DanielPFlorian/Transformers-Github-Issues · Datasets at Hugging Face

https://api.github.com/repos/huggingface/transformers/issues/21990

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21990/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21990/comments

https://api.github.com/repos/huggingface/transformers/issues/21990/events

https://github.com/huggingface/transformers/pull/21990

1,613,066,771

PR_kwDOCUB6oc5LdIO4

21,990

Add Image Completion Transformer (ICT)

{ "login": "sheonhan", "id": 4163701, "node_id": "MDQ6VXNlcjQxNjM3MDE=", "avatar_url": "https://avatars.githubusercontent.com/u/4163701?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sheonhan", "html_url": "https://github.com/sheonhan", "followers_url": "https://api.github.com/users/sheonhan/followers", "following_url": "https://api.github.com/users/sheonhan/following{/other_user}", "gists_url": "https://api.github.com/users/sheonhan/gists{/gist_id}", "starred_url": "https://api.github.com/users/sheonhan/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sheonhan/subscriptions", "organizations_url": "https://api.github.com/users/sheonhan/orgs", "repos_url": "https://api.github.com/users/sheonhan/repos", "events_url": "https://api.github.com/users/sheonhan/events{/privacy}", "received_events_url": "https://api.github.com/users/sheonhan/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/transformers/pr_21990). All of your documentation changes will be reflected on that endpoint.", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md) are likely to be ignored." ]

1,678

1,689

CONTRIBUTOR

null

# What does this PR do?   Fixes # (issue) ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [ ] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [ ] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [ ] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [ ] Did you write any new necessary tests? ## Who can review? Anyone in the community is free to review the PR once the tests have passed. Feel free to tag members/contributors who may be interested in your PR.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21990/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21990/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21990", "html_url": "https://github.com/huggingface/transformers/pull/21990", "diff_url": "https://github.com/huggingface/transformers/pull/21990.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21990.patch", "merged_at": null }

https://api.github.com/repos/huggingface/transformers/issues/21989

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21989/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21989/comments

https://api.github.com/repos/huggingface/transformers/issues/21989/events

https://github.com/huggingface/transformers/issues/21989

1,612,715,039

I_kwDOCUB6oc5gIBQf

21,989

RuntimeError: "LayerNormKernelImpl" not implemented for 'Half'

{ "login": "realSAH", "id": 98207838, "node_id": "U_kgDOBdqIXg", "avatar_url": "https://avatars.githubusercontent.com/u/98207838?v=4", "gravatar_id": "", "url": "https://api.github.com/users/realSAH", "html_url": "https://github.com/realSAH", "followers_url": "https://api.github.com/users/realSAH/followers", "following_url": "https://api.github.com/users/realSAH/following{/other_user}", "gists_url": "https://api.github.com/users/realSAH/gists{/gist_id}", "starred_url": "https://api.github.com/users/realSAH/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/realSAH/subscriptions", "organizations_url": "https://api.github.com/users/realSAH/orgs", "repos_url": "https://api.github.com/users/realSAH/repos", "events_url": "https://api.github.com/users/realSAH/events{/privacy}", "received_events_url": "https://api.github.com/users/realSAH/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "You need to execute a model loaded in half precision on a GPU, the operations are not implemented in half on the CPU.", "@sgugger Then how come that this example works on cpu?\r\n\r\n```\r\nfrom transformers import GPTJForCausalLM\r\nimport torch\r\n\r\nmodel = GPTJForCausalLM.from_pretrained(\r\n \"EleutherAI/gpt-j-6B\", revision=\"float16\", torch_dtype=torch.float16, low_cpu_mem_usage=True\r\n)\r\n```\r\n", "What code are you using exactly to get the error?\r\n```\r\nimport torch\r\nfrom transformers import AutoModelForCausalLM\r\n\r\nmodel = AutoModelForCausalLM.from_pretrained('bigscience/bloomz-7b1', torch_dtype=torch.float16)\r\n```\r\nworks perfectly fine.", "@sgugger \r\n\r\nYes, it **loads up perfectly fine** but if you proceed to build the pipeline and generate text, you get the `Half` implementation error. \r\nI just tried your code again\r\n\r\n```\r\n\r\nimport torch\r\nfrom transformers import AutoModelForCausalLM, pipeline\r\nmodel = AutoModelForCausalLM.from_pretrained('bigscience/bloomz-7b1', torch_dtype=torch.float16, low_cpu_mem_usage=True)\r\ng = pipeline(task='text-generation', model=model, tokenizer='bigscience/bloomz-7b1')\r\ng(\"Hi, \")\r\n\r\n```\r\n\r\n\r\nI got this traceback:\r\n\r\n\r\n```\r\n\r\n\r\nIn [1]:\r\n ...: import torch\r\n ...: from transformers import AutoModelForCausalLM, pipeline\r\n ...: model = AutoModelForCausalLM.from_pretrained('bigscience/bloomz-7b1', torch_dtype=torch.float16, low_cpu_mem_usage=True)\r\n ...: g = pipeline(task='text-generation', model=model, tokenizer='bigscience/bloomz-7b1')\r\n ...: g(\"Hi, \")\r\n ...:\r\nC:\\Users\\aalsaf01\\venvs\\nlp\\lib\\site-packages\\transformers\\generation\\utils.py:1273: UserWarning: Neither `max_length` nor `max_new_tokens` has been set, `max\r\n_length` will default to 20 (`generation_config.max_length`). Controlling `max_length` via the config is deprecated and `max_length` will be removed from the\r\nconfig in v5 of Transformers -- we recommend using `max_new_tokens` to control the maximum length of the generation.\r\n warnings.warn(\r\n╭─────────────────────────────── Traceback (most recent call last) ────────────────────────────────╮\r\n│ in <module> │\r\n│ │\r\n│ C:\\Users\\aalsaf01\\venvs\\nlp\\lib\\site-packages\\transformers\\pipelines\\text_generation.py:210 in │\r\n│ __call__ │\r\n│ │\r\n│ 207 │ │ │ - **generated_token_ids** (`torch.Tensor` or `tf.Tensor`, present when `retu │\r\n│ 208 │ │ │ ids of the generated text. │\r\n│ 209 │ │ \"\"\" │\r\n│ ❱ 210 │ │ return super().__call__(text_inputs, **kwargs) │\r\n│ 211 │ │\r\n│ 212 │ def preprocess(self, prompt_text, prefix=\"\", handle_long_generation=None, **generate │\r\n│ 213 │ │ inputs = self.tokenizer( │\r\n│ │\r\n│ C:\\Users\\aalsaf01\\venvs\\nlp\\lib\\site-packages\\transformers\\pipelines\\base.py:1084 in __call__ │\r\n│ │\r\n│ 1081 │ │ │ │ ) │\r\n│ 1082 │ │ │ ) │\r\n│ 1083 │ │ else: │\r\n│ ❱ 1084 │ │ │ return self.run_single(inputs, preprocess_params, forward_params, postproces │\r\n│ 1085 │ │\r\n│ 1086 │ def run_multi(self, inputs, preprocess_params, forward_params, postprocess_params): │\r\n│ 1087 │ │ return [self.run_single(item, preprocess_params, forward_params, postprocess_par │\r\n│ │\r\n│ C:\\Users\\aalsaf01\\venvs\\nlp\\lib\\site-packages\\transformers\\pipelines\\base.py:1091 in run_single │\r\n│ │\r\n│ 1088 │ │\r\n│ 1089 │ def run_single(self, inputs, preprocess_params, forward_params, postprocess_params): │\r\n│ 1090 │ │ model_inputs = self.preprocess(inputs, **preprocess_params) │\r\n│ ❱ 1091 │ │ model_outputs = self.forward(model_inputs, **forward_params) │\r\n│ 1092 │ │ outputs = self.postprocess(model_outputs, **postprocess_params) │\r\n│ 1093 │ │ return outputs │\r\n│ 1094 │\r\n│ │\r\n│ C:\\Users\\aalsaf01\\venvs\\nlp\\lib\\site-packages\\transformers\\pipelines\\base.py:992 in forward │\r\n│ │\r\n│ 989 │ │ │ │ inference_context = self.get_inference_context() │\r\n│ 990 │ │ │ │ with inference_context(): │\r\n│ 991 │ │ │ │ │ model_inputs = self._ensure_tensor_on_device(model_inputs, device=se │\r\n│ ❱ 992 │ │ │ │ │ model_outputs = self._forward(model_inputs, **forward_params) │\r\n│ 993 │ │ │ │ │ model_outputs = self._ensure_tensor_on_device(model_outputs, device= │\r\n│ 994 │ │ │ else: │\r\n│ 995 │ │ │ │ raise ValueError(f\"Framework {self.framework} is not supported\") │\r\n│ │\r\n│ C:\\Users\\aalsaf01\\venvs\\nlp\\lib\\site-packages\\transformers\\pipelines\\text_generation.py:252 in │\r\n│ _forward │\r\n│ │\r\n│ 249 │ │ │ in_b = input_ids.shape[0] │\r\n│ 250 │ │ prompt_text = model_inputs.pop(\"prompt_text\") │\r\n│ 251 │ │ # BS x SL │\r\n│ ❱ 252 │ │ generated_sequence = self.model.generate(input_ids=input_ids, attention_mask=att │\r\n│ 253 │ │ out_b = generated_sequence.shape[0] │\r\n│ 254 │ │ if self.framework == \"pt\": │\r\n│ 255 │ │ │ generated_sequence = generated_sequence.reshape(in_b, out_b // in_b, *genera │\r\n│ │\r\n│ C:\\Users\\aalsaf01\\venvs\\nlp\\lib\\site-packages\\torch\\autograd\\grad_mode.py:27 in decorate_context │\r\n│ │\r\n│ 24 │ │ @functools.wraps(func) │\r\n│ 25 │ │ def decorate_context(*args, **kwargs): │\r\n│ 26 │ │ │ with self.clone(): │\r\n│ ❱ 27 │ │ │ │ return func(*args, **kwargs) │\r\n│ 28 │ │ return cast(F, decorate_context) │\r\n│ 29 │ │\r\n│ 30 │ def _wrap_generator(self, func): │\r\n│ │\r\n│ C:\\Users\\aalsaf01\\venvs\\nlp\\lib\\site-packages\\transformers\\generation\\utils.py:1391 in generate │\r\n│ │\r\n│ 1388 │ │ │ │ ) │\r\n│ 1389 │ │ │ │\r\n│ 1390 │ │ │ # 11. run greedy search │\r\n│ ❱ 1391 │ │ │ return self.greedy_search( │\r\n│ 1392 │ │ │ │ input_ids, │\r\n│ 1393 │ │ │ │ logits_processor=logits_processor, │\r\n│ 1394 │ │ │ │ stopping_criteria=stopping_criteria, │\r\n│ │\r\n│ C:\\Users\\aalsaf01\\venvs\\nlp\\lib\\site-packages\\transformers\\generation\\utils.py:2179 in │\r\n│ greedy_search │\r\n│ │\r\n│ 2176 │ │ │ model_inputs = self.prepare_inputs_for_generation(input_ids, **model_kwargs) │\r\n│ 2177 │ │ │ │\r\n│ 2178 │ │ │ # forward pass to get next token │\r\n│ ❱ 2179 │ │ │ outputs = self( │\r\n│ 2180 │ │ │ │ **model_inputs, │\r\n│ 2181 │ │ │ │ return_dict=True, │\r\n│ 2182 │ │ │ │ output_attentions=output_attentions, │\r\n│ │\r\n│ C:\\Users\\aalsaf01\\venvs\\nlp\\lib\\site-packages\\torch\\nn\\modules\\module.py:1194 in _call_impl │\r\n│ │\r\n│ 1191 │ │ # this function, and just call forward. │\r\n│ 1192 │ │ if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks o │\r\n│ 1193 │ │ │ │ or _global_forward_hooks or _global_forward_pre_hooks): │\r\n│ ❱ 1194 │ │ │ return forward_call(*input, **kwargs) │\r\n│ 1195 │ │ # Do not call functions when jit is used │\r\n│ 1196 │ │ full_backward_hooks, non_full_backward_hooks = [], [] │\r\n│ 1197 │ │ if self._backward_hooks or _global_backward_hooks: │\r\n│ │\r\n│ C:\\Users\\aalsaf01\\venvs\\nlp\\lib\\site-packages\\transformers\\models\\bloom\\modeling_bloom.py:900 in │\r\n│ forward │\r\n│ │\r\n│ 897 │ │ │\r\n│ 898 │ │ return_dict = return_dict if return_dict is not None else self.config.use_return │\r\n│ 899 │ │ │\r\n│ ❱ 900 │ │ transformer_outputs = self.transformer( │\r\n│ 901 │ │ │ input_ids, │\r\n│ 902 │ │ │ past_key_values=past_key_values, │\r\n│ 903 │ │ │ attention_mask=attention_mask, │\r\n│ │\r\n│ C:\\Users\\aalsaf01\\venvs\\nlp\\lib\\site-packages\\torch\\nn\\modules\\module.py:1194 in _call_impl │\r\n│ │\r\n│ 1191 │ │ # this function, and just call forward. │\r\n│ 1192 │ │ if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks o │\r\n│ 1193 │ │ │ │ or _global_forward_hooks or _global_forward_pre_hooks): │\r\n│ ❱ 1194 │ │ │ return forward_call(*input, **kwargs) │\r\n│ 1195 │ │ # Do not call functions when jit is used │\r\n│ 1196 │ │ full_backward_hooks, non_full_backward_hooks = [], [] │\r\n│ 1197 │ │ if self._backward_hooks or _global_backward_hooks: │\r\n│ │\r\n│ C:\\Users\\aalsaf01\\venvs\\nlp\\lib\\site-packages\\transformers\\models\\bloom\\modeling_bloom.py:729 in │\r\n│ forward │\r\n│ │\r\n│ 726 │ │ if inputs_embeds is None: │\r\n│ 727 │ │ │ inputs_embeds = self.word_embeddings(input_ids) │\r\n│ 728 │ │ │\r\n│ ❱ 729 │ │ hidden_states = self.word_embeddings_layernorm(inputs_embeds) │\r\n│ 730 │ │ │\r\n│ 731 │ │ presents = () if use_cache else None │\r\n│ 732 │ │ all_self_attentions = () if output_attentions else None │\r\n│ │\r\n│ C:\\Users\\aalsaf01\\venvs\\nlp\\lib\\site-packages\\torch\\nn\\modules\\module.py:1194 in _call_impl │\r\n│ │\r\n│ 1191 │ │ # this function, and just call forward. │\r\n│ 1192 │ │ if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks o │\r\n│ 1193 │ │ │ │ or _global_forward_hooks or _global_forward_pre_hooks): │\r\n│ ❱ 1194 │ │ │ return forward_call(*input, **kwargs) │\r\n│ 1195 │ │ # Do not call functions when jit is used │\r\n│ 1196 │ │ full_backward_hooks, non_full_backward_hooks = [], [] │\r\n│ 1197 │ │ if self._backward_hooks or _global_backward_hooks: │\r\n│ │\r\n│ C:\\Users\\aalsaf01\\venvs\\nlp\\lib\\site-packages\\torch\\nn\\modules\\normalization.py:190 in forward │\r\n│ │\r\n│ 187 │ │ │ init.zeros_(self.bias) │\r\n│ 188 │ │\r\n│ 189 │ def forward(self, input: Tensor) -> Tensor: │\r\n│ ❱ 190 │ │ return F.layer_norm( │\r\n│ 191 │ │ │ input, self.normalized_shape, self.weight, self.bias, self.eps) │\r\n│ 192 │ │\r\n│ 193 │ def extra_repr(self) -> str: │\r\n│ │\r\n│ C:\\Users\\aalsaf01\\venvs\\nlp\\lib\\site-packages\\torch\\nn\\functional.py:2515 in layer_norm │\r\n│ │\r\n│ 2512 │ │ return handle_torch_function( │\r\n│ 2513 │ │ │ layer_norm, (input, weight, bias), input, normalized_shape, weight=weight, b │\r\n│ 2514 │ │ ) │\r\n│ ❱ 2515 │ return torch.layer_norm(input, normalized_shape, weight, bias, eps, torch.backends.c │\r\n│ 2516 │\r\n│ 2517 │\r\n│ 2518 def group_norm( │\r\n╰──────────────────────────────────────────────────────────────────────────────────────────────────╯\r\nRuntimeError: \"LayerNormKernelImpl\" not implemented for 'Half'\r\n```\r\n\r\n\r\nBy the way, it also complained about `accelerate` library not being installed saying that its crucial for `low_cpu` and half precision. Then after installation, the loading works fine, but text generation still fails.\r\n\r\nSo the question would be, why does it still work with GPT-J as per the official example on huggingface docs.\r\n", "also fault when trying the blog of transformers in https://mp.weixin.qq.com/s/k8rE9GrF97E-0TKJhih9kw.", "As I said before, you need to **run** your model on the GPU as the operations are not all implemented on the CPU in float16. On CPU you can only run models in float32.", "Okay, thanks for explaining that. I think an update for docs would be appropriate.\r\n\r\nhttps://huggingface.co/docs/transformers/model_doc/gptj\r\n\r\nOne can indicate that low precision example that works on CPU is just a coincidence as the operations happen to be implemented for CPU. In general, this requires acceleration device.\r\n\r\nI'm not sure if Pytorch have cpu implementation on their agenda.", "Thanks for pointing this example out! It indeed needs to be add a GPU to work. cc @stevhliu or @MKhalusova if you want to fix it (it's the example just before GPTJConfig on the page linked above that loads the model in float16).", "Tesla P40 not support Half..." ]

