment is ok, because I tried bert + fp16 and it... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/802/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/802/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/801 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/801/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/801/comments | https://api.github.com/repos/huggingface/transformers/issues/801/events | https://github.com/huggingface/transformers/pull/801 | 469,077,120 | MDExOlB1bGxSZXF1ZXN0Mjk4Mzc1OTk1 | 801 | import sys twice | {
"login": "bzantium",
"id": 19511788,
"node_id": "MDQ6VXNlcjE5NTExNzg4",
"avatar_url": "https://avatars.githubusercontent.com/u/19511788?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/bzantium",
"html_url": "https://github.com/bzantium",
"followers_url": "https://api.github.com/users/bza... | [] | closed | false | null | [] | [
"Thanks!"
] | 1,563 | 1,563 | 1,563 | CONTRIBUTOR | null | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/801/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/801/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/801",
"html_url": "https://github.com/huggingface/transformers/pull/801",
"diff_url": "https://github.com/huggingface/transformers/pull/801.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/801.patch",
"merged_at": 15633... | |
https://api.github.com/repos/huggingface/transformers/issues/800 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/800/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/800/comments | https://api.github.com/repos/huggingface/transformers/issues/800/events | https://github.com/huggingface/transformers/issues/800 | 469,011,508 | MDU6SXNzdWU0NjkwMTE1MDg= | 800 | attention_mask at run_squad.py | {
"login": "seanie12",
"id": 19561061,
"node_id": "MDQ6VXNlcjE5NTYxMDYx",
"avatar_url": "https://avatars.githubusercontent.com/u/19561061?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/seanie12",
"html_url": "https://github.com/seanie12",
"followers_url": "https://api.github.com/users/sea... | [] | closed | false | null | [] | [
"I have same issue. ",
"Thanks @seanie12!"
] | 1,563 | 1,563 | 1,563 | NONE | null | I think there's minor mistake in [run_squad.py](https://github.com/huggingface/pytorch-transformers/blob/5fe0b378d8/examples/run_squad.py#L298) at line 298
```
inputs = {'input_ids': batch[0],
'token_type_ids': None if args.model_type == 'xlm' else batch[1],
'attention_mask': batch[2],
'start_positions': bat... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/800/reactions",
"total_count": 1,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 1
} | https://api.github.com/repos/huggingface/transformers/issues/800/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/799 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/799/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/799/comments | https://api.github.com/repos/huggingface/transformers/issues/799/events | https://github.com/huggingface/transformers/issues/799 | 469,008,302 | MDU6SXNzdWU0NjkwMDgzMDI= | 799 | Error while adding new tokens to GPT2 tokenizer | {
"login": "ZHAOTING",
"id": 5592709,
"node_id": "MDQ6VXNlcjU1OTI3MDk=",
"avatar_url": "https://avatars.githubusercontent.com/u/5592709?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ZHAOTING",
"html_url": "https://github.com/ZHAOTING",
"followers_url": "https://api.github.com/users/ZHAOT... | [] | closed | false | null | [] | [
"Added:\r\n\r\nI also found that `_convert_token_to_id()` in tokenization_gpt2.py (line 182) uses `unk_token`, which is initially `None` in GPT2 tokenizer. This line of code can also lead to bugs.\r\n~~~~\r\ndef _convert_token_to_id(self, token):\r\n \"\"\" Converts a token (str/unicode) in an id using the vocab... | 1,563 | 1,606 | 1,565 | CONTRIBUTOR | null | A **NoneType Error** is encountered when I call `add_tokens()` to add new tokens to **GPT2 tokenizer** and the error is as following:
~~~~
File ".../pytorch_transformers/tokenization_utils.py", line 311, in add_tokens
if self.convert_tokens_to_ids(token) == self.convert_tokens_to_ids(self.unk_token):
File "..... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/799/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/799/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/798 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/798/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/798/comments | https://api.github.com/repos/huggingface/transformers/issues/798/events | https://github.com/huggingface/transformers/issues/798 | 469,001,218 | MDU6SXNzdWU0NjkwMDEyMTg= | 798 | [bug]BertAdam change to AdamW in example | {
"login": "shibing624",
"id": 10249622,
"node_id": "MDQ6VXNlcjEwMjQ5NjIy",
"avatar_url": "https://avatars.githubusercontent.com/u/10249622?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/shibing624",
"html_url": "https://github.com/shibing624",
"followers_url": "https://api.github.com/use... | [] | closed | false | null | [] | [
"Changing that line still causes an error in line 568, BertAdam has to be changed to AdamW as well and the warmup kwarg has to be removed.",
"#797 (specifically d6522e28732fd14a926440ef5f315e6a8e13792c) ",
"have fix the error! I tested it on toy dataset.",
"@shibing624 this bug can be closed"
] | 1,563 | 1,563 | 1,563 | NONE | null | https://github.com/huggingface/pytorch-transformers/blob/master/examples/lm_finetuning/simple_lm_finetuning.py#L35 BertAdam change to AdamW | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/798/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/798/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/797 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/797/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/797/comments | https://api.github.com/repos/huggingface/transformers/issues/797/events | https://github.com/huggingface/transformers/pull/797 | 468,939,227 | MDExOlB1bGxSZXF1ZXN0Mjk4MjcwMzY0 | 797 | fix some errors for distributed lm_finetuning | {
"login": "yzy5630",
"id": 9417680,
"node_id": "MDQ6VXNlcjk0MTc2ODA=",
"avatar_url": "https://avatars.githubusercontent.com/u/9417680?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/yzy5630",
"html_url": "https://github.com/yzy5630",
"followers_url": "https://api.github.com/users/yzy5630/... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/pytorch-transformers/pull/797?src=pr&el=h1) Report\n> Merging [#797](https://codecov.io/gh/huggingface/pytorch-transformers/pull/797?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/pytorch-transformers/commit/5fe0b378d899f81eb0a7f2db0c4eb0234748e915?src... | 1,563 | 1,563 | 1,563 | CONTRIBUTOR | null | 1. makedirs
2. save models | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/797/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/797/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/797",
"html_url": "https://github.com/huggingface/transformers/pull/797",
"diff_url": "https://github.com/huggingface/transformers/pull/797.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/797.patch",
"merged_at": 15634... |
https://api.github.com/repos/huggingface/transformers/issues/796 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/796/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/796/comments | https://api.github.com/repos/huggingface/transformers/issues/796/events | https://github.com/huggingface/transformers/pull/796 | 468,879,150 | MDExOlB1bGxSZXF1ZXN0Mjk4MjIzNzQz | 796 | Minor documentation updates | {
"login": "stefan-it",
"id": 20651387,
"node_id": "MDQ6VXNlcjIwNjUxMzg3",
"avatar_url": "https://avatars.githubusercontent.com/u/20651387?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/stefan-it",
"html_url": "https://github.com/stefan-it",
"followers_url": "https://api.github.com/users/... | [] | closed | false | null | [] | [
"Thanks Stefan!"
] | 1,563 | 1,563 | 1,563 | COLLABORATOR | null | Hi,
this PR just updates some urls in the documentation :)
---
Thanks for your great work on PyTorch-Transformers 🤗 | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/796/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/796/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/796",
"html_url": "https://github.com/huggingface/transformers/pull/796",
"diff_url": "https://github.com/huggingface/transformers/pull/796.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/796.patch",
"merged_at": 15633... |
https://api.github.com/repos/huggingface/transformers/issues/795 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/795/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/795/comments | https://api.github.com/repos/huggingface/transformers/issues/795/events | https://github.com/huggingface/transformers/issues/795 | 468,878,037 | MDU6SXNzdWU0Njg4NzgwMzc= | 795 | XLNet-large-cased: hyper-parameters for fine-tuning on SST-2 | {
"login": "avostryakov",
"id": 174194,
"node_id": "MDQ6VXNlcjE3NDE5NA==",
"avatar_url": "https://avatars.githubusercontent.com/u/174194?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/avostryakov",
"html_url": "https://github.com/avostryakov",
"followers_url": "https://api.github.com/user... | [
{
"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 tried to finetune xlnet base on squad 2.0 but the numbers on dev are pretty bad\r\n`Results: {'exact': 3.0405120862461046, 'f1': 6.947601433150003, 'total': 11873, 'HasAns_exact': 6.056005398110662, 'HasAns_f1': 13.881388632893048, 'HasAns_total': 5928, 'NoAns_exact': 0.0336417157275021, 'NoAns_f1': 0.03364... | 1,563 | 1,569 | 1,569 | NONE | null | I tried to finetune XLNet on one of the classification tasks from GLUE (Ubuntu, GPU Titan RTX, CUDA 10.0, pytorch 1.1):
export GLUE_DIR=/path/to/glue
python ./examples/run_glue.py \
--model_type xlnet \
--model_name_or_path xlnet-large-cased \
--do_train \
--do_eval \
--task_name=sst-2... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/795/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/795/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/794 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/794/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/794/comments | https://api.github.com/repos/huggingface/transformers/issues/794/events | https://github.com/huggingface/transformers/pull/794 | 468,805,569 | MDExOlB1bGxSZXF1ZXN0Mjk4MTYzODY4 | 794 | Adding additional model loading functionality | {
"login": "zphang",
"id": 1668462,
"node_id": "MDQ6VXNlcjE2Njg0NjI=",
"avatar_url": "https://avatars.githubusercontent.com/u/1668462?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/zphang",
"html_url": "https://github.com/zphang",
"followers_url": "https://api.github.com/users/zphang/foll... | [] | closed | false | null | [] | [
"Hi Jason,\r\n\r\nCan you give me a little more information on the model-loading workflow you are using so I can understand the whys and wherefores of these proposed modifications?",
"Hey Thomas,\r\n\r\nSorry for the delay. My thinking is this: the `from_pretrained` method current does two things: resolve the pat... | 1,563 | 1,567 | 1,567 | CONTRIBUTOR | null | (Porting over some functionality from my old fork)
This PR adds additional methods to `PreTrainedModel` for loading models for `state_dict`s. Currently, `from_pretrained()` does a lot of the heavy lifting, but is primarily designed to load from file/folders. This adds additional options for users with different mode... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/794/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/794/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/794",
"html_url": "https://github.com/huggingface/transformers/pull/794",
"diff_url": "https://github.com/huggingface/transformers/pull/794.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/794.patch",
"merged_at": null
... |
https://api.github.com/repos/huggingface/transformers/issues/793 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/793/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/793/comments | https://api.github.com/repos/huggingface/transformers/issues/793/events | https://github.com/huggingface/transformers/issues/793 | 468,792,527 | MDU6SXNzdWU0Njg3OTI1Mjc= | 793 | BertModel docstring missing pooled_output | {
"login": "sleepinyourhat",
"id": 1284441,
"node_id": "MDQ6VXNlcjEyODQ0NDE=",
"avatar_url": "https://avatars.githubusercontent.com/u/1284441?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sleepinyourhat",
"html_url": "https://github.com/sleepinyourhat",
"followers_url": "https://api.gith... | [] | closed | false | null | [] | [
"Damned, missed that one, you are right.\r\nAdding the missing doc-string:\r\n```\r\n**pooler_output**: ``torch.FloatTensor`` of shape ``(batch_size, hidden_size)``\r\n Last layer hidden-state of the first token of the sequence (classification token)\r\n further processed by a Linear layer and a Tanh activati... | 1,563 | 1,564 | 1,563 | NONE | null | The BERT docstring describes three outputs here:
https://github.com/huggingface/pytorch-transformers/blob/master/pytorch_transformers/modeling_bert.py#L626
But none of these correspond to the pooled_output output that's added here:
https://github.com/huggingface/pytorch-transformers/blob/master/pytorch_transfo... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/793/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/793/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/792 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/792/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/792/comments | https://api.github.com/repos/huggingface/transformers/issues/792/events | https://github.com/huggingface/transformers/issues/792 | 468,790,463 | MDU6SXNzdWU0Njg3OTA0NjM= | 792 | Issue running run_transfo_xl.py | {
"login": "korymath",
"id": 178099,
"node_id": "MDQ6VXNlcjE3ODA5OQ==",
"avatar_url": "https://avatars.githubusercontent.com/u/178099?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/korymath",
"html_url": "https://github.com/korymath",
"followers_url": "https://api.github.com/users/korymat... | [] | closed | false | null | [] | [] | 1,563 | 1,563 | 1,563 | NONE | null | Run code:
```
python run_transfo_xl.py --work_dir ../log
```
Output
```
07/16/2019 18:01:46 - INFO - __main__ - device: cuda
07/16/2019 18:01:46 - INFO - pytorch_transformers.tokenization_utils - loading file https://s3.amazonaws.com/models.huggingface.co/bert/transfo-xl-wt103-vocab.bin from cache at /hom... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/792/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/792/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/791 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/791/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/791/comments | https://api.github.com/repos/huggingface/transformers/issues/791/events | https://github.com/huggingface/transformers/pull/791 | 468,738,423 | MDExOlB1bGxSZXF1ZXN0Mjk4MTA5NTg2 | 791 | RestructuredText table for pretrained models. | {
"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/pytorch-transformers/pull/791?src=pr&el=h1) Report\n> Merging [#791](https://codecov.io/gh/huggingface/pytorch-transformers/pull/791?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/pytorch-transformers/commit/b33a385091de604afb566155ec03329b84c96926?src... | 1,563 | 1,565 | 1,565 | MEMBER | null | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/791/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/791/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/791",
"html_url": "https://github.com/huggingface/transformers/pull/791",
"diff_url": "https://github.com/huggingface/transformers/pull/791.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/791.patch",
"merged_at": 15650... | |
https://api.github.com/repos/huggingface/transformers/issues/790 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/790/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/790/comments | https://api.github.com/repos/huggingface/transformers/issues/790/events | https://github.com/huggingface/transformers/issues/790 | 468,574,882 | MDU6SXNzdWU0Njg1NzQ4ODI= | 790 | XLNet Embeddings | {
"login": "kushalj001",
"id": 32245327,
"node_id": "MDQ6VXNlcjMyMjQ1MzI3",
"avatar_url": "https://avatars.githubusercontent.com/u/32245327?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/kushalj001",
"html_url": "https://github.com/kushalj001",
"followers_url": "https://api.github.com/use... | [] | closed | false | null | [] | [
"I'm currently finishing to add the documentation but just use `XLNetModel` instead of `BertModel` in the usage example with `BertModel`",
"Thanks a lot, @thomwolf for the quick reply. I'll try it out.",
"Here is an example now: https://huggingface.co/pytorch-transformers/model_doc/xlnet.html#pytorch_transforme... | 1,563 | 1,566 | 1,566 | CONTRIBUTOR | null | How can I retrieve contextual word vectors for my dataset using XLNet ?