1,678

1,682

1,678

NONE

null

### System Info I'm feeding flags of low memory and half precision data type to `AutoModelForCausalLM.from_pretrained('bigscience\bloomz7b1')` and I'm receiving the error above. I'm not sure if this is a bug, is it like those flags are only meant to be passed for specific models for which half precision is implemented? If so, how can one tell in a graceful way? Those low memory flags seem to work like a dream with other models like `EleutherAI/gpt-j-6B`. Thanks ### Who can help? _No response_ ### Information - [ ] The official example scripts - [X] My own modified scripts ### Tasks - [ ] An officially supported task in the `examples` folder (such as GLUE/SQuAD, ...) - [ ] My own task or dataset (give details below) ### Reproduction as above. ### Expected behavior model loaded in half precision.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21989/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21989/timeline

completed

null

https://api.github.com/repos/huggingface/transformers/issues/21988

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21988/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21988/comments

https://api.github.com/repos/huggingface/transformers/issues/21988/events

https://github.com/huggingface/transformers/pull/21988

1,612,684,946

PR_kwDOCUB6oc5Lb31I

21,988

Fix broken link

{ "login": "ngoquanghuy99", "id": 36761076, "node_id": "MDQ6VXNlcjM2NzYxMDc2", "avatar_url": "https://avatars.githubusercontent.com/u/36761076?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ngoquanghuy99", "html_url": "https://github.com/ngoquanghuy99", "followers_url": "https://api.github.com/users/ngoquanghuy99/followers", "following_url": "https://api.github.com/users/ngoquanghuy99/following{/other_user}", "gists_url": "https://api.github.com/users/ngoquanghuy99/gists{/gist_id}", "starred_url": "https://api.github.com/users/ngoquanghuy99/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ngoquanghuy99/subscriptions", "organizations_url": "https://api.github.com/users/ngoquanghuy99/orgs", "repos_url": "https://api.github.com/users/ngoquanghuy99/repos", "events_url": "https://api.github.com/users/ngoquanghuy99/events{/privacy}", "received_events_url": "https://api.github.com/users/ngoquanghuy99/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/transformers/pr_21988). All of your documentation changes will be reflected on that endpoint.", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md) are likely to be ignored." ]

1,678

1,680

CONTRIBUTOR

null

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21988/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21988/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21988", "html_url": "https://github.com/huggingface/transformers/pull/21988", "diff_url": "https://github.com/huggingface/transformers/pull/21988.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21988.patch", "merged_at": null }

https://api.github.com/repos/huggingface/transformers/issues/21987

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21987/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21987/comments

https://api.github.com/repos/huggingface/transformers/issues/21987/events

https://github.com/huggingface/transformers/issues/21987

1,612,604,545

I_kwDOCUB6oc5gHmSB

21,987

Long inputs to Flan-T5/UL2 text generation with load_in_8bit=True outputs <pad> tokens repeatedly

{ "login": "akkikiki", "id": 1423362, "node_id": "MDQ6VXNlcjE0MjMzNjI=", "avatar_url": "https://avatars.githubusercontent.com/u/1423362?v=4", "gravatar_id": "", "url": "https://api.github.com/users/akkikiki", "html_url": "https://github.com/akkikiki", "followers_url": "https://api.github.com/users/akkikiki/followers", "following_url": "https://api.github.com/users/akkikiki/following{/other_user}", "gists_url": "https://api.github.com/users/akkikiki/gists{/gist_id}", "starred_url": "https://api.github.com/users/akkikiki/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/akkikiki/subscriptions", "organizations_url": "https://api.github.com/users/akkikiki/orgs", "repos_url": "https://api.github.com/users/akkikiki/repos", "events_url": "https://api.github.com/users/akkikiki/events{/privacy}", "received_events_url": "https://api.github.com/users/akkikiki/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "Thanks a lot for the issue @akkikiki !\r\nWhat is the hardware you are using + bnb version?", "Thanks a lot for the reply!\r\nThe hardware is 8 V100 (16GB) GPUs and the bnb version is 0.37.0.", "I think sadly there is indeed an issue with V100 right now as stated by @TimDettmers here: https://github.com/huggingface/transformers/pull/21955#issuecomment-1455235281 \r\nIt should be fixed somehow soon, also as stated in this comment, more universal methods (that cover most of GPU hardware) should be published soon!", "Thanks @younesbelkada!\r\nInteresting, so some smart workaround for GPUs without hardware-level support on int8.\r\n\r\nFYI, I actually played around with `BitsAndBytesConfig`, and seems like `quantization_config = BitsAndBytesConfig(llm_int8_threshold=5.0)` resolved the issue.\r\n\r\nOutput result with `quantization_config = BitsAndBytesConfig(llm_int8_threshold=5.0)`:\r\n```\r\n<pad> A Haiku is a Japanese poetry form that uses a 5-7-5 syllable structure. A typical tweet is limited to 140 characters. The answer is no.</s>\r\n```\r\n\r\nWill just close this thread for now. Thanks again for the heads up on V100 issue!", "This is great! Thanks for the advice! Would you mind posting it in #21955 so that people can be aware of this hack 🙏 ?", "> This is great! Thanks for the advice! Would you mind posting it in #21955 so that people can be aware of this hack 🙏 ?\r\n\r\nWill do!", "Thanks a lot @akkikiki ! Much apprciated!" ]

1,678

CONTRIBUTOR

null

### System Info - `transformers` version: 4.27.0.dev0 - Platform: Linux-5.4.228-141.415.amzn2int.x86_64-x86_64-with-glibc2.17 - Python version: 3.8.16 - Huggingface_hub version: 0.12.1 - PyTorch version (GPU?): 1.13.1 (True) - Tensorflow version (GPU?): not installed (NA) - Flax version (CPU?/GPU?/TPU?): not installed (NA) - Jax version: not installed - JaxLib version: not installed - Using GPU in script?: Yes - Using distributed or parallel set-up in script?: No ### Who can help? @younesbelkada ### Information - [ ] The official example scripts - [X] My own modified scripts ### Tasks - [X] An officially supported task in the `examples` folder (such as GLUE/SQuAD, ...) - [ ] My own task or dataset (give details below) ### Reproduction When input texts are short, the generated texts look good. But when input texts are long e.g., the following, then it produces <pad> tokens. Input ``` model = T5ForConditionalGeneration.from_pretrained("google/flan-ul2", device_map=device_map) # same with "google/flan-t5-xxl" load_in_8bit=True) input_text = """Q: Answer the following yes/no question by reasoning step-by-step. Could a dandelion suffer from hepatitis? A: Hepatitis only affects organisms with livers. Dandelions don’t have a liver. The answer is yes. Q: Answer the following yes/no question by reasoning step-by-step. Can you write a whole Haiku in a single tweet? A: """ input_ids = tokenizer(input_text, return_tensors="pt").input_ids.to("cuda") outputs = model.generate(input_ids, max_length=100) print(tokenizer.decode(outputs[0])) ``` Output: ``` <pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad> ``` ### Expected behavior This is the result when loaded with `load_in_8bit=False` ``` <pad> A Haiku is a Japanese poetry form that uses a 5-7-5 syllable structure. A typical tweet is limited to 140 characters. The answer is no.</s> ```

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21987/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21987/timeline

completed

null

https://api.github.com/repos/huggingface/transformers/issues/21986

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21986/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21986/comments

https://api.github.com/repos/huggingface/transformers/issues/21986/events

https://github.com/huggingface/transformers/pull/21986

1,612,419,400

PR_kwDOCUB6oc5La-4U

21,986

save_pretrained crashes when torch_dtype is passed

{ "login": "NikolaBorisov", "id": 184322, "node_id": "MDQ6VXNlcjE4NDMyMg==", "avatar_url": "https://avatars.githubusercontent.com/u/184322?v=4", "gravatar_id": "", "url": "https://api.github.com/users/NikolaBorisov", "html_url": "https://github.com/NikolaBorisov", "followers_url": "https://api.github.com/users/NikolaBorisov/followers", "following_url": "https://api.github.com/users/NikolaBorisov/following{/other_user}", "gists_url": "https://api.github.com/users/NikolaBorisov/gists{/gist_id}", "starred_url": "https://api.github.com/users/NikolaBorisov/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/NikolaBorisov/subscriptions", "organizations_url": "https://api.github.com/users/NikolaBorisov/orgs", "repos_url": "https://api.github.com/users/NikolaBorisov/repos", "events_url": "https://api.github.com/users/NikolaBorisov/events{/privacy}", "received_events_url": "https://api.github.com/users/NikolaBorisov/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/transformers/pr_21986). All of your documentation changes will be reflected on that endpoint.", "My understanding that the following code should work\r\n\r\n```\r\nfrom transformers import pipeline\r\np = pipeline(\"some/model\", torch_dtype=torch.bfloat16)\r\n``` \r\n\r\nfrom what I can see the tokenizer kwargs basically start with copy of the model_kwargs? \r\n\r\nhttps://github.com/huggingface/transformers/blob/main/src/transformers/pipelines/__init__.py#L869", "Indeed, this means that the problem is bigger! Popping the argument is good if we want to keep the pipeline that way, I think we agreed to properly handle the extra tokenizer kwargs @Narsil if you want to take care of it, it will fix this! ", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md) are likely to be ignored.", "@ArthurZucker @Narsil What do you want to do here. I think this PR does the right thing at the moment because extra arguments to the pipeline are passed around to the tokenizer and then the tokenizer crashes on save_pretrained. \r\nI think merging this is strictly better than the current state. Sure the code could be improved internally. Conceptually there should be something that decides which kwargs to the pipeline should be passed to each of the components in the pipeline.", "Hey! Thanks for reporting the bug, I merged something that seemed more aligned with what we want: torch dtype should just not be passed to the tokenizer, so poping it outside" ]

1,678

1,680

NONE

null

If you have the following code ``` p = pipeline(... torch_dtype=torch.bfloat16) p.save_pretrained() ``` you get a crash because `torch.bfloat16` is not json serializable in the `tokenizer.save_pretrained()` method. This PR fixes this. @Narsil @ArthurZucker

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21986/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21986/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21986", "html_url": "https://github.com/huggingface/transformers/pull/21986", "diff_url": "https://github.com/huggingface/transformers/pull/21986.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21986.patch", "merged_at": null }

https://api.github.com/repos/huggingface/transformers/issues/21985

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21985/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21985/comments

https://api.github.com/repos/huggingface/transformers/issues/21985/events

https://github.com/huggingface/transformers/issues/21985

1,612,279,178

I_kwDOCUB6oc5gGW2K

21,985

Ranking a pre-defined list of output candidates for LMs

{ "login": "yuchenlin", "id": 10104354, "node_id": "MDQ6VXNlcjEwMTA0MzU0", "avatar_url": "https://avatars.githubusercontent.com/u/10104354?v=4", "gravatar_id": "", "url": "https://api.github.com/users/yuchenlin", "html_url": "https://github.com/yuchenlin", "followers_url": "https://api.github.com/users/yuchenlin/followers", "following_url": "https://api.github.com/users/yuchenlin/following{/other_user}", "gists_url": "https://api.github.com/users/yuchenlin/gists{/gist_id}", "starred_url": "https://api.github.com/users/yuchenlin/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/yuchenlin/subscriptions", "organizations_url": "https://api.github.com/users/yuchenlin/orgs", "repos_url": "https://api.github.com/users/yuchenlin/repos", "events_url": "https://api.github.com/users/yuchenlin/events{/privacy}", "received_events_url": "https://api.github.com/users/yuchenlin/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "Hey @gante , I was wondering you may will also have great suggestions on this. Thanks a lot in advance! :D. ", "Hey @yuchenlin 👋 We usually don't add those sorts of tools to `transformers`, but I'd be happy to guide you.\r\n\r\nFirst of all, I'd suggest to check [the documentation and the examples for this function](https://huggingface.co/docs/transformers/v4.26.1/en/main_classes/text_generation#transformers.GenerationMixin.compute_transition_scores). Then, by invoking it for several models and/or inputs, you will be able to re-rank candidate options according to your needs :)\r\n\r\nIf you find the end results satisfying, I'd suggest opening a Spaces so it can be shared with the world 🌍 ", "Hi @gante thank you so much for the help! I was reading this before but it seems that this method can only output the transition scores for tokens that are considered by the model.generate process. However, the more general case for this application is that we have some external output candidates that may not be generated by \"model.generate()\". \r\n\r\nAnd sure I will put a solution to a Spaces if I manage to do this and I believe this will help many others! :D ", "I see. We have in our plans to build a function that returns the score for any candidate sequence, but have other competing priorities :) Feel free to have a go at it, if you're interested!", "Hi @gante , thanks for letting me know. \r\n\r\nI managed to develop a solution here: https://github.com/yuchenlin/rank_outputs/blob/main/rank_outputs/main.py \r\n\r\nWill try to wrap it up as a more general tool. \r\n\r\nThanks! \r\n\r\n" ]

1,678

CONTRIBUTOR

null

### Feature request Can we support a feature that given an input and a list of output candidates for a LM (say fine-tuned T5 or GPT-2), we can get a score for each output candidate and then return the top K of them? For example, I fine-tuned a seq2seq model (e.g., BART or T5) for summarization, and now given an input doc and a list of candidate summaries, I want to know which ones are the best. I do not want the fine-tuned model to decode and generate its summaries as we normally do with them. I believe we can do this by computing the token-by-token log-likelihood (with or without length normalization), and then return the top K from the candidate list. ### Motivation I was reading the paper of T0, and it mentioned that they used such a method to evaluate tasks with multiple choices: The paper describes it like this on Page 6 from https://arxiv.org/pdf/2110.08207.pdf: > For tasks that involve choosing the correct completion from several options (e.g. multiple choice question answering), we follow Brown et al. (2020) and use **rank classification** to evaluate our model: we compute the log-likelihood of each of the target options under the fine-tuned model and select the option with the highest log-likelihood as the prediction. For simplicity, we do not apply length normalization to the log-likelihoods of the target options. ### Your contribution I found their code that maybe helpful but I feel that they are a bit hard to use directly. https://github.com/bigscience-workshop/t-zero/blob/25c0761427f3894a8ec5a062a075b96037fb1492/t0/model.py#L67

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21985/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21985/timeline

completed

null

https://api.github.com/repos/huggingface/transformers/issues/21984

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21984/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21984/comments

https://api.github.com/repos/huggingface/transformers/issues/21984/events

https://github.com/huggingface/transformers/pull/21984

1,612,123,859

PR_kwDOCUB6oc5LZ_EQ

21,984

Update Jukebox tests

{ "login": "ydshieh", "id": 2521628, "node_id": "MDQ6VXNlcjI1MjE2Mjg=", "avatar_url": "https://avatars.githubusercontent.com/u/2521628?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ydshieh", "html_url": "https://github.com/ydshieh", "followers_url": "https://api.github.com/users/ydshieh/followers", "following_url": "https://api.github.com/users/ydshieh/following{/other_user}", "gists_url": "https://api.github.com/users/ydshieh/gists{/gist_id}", "starred_url": "https://api.github.com/users/ydshieh/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ydshieh/subscriptions", "organizations_url": "https://api.github.com/users/ydshieh/orgs", "repos_url": "https://api.github.com/users/ydshieh/repos", "events_url": "https://api.github.com/users/ydshieh/events{/privacy}", "received_events_url": "https://api.github.com/users/ydshieh/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "test failure irrelevant to this PR." ]

1,678

COLLABORATOR

null

# What does this PR do? Updage Jukebox tests, for PyTorch 2.0. The reason is the same as in #21975: tiny diff in scores, but sampling can give different results.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21984/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21984/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21984", "html_url": "https://github.com/huggingface/transformers/pull/21984", "diff_url": "https://github.com/huggingface/transformers/pull/21984.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21984.patch", "merged_at": 1678159215000 }

https://api.github.com/repos/huggingface/transformers/issues/21983

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21983/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21983/comments

https://api.github.com/repos/huggingface/transformers/issues/21983/events

https://github.com/huggingface/transformers/pull/21983

1,612,014,592

PR_kwDOCUB6oc5LZnmy

21,983

Remove unneeded casts to bool

{ "login": "regisss", "id": 15324346, "node_id": "MDQ6VXNlcjE1MzI0MzQ2", "avatar_url": "https://avatars.githubusercontent.com/u/15324346?v=4", "gravatar_id": "", "url": "https://api.github.com/users/regisss", "html_url": "https://github.com/regisss", "followers_url": "https://api.github.com/users/regisss/followers", "following_url": "https://api.github.com/users/regisss/following{/other_user}", "gists_url": "https://api.github.com/users/regisss/gists{/gist_id}", "starred_url": "https://api.github.com/users/regisss/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/regisss/subscriptions", "organizations_url": "https://api.github.com/users/regisss/orgs", "repos_url": "https://api.github.com/users/regisss/repos", "events_url": "https://api.github.com/users/regisss/events{/privacy}", "received_events_url": "https://api.github.com/users/regisss/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,678