The usage and examples in the documentation do not include any guide to use XLNet.
Thanks. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/790/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/790/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/789 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/789/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/789/comments | https://api.github.com/repos/huggingface/transformers/issues/789/events | https://github.com/huggingface/transformers/issues/789 | 468,390,083 | MDU6SXNzdWU0NjgzOTAwODM= | 789 | XLNet text generation ability : inference is slow | {
"login": "astariul",
"id": 43774355,
"node_id": "MDQ6VXNlcjQzNzc0MzU1",
"avatar_url": "https://avatars.githubusercontent.com/u/43774355?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/astariul",
"html_url": "https://github.com/astariul",
"followers_url": "https://api.github.com/users/ast... | [
{
"id": 1834059054,
"node_id": "MDU6TGFiZWwxODM0MDU5MDU0",
"url": "https://api.github.com/repos/huggingface/transformers/labels/Ex:%20Generation",
"name": "Ex: Generation",
"color": "06EFF8",
"default": false,
"description": "Natural Language Generation"
}
] | closed | false | {
"login": "patrickvonplaten",
"id": 23423619,
"node_id": "MDQ6VXNlcjIzNDIzNjE5",
"avatar_url": "https://avatars.githubusercontent.com/u/23423619?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/patrickvonplaten",
"html_url": "https://github.com/patrickvonplaten",
"followers_url": "https://... | [
{
"login": "patrickvonplaten",
"id": 23423619,
"node_id": "MDQ6VXNlcjIzNDIzNjE5",
"avatar_url": "https://avatars.githubusercontent.com/u/23423619?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/patrickvonplaten",
"html_url": "https://github.com/patrickvonplaten",
"follo... | [
"I tried it, but text quality is lowered a lot and inference time does not change at all.\r\n\r\nI simply changed `perm_mask` to be 0 over initial context and 1 over generated tokens.\r\n\r\n---\r\n\r\nInput :\r\n\r\n> In Seoul, you can do a lot of things ! For example you can\r\n\r\nGenerated text with full bidire... | 1,563 | 1,591 | 1,591 | CONTRIBUTOR | null | I compared the inference time for generating text with the given [example script](https://github.com/huggingface/pytorch-pretrained-BERT/blob/xlnet/examples/run_generation.py) between XLNet & GPT-2, on CPU.
To generate 100 tokens, XLNet takes **3m22s** while GPT-2 takes **14s**. And it grows exponentially : for 500 ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/789/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/789/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/788 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/788/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/788/comments | https://api.github.com/repos/huggingface/transformers/issues/788/events | https://github.com/huggingface/transformers/issues/788 | 468,140,681 | MDU6SXNzdWU0NjgxNDA2ODE= | 788 | bert-large config file | {
"login": "desperadoola",
"id": 30496727,
"node_id": "MDQ6VXNlcjMwNDk2NzI3",
"avatar_url": "https://avatars.githubusercontent.com/u/30496727?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/desperadoola",
"html_url": "https://github.com/desperadoola",
"followers_url": "https://api.github.c... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"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"
] | 1,563 | 1,568 | 1,568 | NONE | null | Here is the config file I download from path in modelling for bert large,
{
"attention_probs_dropout_prob": 0.1,
"directionality": "bidi",
"hidden_act": "gelu",
"hidden_dropout_prob": 0.1,
"hidden_size": 1024,
"initializer_range": 0.02,
"intermediate_size": 4096,
"max_position_embedding... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/788/reactions",
"total_count": 6,
"+1": 6,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/788/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/787 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/787/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/787/comments | https://api.github.com/repos/huggingface/transformers/issues/787/events | https://github.com/huggingface/transformers/issues/787 | 467,904,319 | MDU6SXNzdWU0Njc5MDQzMTk= | 787 | How to use Bert QA model for predictions? | {
"login": "Swathygsb",
"id": 23665054,
"node_id": "MDQ6VXNlcjIzNjY1MDU0",
"avatar_url": "https://avatars.githubusercontent.com/u/23665054?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Swathygsb",
"html_url": "https://github.com/Swathygsb",
"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 | [] | [
"You can write your own code like the prediction phase [here](https://github.com/huggingface/pytorch-pretrained-BERT/blob/78462aad6113d50063d8251e27dbaadb7f44fbf0/examples/run_squad.py#L345) ",
"@Swathygsb have you figured it out? I have the same use case as you and I'm struggling to understand the source code. "... | 1,563 | 1,569 | 1,569 | NONE | null | Hi,
Can you give sample codes for how to use Bert QA model for predicting an answer given a text corpus and a question? | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/787/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/787/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/786 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/786/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/786/comments | https://api.github.com/repos/huggingface/transformers/issues/786/events | https://github.com/huggingface/transformers/pull/786 | 467,324,508 | MDExOlB1bGxSZXF1ZXN0Mjk2OTk4NDUy | 786 | New documentation for pytorch-transformers | {
"login": "thomwolf",
"id": 7353373,
"node_id": "MDQ6VXNlcjczNTMzNzM=",
"avatar_url": "https://avatars.githubusercontent.com/u/7353373?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/thomwolf",
"html_url": "https://github.com/thomwolf",
"followers_url": "https://api.github.com/users/thomw... | [] | closed | false | null | [] | [] | 1,562 | 1,566 | 1,563 | MEMBER | null | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/786/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/786/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/786",
"html_url": "https://github.com/huggingface/transformers/pull/786",
"diff_url": "https://github.com/huggingface/transformers/pull/786.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/786.patch",
"merged_at": 15630... | |
https://api.github.com/repos/huggingface/transformers/issues/785 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/785/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/785/comments | https://api.github.com/repos/huggingface/transformers/issues/785/events | https://github.com/huggingface/transformers/issues/785 | 467,272,230 | MDU6SXNzdWU0NjcyNzIyMzA= | 785 | Implementation of 15% words masking would cause the drop of performance in short text | {
"login": "zhangsh950618",
"id": 39693134,
"node_id": "MDQ6VXNlcjM5NjkzMTM0",
"avatar_url": "https://avatars.githubusercontent.com/u/39693134?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/zhangsh950618",
"html_url": "https://github.com/zhangsh950618",
"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"
] | 1,562 | 1,568 | 1,568 | NONE | null | I found the same problem that the implementation is different from tensorflow. If we use the implementation of pytorch will produce two extreme case especially for short sentences like article title,usually 10-20 characters.
case 1. sentence with too much '[MASK]'
case 2. sentence with none '[MASK]'
both case1 and c... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/785/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/785/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/784 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/784/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/784/comments | https://api.github.com/repos/huggingface/transformers/issues/784/events | https://github.com/huggingface/transformers/issues/784 | 467,226,420 | MDU6SXNzdWU0NjcyMjY0MjA= | 784 | [bug] from_pretrained error with from_tf | {
"login": "shibing624",
"id": 10249622,
"node_id": "MDQ6VXNlcjEwMjQ5NjIy",
"avatar_url": "https://avatars.githubusercontent.com/u/10249622?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/shibing624",
"html_url": "https://github.com/shibing624",
"followers_url": "https://api.github.com/use... | [] | closed | false | null | [] | [
"Yeah this is solved in the coming release"
] | 1,562 | 1,563 | 1,563 | NONE | null | https://github.com/huggingface/pytorch-pretrained-BERT/blob/master/pytorch_pretrained_bert/modeling.py#L721 , the weights_path should be archive_file, and set from_tf to str is better to load finetuned model, like model name is model.ckpt-25000.meta. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/784/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/784/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/783 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/783/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/783/comments | https://api.github.com/repos/huggingface/transformers/issues/783/events | https://github.com/huggingface/transformers/issues/783 | 467,181,929 | MDU6SXNzdWU0NjcxODE5Mjk= | 783 | how to get the word vector from bert pretrain model ? | {
"login": "zhangyu68",
"id": 36838019,
"node_id": "MDQ6VXNlcjM2ODM4MDE5",
"avatar_url": "https://avatars.githubusercontent.com/u/36838019?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/zhangyu68",
"html_url": "https://github.com/zhangyu68",
"followers_url": "https://api.github.com/users/... | [] | closed | false | null | [] | [
"This will be possible in the new release out soon.",
"I find a method that can get the words embeddings.Thank you all the same!\r\nself.model = BertModel.from_pretrained(config.bert_path)\r\nself.word_emb = self.model.embeddings"
] | 1,562 | 1,562 | 1,562 | NONE | null | Could you please help me?
I just want to get bert's word vector,but I only can get the encoder's result. How can I get the word vector before data inputing the encoder model ?
Thank you ! | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/783/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/783/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/782 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/782/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/782/comments | https://api.github.com/repos/huggingface/transformers/issues/782/events | https://github.com/huggingface/transformers/issues/782 | 467,175,431 | MDU6SXNzdWU0NjcxNzU0MzE= | 782 | Why the activation function is tanh in BertPooler | {
"login": "xinliweiyuan",
"id": 5919883,
"node_id": "MDQ6VXNlcjU5MTk4ODM=",
"avatar_url": "https://avatars.githubusercontent.com/u/5919883?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/xinliweiyuan",
"html_url": "https://github.com/xinliweiyuan",
"followers_url": "https://api.github.com... | [] | closed | false | null | [] | [
"Because that's what Bert's authors do in the official TF code:\r\nhttps://github.com/google-research/bert/blob/bee6030e31e42a9394ac567da170a89a98d2062f/modeling.py#L231",
"Just wanted to point out for future reference the motivation has been answered by the original BERT authors in [[this GitHub issue]](https://... | 1,562 | 1,591 | 1,563 | NONE | null | I found the activation function in the BertPooler layer is tanh, but Bert never mentions that it uses the tanh. It says gelu activation function is applied in the paper.
So why there is a tanh here ? Waiting for some explanation. Thanks.
```
class BertPooler(nn.Module):
def __init__(self, config):
... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/782/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/782/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/781 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/781/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/781/comments | https://api.github.com/repos/huggingface/transformers/issues/781/events | https://github.com/huggingface/transformers/pull/781 | 467,128,858 | MDExOlB1bGxSZXF1ZXN0Mjk2ODQzOTgy | 781 | Clean up input embeddings resizing and weights tying | {
"login": "thomwolf",
"id": 7353373,
"node_id": "MDQ6VXNlcjczNTMzNzM=",
"avatar_url": "https://avatars.githubusercontent.com/u/7353373?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/thomwolf",
"html_url": "https://github.com/thomwolf",
"followers_url": "https://api.github.com/users/thomw... | [] | closed | false | null | [] | [
"I have added a test suite that tests both the `tie_weights` function as well as the `resize_token_embeddings`",
"# [Codecov](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/pull/781?src=pr&el=h1) Report\n> Merging [#781](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/pull/781?src=pr&el=desc)... | 1,562 | 1,566 | 1,562 | MEMBER | null | Still need to add tests on these features | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/781/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/781/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/781",
"html_url": "https://github.com/huggingface/transformers/pull/781",
"diff_url": "https://github.com/huggingface/transformers/pull/781.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/781.patch",
"merged_at": 15629... |
https://api.github.com/repos/huggingface/transformers/issues/780 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/780/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/780/comments | https://api.github.com/repos/huggingface/transformers/issues/780/events | https://github.com/huggingface/transformers/issues/780 | 467,127,635 | MDU6SXNzdWU0NjcxMjc2MzU= | 780 | Fail to run finetune_on_pregenerated.py | {
"login": "allisonyw",
"id": 38665667,
"node_id": "MDQ6VXNlcjM4NjY1NjY3",
"avatar_url": "https://avatars.githubusercontent.com/u/38665667?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/allisonyw",
"html_url": "https://github.com/allisonyw",
"followers_url": "https://api.github.com/users/... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.\n"
] | 1,562 | 1,568 | 1,568 | NONE | null | Hi,
I am fine tuning BERT for my own data set. Pregenerate training data was smooth but when I run finetune_on_pregenerated.py I got the following KeyError:
2019-07-11 22:53:04,151: ***** Running training *****
2019-07-11 22:53:04,151: Num examples = 35832
2019-07-11 22:53:04,151: Batch size = 32
2019-07-... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/780/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/780/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/779 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/779/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/779/comments | https://api.github.com/repos/huggingface/transformers/issues/779/events | https://github.com/huggingface/transformers/issues/779 | 467,084,782 | MDU6SXNzdWU0NjcwODQ3ODI= | 779 | Should close the SummaryWriter after using it | {
"login": "t-yaxli",
"id": 51250153,
"node_id": "MDQ6VXNlcjUxMjUwMTUz",
"avatar_url": "https://avatars.githubusercontent.com/u/51250153?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/t-yaxli",
"html_url": "https://github.com/t-yaxli",
"followers_url": "https://api.github.com/users/t-yaxl... | [] | closed | false | null | [] | [
"Oh yes you are right, thanks it's fixed in the coming release."
] | 1,562 | 1,563 | 1,563 | NONE | null | Really appreciate the good work to implement this package!