CONTRIBUTOR

null

# What does this PR do?   This PR removes some conversions to `torch.bool` which are not needed anymore following #21384. ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [x] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [ ] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [ ] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [ ] Did you write any new necessary tests? ## Who can review? Anyone in the community is free to review the PR once the tests have passed. Feel free to tag members/contributors who may be interested in your PR.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21983/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21983/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21983", "html_url": "https://github.com/huggingface/transformers/pull/21983", "diff_url": "https://github.com/huggingface/transformers/pull/21983.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21983.patch", "merged_at": 1678192550000 }

https://api.github.com/repos/huggingface/transformers/issues/21982

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21982/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21982/comments

https://api.github.com/repos/huggingface/transformers/issues/21982/events

https://github.com/huggingface/transformers/pull/21982

1,611,948,101

PR_kwDOCUB6oc5LZZNB

21,982

docs: New terms and updates to glossary

{ "login": "MichaelRipa", "id": 51883134, "node_id": "MDQ6VXNlcjUxODgzMTM0", "avatar_url": "https://avatars.githubusercontent.com/u/51883134?v=4", "gravatar_id": "", "url": "https://api.github.com/users/MichaelRipa", "html_url": "https://github.com/MichaelRipa", "followers_url": "https://api.github.com/users/MichaelRipa/followers", "following_url": "https://api.github.com/users/MichaelRipa/following{/other_user}", "gists_url": "https://api.github.com/users/MichaelRipa/gists{/gist_id}", "starred_url": "https://api.github.com/users/MichaelRipa/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/MichaelRipa/subscriptions", "organizations_url": "https://api.github.com/users/MichaelRipa/orgs", "repos_url": "https://api.github.com/users/MichaelRipa/repos", "events_url": "https://api.github.com/users/MichaelRipa/events{/privacy}", "received_events_url": "https://api.github.com/users/MichaelRipa/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "Thanks for the detailed edits, they are helpful! 🙂 I will refresh permissions and also add a link to the pipeline for inference doc tomorrow ", "Added my update, and I followed the instructions to refresh permissions. I don't seem to have permissions to manually restart the CircleCI Pipeline, so not sure whether the steps I took were reflected in my last commit (I did my updates before refreshing permissions). ", "You can push an empty commit with `git commit -m \"Trigger CI\" --allow-empty\" and then push to your branch.", "Thanks for the review! I'll start going over the changes / implementing your suggestions over the weekend ", "> Thanks for your PR. I don't think we should remove entries from the glossary. Linking to other entries is better. Could you also add an entry for \"self-supervised learning\" since most pretraining of Transformer models use that technique?\r\n\r\nI will add back in the old entries and add links between them, and yeah that's a great idea!", "Awesome! I committed the suggestion, looks to have merged successfully🙂 Thanks again for the help with the edits/reviews! (also feel free to ping me if you see any further edits/additions to be done)", "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/transformers/pr_21982). All of your documentation changes will be reflected on that endpoint." ]

1,678

CONTRIBUTOR

null

Updates to the glossary as proposed in #21801 . All terms suggested in the issue were added, except for Encoder/Decoder. Acronyms were also added for terms in which it made sense (e.g. NLP) Note that this PR replaces ***autoencoding models*** with ***encoder models*** (and provides a more detailed definition) as well as merging ***autoregressive models*** and ***causal language modeling*** with ***decoder models***. This is my first draft and would likely benefit from additional edits/revision. Any suggestions in terms of updates or areas to expand further on are welcome 🙂 # What does this PR do?   Fixes #21801 ## Before submitting - [x ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [ ] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [ ] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [ ] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [ ] Did you write any new necessary tests? ## Who can review? Anyone in the community is free to review the PR once the tests have passed. Feel free to tag members/contributors who may be interested in your PR. @stevhliu

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21982/reactions", "total_count": 1, "+1": 0, "-1": 0, "laugh": 0, "hooray": 1, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21982/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21982", "html_url": "https://github.com/huggingface/transformers/pull/21982", "diff_url": "https://github.com/huggingface/transformers/pull/21982.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21982.patch", "merged_at": 1678748978000 }

https://api.github.com/repos/huggingface/transformers/issues/21981

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21981/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21981/comments

https://api.github.com/repos/huggingface/transformers/issues/21981/events

https://github.com/huggingface/transformers/issues/21981

1,611,917,272

I_kwDOCUB6oc5gE-fY

21,981

Function infer_channel_dimension_format has a bug

{ "login": "aleksmirosh", "id": 74064180, "node_id": "MDQ6VXNlcjc0MDY0MTgw", "avatar_url": "https://avatars.githubusercontent.com/u/74064180?v=4", "gravatar_id": "", "url": "https://api.github.com/users/aleksmirosh", "html_url": "https://github.com/aleksmirosh", "followers_url": "https://api.github.com/users/aleksmirosh/followers", "following_url": "https://api.github.com/users/aleksmirosh/following{/other_user}", "gists_url": "https://api.github.com/users/aleksmirosh/gists{/gist_id}", "starred_url": "https://api.github.com/users/aleksmirosh/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/aleksmirosh/subscriptions", "organizations_url": "https://api.github.com/users/aleksmirosh/orgs", "repos_url": "https://api.github.com/users/aleksmirosh/repos", "events_url": "https://api.github.com/users/aleksmirosh/events{/privacy}", "received_events_url": "https://api.github.com/users/aleksmirosh/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "Hi @aleksmirosh - you've tagged the right person :) Thanks for raising this issue. \r\n\r\nThe functions in the image transforms library currently only support grayscale (1 channel) and RGB (3 channel) images. What format are the images you're trying to use? ", "@amyeroberts \r\nThank you for your quick response and sorry for not complet describing of the issue.\r\nSo I use 6 channels. \r\nI used version 4.24.0 before and tried to update it to the latest.\r\nDo you plan to update to any channel number in the future? \r\n", "There aren't any immediate plans to accept an arbitrary number of channels. I agree it would be useful and will add it to the potential extensions of the library. \r\n\r\nFor the 6 channel images - are these concatenations of 2, 3-channel images or a single image with 6 channels? If the former, the simplest way to get this working quickly would be to pass each of 3 RBG images to the image processor and then concatenate. However, this will likely be quite inefficient. ", "Hello! Just wanted to add that I've come across this issue as well, but using the CLIPProcessor which uses https://github.com/huggingface/transformers/blob/2f320661f364557c821c285729dab3881e977363/src/transformers/image_transforms.py#L304\r\n\r\nBasically I pass in images that always have the dimension (H,W,C), but occasionally I'll get images that are (1,1,3) or (3,*,3), but in both cases the first dimension is inferred as the channel dimension, which is not what I intended. The (3,1,3) case will not error, but silently proceed, but the (1,1,3) case errors for me b/c the mean is of length 3 but the inferred num_channels of the image is 1", "> For the 6 channel images - are these concatenations of 2, 3-channel images or a single image with 6 channels? If the former, the simplest way to get this working quickly would be to pass each of 3 RBG images to the image processor and then concatenate. However, this will likely be quite inefficient.\r\n\r\nthis is 6 channels single image, but I will try to use concatenation. thank you for the advice.\r\nAlso if using the parameter 'data_format=None' could it help?\r\n\r\n@terrykong maybe data_format=None could help with your case?", "Thanks for the suggestion @aleksmirosh. Unfortunately, that kwarg only affects the output format: https://github.com/huggingface/transformers/blob/2f320661f364557c821c285729dab3881e977363/src/transformers/image_transforms.py#L322-L323\r\n\r\nThe input format is not configurable and is inferred: https://github.com/huggingface/transformers/blob/2f320661f364557c821c285729dab3881e977363/src/transformers/image_transforms.py#L340\r\n\r\nIt would be nice if the input format could be specified.", "Hi @aleksmirosh , `data_format` specifies the desired output format for the images. I'm working on a PR to add the option to specify for the input format @terrykong ", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md) are likely to be ignored." ]

1,678

1,693

NONE

null

### System Info Hi, I am working on my own DataLoader. So my input to class transformers.SegformerImageProcessor has shape (6, 512, 512). So channel first. As I realized the problem is **image.shape[first_dim]** for me is 0, but your if construction said channel dimension should be always 1 or 3, which does not make sense for 3 dim images, because above you assigned first_dim, last_dim = 0, 2. @amyeroberts not sure I tagged right person, sorry ### Who can help? _No response_ ### Information - [X] The official example scripts - [ ] My own modified scripts ### Tasks - [ ] An officially supported task in the `examples` folder (such as GLUE/SQuAD, ...) - [X] My own task or dataset (give details below) ### Reproduction ``` def infer_channel_dimension_format(image: np.ndarray) -> ChannelDimension: if image.ndim == 3: first_dim, last_dim = 0, 2 elif image.ndim == 4: first_dim, last_dim = 1, 3 else: raise ValueError(f"Unsupported number of image dimensions: {image.ndim}") if image.shape[first_dim] in (1, 3): return ChannelDimension.FIRST elif image.shape[last_dim] in (1, 3): return ChannelDimension.LAST raise ValueError("Unable to infer channel dimension format") ``` any image wish a shape (ch, w, h) infer_channel_dimension_format(image) ### Expected behavior I expected function will return ChannelDimension.FIRST for image in shape (CH, W, H)

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21981/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21981/timeline

completed

null

https://api.github.com/repos/huggingface/transformers/issues/21980

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21980/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21980/comments

https://api.github.com/repos/huggingface/transformers/issues/21980/events

https://github.com/huggingface/transformers/pull/21980

1,611,887,155

PR_kwDOCUB6oc5LZMUe

21,980

Fix gradient checkpointing bug in ESM

{ "login": "KMFODA", "id": 35491698, "node_id": "MDQ6VXNlcjM1NDkxNjk4", "avatar_url": "https://avatars.githubusercontent.com/u/35491698?v=4", "gravatar_id": "", "url": "https://api.github.com/users/KMFODA", "html_url": "https://github.com/KMFODA", "followers_url": "https://api.github.com/users/KMFODA/followers", "following_url": "https://api.github.com/users/KMFODA/following{/other_user}", "gists_url": "https://api.github.com/users/KMFODA/gists{/gist_id}", "starred_url": "https://api.github.com/users/KMFODA/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/KMFODA/subscriptions", "organizations_url": "https://api.github.com/users/KMFODA/orgs", "repos_url": "https://api.github.com/users/KMFODA/repos", "events_url": "https://api.github.com/users/KMFODA/events{/privacy}", "received_events_url": "https://api.github.com/users/KMFODA/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,678

CONTRIBUTOR

null

This PR fixes a bug that a user can encounter while using generate and models that use gradient_checkpointing. Fixes Issue https://github.com/huggingface/transformers/issues/21737 cc @younesbelkada or @gante

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21980/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21980/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21980", "html_url": "https://github.com/huggingface/transformers/pull/21980", "diff_url": "https://github.com/huggingface/transformers/pull/21980.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21980.patch", "merged_at": 1678124693000 }

https://api.github.com/repos/huggingface/transformers/issues/21979

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21979/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21979/comments

https://api.github.com/repos/huggingface/transformers/issues/21979/events

https://github.com/huggingface/transformers/pull/21979

1,611,881,466

PR_kwDOCUB6oc5LZLH-

21,979

Fix gradient checkpointing bug in Codegen

{ "login": "KMFODA", "id": 35491698, "node_id": "MDQ6VXNlcjM1NDkxNjk4", "avatar_url": "https://avatars.githubusercontent.com/u/35491698?v=4", "gravatar_id": "", "url": "https://api.github.com/users/KMFODA", "html_url": "https://github.com/KMFODA", "followers_url": "https://api.github.com/users/KMFODA/followers", "following_url": "https://api.github.com/users/KMFODA/following{/other_user}", "gists_url": "https://api.github.com/users/KMFODA/gists{/gist_id}", "starred_url": "https://api.github.com/users/KMFODA/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/KMFODA/subscriptions", "organizations_url": "https://api.github.com/users/KMFODA/orgs", "repos_url": "https://api.github.com/users/KMFODA/repos", "events_url": "https://api.github.com/users/KMFODA/events{/privacy}", "received_events_url": "https://api.github.com/users/KMFODA/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,678

CONTRIBUTOR

null

This PR fixes a bug that a user can encounter while using generate and models that use gradient_checkpointing. Fixes Issue https://github.com/huggingface/transformers/issues/21737 cc @younesbelkada or @gante

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21979/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21979/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21979", "html_url": "https://github.com/huggingface/transformers/pull/21979", "diff_url": "https://github.com/huggingface/transformers/pull/21979.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21979.patch", "merged_at": 1678124672000 }

https://api.github.com/repos/huggingface/transformers/issues/21978

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21978/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21978/comments

https://api.github.com/repos/huggingface/transformers/issues/21978/events

https://github.com/huggingface/transformers/pull/21978

1,611,876,051

PR_kwDOCUB6oc5LZJ-W

21,978

Fix gradient checkpointing bug in BlipText

{ "login": "KMFODA", "id": 35491698, "node_id": "MDQ6VXNlcjM1NDkxNjk4", "avatar_url": "https://avatars.githubusercontent.com/u/35491698?v=4", "gravatar_id": "", "url": "https://api.github.com/users/KMFODA", "html_url": "https://github.com/KMFODA", "followers_url": "https://api.github.com/users/KMFODA/followers", "following_url": "https://api.github.com/users/KMFODA/following{/other_user}", "gists_url": "https://api.github.com/users/KMFODA/gists{/gist_id}", "starred_url": "https://api.github.com/users/KMFODA/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/KMFODA/subscriptions", "organizations_url": "https://api.github.com/users/KMFODA/orgs", "repos_url": "https://api.github.com/users/KMFODA/repos", "events_url": "https://api.github.com/users/KMFODA/events{/privacy}", "received_events_url": "https://api.github.com/users/KMFODA/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,678

CONTRIBUTOR

null

This PR fixes a bug that a user can encounter while using generate and models that use gradient_checkpointing. Fixes Issue https://github.com/huggingface/transformers/issues/21737 cc @younesbelkada or @gante

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21978/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21978/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21978", "html_url": "https://github.com/huggingface/transformers/pull/21978", "diff_url": "https://github.com/huggingface/transformers/pull/21978.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21978.patch", "merged_at": 1678124633000 }

https://api.github.com/repos/huggingface/transformers/issues/21977

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21977/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21977/comments

https://api.github.com/repos/huggingface/transformers/issues/21977/events

https://github.com/huggingface/transformers/pull/21977

1,611,867,159

PR_kwDOCUB6oc5LZIBr

21,977

Fix gradient checkpointing bug in Blenderbot Small

{ "login": "KMFODA", "id": 35491698, "node_id": "MDQ6VXNlcjM1NDkxNjk4", "avatar_url": "https://avatars.githubusercontent.com/u/35491698?v=4", "gravatar_id": "", "url": "https://api.github.com/users/KMFODA", "html_url": "https://github.com/KMFODA", "followers_url": "https://api.github.com/users/KMFODA/followers", "following_url": "https://api.github.com/users/KMFODA/following{/other_user}", "gists_url": "https://api.github.com/users/KMFODA/gists{/gist_id}", "starred_url": "https://api.github.com/users/KMFODA/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/KMFODA/subscriptions", "organizations_url": "https://api.github.com/users/KMFODA/orgs", "repos_url": "https://api.github.com/users/KMFODA/repos", "events_url": "https://api.github.com/users/KMFODA/events{/privacy}", "received_events_url": "https://api.github.com/users/KMFODA/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,678

CONTRIBUTOR

null

This PR fixes a bug that a user can encounter while using generate and models that use gradient_checkpointing. Fixes Issue https://github.com/huggingface/transformers/issues/21737 cc @younesbelkada or @gante

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21977/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21977/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21977", "html_url": "https://github.com/huggingface/transformers/pull/21977", "diff_url": "https://github.com/huggingface/transformers/pull/21977.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21977.patch", "merged_at": 1678124605000 }

https://api.github.com/repos/huggingface/transformers/issues/21976

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21976/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21976/comments

https://api.github.com/repos/huggingface/transformers/issues/21976/events

https://github.com/huggingface/transformers/pull/21976

1,611,858,033

PR_kwDOCUB6oc5LZF-0

21,976

Fix gradient checkpointing bug in BigBird Pegasus

{ "login": "KMFODA", "id": 35491698, "node_id": "MDQ6VXNlcjM1NDkxNjk4", "avatar_url": "https://avatars.githubusercontent.com/u/35491698?v=4", "gravatar_id": "", "url": "https://api.github.com/users/KMFODA", "html_url": "https://github.com/KMFODA", "followers_url": "https://api.github.com/users/KMFODA/followers", "following_url": "https://api.github.com/users/KMFODA/following{/other_user}", "gists_url": "https://api.github.com/users/KMFODA/gists{/gist_id}", "starred_url": "https://api.github.com/users/KMFODA/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/KMFODA/subscriptions", "organizations_url": "https://api.github.com/users/KMFODA/orgs", "repos_url": "https://api.github.com/users/KMFODA/repos", "events_url": "https://api.github.com/users/KMFODA/events{/privacy}", "received_events_url": "https://api.github.com/users/KMFODA/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,678

CONTRIBUTOR

null

This PR fixes a bug that a user can encounter while using generate and models that use gradient_checkpointing. Fixes Issue https://github.com/huggingface/transformers/issues/21737 cc @younesbelkada or @gante

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21976/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21976/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21976", "html_url": "https://github.com/huggingface/transformers/pull/21976", "diff_url": "https://github.com/huggingface/transformers/pull/21976.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21976.patch", "merged_at": 1678124573000 }

https://api.github.com/repos/huggingface/transformers/issues/21975

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21975/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21975/comments

https://api.github.com/repos/huggingface/transformers/issues/21975/events

https://github.com/huggingface/transformers/pull/21975

1,611,831,322

PR_kwDOCUB6oc5LZAJQ

21,975

Update expected values for `test_xglm_sample`

{ "login": "ydshieh", "id": 2521628, "node_id": "MDQ6VXNlcjI1MjE2Mjg=", "avatar_url": "https://avatars.githubusercontent.com/u/2521628?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ydshieh", "html_url": "https://github.com/ydshieh", "followers_url": "https://api.github.com/users/ydshieh/followers", "following_url": "https://api.github.com/users/ydshieh/following{/other_user}", "gists_url": "https://api.github.com/users/ydshieh/gists{/gist_id}", "starred_url": "https://api.github.com/users/ydshieh/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ydshieh/subscriptions", "organizations_url": "https://api.github.com/users/ydshieh/orgs", "repos_url": "https://api.github.com/users/ydshieh/repos", "events_url": "https://api.github.com/users/ydshieh/events{/privacy}", "received_events_url": "https://api.github.com/users/ydshieh/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,678

COLLABORATOR

null

# What does this PR do? Despite the `probs` value has only a difference of `7.59e-07`, the `torch.multinomial(probs, num_samples=1)` still gives different `next_tokens`: - 3967 (`sun`) in `torch 1.13.1` - 4565 (`water`) in `torch 2.0` Currently I keep both values, but we can remove the one for `torch 1.13` soon.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21975/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21975/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21975", "html_url": "https://github.com/huggingface/transformers/pull/21975", "diff_url": "https://github.com/huggingface/transformers/pull/21975.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21975.patch", "merged_at": 1678122452000 }

https://api.github.com/repos/huggingface/transformers/issues/21974

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21974/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21974/comments

https://api.github.com/repos/huggingface/transformers/issues/21974/events

https://github.com/huggingface/transformers/pull/21974

1,611,828,710

PR_kwDOCUB6oc5LY_k4

21,974

[DETR, YOLOS] Fix device bug

{ "login": "NielsRogge", "id": 48327001, "node_id": "MDQ6VXNlcjQ4MzI3MDAx", "avatar_url": "https://avatars.githubusercontent.com/u/48327001?v=4", "gravatar_id": "", "url": "https://api.github.com/users/NielsRogge", "html_url": "https://github.com/NielsRogge", "followers_url": "https://api.github.com/users/NielsRogge/followers", "following_url": "https://api.github.com/users/NielsRogge/following{/other_user}", "gists_url": "https://api.github.com/users/NielsRogge/gists{/gist_id}", "starred_url": "https://api.github.com/users/NielsRogge/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/NielsRogge/subscriptions", "organizations_url": "https://api.github.com/users/NielsRogge/orgs", "repos_url": "https://api.github.com/users/NielsRogge/repos", "events_url": "https://api.github.com/users/NielsRogge/events{/privacy}", "received_events_url": "https://api.github.com/users/NielsRogge/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,678

CONTRIBUTOR

null

# What does this PR do? This PR fixes a device bug for DETR and YOLOS's `post_process_object_detection` methods, which currently give a device mismatch between CPU and CUDA when running the model on CUDA. The PR also makes sure the postprocess methods are tested in the integration tests.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21974/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21974/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21974", "html_url": "https://github.com/huggingface/transformers/pull/21974", "diff_url": "https://github.com/huggingface/transformers/pull/21974.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21974.patch", "merged_at": 1678192445000 }

https://api.github.com/repos/huggingface/transformers/issues/21973

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21973/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21973/comments

https://api.github.com/repos/huggingface/transformers/issues/21973/events

https://github.com/huggingface/transformers/issues/21973

1,611,766,011

I_kwDOCUB6oc5gEZj7

21,973

GitHub Private Vulnerability reporting

{ "login": "Sim4n6", "id": 13036531, "node_id": "MDQ6VXNlcjEzMDM2NTMx", "avatar_url": "https://avatars.githubusercontent.com/u/13036531?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Sim4n6", "html_url": "https://github.com/Sim4n6", "followers_url": "https://api.github.com/users/Sim4n6/followers", "following_url": "https://api.github.com/users/Sim4n6/following{/other_user}", "gists_url": "https://api.github.com/users/Sim4n6/gists{/gist_id}", "starred_url": "https://api.github.com/users/Sim4n6/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Sim4n6/subscriptions", "organizations_url": "https://api.github.com/users/Sim4n6/orgs", "repos_url": "https://api.github.com/users/Sim4n6/repos", "events_url": "https://api.github.com/users/Sim4n6/events{/privacy}", "received_events_url": "https://api.github.com/users/Sim4n6/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "cc @Michellehbn ", "Hi @Sim4n6, Thanks for reaching out to us! 🤗 We have a bug bounty program with HackerOne and would love for you to submit security vulnerability reports to https://hackerone.com/hugging_face. This is a private program so we will need to invite you. Do you happen to have an H1 username? Or you can send security@huggingface.co an email and we'll send you an invite! ", "it is [Sim4n6](https://hackerone.com/sim4n6?type=user). No problem.", "Invite sent! Thanks again! ", "Done." ]