I have tried to run the script: [run_glue.py](https://github.com/huggingface/pytorch-pretrained-BERT/blob/xlnet/examples/run_glue.py). When I test with this script, I found some of the scalars add into the SummaryWriter did not appears in TensorBoard. I thi... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/779/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/779/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/778 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/778/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/778/comments | https://api.github.com/repos/huggingface/transformers/issues/778/events | https://github.com/huggingface/transformers/issues/778 | 466,912,862 | MDU6SXNzdWU0NjY5MTI4NjI= | 778 | Order of tokens in vocabulary of German model | {
"login": "schoennenbeck",
"id": 22288048,
"node_id": "MDQ6VXNlcjIyMjg4MDQ4",
"avatar_url": "https://avatars.githubusercontent.com/u/22288048?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/schoennenbeck",
"html_url": "https://github.com/schoennenbeck",
"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 | [] | [
"@tholor and @timoeller may have some insights on these",
"Hey Sebastian, thanks for using the German Bert and digging into its details. The mysterious [unused3001] token was actually a special comma symbol to get rid of [UNK] tokens in some of our training texts. But we covered it up later on in the process + di... | 1,562 | 1,569 | 1,569 | NONE | null | The vocabulary for the German model ('bert-base-german-cased') has the token '[unused3001]' at position 0 (and the '[PAD]' token at position 1). However, the BertEmbedding has padding_idx=0 as usual.
Is this behaviour intended and if so would it be possible to get some insight into the rationale behind it?
| {
"url": "https://api.github.com/repos/huggingface/transformers/issues/778/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/778/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/777 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/777/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/777/comments | https://api.github.com/repos/huggingface/transformers/issues/777/events | https://github.com/huggingface/transformers/pull/777 | 466,901,688 | MDExOlB1bGxSZXF1ZXN0Mjk2NjU3NzA1 | 777 | Working GLUE Example for XLNet (STS-B) | {
"login": "thomwolf",
"id": 7353373,
"node_id": "MDQ6VXNlcjczNTMzNzM=",
"avatar_url": "https://avatars.githubusercontent.com/u/7353373?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/thomwolf",
"html_url": "https://github.com/thomwolf",
"followers_url": "https://api.github.com/users/thomw... | [] | closed | false | null | [] | [] | 1,562 | 1,566 | 1,562 | MEMBER | null | Same as #776 but let's merge it on XLNet for the moment.
`run_glue.py` is now a single script able to train BERT, XLNet and XLM on all GLUE tasks.
Example for XLNet:
```bash
CUDA_VISIBLE_DEVICES=0,1,2,3 python ./examples/run_glue.py --do_train --task_name=sts-b --data_dir=${GLUE_DIR}/STS-B --output_dir... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/777/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/777/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/777",
"html_url": "https://github.com/huggingface/transformers/pull/777",
"diff_url": "https://github.com/huggingface/transformers/pull/777.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/777.patch",
"merged_at": 15628... |
https://api.github.com/repos/huggingface/transformers/issues/776 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/776/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/776/comments | https://api.github.com/repos/huggingface/transformers/issues/776/events | https://github.com/huggingface/transformers/pull/776 | 466,900,653 | MDExOlB1bGxSZXF1ZXN0Mjk2NjU2ODQ5 | 776 | Working GLUE Example for XLNet (STS-B) | {
"login": "thomwolf",
"id": 7353373,
"node_id": "MDQ6VXNlcjczNTMzNzM=",
"avatar_url": "https://avatars.githubusercontent.com/u/7353373?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/thomwolf",
"html_url": "https://github.com/thomwolf",
"followers_url": "https://api.github.com/users/thomw... | [] | closed | false | null | [] | [] | 1,562 | 1,566 | 1,562 | MEMBER | null | `run_glue.py` is now a single script able to train BERT, XLNet and XLM on all GLUE tasks.
Example for XLNet:
```bash
CUDA_VISIBLE_DEVICES=0,1,2,3 python ./examples/run_glue.py --do_train --task_name=sts-b --data_dir=${GLUE_DIR}/STS-B --output_dir=./proc_data/sts-b-110 --max_seq_length=128 --per_gpu_ev... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/776/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/776/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/776",
"html_url": "https://github.com/huggingface/transformers/pull/776",
"diff_url": "https://github.com/huggingface/transformers/pull/776.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/776.patch",
"merged_at": null
... |
https://api.github.com/repos/huggingface/transformers/issues/775 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/775/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/775/comments | https://api.github.com/repos/huggingface/transformers/issues/775/events | https://github.com/huggingface/transformers/pull/775 | 466,838,926 | MDExOlB1bGxSZXF1ZXN0Mjk2NjA2MDQx | 775 | fix typo in readme: extract_classif.py ==> extract_features.py | {
"login": "xinfeng1i",
"id": 2620608,
"node_id": "MDQ6VXNlcjI2MjA2MDg=",
"avatar_url": "https://avatars.githubusercontent.com/u/2620608?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/xinfeng1i",
"html_url": "https://github.com/xinfeng1i",
"followers_url": "https://api.github.com/users/xi... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/pull/775?src=pr&el=h1) Report\n> Merging [#775](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/pull/775?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/commit/78462aad6113d50063d8251e27dbaadb7f4... | 1,562 | 1,563 | 1,563 | NONE | null | There seems to be a typo in the `README.md` file in Section `Example` (as shown in the following figure), I guess the script name should be `extract_features.py`.
| {
"url": "https://api.github.com/repos/huggingface/transformers/issues/775/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/775/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/775",
"html_url": "https://github.com/huggingface/transformers/pull/775",
"diff_url": "https://github.com/huggingface/transformers/pull/775.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/775.patch",
"merged_at": null
... |
https://api.github.com/repos/huggingface/transformers/issues/774 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/774/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/774/comments | https://api.github.com/repos/huggingface/transformers/issues/774/events | https://github.com/huggingface/transformers/issues/774 | 466,632,277 | MDU6SXNzdWU0NjY2MzIyNzc= | 774 | XLNet text generation ability | {
"login": "astariul",
"id": 43774355,
"node_id": "MDQ6VXNlcjQzNzc0MzU1",
"avatar_url": "https://avatars.githubusercontent.com/u/43774355?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/astariul",
"html_url": "https://github.com/astariul",
"followers_url": "https://api.github.com/users/ast... | [] | closed | false | null | [] | [
"Indeed, I've now added the text padding trick of Aman (add some padding text to have longer inputs) and the quality is really a lot higher.\r\n\r\nWill merge the xlnet branch in master and release on Monday."
] | 1,562 | 1,563 | 1,563 | CONTRIBUTOR | null | Really appreciate the good work to implement XLNet !
I tried running the [XLNet text generation example](https://github.com/huggingface/pytorch-pretrained-BERT/blob/xlnet/examples/generation_xlnet.py)
But the generated text quality is really low.
Tricks used by https://github.com/rusiaaman/XLnet-gen needs to ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/774/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/774/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/773 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/773/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/773/comments | https://api.github.com/repos/huggingface/transformers/issues/773/events | https://github.com/huggingface/transformers/pull/773 | 466,566,367 | MDExOlB1bGxSZXF1ZXN0Mjk2Mzg0NTM3 | 773 | Sphinx doc, XLM Checkpoints | {
"login": "LysandreJik",
"id": 30755778,
"node_id": "MDQ6VXNlcjMwNzU1Nzc4",
"avatar_url": "https://avatars.githubusercontent.com/u/30755778?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/LysandreJik",
"html_url": "https://github.com/LysandreJik",
"followers_url": "https://api.github.com/... | [] | closed | false | null | [] | [] | 1,562 | 1,562 | 1,562 | MEMBER | null | The updated sphinx documentation with additional pages, fixed links, an added a whole new HuggingFace-based theme.
Additionally, patched the XLM weights conversion script and added 5 new checkpoints for XLM. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/773/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/773/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/773",
"html_url": "https://github.com/huggingface/transformers/pull/773",
"diff_url": "https://github.com/huggingface/transformers/pull/773.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/773.patch",
"merged_at": 15628... |
https://api.github.com/repos/huggingface/transformers/issues/772 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/772/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/772/comments | https://api.github.com/repos/huggingface/transformers/issues/772/events | https://github.com/huggingface/transformers/issues/772 | 466,560,805 | MDU6SXNzdWU0NjY1NjA4MDU= | 772 | Cannot load 'bert-base-german-cased' | {
"login": "laifi",
"id": 34584914,
"node_id": "MDQ6VXNlcjM0NTg0OTE0",
"avatar_url": "https://avatars.githubusercontent.com/u/34584914?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/laifi",
"html_url": "https://github.com/laifi",
"followers_url": "https://api.github.com/users/laifi/follow... | [] | closed | false | null | [] | [
"Hi @laifi, \r\n\r\nI cannot reproduce this issue. Are you sure that you run with the latest code from master branch? It looks suspicious to me that `tokenizer = BertTokenizer.from_pretrained('bert-base-german-cased')` doesn't find the model. \r\nCan you please check if you have [the according line](https://github.... | 1,562 | 1,605 | 1,562 | NONE | null | `tokenizer = BertTokenizer.from_pretrained('bert-base-german-cased')`
**Output:**
> Model name 'bert-base-german-cased' was not found in model name list (bert-base-uncased, bert-large-uncased, bert-base-cased, bert-large-cased, bert-base-multilingual-uncased, bert-base-multilingual-cased, bert-base-chinese). We a... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/772/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/772/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/771 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/771/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/771/comments | https://api.github.com/repos/huggingface/transformers/issues/771/events | https://github.com/huggingface/transformers/issues/771 | 466,475,280 | MDU6SXNzdWU0NjY0NzUyODA= | 771 | Performance dramatically drops down without training. | {
"login": "DariaD",
"id": 10706920,
"node_id": "MDQ6VXNlcjEwNzA2OTIw",
"avatar_url": "https://avatars.githubusercontent.com/u/10706920?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/DariaD",
"html_url": "https://github.com/DariaD",
"followers_url": "https://api.github.com/users/DariaD/fo... | [] | closed | false | null | [] | [
"If you want to evaluate only, you have to set `--output_dir` to the path of your previously trained model. Otherwise, the script will use the original model."
] | 1,562 | 1,563 | 1,563 | NONE | null | I use run_classivier and run_squad as it is shown in README.
If I remove `--do_train` (I already tuned the model and just want to evaluate one more time or with different development set) I expect that result would be the same but performance drops down.
For example, SQuAD:
with training: `{"exact_match": 81.35... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/771/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/771/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/770 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/770/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/770/comments | https://api.github.com/repos/huggingface/transformers/issues/770/events | https://github.com/huggingface/transformers/issues/770 | 466,394,359 | MDU6SXNzdWU0NjYzOTQzNTk= | 770 | How can I load a fine-tuned model? | {
"login": "daz261",
"id": 35951484,
"node_id": "MDQ6VXNlcjM1OTUxNDg0",
"avatar_url": "https://avatars.githubusercontent.com/u/35951484?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/daz261",
"html_url": "https://github.com/daz261",
"followers_url": "https://api.github.com/users/daz261/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 | [] | [
"you can use the path to the folder containing your fine-tuned model as `--bert_model`.",
"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"
] | 1,562 | 1,568 | 1,568 | NONE | null | I finetuned a new model by running pregenerate_training_data.py and finetune_on_pregenerated.py and the output is saved as pytorch_model.bin.
How do I load the model to run the regular run_classfier,py predictions? To which files do I have to add code? | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/770/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/770/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/769 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/769/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/769/comments | https://api.github.com/repos/huggingface/transformers/issues/769/events | https://github.com/huggingface/transformers/issues/769 | 466,099,337 | MDU6SXNzdWU0NjYwOTkzMzc= | 769 | XLNet tensor at wrong device issuse | {
"login": "boy2000-007man",
"id": 4197489,
"node_id": "MDQ6VXNlcjQxOTc0ODk=",
"avatar_url": "https://avatars.githubusercontent.com/u/4197489?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/boy2000-007man",
"html_url": "https://github.com/boy2000-007man",
"followers_url": "https://api.gith... | [] | closed | false | null | [] | [
"This model was WIP. Fixed now."