1,678

NONE

null

### Feature request Enable Private vulnerability reporting in the GitHub repository, please. https://docs.github.com/en/code-security/security-advisories/repository-security-advisories/configuring-private-vulnerability-reporting-for-a-repository ### Motivation I may have identified something low-impact vulnerability, but I would like to report it privately via the usual channel and request a CVE in that case. ### Your contribution The report that I will submit, when the feature is enabled, please.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21973/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21973/timeline

completed

null

https://api.github.com/repos/huggingface/transformers/issues/21972

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21972/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21972/comments

https://api.github.com/repos/huggingface/transformers/issues/21972/events

https://github.com/huggingface/transformers/issues/21972

1,611,752,143

I_kwDOCUB6oc5gEWLP

21,972

Support for `Flax` Trainer

{ "login": "Shubhamai", "id": 51819922, "node_id": "MDQ6VXNlcjUxODE5OTIy", "avatar_url": "https://avatars.githubusercontent.com/u/51819922?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Shubhamai", "html_url": "https://github.com/Shubhamai", "followers_url": "https://api.github.com/users/Shubhamai/followers", "following_url": "https://api.github.com/users/Shubhamai/following{/other_user}", "gists_url": "https://api.github.com/users/Shubhamai/gists{/gist_id}", "starred_url": "https://api.github.com/users/Shubhamai/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Shubhamai/subscriptions", "organizations_url": "https://api.github.com/users/Shubhamai/orgs", "repos_url": "https://api.github.com/users/Shubhamai/repos", "events_url": "https://api.github.com/users/Shubhamai/events{/privacy}", "received_events_url": "https://api.github.com/users/Shubhamai/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "This is also something we do not want to add or maintain for Flax, as researchers usually dislike Trainer classes very much.", "Understood, thanks for such a quick reply :hugs: " ]

1,678

CONTRIBUTOR

null

### Feature request A `Trainer` class similar to PyTorch/Tensorflow (deprecating in v5) . ### Motivation The process of training HuggingFace models in `PyTorch` has become more accessible due to the availability of the `Trainer` class and extensive documentation and tutorials. Similarly, in `Tensorflow`, training models require only a single line of code (`model.fit`). and it makes sense to deprecate the current trainer to avoid redundancy. However, the process of training `Flax` models currently requires boilerplate code similar to `PyTorch` which a `Trainer` class would be helpful to eliminate. Making flax `Trainer` available will allow for really fast training in GPU/TPUs and with the support of weights conversion, one can instantly convert the model to `Tensorflow`/`PyTorch` for inference and deployment. ### Your contribution I can submit the PR. cc @sanchit-gandhi since it's related to flax.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21972/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21972/timeline

completed

null

https://api.github.com/repos/huggingface/transformers/issues/21971

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21971/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21971/comments

https://api.github.com/repos/huggingface/transformers/issues/21971/events

https://github.com/huggingface/transformers/issues/21971

1,611,636,612

I_kwDOCUB6oc5gD5-E

21,971

No clear documentation for enabling padding in FeatureExtractionPipeline

{ "login": "anruijian", "id": 115125339, "node_id": "U_kgDOBtysWw", "avatar_url": "https://avatars.githubusercontent.com/u/115125339?v=4", "gravatar_id": "", "url": "https://api.github.com/users/anruijian", "html_url": "https://github.com/anruijian", "followers_url": "https://api.github.com/users/anruijian/followers", "following_url": "https://api.github.com/users/anruijian/following{/other_user}", "gists_url": "https://api.github.com/users/anruijian/gists{/gist_id}", "starred_url": "https://api.github.com/users/anruijian/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/anruijian/subscriptions", "organizations_url": "https://api.github.com/users/anruijian/orgs", "repos_url": "https://api.github.com/users/anruijian/repos", "events_url": "https://api.github.com/users/anruijian/events{/privacy}", "received_events_url": "https://api.github.com/users/anruijian/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "I simplified a bit for future readers:\r\n\r\n```python\r\nexample = \"After stealing money from the bank vault, the bank robber was seen fishing on the Mississippi river bank.\"\r\n\r\npipeline_without_padding_as_an_argument = pipeline(\r\n \"feature-extraction\",\r\n model=\"bert-base-uncased\",\r\n return_tensors=True,\r\n)\r\npipeline_with_padding_in_kwarg = pipeline(\r\n \"feature-extraction\",\r\n model=\"bert-base-uncased\",\r\n tokenize_kwargs={\"padding\": \"max_length\"},\r\n return_tensors=True,\r\n)\r\n```\r\n\r\nAlso please note that `padding: \"max_length\"` makes unnecessary long tensors in most cases, slowing down the overal inference of your model.\r\n\r\nI would refrain heavily from using it in a pipeline. `{\"padding\": True}` should be better.", "That makes sense. Do we need to make it clear in the documentation about how to use `tokenize_kwargs`?", "PRs to make it clearer are welcome for sure ! ", "@Narsil Submitted PR #22031 that adds the `tokenize_kwargs` definition in `FeatureExtrationPipeline`. Thanks!", "Thanks it looks great." ]

1,678

CONTRIBUTOR

null

### System Info - `transformers` version: 4.26.0 - Platform: Linux-4.19.0-23-cloud-amd64-x86_64-with-debian-10.13 - Python version: 3.7.12 - Huggingface_hub version: 0.12.0 - PyTorch version (GPU?): 1.13.1+cu116 (True) - Tensorflow version (GPU?): not installed (NA) - Flax version (CPU?/GPU?/TPU?): not installed (NA) - Jax version: not installed - JaxLib version: not installed - Using GPU in script?: <fill in> - Using distributed or parallel set-up in script?: <fill in> ### Who can help? @Narsil ### Information - [ ] The official example scripts - [ ] My own modified scripts ### Tasks - [ ] An officially supported task in the `examples` folder (such as GLUE/SQuAD, ...) - [ ] My own task or dataset (give details below) ### Reproduction The documentation for the [FeatureExtractionPipeline](https://huggingface.co/docs/transformers/v4.26.1/en/main_classes/pipelines#transformers.FeatureExtractionPipeline) does not currently provide instructions on how to use the truncation and padding arguments. These arguments can be passed in the `tokenize_kwargs` parameter, which is parsed by the [self._sanitize_parameters](https://github.com/huggingface/transformers/blob/v4.26.1/src/transformers/pipelines/feature_extraction.py#L58) method. While the `truncation` argument can also be passed as a separate keyword argument, the `padding` argument can only be recognized if it is included in `tokenize_kwargs`. To improve clarity, it would be beneficial for the documentation to explicitly state that the existence of `tokenize_kwargs` parameter for passing tokenizer arguments and add that only the `truncation` argument can be used as a keyword argument, while other tokenizer parameters should be included in `tokenize_kwargs`. I can submit a PR to add the documentation if it sounds good to you! ### Expected behavior Below is the code to show that how padding should be used in FeatureExtractionPipeline. ```python example = "After stealing money from the bank vault, the bank robber was seen fishing on the Mississippi river bank." tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased", use_fast=False) pipeline_without_padding_as_an_argument = pipeline( "feature-extraction", model="bert-base-uncased", tokenizer=tokenizer, return_tensors=True, ) pipeline_with_padding_as_an_argument = pipeline( "feature-extraction", model="bert-base-uncased", tokenizer=tokenizer, padding="max_length", return_tensors=True, ) pipeline_with_padding_in_kwarg = pipeline( "feature-extraction", model="bert-base-uncased", tokenizer=tokenizer, tokenize_kwargs={"padding": "max_length"}, return_tensors=True, ) print( pipeline_without_padding_as_an_argument(example).shape ) # torch.Size([1, 22, 768]) print( pipeline_with_padding_as_an_argument(example).shape ) # torch.Size([1, 22, 768]) padding = max_length not working print( pipeline_with_padding_in_kwarg(example).shape ) # torch.Size([1, 512, 768]) padding = max_length working ```

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21971/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21971/timeline

completed

null

https://api.github.com/repos/huggingface/transformers/issues/21970

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21970/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21970/comments

https://api.github.com/repos/huggingface/transformers/issues/21970/events

https://github.com/huggingface/transformers/issues/21970

1,611,579,178

I_kwDOCUB6oc5gDr8q

21,970

Unable to load a pretrained model

{ "login": "poojitharamachandra", "id": 39840406, "node_id": "MDQ6VXNlcjM5ODQwNDA2", "avatar_url": "https://avatars.githubusercontent.com/u/39840406?v=4", "gravatar_id": "", "url": "https://api.github.com/users/poojitharamachandra", "html_url": "https://github.com/poojitharamachandra", "followers_url": "https://api.github.com/users/poojitharamachandra/followers", "following_url": "https://api.github.com/users/poojitharamachandra/following{/other_user}", "gists_url": "https://api.github.com/users/poojitharamachandra/gists{/gist_id}", "starred_url": "https://api.github.com/users/poojitharamachandra/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/poojitharamachandra/subscriptions", "organizations_url": "https://api.github.com/users/poojitharamachandra/orgs", "repos_url": "https://api.github.com/users/poojitharamachandra/repos", "events_url": "https://api.github.com/users/poojitharamachandra/events{/privacy}", "received_events_url": "https://api.github.com/users/poojitharamachandra/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "Could you paste the code you are using?\r\n```py\r\nfrom transformers import AutoTokenizer, AutoModel\r\n\r\ntokenizer = AutoTokenizer.from_pretrained(\"microsoft/unixcoder-base\")\r\nmodel = AutoModel.from_pretrained(\"microsoft/unixcoder-base\")\r\n```\r\nworks perfectly fine.", "from transformers import (WEIGHTS_NAME, AdamW, get_linear_schedule_with_warmup,\r\n RobertaConfig, RobertaModel, RobertaTokenizer) \r\n\r\n tokenizer = RobertaTokenizer.from_pretrained(args.model_name_or_path)\r\n config = RobertaConfig.from_pretrained(args.model_name_or_path)\r\n model = RobertaModel.from_pretrained(args.model_name_or_path) \r\n.............................\r\n\r\n\r\n\r\n\r\n\r\n\r\n model = RobertaModel.from_pretrained('./changesets_model')\r\n model = Model(model)\r\n model.to(args.device)", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md) are likely to be ignored." ]

1,678

1,681

NONE

null

### System Info - `transformers` version: 4.18.0 - Platform: Linux-4.18.0-425.10.1.el8_7.x86_64-x86_64-with-glibc2.28 - Python version: 3.10.4 - Huggingface_hub version: 0.2.1 - PyTorch version (GPU?): 1.11.0 (True) - Tensorflow version (GPU?): not installed (NA) - Flax version (CPU?/GPU?/TPU?): not installed (NA) - Jax version: not installed - JaxLib version: not installed - Using GPU in script?: yes - Using distributed or parallel set-up in script?: no OSError: Can't load config for '~/Unixcoder/model/changesets_model'. If you were trying to load it from 'https://huggingface.co/models', make sure you don't have a local directory with the same name. Otherwise, make sure '~/Unixcoder/model/changesets_model' is the correct path to a directory containing a config.json file I am trying to load a second model during the course of training the first model. I am able to load the model for the first time, but not for the second time(even though all the config files are present) I am trying to load the model downloaded from https://huggingface.co/microsoft/unixcoder-base/tree/main ### Who can help? _No response_ ### Information - [X] The official example scripts - [ ] My own modified scripts ### Tasks - [ ] An officially supported task in the `examples` folder (such as GLUE/SQuAD, ...) - [X] My own task or dataset (give details below) ### Reproduction 1. load the model from downloaded version of https://huggingface.co/microsoft/unixcoder-base/tree/main 2. train the model on custom dataset 3. load the similar model that is pretrained on different dataset ### Expected behavior model is successfully loaded