] | 1,562 | 1,563 | 1,563 | CONTRIBUTOR | null | ```bash
File "env.xlnet/lib/python3.6/site-packages/pytorch_transformers/modeling_xlnet.py", line 397, in rel_shift
x = torch.index_select(x, 1, torch.arange(klen))
RuntimeError: Expected object of backend CUDA but got backend CPU for argument #3 'index'
```
I meet this issue when using `pytorch-transformers... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/769/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/769/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/768 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/768/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/768/comments | https://api.github.com/repos/huggingface/transformers/issues/768/events | https://github.com/huggingface/transformers/issues/768 | 465,861,420 | MDU6SXNzdWU0NjU4NjE0MjA= | 768 | GPT-2 language model decoding method | {
"login": "haozheji",
"id": 25786613,
"node_id": "MDQ6VXNlcjI1Nzg2NjEz",
"avatar_url": "https://avatars.githubusercontent.com/u/25786613?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/haozheji",
"html_url": "https://github.com/haozheji",
"followers_url": "https://api.github.com/users/hao... | [] | closed | false | null | [] | [
"`run_gpt2` has top-K which is better than beam-search for high-entropy tasks like open-domain generation. The coming release example (currently on the xlnet branch to be merged with master on Monday) will have top-K and Nucleus sampling (see Holtzman et al. http://arxiv.org/abs/1904.09751)",
"Hi,\r\nIs it possib... | 1,562 | 1,576 | 1,563 | CONTRIBUTOR | null | I am wondering what is the official decoding method when evaluating the language model? The doc says `run_gpt2.py` implement the beam-search. While to me, it seems it's still greedy search with sampling. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/768/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/768/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/767 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/767/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/767/comments | https://api.github.com/repos/huggingface/transformers/issues/767/events | https://github.com/huggingface/transformers/pull/767 | 465,828,120 | MDExOlB1bGxSZXF1ZXN0Mjk1NzkyNzM2 | 767 | 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 | [] | [] | 1,562 | 1,562 | 1,562 | MEMBER | null | Sphinx based documentation with Google style comments. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/767/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/767/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/767",
"html_url": "https://github.com/huggingface/transformers/pull/767",
"diff_url": "https://github.com/huggingface/transformers/pull/767.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/767.patch",
"merged_at": 15626... |
https://api.github.com/repos/huggingface/transformers/issues/766 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/766/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/766/comments | https://api.github.com/repos/huggingface/transformers/issues/766/events | https://github.com/huggingface/transformers/issues/766 | 465,778,432 | MDU6SXNzdWU0NjU3Nzg0MzI= | 766 | Fine tune Xlnet | {
"login": "AhmedBahaaElDinMohammed",
"id": 51789113,
"node_id": "MDQ6VXNlcjUxNzg5MTEz",
"avatar_url": "https://avatars.githubusercontent.com/u/51789113?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/AhmedBahaaElDinMohammed",
"html_url": "https://github.com/AhmedBahaaElDinMohammed",
"foll... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"As far as I know, the pytorch code of XLNet is not completely ready now. But you could find it in the branch `xlnet` and the classifier code is nearly ready in the file `example/run_xlnet_classifier.py`. I have successfully fine-tuned it on the SST-2 task (which belongs to GLUE) with following args:\r\n\r\n```shel... | 1,562 | 1,569 | 1,569 | NONE | null | Can anybody guide me on how to fine tune xlnet for simple text classification task or any reference code because i am lost. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/766/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/766/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/765 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/765/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/765/comments | https://api.github.com/repos/huggingface/transformers/issues/765/events | https://github.com/huggingface/transformers/issues/765 | 465,511,861 | MDU6SXNzdWU0NjU1MTE4NjE= | 765 | Is is possible to fine-tune GPT2 on downstream tasks currently? | {
"login": "shizhediao",
"id": 18120087,
"node_id": "MDQ6VXNlcjE4MTIwMDg3",
"avatar_url": "https://avatars.githubusercontent.com/u/18120087?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/shizhediao",
"html_url": "https://github.com/shizhediao",
"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 | [] | [
"Yes we could add this. You mean tasks like GLUE or SQuAD?",
"> Yes we could add this. You mean tasks like GLUE or SQuAD?\r\n\r\nYes! exactly!\r\nPlease add this, thanks!",
"@thomwolf Are you still working on the code to finetune the GPT2 language model (not classification task)? Thanks.",
"@experiencor @thom... | 1,562 | 1,625 | 1,570 | NONE | null | Is is possible to fine-tune GPT2 on downstream tasks currently? | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/765/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/765/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/764 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/764/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/764/comments | https://api.github.com/repos/huggingface/transformers/issues/764/events | https://github.com/huggingface/transformers/issues/764 | 465,337,246 | MDU6SXNzdWU0NjUzMzcyNDY= | 764 | Adding extra inputs when fine-tuning BERT | {
"login": "nadavborenstein",
"id": 15877500,
"node_id": "MDQ6VXNlcjE1ODc3NTAw",
"avatar_url": "https://avatars.githubusercontent.com/u/15877500?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/nadavborenstein",
"html_url": "https://github.com/nadavborenstein",
"followers_url": "https://api... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"You could try stacking a linear layer over-top of BERT that takes as input the BERT sequence representation + your features. You would have to fine-tune through all of BERT.",
"This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occur... | 1,562 | 1,601 | 1,568 | NONE | null | I am trying to fine-tune BERT for a sequence classification task where in addition to the sequences, I have extra features such as the writer age, tags, etc. I want to use those extra features, and I was thinking about concatenating them to the input of the final linear layer.
Is there a way of doing such a thing? If ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/764/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/764/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/763 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/763/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/763/comments | https://api.github.com/repos/huggingface/transformers/issues/763/events | https://github.com/huggingface/transformers/issues/763 | 465,149,081 | MDU6SXNzdWU0NjUxNDkwODE= | 763 | ''bert-large-uncased-whole-word-masking-finetuned-squad' CAN'T be reached. | {
"login": "sinboyxx",
"id": 32317326,
"node_id": "MDQ6VXNlcjMyMzE3MzI2",
"avatar_url": "https://avatars.githubusercontent.com/u/32317326?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sinboyxx",
"html_url": "https://github.com/sinboyxx",
"followers_url": "https://api.github.com/users/sin... | [
{
"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 | [] | [
"also ran into this. I think they forgot to upload the file/make it public. You can find the vocab file on the original google repo\r\n\r\nhttps://github.com/google-research/bert",
"This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity ... | 1,562 | 1,568 | 1,568 | NONE | null | 'bert-large-uncased-whole-word-masking-finetuned-squad' can't be reached from the addr in tokenization.py:
https://s3.amazonaws.com/models.huggingface.co/bert/bert-large-uncased-whole-word-masking-finetuned-squad-vocab.txt
| {
"url": "https://api.github.com/repos/huggingface/transformers/issues/763/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/763/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/762 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/762/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/762/comments | https://api.github.com/repos/huggingface/transformers/issues/762/events | https://github.com/huggingface/transformers/issues/762 | 465,051,890 | MDU6SXNzdWU0NjUwNTE4OTA= | 762 | randrange() error when running pregenerate_training_data.py code in lm_finetuning | {
"login": "KavyaGujjala",
"id": 28920687,
"node_id": "MDQ6VXNlcjI4OTIwNjg3",
"avatar_url": "https://avatars.githubusercontent.com/u/28920687?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/KavyaGujjala",
"html_url": "https://github.com/KavyaGujjala",
"followers_url": "https://api.github.c... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"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"
] | 1,562 | 1,568 | 1,568 | NONE | null | Hi,
I am trying to run pregenerate_training_data.py code in lm_finetuning using a text file which has two documents ( each document has around 200 sentences )
I ran into this error:
```
Better speed can be achieved with apex installed from https://www.github.com/nvidia/apex.
Loading Dataset: 399 lines [00:00, ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/762/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/762/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/761 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/761/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/761/comments | https://api.github.com/repos/huggingface/transformers/issues/761/events | https://github.com/huggingface/transformers/issues/761 | 465,004,095 | MDU6SXNzdWU0NjUwMDQwOTU= | 761 | Help loading BioBERT weights | {
"login": "happypanda5",
"id": 48689790,
"node_id": "MDQ6VXNlcjQ4Njg5Nzkw",
"avatar_url": "https://avatars.githubusercontent.com/u/48689790?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/happypanda5",
"html_url": "https://github.com/happypanda5",
"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 | [] | [
"Not really providing a solution here, but have you considered https://github.com/allenai/scibert instead?\r\nAllenAI provides PyTorch weights, and through tests they claim their model is superior https://arxiv.org/pdf/1903.10676.pdf on their suite of tasks. For that and for ease of use, it may be a valid alternati... | 1,562 | 1,570 | 1,570 | NONE | null | I have completed the following:
**1. Downloaded pretrained BioBERT weights from their current release**
**2. Convert TensorFlow checkpoints into Pytorch weights bin file using the following code**
import os
os.system( ' pytorch_pretrained_bert convert_tf_checkpoint_to_pytorch \
"/content/biobert_v1.1_pub... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/761/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/761/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/760 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/760/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/760/comments | https://api.github.com/repos/huggingface/transformers/issues/760/events | https://github.com/huggingface/transformers/issues/760 | 464,768,544 | MDU6SXNzdWU0NjQ3Njg1NDQ= | 760 | Simple LM finetuning falls with RunTime Error: CUDA out of memory | {
"login": "MNCTTY",
"id": 37251686,
"node_id": "MDQ6VXNlcjM3MjUxNjg2",
"avatar_url": "https://avatars.githubusercontent.com/u/37251686?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/MNCTTY",
"html_url": "https://github.com/MNCTTY",
"followers_url": "https://api.github.com/users/MNCTTY/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 | [] | [
"How much memory does your GPU have. You can check this by running `nvidia-smi`.",
"I also has same phenomena. Also, the learning time become slower and much GPU consumption occur, both of which I think is natural, regarding parameters BERT has.\r\n\r\nThe substitutional way is that, no fine-tuning and dump. \r\n... | 1,562 | 1,569 | 1,569 | NONE | null | I tried to run simple_lm_finetuning.py on my own data with multi lingual uncased model, and the script breaks down with error 'CUDA out of memory'. Can anyone say what should I do in this situation?
I've already decreased batch size from 32 to 2, but even then I get this error.
![image](https://user-images.githubuser... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/760/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/760/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/759 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/759/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/759/comments | https://api.github.com/repos/huggingface/transformers/issues/759/events | https://github.com/huggingface/transformers/pull/759 | 464,576,856 | MDExOlB1bGxSZXF1ZXN0Mjk0ODI0NzIw | 759 | Release 0.7: pytorch-pretrained-bert => pytorch-transformers | {
"login": "thomwolf",
"id": 7353373,
"node_id": "MDQ6VXNlcjczNTMzNzM=",
"avatar_url": "https://avatars.githubusercontent.com/u/7353373?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/thomwolf",
"html_url": "https://github.com/thomwolf",
"followers_url": "https://api.github.com/users/thomw... | [] | closed | false | null | [] | [] | 1,562 | 1,566 | 1,562 | MEMBER | null | Name change: `pytorch-pretrained-bert` => `pytorch-transformers`
Standardize tokenization + tests
Refactor examples and add tests for examples as well | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/759/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/759/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/759",
"html_url": "https://github.com/huggingface/transformers/pull/759",
"diff_url": "https://github.com/huggingface/transformers/pull/759.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/759.patch",
"merged_at": 15626... |
https://api.github.com/repos/huggingface/transformers/issues/758 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/758/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/758/comments | https://api.github.com/repos/huggingface/transformers/issues/758/events | https://github.com/huggingface/transformers/pull/758 | 464,313,727 | MDExOlB1bGxSZXF1ZXN0Mjk0NjIwMTcx | 758 | Release 0.7 - Add doc | {
"login": "thomwolf",
"id": 7353373,
"node_id": "MDQ6VXNlcjczNTMzNzM=",
"avatar_url": "https://avatars.githubusercontent.com/u/7353373?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/thomwolf",
"html_url": "https://github.com/thomwolf",
"followers_url": "https://api.github.com/users/thomw... | [] | closed | false | null | [] | [] | 1,562 | 1,562 | 1,562 | MEMBER | null | Like #757 but let's point on the `xlnet` branch for now. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/758/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/758/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/758",
"html_url": "https://github.com/huggingface/transformers/pull/758",
"diff_url": "https://github.com/huggingface/transformers/pull/758.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/758.patch",
"merged_at": 15623... |
https://api.github.com/repos/huggingface/transformers/issues/757 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/757/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/757/comments | https://api.github.com/repos/huggingface/transformers/issues/757/events | https://github.com/huggingface/transformers/pull/757 | 464,312,700 | MDExOlB1bGxSZXF1ZXN0Mjk0NjE5MzM1 | 757 | Release 0.7 - Add a real doc | {
"login": "thomwolf",
"id": 7353373,
"node_id": "MDQ6VXNlcjczNTMzNzM=",
"avatar_url": "https://avatars.githubusercontent.com/u/7353373?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/thomwolf",
"html_url": "https://github.com/thomwolf",
"followers_url": "https://api.github.com/users/thomw... | [] | closed | false | null | [] | [] | 1,562 | 1,562 | 1,562 | MEMBER | null | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/757/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/757/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/757",
"html_url": "https://github.com/huggingface/transformers/pull/757",
"diff_url": "https://github.com/huggingface/transformers/pull/757.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/757.patch",
"merged_at": null
... | |
https://api.github.com/repos/huggingface/transformers/issues/756 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/756/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/756/comments | https://api.github.com/repos/huggingface/transformers/issues/756/events | https://github.com/huggingface/transformers/issues/756 | 464,282,735 | MDU6SXNzdWU0NjQyODI3MzU= | 756 | Invalid Syntax Error trying to run pregenerate_training_data.py | {
"login": "MNCTTY",
"id": 37251686,
"node_id": "MDQ6VXNlcjM3MjUxNjg2",
"avatar_url": "https://avatars.githubusercontent.com/u/37251686?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/MNCTTY",
"html_url": "https://github.com/MNCTTY",
"followers_url": "https://api.github.com/users/MNCTTY/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 | [] | [
"Your snippet is too short to see what type of error there is, can you extract a larger one?",
"Hi \r\nyes\r\n\r\n",
"Pull the latest changes from master and report if that helps.",
"it has been disap... | 1,562 | 1,568 | 1,568 | NONE | null | I'm getting such error, can't understand what's wrong
Plus, is it possible to further fine tune once fine tuned model, that appears after simple finetuning .py in corresponding folder (pytorch_model.bin) ? | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/756/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/756/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/755 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/755/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/755/comments | https://api.github.com/repos/huggingface/transformers/issues/755/events | https://github.com/huggingface/transformers/pull/755 | 463,973,000 | MDExOlB1bGxSZXF1ZXN0Mjk0MzQ2NzA5 | 755 | TorchScript trace comparison with different sizes | {
"login": "LysandreJik",
"id": 30755778,
"node_id": "MDQ6VXNlcjMwNzU1Nzc4",
"avatar_url": "https://avatars.githubusercontent.com/u/30755778?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/LysandreJik",
"html_url": "https://github.com/LysandreJik",
"followers_url": "https://api.github.com/... | [] | closed | false | null | [] | [] | 1,562 | 1,566 | 1,565 | MEMBER | null | Adds a test to compare TorchScript traces with different batch sizes and sequences lengths. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/755/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/755/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/755",
"html_url": "https://github.com/huggingface/transformers/pull/755",
"diff_url": "https://github.com/huggingface/transformers/pull/755.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/755.patch",
"merged_at": null
... |
https://api.github.com/repos/huggingface/transformers/issues/754 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/754/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/754/comments | https://api.github.com/repos/huggingface/transformers/issues/754/events | https://github.com/huggingface/transformers/issues/754 | 463,967,141 | MDU6SXNzdWU0NjM5NjcxNDE= | 754 | Get Attention Values for Pretrained Model | {
"login": "Sparkier",
"id": 5690524,
"node_id": "MDQ6VXNlcjU2OTA1MjQ=",
"avatar_url": "https://avatars.githubusercontent.com/u/5690524?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Sparkier",
"html_url": "https://github.com/Sparkier",
"followers_url": "https://api.github.com/users/Spark... | [
{
"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 | [] | [
"You need to install the master version (not with pip or conda) : \r\n```\r\ngit clone https://github.com/huggingface/pytorch-pretrained-BERT.git\r\ncd pytorch-pretrained-BERT\r\npython setup.py install\r\n```\r\n\r\nThen you can use it like this : \r\n```\r\nmodel = BertModel.from_pretrained('bert-base-uncased',\r... | 1,562 | 1,568 | 1,568 | NONE | null | When using BertModel.from_pretrained, I am not able to have it also return the attention layers. Why does that not word? Am I doing something wrong? | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/754/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/754/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/753 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/753/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/753/comments | https://api.github.com/repos/huggingface/transformers/issues/753/events | https://github.com/huggingface/transformers/issues/753 | 463,959,301 | MDU6SXNzdWU0NjM5NTkzMDE= | 753 | `bert-base-uncased` works for CoLA, `bert-large-uncased` always predicts one class | {
"login": "rococode",
"id": 32279130,
"node_id": "MDQ6VXNlcjMyMjc5MTMw",
"avatar_url": "https://avatars.githubusercontent.com/u/32279130?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/rococode",
"html_url": "https://github.com/rococode",
"followers_url": "https://api.github.com/users/roc... | [
{
"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"
] | 1,562 | 1,567 | 1,567 | CONTRIBUTOR | null | I'm having an issue with CoLA where finetuning off of bert-large results in a model that only predicts one class.