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21970/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21970/timeline

completed

null

https://api.github.com/repos/huggingface/transformers/issues/21969

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21969/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21969/comments

https://api.github.com/repos/huggingface/transformers/issues/21969/events

https://github.com/huggingface/transformers/pull/21969

1,611,573,027

PR_kwDOCUB6oc5LYHbj

21,969

Add check before int casting for PIL conversion

{ "login": "amyeroberts", "id": 22614925, "node_id": "MDQ6VXNlcjIyNjE0OTI1", "avatar_url": "https://avatars.githubusercontent.com/u/22614925?v=4", "gravatar_id": "", "url": "https://api.github.com/users/amyeroberts", "html_url": "https://github.com/amyeroberts", "followers_url": "https://api.github.com/users/amyeroberts/followers", "following_url": "https://api.github.com/users/amyeroberts/following{/other_user}", "gists_url": "https://api.github.com/users/amyeroberts/gists{/gist_id}", "starred_url": "https://api.github.com/users/amyeroberts/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/amyeroberts/subscriptions", "organizations_url": "https://api.github.com/users/amyeroberts/orgs", "repos_url": "https://api.github.com/users/amyeroberts/repos", "events_url": "https://api.github.com/users/amyeroberts/events{/privacy}", "received_events_url": "https://api.github.com/users/amyeroberts/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,678

COLLABORATOR

null

# What does this PR do? Adds safeguards when images are converted to PIL images: * Additional condition if inferring `do_rescale` * Raise an error if values cannot be cast to `uint8` The PIL library is used for resizing images in image processors. If not explicitly set, whether or not to rescale pixel values is inferred based on the input type: if float then values are multiplied by 255. If the input image has integer values between 0-255, but are of floating type, then these pixel are rescaled to values between [0, 65025]. This results in overflow errors when cast to `uint` [here](https://github.com/huggingface/transformers/blob/bc33fbf956eef62d0ba8d3cd67ee955ad5defcdb/src/transformers/image_transforms.py#L162) before converting to a `PIL.Image.Image`. Fixes #21915 ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [x] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [ ] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [x] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [x] Did you write any new necessary tests?

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21969/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21969/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21969", "html_url": "https://github.com/huggingface/transformers/pull/21969", "diff_url": "https://github.com/huggingface/transformers/pull/21969.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21969.patch", "merged_at": 1678187650000 }

https://api.github.com/repos/huggingface/transformers/issues/21968

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21968/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21968/comments

https://api.github.com/repos/huggingface/transformers/issues/21968/events

https://github.com/huggingface/transformers/pull/21968

1,611,560,310

PR_kwDOCUB6oc5LYElM

21,968

[TF] Fix creating a PR while pushing in TF framework

{ "login": "ArthurZucker", "id": 48595927, "node_id": "MDQ6VXNlcjQ4NTk1OTI3", "avatar_url": "https://avatars.githubusercontent.com/u/48595927?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ArthurZucker", "html_url": "https://github.com/ArthurZucker", "followers_url": "https://api.github.com/users/ArthurZucker/followers", "following_url": "https://api.github.com/users/ArthurZucker/following{/other_user}", "gists_url": "https://api.github.com/users/ArthurZucker/gists{/gist_id}", "starred_url": "https://api.github.com/users/ArthurZucker/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ArthurZucker/subscriptions", "organizations_url": "https://api.github.com/users/ArthurZucker/orgs", "repos_url": "https://api.github.com/users/ArthurZucker/repos", "events_url": "https://api.github.com/users/ArthurZucker/events{/privacy}", "received_events_url": "https://api.github.com/users/ArthurZucker/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "Yep, adding this", "I had a question, the `create_pr` function parameter is not available in the corresponding `PyTorch` or `Flax` implementation, I was wondering if this is intended. ", "Not sure I follow, torch already has this parameter. " ]

1,678

COLLABORATOR

null

# What does this PR do? Fixes #21967, where models in TF can't push and open a PR. A test should probably be added

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21968/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21968/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21968", "html_url": "https://github.com/huggingface/transformers/pull/21968", "diff_url": "https://github.com/huggingface/transformers/pull/21968.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21968.patch", "merged_at": 1678206728000 }

https://api.github.com/repos/huggingface/transformers/issues/21967

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21967/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21967/comments

https://api.github.com/repos/huggingface/transformers/issues/21967/events

https://github.com/huggingface/transformers/issues/21967

1,611,559,489

I_kwDOCUB6oc5gDnJB

21,967

[TF] Can't open a pr when pushing

{ "login": "ArthurZucker", "id": 48595927, "node_id": "MDQ6VXNlcjQ4NTk1OTI3", "avatar_url": "https://avatars.githubusercontent.com/u/48595927?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ArthurZucker", "html_url": "https://github.com/ArthurZucker", "followers_url": "https://api.github.com/users/ArthurZucker/followers", "following_url": "https://api.github.com/users/ArthurZucker/following{/other_user}", "gists_url": "https://api.github.com/users/ArthurZucker/gists{/gist_id}", "starred_url": "https://api.github.com/users/ArthurZucker/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ArthurZucker/subscriptions", "organizations_url": "https://api.github.com/users/ArthurZucker/orgs", "repos_url": "https://api.github.com/users/ArthurZucker/repos", "events_url": "https://api.github.com/users/ArthurZucker/events{/privacy}", "received_events_url": "https://api.github.com/users/ArthurZucker/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "cc @gante " ]

1,678

COLLABORATOR

null

Creating a PR when uploading a model should work. ```python >>> from transformers import FlaxT5ForConditionalGeneration, TFT5ForConditionalGeneration >>> import jax.numpy as jnp >>> model = FlaxT5ForConditionalGeneration.from_pretrained("./art/flan-ul2", dtype = jnp.bfloat16, from_pt = True) >>> model.push_to_hub("google/flan-ul2", use_auth_token = "XXXXX",create_pr = True) >>> del model >>> model = TFT5ForConditionalGeneration.from_pretrained("./art/flan-ul2", from_pt = True) >>> model.push_to_hub("google/flan-ul2", use_auth_token = "XXXXX",create_pr = True) File "/home/arthur_huggingface_co/transformers/src/transformers/modeling_tf_utils.py", line 2986, in push_to_hub self.create_model_card(**base_model_card_args) TypeError: create_model_card() got an unexpected keyword argument 'create_pr' ```

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21967/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/21967/timeline

completed

null

https://api.github.com/repos/huggingface/transformers/issues/21966

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21966/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21966/comments

https://api.github.com/repos/huggingface/transformers/issues/21966/events

https://github.com/huggingface/transformers/pull/21966

1,611,555,392

PR_kwDOCUB6oc5LYDg6

21,966

Use larger atol in `torch.allclose` for some tests

{ "login": "ydshieh", "id": 2521628, "node_id": "MDQ6VXNlcjI1MjE2Mjg=", "avatar_url": "https://avatars.githubusercontent.com/u/2521628?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ydshieh", "html_url": "https://github.com/ydshieh", "followers_url": "https://api.github.com/users/ydshieh/followers", "following_url": "https://api.github.com/users/ydshieh/following{/other_user}", "gists_url": "https://api.github.com/users/ydshieh/gists{/gist_id}", "starred_url": "https://api.github.com/users/ydshieh/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ydshieh/subscriptions", "organizations_url": "https://api.github.com/users/ydshieh/orgs", "repos_url": "https://api.github.com/users/ydshieh/repos", "events_url": "https://api.github.com/users/ydshieh/events{/privacy}", "received_events_url": "https://api.github.com/users/ydshieh/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,678

COLLABORATOR

null

# What does this PR do? Running CI against torch 2.0 (and therefore CUDA 11.7), some tests for `BridgeTowerModel` failed: - test_disk_offload - test_cpu_offload - test_model_parallelism While with torch `1.13.1` (with CUDA 11.6), the difference between `base_output` and `new_output` in these 3 tests are 0.0, we get `1e-7 ~ 3e-6` as difference for torch `2.0`. **Deep debugging reveals that the first non-zero difference occurs when `nn.MultiheadAttention` is called.** This PR increase the atol to `1e-5` (the default one is `1e-8`) in `ModelTesterMixin`. If we don't feel super good with this for all model tests, we can override these 3 tests in `BridgeTowerModelTest`. (But when more models start using `nn.MultiheadAttention`, it's best to keep this larger value in common testing file)

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21966/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21966/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21966", "html_url": "https://github.com/huggingface/transformers/pull/21966", "diff_url": "https://github.com/huggingface/transformers/pull/21966.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21966.patch", "merged_at": 1678120861000 }

https://api.github.com/repos/huggingface/transformers/issues/21965

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21965/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21965/comments

https://api.github.com/repos/huggingface/transformers/issues/21965/events

https://github.com/huggingface/transformers/pull/21965

1,611,092,004

PR_kwDOCUB6oc5LWfE4

21,965

[🛠️] Fix-whisper-breaking-changes

{ "login": "ArthurZucker", "id": 48595927, "node_id": "MDQ6VXNlcjQ4NTk1OTI3", "avatar_url": "https://avatars.githubusercontent.com/u/48595927?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ArthurZucker", "html_url": "https://github.com/ArthurZucker", "followers_url": "https://api.github.com/users/ArthurZucker/followers", "following_url": "https://api.github.com/users/ArthurZucker/following{/other_user}", "gists_url": "https://api.github.com/users/ArthurZucker/gists{/gist_id}", "starred_url": "https://api.github.com/users/ArthurZucker/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ArthurZucker/subscriptions", "organizations_url": "https://api.github.com/users/ArthurZucker/orgs", "repos_url": "https://api.github.com/users/ArthurZucker/repos", "events_url": "https://api.github.com/users/ArthurZucker/events{/privacy}", "received_events_url": "https://api.github.com/users/ArthurZucker/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "Test are failing because we do not check the `model.config` updating and will look good! ", "The same should now be applied to both the `TF` and the `Flax` version as the overwriting of the `generate` function is also supported. Will open a follow up PR for these ", "For TF and flax #21334" ]

1,678

1,706

1,678

COLLABORATOR

null

# What does this PR do? Should fix the backward compatibility issue with `model.config.forced_decoder_ids = ...` and should help users who want to generate with timestamps. Fixes #21937 and #21878