I make one change in configs to train the large model - I set `train_batch_size` to `16` for `bert-large-uncased`.
These are the two training commands I use (missing do_lowercase, I know, but it's f... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/753/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/753/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/752 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/752/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/752/comments | https://api.github.com/repos/huggingface/transformers/issues/752/events | https://github.com/huggingface/transformers/issues/752 | 463,578,039 | MDU6SXNzdWU0NjM1NzgwMzk= | 752 | how to set the init learning rate when use bertAdam? | {
"login": "mmxuan18",
"id": 6283983,
"node_id": "MDQ6VXNlcjYyODM5ODM=",
"avatar_url": "https://avatars.githubusercontent.com/u/6283983?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/mmxuan18",
"html_url": "https://github.com/mmxuan18",
"followers_url": "https://api.github.com/users/mmxua... | [
{
"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"
] | 1,562 | 1,567 | 1,567 | NONE | null | i set the BertAdam learning rate as the default value of args (3e-5), and i step in to the BertAdam step by step , and print lr_scheduled see that the acturly lr is very small over all the training process (between <0 ~ 1> * 3e-5), this cause the loss decrease very slow, when i set the init learning rate as 0.1, the lo... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/752/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/752/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/751 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/751/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/751/comments | https://api.github.com/repos/huggingface/transformers/issues/751/events | https://github.com/huggingface/transformers/issues/751 | 463,490,748 | MDU6SXNzdWU0NjM0OTA3NDg= | 751 | Slower and more memory hungry than the TensorFlow BERT? | {
"login": "nuwapi",
"id": 26151903,
"node_id": "MDQ6VXNlcjI2MTUxOTAz",
"avatar_url": "https://avatars.githubusercontent.com/u/26151903?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/nuwapi",
"html_url": "https://github.com/nuwapi",
"followers_url": "https://api.github.com/users/nuwapi/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 | [] | [
"Yes, this library is not made for training a model from scratch.\r\n\r\nYou should use one of the libraries I referred to here: https://github.com/huggingface/pytorch-pretrained-BERT/issues/543#issuecomment-491207121\r\n\r\nI might give it a look one day but not in the short-term.",
"@thomwolf Thank you so much ... | 1,562 | 1,615 | 1,573 | NONE | null | Hi pytorch-pretrained-BERT developers,
I have been using TensorFlow BERT since it came out, recently I wanted to switch to PyTorch because it is a great library. For this, I did a bunch of tests to compare training specs between Google's TF BERT and your implementation. To my surprise, this is a lot slower and can o... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/751/reactions",
"total_count": 13,
"+1": 13,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/751/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/750 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/750/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/750/comments | https://api.github.com/repos/huggingface/transformers/issues/750/events | https://github.com/huggingface/transformers/issues/750 | 463,425,410 | MDU6SXNzdWU0NjM0MjU0MTA= | 750 | Incorrect training loss scaling factor in examples/run_classifier.py? | {
"login": "ethanjperez",
"id": 6402205,
"node_id": "MDQ6VXNlcjY0MDIyMDU=",
"avatar_url": "https://avatars.githubusercontent.com/u/6402205?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ethanjperez",
"html_url": "https://github.com/ethanjperez",
"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 | [] | [
"You are right Ethan.\r\nI'm refactoring the examples which were a bit rotten, let's include this fix as well.",
"Great! Either way the examples are a great starting point :)\r\n\r\nI'm also wondering if tensorboard is [only logging](https://github.com/huggingface/pytorch-pretrained-BERT/blob/master/examples/run_... | 1,562 | 1,568 | 1,568 | CONTRIBUTOR | null | In [examples/run_classifier.py](https://github.com/huggingface/pytorch-pretrained-BERT/commit/87b9ec3843f7f9a81253075f92c9e6537ecefe1c), the overall 'loss' is produce as 'tr_loss/global_step' (instead of 'tr_loss/nb_tr_steps'). Is this behavior correct? @mprouveur made the change in this [commit](https://github.com/hug... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/750/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/750/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/749 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/749/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/749/comments | https://api.github.com/repos/huggingface/transformers/issues/749/events | https://github.com/huggingface/transformers/issues/749 | 463,356,518 | MDU6SXNzdWU0NjMzNTY1MTg= | 749 | Attribute Error : 'BertModel' object has no attribute 'bert' | {
"login": "PradyumnaGupta",
"id": 39255758,
"node_id": "MDQ6VXNlcjM5MjU1NzU4",
"avatar_url": "https://avatars.githubusercontent.com/u/39255758?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/PradyumnaGupta",
"html_url": "https://github.com/PradyumnaGupta",
"followers_url": "https://api.gi... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"You can only load from a tensorflow checkpoint in a `BertForPretraing` model.\r\nI will add a check.\r\nAlternatively, you should use the conversion script to make a pytorch model and then you can import the resulting pytorch model in any type of Bert model.",
"I use the BertForPretraining.from_pretrained().bert... | 1,562 | 1,666 | 1,568 | NONE | null | I am using Google's Bert tensorflow checkpoints to create a model from .from_pretrained as shown below-
`
model = BertModel.from_pretrained('/content/uncased_L-12_H-768_A-12',from_tf=True)
`
But I am getting the following error-
`
---------------------------------------------------------------------... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/749/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/749/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/748 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/748/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/748/comments | https://api.github.com/repos/huggingface/transformers/issues/748/events | https://github.com/huggingface/transformers/pull/748 | 463,270,020 | MDExOlB1bGxSZXF1ZXN0MjkzNzgyNjkz | 748 | Release 0.7 - Add Torchscript capabilities | {
"login": "thomwolf",
"id": 7353373,
"node_id": "MDQ6VXNlcjczNTMzNzM=",
"avatar_url": "https://avatars.githubusercontent.com/u/7353373?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/thomwolf",
"html_url": "https://github.com/thomwolf",
"followers_url": "https://api.github.com/users/thomw... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/pull/748?src=pr&el=h1) Report\n> Merging [#748](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/pull/748?src=pr&el=desc) into [xlnet](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/commit/708877958a308a0f0e8fd199f8f327e4797f... | 1,562 | 1,562 | 1,562 | MEMBER | null | Add Torchscript capabilities to all models. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/748/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/748/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/748",
"html_url": "https://github.com/huggingface/transformers/pull/748",
"diff_url": "https://github.com/huggingface/transformers/pull/748.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/748.patch",
"merged_at": 15621... |
https://api.github.com/repos/huggingface/transformers/issues/747 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/747/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/747/comments | https://api.github.com/repos/huggingface/transformers/issues/747/events | https://github.com/huggingface/transformers/issues/747 | 463,143,072 | MDU6SXNzdWU0NjMxNDMwNzI= | 747 | BERT pretraining routine | {
"login": "yhalk",
"id": 13349982,
"node_id": "MDQ6VXNlcjEzMzQ5OTgy",
"avatar_url": "https://avatars.githubusercontent.com/u/13349982?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/yhalk",
"html_url": "https://github.com/yhalk",
"followers_url": "https://api.github.com/users/yhalk/follow... | [] | closed | false | null | [] | [
"I would not advise to use them for training from scratch. See #751 for discussion and links."
] | 1,562 | 1,563 | 1,563 | NONE | null | Hi,
I was wondering whether the scripts for finetuning can be used to pretrain BERT from scratch on a small dataset that does not require TPUs - is there any difference with the TF pretrain code (different batch sampling or train loss evaluation) other than the TPU support?
Thank you very much in advance!
| {
"url": "https://api.github.com/repos/huggingface/transformers/issues/747/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/747/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/746 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/746/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/746/comments | https://api.github.com/repos/huggingface/transformers/issues/746/events | https://github.com/huggingface/transformers/issues/746 | 463,118,282 | MDU6SXNzdWU0NjMxMTgyODI= | 746 | GPT2Tokenizer for Hindi Data | {
"login": "DEBADRIBASAK",
"id": 32904247,
"node_id": "MDQ6VXNlcjMyOTA0MjQ3",
"avatar_url": "https://avatars.githubusercontent.com/u/32904247?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/DEBADRIBASAK",
"html_url": "https://github.com/DEBADRIBASAK",
"followers_url": "https://api.github.c... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"I might be wrong, but I think GPT2Tokenizer uses byte pair encoding, a form of subword-level encoding. On an intuitive level, this is a between character-level and word level, and akin to breaking the word apart by syllable (in reality it's breaking the word apart by the highest frequency patterns). I know some pe... | 1,562 | 1,572 | 1,572 | NONE | null | I was trying to fine-tune GPT2LMHeadModel with Hindi data corpus. It is performing well. But when I looked at the tokens that are generated from the GPT2Tokenizer, I saw that they are containing tokens of almost character level. I am not understanding how is this kind of encoding handling Hindi data, or any form of non... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/746/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/746/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/745 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/745/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/745/comments | https://api.github.com/repos/huggingface/transformers/issues/745/events | https://github.com/huggingface/transformers/pull/745 | 462,929,669 | MDExOlB1bGxSZXF1ZXN0MjkzNTEyODA5 | 745 | fix evaluation bug | {
"login": "leimao",
"id": 17606112,
"node_id": "MDQ6VXNlcjE3NjA2MTEy",
"avatar_url": "https://avatars.githubusercontent.com/u/17606112?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/leimao",
"html_url": "https://github.com/leimao",
"followers_url": "https://api.github.com/users/leimao/fo... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/pull/745?src=pr&el=h1) Report\n> Merging [#745](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/pull/745?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/commit/dad3c7a485b7ffc6fd2766f349e6ee845ec... | 1,562 | 1,562 | 1,562 | CONTRIBUTOR | null | The original `run_squad.py` has a potential bug. If we only want to run the script to do evaluation, the model will not be properly loaded. The simple fix is provided. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/745/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/745/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/745",
"html_url": "https://github.com/huggingface/transformers/pull/745",
"diff_url": "https://github.com/huggingface/transformers/pull/745.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/745.patch",
"merged_at": 15623... |
https://api.github.com/repos/huggingface/transformers/issues/744 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/744/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/744/comments | https://api.github.com/repos/huggingface/transformers/issues/744/events | https://github.com/huggingface/transformers/issues/744 | 462,712,749 | MDU6SXNzdWU0NjI3MTI3NDk= | 744 | Recommended multilingual bert cased model returns similar embeddings | {
"login": "oleg-yaroshevskiy",
"id": 5859692,
"node_id": "MDQ6VXNlcjU4NTk2OTI=",
"avatar_url": "https://avatars.githubusercontent.com/u/5859692?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/oleg-yaroshevskiy",
"html_url": "https://github.com/oleg-yaroshevskiy",
"followers_url": "https:/... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"I second this issue #735 ",
"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"
] | 1,561 | 1,567 | 1,567 | NONE | null | I'm trying to get embeddings for multilingual input:
```
tokenizer = BertTokenizer.from_pretrained("bert-base-multilingual-cased", do_lower_case=False)
class NeuralNet(BertPreTrainedModel):
def __init__(self, config):
super(NeuralNet, self).__init__(config)
self.bert = BertModel(config)
... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/744/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/744/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/743 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/743/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/743/comments | https://api.github.com/repos/huggingface/transformers/issues/743/events | https://github.com/huggingface/transformers/issues/743 | 462,410,393 | MDU6SXNzdWU0NjI0MTAzOTM= | 743 | Cannot reproduce results from version 0.4.0 | {
"login": "hguan6",
"id": 19914123,
"node_id": "MDQ6VXNlcjE5OTE0MTIz",
"avatar_url": "https://avatars.githubusercontent.com/u/19914123?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/hguan6",
"html_url": "https://github.com/hguan6",
"followers_url": "https://api.github.com/users/hguan6/fo... | [] | closed | false | null | [] | [
"`pip install pytorch-pretrained-bert==0.4.0` should work normally",
"Though if you did it with the latest release in March 2019 it was probably more 0.6.1 (see the list and dates here: https://github.com/huggingface/pytorch-pretrained-BERT/releases) so `pip install pytorch-pretrained-bert==0.6.1`",
"Thank you!... | 1,561 | 1,561 | 1,561 | NONE | null | Hi, I have a research project that I did a few months ago. Now I have problem reproducing results of 0.4.0, and unfortunately, I lost version 0.4.0. Can you please send me the code of this version to hguan6@asu.edu? In fact, I am not quite sure it's 0.4.0, but I remember I did it in March 2019. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/743/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/743/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/742 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/742/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/742/comments | https://api.github.com/repos/huggingface/transformers/issues/742/events | https://github.com/huggingface/transformers/pull/742 | 462,306,120 | MDExOlB1bGxSZXF1ZXN0MjkzMDQwNzA1 | 742 | When not loading a pretrained model, all layers are initialized with copies of the same weights | {
"login": "Rocketknight1",
"id": 12866554,
"node_id": "MDQ6VXNlcjEyODY2NTU0",
"avatar_url": "https://avatars.githubusercontent.com/u/12866554?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Rocketknight1",
"html_url": "https://github.com/Rocketknight1",
"followers_url": "https://api.githu... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/pull/742?src=pr&el=h1) Report\n> Merging [#742](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/pull/742?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/commit/dad3c7a485b7ffc6fd2766f349e6ee845ec... | 1,561 | 1,561 | 1,561 | MEMBER | null | Although this repo is mostly used for loading and training pre-trained BERT models, the code does support model initialization too! However, I found an issue with the initialization code - because it just makes one layer and copies it, the weights will be identical across all layers at initialization. This probably isn... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/742/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/742/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/742",
"html_url": "https://github.com/huggingface/transformers/pull/742",
"diff_url": "https://github.com/huggingface/transformers/pull/742.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/742.patch",
"merged_at": null
... |
https://api.github.com/repos/huggingface/transformers/issues/741 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/741/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/741/comments | https://api.github.com/repos/huggingface/transformers/issues/741/events | https://github.com/huggingface/transformers/issues/741 | 462,304,734 | MDU6SXNzdWU0NjIzMDQ3MzQ= | 741 | Using BertForNextSentencePrediction and GPT2LMHeadModel in a GAN setup. | {
"login": "jroakes",
"id": 10191545,
"node_id": "MDQ6VXNlcjEwMTkxNTQ1",
"avatar_url": "https://avatars.githubusercontent.com/u/10191545?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jroakes",
"html_url": "https://github.com/jroakes",
"followers_url": "https://api.github.com/users/jroake... | [] | closed | false | null | [] | [] | 1,561 | 1,562 | 1,562 | NONE | null | I am using the following code (**Training Loop**) as the meat of the training loop whereby the discriminator is BertForNextSentencePrediction and the generator is GPT2LMHeadModel. I have also included the structure of the training data (**Input data:**). The loss in the generator and discriminator appear to be falling... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/741/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/741/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/740 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/740/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/740/comments | https://api.github.com/repos/huggingface/transformers/issues/740/events | https://github.com/huggingface/transformers/issues/740 | 462,293,724 | MDU6SXNzdWU0NjIyOTM3MjQ= | 740 | How to get perplexity score of a sentence using anyone of the given Language Models? | {
"login": "vikas-baghel",
"id": 36075292,
"node_id": "MDQ6VXNlcjM2MDc1Mjky",
"avatar_url": "https://avatars.githubusercontent.com/u/36075292?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/vikas-baghel",
"html_url": "https://github.com/vikas-baghel",
"followers_url": "https://api.github.c... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"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"
] | 1,561 | 1,567 | 1,567 | NONE | null | I want to find the perlexity score of sentence. I know that we can find the perplexity if we have the loss as perplexity = 2^(entropy loss). Can you tell me how to do it with the models you have listed?
It will be of great help. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/740/reactions",
"total_count": 6,
"+1": 6,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/740/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/739 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/739/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/739/comments | https://api.github.com/repos/huggingface/transformers/issues/739/events | https://github.com/huggingface/transformers/issues/739 | 462,080,910 | MDU6SXNzdWU0NjIwODA5MTA= | 739 | where is "pytorch_model.bin"? | {
"login": "jufengada",
"id": 33510761,
"node_id": "MDQ6VXNlcjMzNTEwNzYx",
"avatar_url": "https://avatars.githubusercontent.com/u/33510761?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jufengada",
"html_url": "https://github.com/jufengada",
"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 | [] | [
"@jufengada \r\n\r\nAssuming that you've installed pytorch_pretrained_bert package properly. If you load any of the `BERT` models ex: `BertForSequenceClassification` with `.from_pretrained` method with arguments for type of Bert architectures say `bert-base-uncased`; pytorch_model.bin will be downloaded from an s3 ... | 1,561 | 1,567 | 1,567 | NONE | null | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/739/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/739/timeline | completed | null | null | |
https://api.github.com/repos/huggingface/transformers/issues/738 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/738/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/738/comments | https://api.github.com/repos/huggingface/transformers/issues/738/events | https://github.com/huggingface/transformers/issues/738 | 461,845,230 | MDU6SXNzdWU0NjE4NDUyMzA= | 738 | BertTokenizer never_split issue | {
"login": "ardellelee",
"id": 19727258,
"node_id": "MDQ6VXNlcjE5NzI3MjU4",
"avatar_url": "https://avatars.githubusercontent.com/u/19727258?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ardellelee",
"html_url": "https://github.com/ardellelee",
"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 | [] | [
"which version of python do you use in these environments?",
"> which version of python do you use in these environments?\r\n\r\nHi Thomwolf,\r\n\r\nI'm using Python 3.6.8 for all these environments.\r\n\r\n",
"In case it's helpful, I create a gist to include some details of this issue: https://gist.github.com/... | 1,561 | 1,567 | 1,567 | NONE | null | Hi,
I'm using the BertTokenizer to tokenize a piece of text where I use some entity markers to mark the beginning and end of entities, e.g.:
> This was among a batch of paperback [E1] Oxford World [/E1] ' s Classics
I've manually added such entity markers to _vocab file_ and the _never_split_ tuple in _BertTok... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/738/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/738/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/737 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/737/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/737/comments | https://api.github.com/repos/huggingface/transformers/issues/737/events | https://github.com/huggingface/transformers/issues/737 | 461,815,202 | MDU6SXNzdWU0NjE4MTUyMDI= | 737 | gpt-2 model doesn't output hidden states of all layers. | {
"login": "fabrahman",
"id": 22799593,
"node_id": "MDQ6VXNlcjIyNzk5NTkz",
"avatar_url": "https://avatars.githubusercontent.com/u/22799593?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/fabrahman",
"html_url": "https://github.com/fabrahman",
"followers_url": "https://api.github.com/users/... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"Currently not indeed. This option will be in the coming release.",
"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"
] | 1,561 | 1,567 | 1,567 | NONE | null | Using GPT2Model , it seems like it outputs the hidden states of only 1 layer. However, according to code and documentation it is expected to output hidden states features for each layer.
Am I making a mistake?
Thanks for the advise, | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/737/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/737/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/736 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/736/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/736/comments | https://api.github.com/repos/huggingface/transformers/issues/736/events | https://github.com/huggingface/transformers/issues/736 | 461,789,234 | MDU6SXNzdWU0NjE3ODkyMzQ= | 736 | Question regarding crossentropy loss function for BERTMaskedLM | {
"login": "chithangduong",
"id": 22811551,
"node_id": "MDQ6VXNlcjIyODExNTUx",
"avatar_url": "https://avatars.githubusercontent.com/u/22811551?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/chithangduong",
"html_url": "https://github.com/chithangduong",
"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 | [] | [
"30k is ok for a softmax, it's not that much and that because Bert is using a sub-word (open-)vocabulary.\r\n\r\nFull word (and closed-vocabulary) models like word2vec have to handle several 100k words hence the specific speed-ups. They are also older and the computation power available at the time was more constra... | 1,561 | 1,568 | 1,568 | NONE | null | How does BERT handle large number of classes to predict? The number of classes is essentially the vocabulary size which is 30522 for the BERT-base model. When BERT tries to predict a word using CrossEntropy loss, it needs to compute the softmax for a large number of classes.
In shallow approach such as word2vec, n... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/736/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/736/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/735 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/735/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/735/comments | https://api.github.com/repos/huggingface/transformers/issues/735/events | https://github.com/huggingface/transformers/issues/735 | 461,648,786 | MDU6SXNzdWU0NjE2NDg3ODY= | 735 | BERT encoding layer produces same output for all inputs during evaluation | {
"login": "josephvalencia",
"id": 19215694,
"node_id": "MDQ6VXNlcjE5MjE1Njk0",
"avatar_url": "https://avatars.githubusercontent.com/u/19215694?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/josephvalencia",
"html_url": "https://github.com/josephvalencia",
"followers_url": "https://api.gi... | [] | closed | false | null | [] | [
"Unlike #695 and others regarding non-determinism, I am calling model.eval() ",
"Can you share your model initialization code as well?",
"My model is just a slight modification of BertForSequenceClassification for multilabel.\r\n\r\nclass BertForMultiLabelSequenceClassification(BertForSequenceClassification):\r... | 1,561 | 1,695 | 1,562 | NONE | null | I am having issues with differences between the output of the BERT layer during training and evaluation time. I am fine-tuning BertForSequenceClassification, but have traced the problem to the pretrained BertModel. During training, the sequence_output within BertModel.forward() produces sensible output, for example :
... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/735/reactions",
"total_count": 7,
"+1": 4,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 3
} | https://api.github.com/repos/huggingface/transformers/issues/735/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/734 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/734/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/734/comments | https://api.github.com/repos/huggingface/transformers/issues/734/events | https://github.com/huggingface/transformers/issues/734 | 461,342,656 | MDU6SXNzdWU0NjEzNDI2NTY= | 734 | Erroneous Code | {
"login": "LeoLai930603",
"id": 15106070,
"node_id": "MDQ6VXNlcjE1MTA2MDcw",
"avatar_url": "https://avatars.githubusercontent.com/u/15106070?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/LeoLai930603",
"html_url": "https://github.com/LeoLai930603",
"followers_url": "https://api.github.c... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"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"
] | 1,561 | 1,567 | 1,567 | NONE | null | I guess there is a minor mistake in this line.
https://github.com/huggingface/pytorch-pretrained-BERT/blob/80684f6f86c13a89fc1e4feac248ef96b013765c/pytorch_pretrained_bert/modeling_transfo_xl.py#L1385
In `TransfoXLLMHeadModel`, the forward computation requires the target (if available) has the shape [batch_size, ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/734/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/734/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/733 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/733/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/733/comments | https://api.github.com/repos/huggingface/transformers/issues/733/events | https://github.com/huggingface/transformers/pull/733 | 461,222,144 | MDExOlB1bGxSZXF1ZXN0MjkyMTkzMDU5 | 733 | Added option to use multiple workers to create training data | {
"login": "ceremonious",
"id": 6596130,
"node_id": "MDQ6VXNlcjY1OTYxMzA=",
"avatar_url": "https://avatars.githubusercontent.com/u/6596130?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ceremonious",
"html_url": "https://github.com/ceremonious",
"followers_url": "https://api.github.com/us... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/pull/733?src=pr&el=h1) Report\n> Merging [#733](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/pull/733?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/commit/98dc30b21e3df6528d0dd17f0910ffea12b... | 1,561 | 1,562 | 1,562 | CONTRIBUTOR | null | Added a command line argument to allow using a multiprocessing pool to generate training data for all the epochs at once.
The shelve object isn't pickleable, so it can't be used with the Pool | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/733/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/733/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/733",
"html_url": "https://github.com/huggingface/transformers/pull/733",
"diff_url": "https://github.com/huggingface/transformers/pull/733.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/733.patch",
"merged_at": 15623... |
https://api.github.com/repos/huggingface/transformers/issues/732 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/732/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/732/comments | https://api.github.com/repos/huggingface/transformers/issues/732/events | https://github.com/huggingface/transformers/issues/732 | 461,131,491 | MDU6SXNzdWU0NjExMzE0OTE= | 732 | GPT & GPT2: binary classification fails | {
"login": "epsdg",
"id": 42873462,
"node_id": "MDQ6VXNlcjQyODczNDYy",
"avatar_url": "https://avatars.githubusercontent.com/u/42873462?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/epsdg",
"html_url": "https://github.com/epsdg",
"followers_url": "https://api.github.com/users/epsdg/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"
] | 1,561 | 1,567 | 1,567 | NONE | null | Using `OpenAIGPTDoubleHeadsModel` for binary classification fails.
`CrossEntropyLoss`, requires that the logits dim matches num_classes.
If `input_ids.size()` is (batch x 1 x seq_len) (only one copy of the input sequence) but mc_labels are {0, 1} (two classes), the loss fn returns a shape mismatch. The only way ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/732/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/732/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/731 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/731/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/731/comments | https://api.github.com/repos/huggingface/transformers/issues/731/events | https://github.com/huggingface/transformers/pull/731 | 461,122,810 | MDExOlB1bGxSZXF1ZXN0MjkyMTExNDc3 | 731 | merge | {
"login": "Eurus-Holmes",
"id": 34226570,
"node_id": "MDQ6VXNlcjM0MjI2NTcw",
"avatar_url": "https://avatars.githubusercontent.com/u/34226570?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Eurus-Holmes",
"html_url": "https://github.com/Eurus-Holmes",
"followers_url": "https://api.github.c... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/pull/731?src=pr&el=h1) Report\n> Merging [#731](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/pull/731?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/commit/98dc30b21e3df6528d0dd17f0910ffea12b... | 1,561 | 1,561 | 1,561 | NONE | null | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/731/reactions",
"total_count": 2,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 1,
"confused": 0,
"heart": 0,
"rocket": 1,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/731/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/731",
"html_url": "https://github.com/huggingface/transformers/pull/731",
"diff_url": "https://github.com/huggingface/transformers/pull/731.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/731.patch",
"merged_at": null
... | |
https://api.github.com/repos/huggingface/transformers/issues/730 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/730/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/730/comments | https://api.github.com/repos/huggingface/transformers/issues/730/events | https://github.com/huggingface/transformers/issues/730 | 460,948,702 | MDU6SXNzdWU0NjA5NDg3MDI= | 730 | bertForNextSentencePrediction | {
"login": "ankitsharma1999",
"id": 43710585,
"node_id": "MDQ6VXNlcjQzNzEwNTg1",
"avatar_url": "https://avatars.githubusercontent.com/u/43710585?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ankitsharma1999",
"html_url": "https://github.com/ankitsharma1999",
"followers_url": "https://api... | [] | closed | false | null | [] | [
"It's a classification task - is the given sentence the next sentence? It's not going to generate the next sentence for you, as BERT is not a classical language model",
"So, how do I know from these values whether the next sentence should be classified as the next sentence or not?",
"Softmax over it will give y... | 1,561 | 1,567 | 1,561 | NONE | null | I copied the code from [PyTorch's official site](https://pytorch.org/hub/huggingface_pytorch-pretrained-bert_bert/) for `bertForNextSentencePrediction`. I get the next_sent_classif_logits as `tensor([[ 5.2880, -6.0952]])`. How do I get the next sentence from these values?
| {
"url": "https://api.github.com/repos/huggingface/transformers/issues/730/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/730/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/729 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/729/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/729/comments | https://api.github.com/repos/huggingface/transformers/issues/729/events | https://github.com/huggingface/transformers/issues/729 | 460,841,674 | MDU6SXNzdWU0NjA4NDE2NzQ= | 729 | Grover generator support | {
"login": "asafamr",
"id": 5182534,
"node_id": "MDQ6VXNlcjUxODI1MzQ=",
"avatar_url": "https://avatars.githubusercontent.com/u/5182534?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/asafamr",
"html_url": "https://github.com/asafamr",
"followers_url": "https://api.github.com/users/asafamr/... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"Maybe if they open source a model larger than the current GPT-2 large.\r\n\r\nI'm also happy to welcome PRs to port additional models (as long as they are provided with test/doc/example)",
"This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further a... | 1,561 | 1,567 | 1,567 | NONE | null | Grover released their trained model:
https://github.com/rowanz/grover
I think it should be similar to GPT-2 large. Any plans to support it? | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/729/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/729/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/728 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/728/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/728/comments | https://api.github.com/repos/huggingface/transformers/issues/728/events | https://github.com/huggingface/transformers/issues/728 | 460,836,757 | MDU6SXNzdWU0NjA4MzY3NTc= | 728 | UnicodeDecodeError: | {
"login": "ZhaoxinRuc",
"id": 38198098,
"node_id": "MDQ6VXNlcjM4MTk4MDk4",
"avatar_url": "https://avatars.githubusercontent.com/u/38198098?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ZhaoxinRuc",
"html_url": "https://github.com/ZhaoxinRuc",
"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 | [] | [
"@ZhaoxinRuc \r\n\r\nI'm assuming that some data in your `vocab.txt` file contains bad characters which your `codecs.py` can't decode properly. When I downloaded the pre-trained weights folder, it came with `vocab.txt` which didn't had this issue. Check if you've downloaded a wrong version or some how changed conte... | 1,561 | 1,567 | 1,567 | NONE | null | Traceback (most recent call last):
File "run_classifier_br.py", line 1061, in <module>
main()
File "run_classifier_br.py", line 772, in main
tokenizer = BertTokenizer.from_pretrained(args.bert_model, do_lower_case=args.do_lower_case)
File "/home/luwei/anaconda3/lib/python3.6/site-packages/pytorch_pre... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/728/reactions",
"total_count": 8,
"+1": 7,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 1
} | https://api.github.com/repos/huggingface/transformers/issues/728/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/727 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/727/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/727/comments | https://api.github.com/repos/huggingface/transformers/issues/727/events | https://github.com/huggingface/transformers/issues/727 | 460,747,363 | MDU6SXNzdWU0NjA3NDczNjM= | 727 | Poor Training and evaluation accuracy even with low loss | {
"login": "amithadiraju1694",
"id": 23751321,
"node_id": "MDQ6VXNlcjIzNzUxMzIx",
"avatar_url": "https://avatars.githubusercontent.com/u/23751321?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/amithadiraju1694",
"html_url": "https://github.com/amithadiraju1694",
"followers_url": "https://... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"@amit8121 can you just sent the run_classifier.py python command that you are using in terminal to run this model",
"@himanshututeja1998 \r\n\r\nThanks for the response. This is a similar command to what I used for `run_classifier.py`\r\n\r\n```\r\nexport BERT_BASE_DIR=./path/to/uncasedweightsfolder/\r\n\r\npyt... | 1,561 | 1,567 | 1,567 | NONE | null | @spolu @cynthia @thomwolf @davidefiocco
I initially used command line arguments to run the `run_classifier.py`, using `cola` as a task, for `Sequence Classification` on a custom data set, I was able to execute and get the results , but they were very poor: an evaluation accuracy of 0.0 and loss of close to .9.
... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/727/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/727/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/726 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/726/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/726/comments | https://api.github.com/repos/huggingface/transformers/issues/726/events | https://github.com/huggingface/transformers/issues/726 | 460,725,157 | MDU6SXNzdWU0NjA3MjUxNTc= | 726 | Examples does not work with apex optimizers | {
"login": "tc-yu",
"id": 33182836,
"node_id": "MDQ6VXNlcjMzMTgyODM2",
"avatar_url": "https://avatars.githubusercontent.com/u/33182836?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/tc-yu",
"html_url": "https://github.com/tc-yu",
"followers_url": "https://api.github.com/users/tc-yu/follow... | [] | closed | false | null | [] | [
"You can use # on line 316 and it can solved this problem."
] | 1,561 | 1,563 | 1,563 | NONE | null | Under fp16 option, optimizer is replaced by one from apex, which does not have the attribute ```get_lr()``` .
https://github.com/huggingface/pytorch-pretrained-BERT/blob/98dc30b21e3df6528d0dd17f0910ffea12bc0f33/examples/run_squad.py#L315-L317
Should be able to reproduce the error by running the example [here](htt... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/726/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/726/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/725 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/725/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/725/comments | https://api.github.com/repos/huggingface/transformers/issues/725/events | https://github.com/huggingface/transformers/issues/725 | 460,533,863 | MDU6SXNzdWU0NjA1MzM4NjM= | 725 | BERT Input size reduced to half in forward function | {
"login": "xinsu626",
"id": 30940128,
"node_id": "MDQ6VXNlcjMwOTQwMTI4",
"avatar_url": "https://avatars.githubusercontent.com/u/30940128?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/xinsu626",
"html_url": "https://github.com/xinsu626",
"followers_url": "https://api.github.com/users/xin... | [] | closed | false | null | [] | [
"Maybe you have 2 GPUs?",
"@thomwolf Thanks a lot. I forgot I was running on two gpus. \r\n"
] | 1,561 | 1,561 | 1,561 | NONE | null | I was trying to modify your BertForSequenceClassification class for long sequence classification. Like below:
```
class MyBertForSequenceClassification(BertPreTrainedModel):
def __init__(self, config, num_labels=2, output_attentions=False):
super(MyBertForSequenceClassification, self).__init__(config)... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/725/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/725/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/724 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/724/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/724/comments | https://api.github.com/repos/huggingface/transformers/issues/724/events | https://github.com/huggingface/transformers/pull/724 | 460,461,332 | MDExOlB1bGxSZXF1ZXN0MjkxNTg3MTM3 | 724 | fixing bugs in load_rocstories_dataset in run_openai_gpt.py | {
"login": "sajidrahman",
"id": 4258481,
"node_id": "MDQ6VXNlcjQyNTg0ODE=",
"avatar_url": "https://avatars.githubusercontent.com/u/4258481?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sajidrahman",
"html_url": "https://github.com/sajidrahman",
"followers_url": "https://api.github.com/us... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/pull/724?src=pr&el=h1) Report\n> Merging [#724](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/pull/724?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/commit/98dc30b21e3df6528d0dd17f0910ffea12b... | 1,561 | 1,562 | 1,562 | NONE | null | The csv reader requires a delimiter argument to read .tsv file in the given example dataset. I've also added link for the dataset and provided a sample eval results in comments. Also, the eval dataset needs to be different from the training dataset, which I've also fixed in the given command to run this script. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/724/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/724/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/724",
"html_url": "https://github.com/huggingface/transformers/pull/724",
"diff_url": "https://github.com/huggingface/transformers/pull/724.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/724.patch",
"merged_at": null
... |
https://api.github.com/repos/huggingface/transformers/issues/723 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/723/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/723/comments | https://api.github.com/repos/huggingface/transformers/issues/723/events | https://github.com/huggingface/transformers/pull/723 | 460,283,950 | MDExOlB1bGxSZXF1ZXN0MjkxNDQzMTYz | 723 | Update Adam optimizer to follow pytorch convention for betas parameter (#510) | {
"login": "tonianelope",
"id": 23743176,
"node_id": "MDQ6VXNlcjIzNzQzMTc2",
"avatar_url": "https://avatars.githubusercontent.com/u/23743176?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/tonianelope",
"html_url": "https://github.com/tonianelope",
"followers_url": "https://api.github.com/... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/pull/723?src=pr&el=h1) Report\n> Merging [#723](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/pull/723?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/commit/98dc30b21e3df6528d0dd17f0910ffea12b... | 1,561 | 1,561 | 1,561 | CONTRIBUTOR | null | see #510
Update optimiser to follow pytorch convention ([Adam Optimiser](https://pytorch.org/docs/stable/optim.html#torch.optim.Adam)) instead of tensorflow, to allow for better integration with other pytorch libraries and frameworks. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/723/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/723/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/723",
"html_url": "https://github.com/huggingface/transformers/pull/723",
"diff_url": "https://github.com/huggingface/transformers/pull/723.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/723.patch",
"merged_at": 15617... |
https://api.github.com/repos/huggingface/transformers/issues/722 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/722/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/722/comments | https://api.github.com/repos/huggingface/transformers/issues/722/events | https://github.com/huggingface/transformers/issues/722 | 460,144,005 | MDU6SXNzdWU0NjAxNDQwMDU= | 722 | low accuracy when fine tuning for the MRPC task with large model | {
"login": "syu0000",
"id": 52182461,
"node_id": "MDQ6VXNlcjUyMTgyNDYx",
"avatar_url": "https://avatars.githubusercontent.com/u/52182461?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/syu0000",
"html_url": "https://github.com/syu0000",
"followers_url": "https://api.github.com/users/syu000... | [
{
"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 batch size will be 8 times smaller with only one GPU, increase it by a factor of 8 using gradient accumulation, e.g. `--train_batch_size 96 --gradient_accumulation_steps 8`",
"Thank you for your help. However, when I use this command: \r\npython run_classifier.py --bert_model bert-large-uncased-whole-word... | 1,561 | 1,567 | 1,567 | NONE | null | I noticed that on the website you said:"Here is an example using distributed training on 8 V100 GPUs and Bert Whole Word Masking model to reach a F1 > 92 on MRPC." However, when I fine tuned the model with max_sequence_length=128 and batch_size=12 on a single 11G GPU, it gives the accuracy of 0.68.
acc = 0.68382352941... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/722/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/722/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/721 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/721/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/721/comments | https://api.github.com/repos/huggingface/transformers/issues/721/events | https://github.com/huggingface/transformers/issues/721 | 459,949,826 | MDU6SXNzdWU0NTk5NDk4MjY= | 721 | Usual loss when pretraining? | {
"login": "PedroUria",
"id": 43831167,
"node_id": "MDQ6VXNlcjQzODMxMTY3",
"avatar_url": "https://avatars.githubusercontent.com/u/43831167?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/PedroUria",
"html_url": "https://github.com/PedroUria",
"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 | [] | [
"@PedroUria \r\n\r\nWe're in a similar boat as you are. In our case, the problem was with accuracy though.\r\n\r\nWe used `BertForSequenceClassification` on a multi-label classification task. We've actually written a similar version of `BertForSequenceClassification` as written in `models.py` of this repository c... | 1,561 | 1,568 | 1,568 | NONE | null | We are pretraining on our own corpus using the `pregenerate_training_data.py` and `finetune_on_pregenerated.py` scripts.
The input text to the first script follows the same format as `sample_text.txt` on the samples folder, and contains about 515000 lines of text. We run `finetune_on_pregenerated.py` for 60 epochs w... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/721/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/721/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/720 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/720/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/720/comments | https://api.github.com/repos/huggingface/transformers/issues/720/events | https://github.com/huggingface/transformers/issues/720 | 459,909,407 | MDU6SXNzdWU0NTk5MDk0MDc= | 720 | Import Error: cannot import name 'warmup_linear' | {
"login": "GBR-613",
"id": 17687869,
"node_id": "MDQ6VXNlcjE3Njg3ODY5",
"avatar_url": "https://avatars.githubusercontent.com/u/17687869?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/GBR-613",
"html_url": "https://github.com/GBR-613",
"followers_url": "https://api.github.com/users/GBR-61... | [] | closed | false | null | [] | [
"I probably was wrong and the issue was supposed to be fixed with PR #518... Anyway, the latest was merged a while ago and did not help. ",
"You have to wait for the next release or use the master branch",
"@thomwolf ,\r\n> You have to wait for the next release or use the master branch\r\n\r\nI cloned the git r... | 1,561 | 1,604 | 1,564 | NONE | null | I get the following error:
```
File "/Users/gregory/PROJECTS/MyML/MLClassification/TrainAndTest/Models/controller.py", line 11, in <module>
from Models.bert import BertModel
File "/Users/gregory/PROJECTS/MyML/MLClassification/TrainAndTest/Models/bert.py", line 9, in <module>
from pytorch_pretrained_b... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/720/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/720/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/719 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/719/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/719/comments | https://api.github.com/repos/huggingface/transformers/issues/719/events | https://github.com/huggingface/transformers/issues/719 | 459,734,720 | MDU6SXNzdWU0NTk3MzQ3MjA= | 719 | Embedding and predictions in one forward pass | {
"login": "lcswillems",
"id": 5437552,
"node_id": "MDQ6VXNlcjU0Mzc1NTI=",
"avatar_url": "https://avatars.githubusercontent.com/u/5437552?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lcswillems",
"html_url": "https://github.com/lcswillems",
"followers_url": "https://api.github.com/users... | [] | closed | false | null | [] | [
"Yes, just make your own PyTorch model taking inspiration from BertModel and BertForMaskedLM.\r\nIf you sub-class `BertPreTrainedModel`, you'll be able to load the pretrained weights using the `from_pretrained()` method",
"Okay, thank you :) "
] | 1,561 | 1,561 | 1,561 | NONE | null | Is it possible to mix `BertModel` and `BertForMaskedLM`? i.e. is it possible to get the embedding and the predictions in one forward pass? | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/719/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/719/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/718 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/718/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/718/comments | https://api.github.com/repos/huggingface/transformers/issues/718/events | https://github.com/huggingface/transformers/pull/718 | 459,600,873 | MDExOlB1bGxSZXF1ZXN0MjkwOTA2MDYy | 718 | Incorrect docstring for BertForMaskedLM | {
"login": "Rocketknight1",
"id": 12866554,
"node_id": "MDQ6VXNlcjEyODY2NTU0",
"avatar_url": "https://avatars.githubusercontent.com/u/12866554?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Rocketknight1",
"html_url": "https://github.com/Rocketknight1",
"followers_url": "https://api.githu... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/pull/718?src=pr&el=h1) Report\n> Merging [#718](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/pull/718?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/commit/98dc30b21e3df6528d0dd17f0910ffea12b... | 1,561 | 1,561 | 1,561 | MEMBER | null | The docstring for the head_mask argument to the BertForMaskedLM class is repeated and one is incorrect - I presume it's just a copy-paste mistake. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/718/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/718/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/718",
"html_url": "https://github.com/huggingface/transformers/pull/718",
"diff_url": "https://github.com/huggingface/transformers/pull/718.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/718.patch",
"merged_at": 15617... |
https://api.github.com/repos/huggingface/transformers/issues/717 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/717/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/717/comments | https://api.github.com/repos/huggingface/transformers/issues/717/events | https://github.com/huggingface/transformers/issues/717 | 459,520,306 | MDU6SXNzdWU0NTk1MjAzMDY= | 717 | BPE vocab | {
"login": "sashank06",
"id": 8636933,
"node_id": "MDQ6VXNlcjg2MzY5MzM=",
"avatar_url": "https://avatars.githubusercontent.com/u/8636933?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sashank06",
"html_url": "https://github.com/sashank06",
"followers_url": "https://api.github.com/users/sa... | [
{
"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 | [] | [
"Not speaking for the core developers, but `pytorch-pretrained-BERT` supports it, because:\r\n\r\n* GPT-1 use BPE, see code [here](https://github.com/huggingface/pytorch-pretrained-BERT/blob/master/pytorch_pretrained_bert/tokenization_openai.py#L73)\r\n* GPT-2 use BPE on byte level, see code [here](https://github.c... | 1,561 | 1,567 | 1,567 | NONE | null | Do you guys have the functionality to support BPE with the models? | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/717/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/717/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/716 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/716/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/716/comments | https://api.github.com/repos/huggingface/transformers/issues/716/events | https://github.com/huggingface/transformers/pull/716 | 459,515,735 | MDExOlB1bGxSZXF1ZXN0MjkwODUwODYx | 716 | Add tie_weights to XLNetForSequenceClassification | {
"login": "Strideradu",
"id": 9002118,
"node_id": "MDQ6VXNlcjkwMDIxMTg=",
"avatar_url": "https://avatars.githubusercontent.com/u/9002118?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Strideradu",
"html_url": "https://github.com/Strideradu",
"followers_url": "https://api.github.com/users... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/pull/716?src=pr&el=h1) Report\n> Merging [#716](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/pull/716?src=pr&el=desc) into [xlnet](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/commit/c946bb51a61f67b0c9eaae1c9cf6f164a774... | 1,561 | 1,561 | 1,561 | NONE | null | XLNetForSequenceClassification doesn't have tie_weights() but initialization will call it, or we can made a function in XLNetPretrainedModel? | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/716/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/716/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/716",
"html_url": "https://github.com/huggingface/transformers/pull/716",
"diff_url": "https://github.com/huggingface/transformers/pull/716.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/716.patch",
"merged_at": null
... |
https://api.github.com/repos/huggingface/transformers/issues/715 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/715/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/715/comments | https://api.github.com/repos/huggingface/transformers/issues/715/events | https://github.com/huggingface/transformers/pull/715 | 459,477,532 | MDExOlB1bGxSZXF1ZXN0MjkwODI1Nzc5 | 715 | Include a reference for LM finetuning | {
"login": "Rocketknight1",
"id": 12866554,
"node_id": "MDQ6VXNlcjEyODY2NTU0",
"avatar_url": "https://avatars.githubusercontent.com/u/12866554?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Rocketknight1",
"html_url": "https://github.com/Rocketknight1",
"followers_url": "https://api.githu... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/pull/715?src=pr&el=h1) Report\n> Merging [#715](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/pull/715?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/commit/c304593d8fa93f25febe1458c63497a8467... | 1,561 | 1,561 | 1,561 | MEMBER | null | @lopuhin recently made me aware of a published paper covering domain fine-tuning of BERT models, so I added a reference to the LM finetuning README. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/715/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/715/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/715",
"html_url": "https://github.com/huggingface/transformers/pull/715",
"diff_url": "https://github.com/huggingface/transformers/pull/715.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/715.patch",
"merged_at": 15612... |
https://api.github.com/repos/huggingface/transformers/issues/714 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/714/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/714/comments | https://api.github.com/repos/huggingface/transformers/issues/714/events | https://github.com/huggingface/transformers/pull/714 | 459,465,533 | MDExOlB1bGxSZXF1ZXN0MjkwODE4MDg2 | 714 | Correct a broken link on README | {
"login": "changukshin",
"id": 6247953,
"node_id": "MDQ6VXNlcjYyNDc5NTM=",
"avatar_url": "https://avatars.githubusercontent.com/u/6247953?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/changukshin",
"html_url": "https://github.com/changukshin",
"followers_url": "https://api.github.com/us... | [] | closed | false | null | [] | [
"# [Codecov](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/pull/714?src=pr&el=h1) Report\n> Merging [#714](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/pull/714?src=pr&el=desc) into [master](https://codecov.io/gh/huggingface/pytorch-pretrained-BERT/commit/c304593d8fa93f25febe1458c63497a8467... | 1,561 | 1,561 | 1,561 | CONTRIBUTOR | null | I've correct a broken link and its contexts on README. | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/714/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/714/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/714",
"html_url": "https://github.com/huggingface/transformers/pull/714",
"diff_url": "https://github.com/huggingface/transformers/pull/714.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/714.patch",
"merged_at": 15612... |
https://api.github.com/repos/huggingface/transformers/issues/713 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/713/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/713/comments | https://api.github.com/repos/huggingface/transformers/issues/713/events | https://github.com/huggingface/transformers/issues/713 | 459,452,941 | MDU6SXNzdWU0NTk0NTI5NDE= | 713 | TypeError: expand_as() takes 1 positional argument but 5 were given | {
"login": "anxingle",
"id": 8489818,
"node_id": "MDQ6VXNlcjg0ODk4MTg=",
"avatar_url": "https://avatars.githubusercontent.com/u/8489818?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/anxingle",
"html_url": "https://github.com/anxingle",
"followers_url": "https://api.github.com/users/anxin... | [
{
"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 | [] | [
"Oh yes, this will be fixed in the coming PR #711.\r\nHead mask is an option to explore the model internals, it's not for production.\r\nSee the `bertology.py` example script.",
"This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occu... | 1,561 | 1,566 | 1,566 | NONE | null | [Model.py](https://github.com/huggingface/pytorch-pretrained-BERT/blob/master/pytorch_pretrained_bert/modeling.py) line [870](https://github.com/huggingface/pytorch-pretrained-BERT/blob/master/pytorch_pretrained_bert/modeling.py#L870):
`head_mask = head_mask.expand_as(self.config.num_hidden_layers, -1, -1, -1, -1)` ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/713/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/713/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/712 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/712/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/712/comments | https://api.github.com/repos/huggingface/transformers/issues/712/events | https://github.com/huggingface/transformers/issues/712 | 459,345,851 | MDU6SXNzdWU0NTkzNDU4NTE= | 712 | BERT Tokenizer not working! Failed to load the bert-base-uncased model. | {
"login": "Raghavendra15",
"id": 7957331,
"node_id": "MDQ6VXNlcjc5NTczMzE=",
"avatar_url": "https://avatars.githubusercontent.com/u/7957331?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Raghavendra15",
"html_url": "https://github.com/Raghavendra15",
"followers_url": "https://api.github.... | [
{
"id": 1314768611,
"node_id": "MDU6TGFiZWwxMzE0NzY4NjEx",
"url": "https://api.github.com/repos/huggingface/transformers/labels/wontfix",
"name": "wontfix",
"color": "ffffff",
"default": true,
"description": null
}
] | closed | false | null | [] | [
"Do you have a good internet connection? The error messages will be improved in the coming release but usually, this comes from the library not being able to reach AWS S3 servers to download the pretrained weights.",
"@thomwolf Thank you so much for your quick response! I followed your advice to people on other p... | 1,561 | 1,593 | 1,571 | NONE | null | The sentence that is being tokenized is: "Weather: Summer’s Finally Here. So Where Is It?"
But it gives the following error:
Error message:
AttributeError Traceback (most recent call last)
<ipython-input-78-c51eef61e2b9> in <module>
----> 1 correct_pairs = convert_sentence_pair(df_ful... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/712/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/712/timeline | completed | null | null |
https://api.github.com/repos/huggingface/transformers/issues/711 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/711/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/711/comments | https://api.github.com/repos/huggingface/transformers/issues/711/events | https://github.com/huggingface/transformers/pull/711 | 459,282,406 | MDExOlB1bGxSZXF1ZXN0MjkwNjc1OTEy | 711 | PyTorch-Transformers 1.0 - w. XLNet and XLM model - Standard API - Torchscript compatibility | {
"login": "thomwolf",
"id": 7353373,
"node_id": "MDQ6VXNlcjczNTMzNzM=",
"avatar_url": "https://avatars.githubusercontent.com/u/7353373?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/thomwolf",
"html_url": "https://github.com/thomwolf",
"followers_url": "https://api.github.com/users/thomw... | [] | closed | false | null | [] | [
"@thomwolf I could get the `XLNetLMHeadModel` running, but I have some issues with the \"normal\" `XLNetModel` implementation:\r\n\r\n```python\r\nimport torch\r\nfrom pytorch_pretrained_bert import XLNetTokenizer, XLNetModel\r\n\r\nimport logging\r\nlogging.basicConfig(level=logging.INFO)\r\n\r\ntokenizer = XLNetT... | 1,561 | 1,576 | 1,563 | MEMBER | null | Current status:
- [x] model with commented code and pretrained loading logic
- [x] tokenizer
- [x] tests for model and tokenizer
- [x] checking standard deviation of hidden states with TF model is ok (max dev btw 1e-4 & 1e-5 until last layer, last layer 1e-3, higher than bert but should be ok, investigated this in ... | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/711/reactions",
"total_count": 79,
"+1": 40,
"-1": 0,
"laugh": 0,
"hooray": 13,
"confused": 0,
"heart": 0,
"rocket": 15,
"eyes": 11
} | https://api.github.com/repos/huggingface/transformers/issues/711/timeline | null | false | {
"url": "https://api.github.com/repos/huggingface/transformers/pulls/711",
"html_url": "https://github.com/huggingface/transformers/pull/711",
"diff_url": "https://github.com/huggingface/transformers/pull/711.diff",
"patch_url": "https://github.com/huggingface/transformers/pull/711.patch",
"merged_at": 15632... |
https://api.github.com/repos/huggingface/transformers/issues/710 | https://api.github.com/repos/huggingface/transformers | https://api.github.com/repos/huggingface/transformers/issues/710/labels{/name} | https://api.github.com/repos/huggingface/transformers/issues/710/comments | https://api.github.com/repos/huggingface/transformers/issues/710/events | https://github.com/huggingface/transformers/issues/710 | 459,174,943 | MDU6SXNzdWU0NTkxNzQ5NDM= | 710 | A way to increase input length limitation? | {
"login": "lcswillems",
"id": 5437552,
"node_id": "MDQ6VXNlcjU0Mzc1NTI=",
"avatar_url": "https://avatars.githubusercontent.com/u/5437552?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lcswillems",
"html_url": "https://github.com/lcswillems",
"followers_url": "https://api.github.com/users... | [] | closed | false | null | [] | [
"No way as far as I can tell, this is a fundamental limitation for absolute position pre-trained models (i.e. BERT, GPT, GPT-2)",
"Okay thank you!"
] | 1,561 | 1,561 | 1,561 | NONE | null | Hi,
Is there a way to increase input length limitation of 512 tokens?
Maybe something to change in the code? | {
"url": "https://api.github.com/repos/huggingface/transformers/issues/710/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/huggingface/transformers/issues/710/timeline | completed | null | null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.