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21965/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21965/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21965", "html_url": "https://github.com/huggingface/transformers/pull/21965", "diff_url": "https://github.com/huggingface/transformers/pull/21965.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21965.patch", "merged_at": 1678782229000 }

https://api.github.com/repos/huggingface/transformers/issues/21964

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21964/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21964/comments

https://api.github.com/repos/huggingface/transformers/issues/21964/events

https://github.com/huggingface/transformers/pull/21964

1,610,812,653

PR_kwDOCUB6oc5LVjNb

21,964

Add BridgeTowerForContrastiveLearning

{ "login": "abhiwand", "id": 12353176, "node_id": "MDQ6VXNlcjEyMzUzMTc2", "avatar_url": "https://avatars.githubusercontent.com/u/12353176?v=4", "gravatar_id": "", "url": "https://api.github.com/users/abhiwand", "html_url": "https://github.com/abhiwand", "followers_url": "https://api.github.com/users/abhiwand/followers", "following_url": "https://api.github.com/users/abhiwand/following{/other_user}", "gists_url": "https://api.github.com/users/abhiwand/gists{/gist_id}", "starred_url": "https://api.github.com/users/abhiwand/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/abhiwand/subscriptions", "organizations_url": "https://api.github.com/users/abhiwand/orgs", "repos_url": "https://api.github.com/users/abhiwand/repos", "events_url": "https://api.github.com/users/abhiwand/events{/privacy}", "received_events_url": "https://api.github.com/users/abhiwand/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "@sgugger We have addressed all of your review in the latest commit. We have also added tests for BridgeTowerForContrastiveLearning. Would you please review the latest commit? We are looking forward to having this PR merged into main.\r\nThanks", "Thanks for adding this model @abhiwand @tileintel ! Great to see it merged into the library :) \r\n\r\nFor the model tests, some of the configuration values in `BridgeTowerModelTester` result in large models being created and used in the test suite e.g. `vocab_size = 50265` set [here](https://github.com/huggingface/transformers/blob/bcc8d30affba29c594320fc80e4a4422fb850175/tests/models/bridgetower/test_modeling_bridgetower.py#L97), which results in periodic OOM errors in the CI runs. \r\n\r\nCould you add a follow up PR for `BridgeTowerModelTester` and `BridgeTowerModelTest` to have smaller default values to create lighter tests? A good reference for this would be [CLIP](https://github.com/huggingface/transformers/blob/main/tests/models/clip/test_modeling_clip.py). Ideally we would also have a similar structure of test classes for the different modalities i.e. `BridgeTower[Text|Vision]ModelTest(er)`. " ]

1,678

1,686

1,678

CONTRIBUTOR

null

# What does this PR do?   Fixes # (issue) ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [x] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [ ] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [x] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [ ] Did you write any new necessary tests? ## Who can review? Anyone in the community is free to review the PR once the tests have passed. Feel free to tag members/contributors who may be interested in your PR.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21964/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21964/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21964", "html_url": "https://github.com/huggingface/transformers/pull/21964", "diff_url": "https://github.com/huggingface/transformers/pull/21964.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21964.patch", "merged_at": 1678284055000 }

https://api.github.com/repos/huggingface/transformers/issues/21963

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21963/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21963/comments

https://api.github.com/repos/huggingface/transformers/issues/21963/events

https://github.com/huggingface/transformers/pull/21963

1,610,702,098

PR_kwDOCUB6oc5LVLNL

21,963

Fix bert issue

{ "login": "saswatmeher", "id": 35535056, "node_id": "MDQ6VXNlcjM1NTM1MDU2", "avatar_url": "https://avatars.githubusercontent.com/u/35535056?v=4", "gravatar_id": "", "url": "https://api.github.com/users/saswatmeher", "html_url": "https://github.com/saswatmeher", "followers_url": "https://api.github.com/users/saswatmeher/followers", "following_url": "https://api.github.com/users/saswatmeher/following{/other_user}", "gists_url": "https://api.github.com/users/saswatmeher/gists{/gist_id}", "starred_url": "https://api.github.com/users/saswatmeher/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/saswatmeher/subscriptions", "organizations_url": "https://api.github.com/users/saswatmeher/orgs", "repos_url": "https://api.github.com/users/saswatmeher/repos", "events_url": "https://api.github.com/users/saswatmeher/events{/privacy}", "received_events_url": "https://api.github.com/users/saswatmeher/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "@younesbelkada or @gante, Can anyone of you please help me with the error I am getting after running fix-copies. Running fix-copies made changes to multiple files(19) and now I am getting tests_torch, tests_torch_and_tf error on it.", "Hi @saswatmeher \r\nThanks for the PR! \r\nI would probably try:\r\n1- `pip install --upgrade -e .[\"quality\"]` and then run `make fix-copies`\r\nLet us know if this works!", "@younesbelkada It worked. Thanks! " ]

1,678

CONTRIBUTOR

null

# What does this PR do? This PR fixes a bug that a user can encounter while using generate and models that use gradient_checkpointing. Fixes issue #21737 for Bert.   ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [x] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [x] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case.(#21737) - [ ] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [ ] Did you write any new necessary tests? ## Who can review? @younesbelkada, @gante

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21963/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21963/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21963", "html_url": "https://github.com/huggingface/transformers/pull/21963", "diff_url": "https://github.com/huggingface/transformers/pull/21963.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21963.patch", "merged_at": 1678114532000 }

https://api.github.com/repos/huggingface/transformers/issues/21962

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21962/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21962/comments

https://api.github.com/repos/huggingface/transformers/issues/21962/events

https://github.com/huggingface/transformers/issues/21962

1,610,693,995

I_kwDOCUB6oc5gAT1r

21,962

use datasets streaming mode in trainer ddp mode cause memory leak

{ "login": "gromzhu", "id": 15223544, "node_id": "MDQ6VXNlcjE1MjIzNTQ0", "avatar_url": "https://avatars.githubusercontent.com/u/15223544?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gromzhu", "html_url": "https://github.com/gromzhu", "followers_url": "https://api.github.com/users/gromzhu/followers", "following_url": "https://api.github.com/users/gromzhu/following{/other_user}", "gists_url": "https://api.github.com/users/gromzhu/gists{/gist_id}", "starred_url": "https://api.github.com/users/gromzhu/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gromzhu/subscriptions", "organizations_url": "https://api.github.com/users/gromzhu/orgs", "repos_url": "https://api.github.com/users/gromzhu/repos", "events_url": "https://api.github.com/users/gromzhu/events{/privacy}", "received_events_url": "https://api.github.com/users/gromzhu/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "cc @lhoestq ", "Hi ! The axis 0 in your plot is time. Do you know how many training steps it corresponds to ?\r\n\r\nFYI the `text` loading in `datasets` samples text files line by line (with a buffer of >10MB to avoid small IO calls).\r\nMoreover .shuffle() uses a shuffle buffer of 1,000 examples, and batched `map` uses batches of 1,000 examples as well.\r\n\r\nTherefore unless there is a major leak somewhere, `datasets` doesn't use much RAM in streaming mode.\r\n\r\nFeel free to try profiling memory usage and check where the biggest source of memory comes from in the code, that would be super helpful to diagnose the potential memory leak and fix it.", "> Hi ! The axis 0 in your plot is time. Do you know how many training steps it corresponds to ?\r\n> \r\n> FYI the `text` loading in `datasets` samples text files line by line (with a buffer of >10MB to avoid small IO calls). Moreover .shuffle() uses a shuffle buffer of 1,000 examples, and batched `map` uses batches of 1,000 examples as well.\r\n> \r\n> Therefore unless there is a major leak somewhere, `datasets` doesn't use much RAM in streaming mode.\r\n> \r\n> Feel free to try profiling memory usage and check where the biggest source of memory comes from in the code, that would be super helpful to diagnose the potential memory leak and fix it.\r\n\r\nit is roughly about between 500000 steps - 650000 steps ", "> Hi ! The axis 0 in your plot is time. Do you know how many training steps it corresponds to ?\r\n> \r\n> FYI the `text` loading in `datasets` samples text files line by line (with a buffer of >10MB to avoid small IO calls). Moreover .shuffle() uses a shuffle buffer of 1,000 examples, and batched `map` uses batches of 1,000 examples as well.\r\n> \r\n> Therefore unless there is a major leak somewhere, `datasets` doesn't use much RAM in streaming mode.\r\n> \r\n> Feel free to try profiling memory usage and check where the biggest source of memory comes from in the code, that would be super helpful to diagnose the potential memory leak and fix it.\r\n\r\ni try use memory_profiler to profiling memory ,but profiling memory can only report the master thread memory. do you know which tool can report the dataloader worker thread memory?", "I haven't tried memory_profiler with multiprocessing, but you can already try iterating on the DataLoader without multiprocessing and check if you observe a memory leak.", "> I haven't tried memory_profiler with multiprocessing, but you can already try iterating on the DataLoader without multiprocessing and check if you observe a memory leak.\r\n\r\nset dataloader_num=0 ,no memory leak", "This could be an issue with the torch `DataLoader` then, or python multiprocessing.\r\n\r\nOn the `datasets` side this is the whole code that `yield` example if `num_worker > 0`:\r\n\r\nhttps://github.com/huggingface/datasets/blob/c5ca1d86949ec3a5fdaec03b80500fb822bcfab4/src/datasets/iterable_dataset.py#L843\r\n\r\nwhich is almost identical to the code without multiprocessing:\r\n\r\nhttps://github.com/huggingface/datasets/blob/c5ca1d86949ec3a5fdaec03b80500fb822bcfab4/src/datasets/iterable_dataset.py#L937-L945\r\n\r\nCould you check on another environment that you also observe the memory leak ?", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md) are likely to be ignored." ]

1,678

1,681

NONE

null

### System Info pytorch 1.11.0 py 3.8 cuda 11.3 transformers 4.26.1 datasets 2.9.0 ### Who can help? @sgugger ### Information - [ ] The official example scripts - [X] My own modified scripts ### Tasks - [ ] An officially supported task in the `examples` folder (such as GLUE/SQuAD, ...) - [X] My own task or dataset (give details below) ### Reproduction import os import time import datetime import sys import numpy as np import random import torch from torch.utils.data import Dataset, DataLoader, random_split, RandomSampler, SequentialSampler,DistributedSampler,BatchSampler torch.manual_seed(42) from transformers import GPT2LMHeadModel, GPT2Tokenizer, GPT2Config, GPT2Model,DataCollatorForLanguageModeling,AutoModelForCausalLM from transformers import AdamW, get_linear_schedule_with_warmup hf_model_path ='./Wenzhong-GPT2-110M' tokenizer = GPT2Tokenizer.from_pretrained(hf_model_path) tokenizer.add_special_tokens({'pad_token': '<|pad|>'}) from datasets import load_dataset gpus=8 max_len = 576 batch_size_node = 17 save_step = 5000 gradient_accumulation = 2 dataloader_num = 4 max_step = 351000*1000//batch_size_node//gradient_accumulation//gpus #max_step = -1 print("total_step:%d"%(max_step)) import datasets datasets.__version__ dataset = load_dataset("text", data_files="./gpt_data_v1/*",split='train',cache_dir='./dataset_cache',streaming=True) print('load over') shuffled_dataset = dataset.shuffle(seed=42) print('shuffle over') def dataset_tokener(example,max_lenth=max_len): example['text'] = list(map(lambda x : x.strip()+'<|endoftext|>',example['text'] )) return tokenizer(example['text'], truncation=True, max_length=max_lenth, padding="longest") #return tokenizer(example[0], truncation=True, max_length=max_lenth, padding="max_length") new_new_dataset = shuffled_dataset.map(dataset_tokener, batched=True, remove_columns=["text"]) print('map over') configuration = GPT2Config.from_pretrained(hf_model_path, output_hidden_states=False) model = AutoModelForCausalLM.from_pretrained(hf_model_path) model.resize_token_embeddings(len(tokenizer)) seed_val = 42 random.seed(seed_val) np.random.seed(seed_val) torch.manual_seed(seed_val) torch.cuda.manual_seed_all(seed_val) from transformers import Trainer,TrainingArguments import os print("strat train") training_args = TrainingArguments(output_dir="./test_trainer", num_train_epochs=1.0, report_to="none", do_train=True, dataloader_num_workers=dataloader_num, local_rank=int(os.environ.get('LOCAL_RANK', -1)), overwrite_output_dir=True, logging_strategy='steps', logging_first_step=True, logging_dir="./logs", log_on_each_node=False, per_device_train_batch_size=batch_size_node, warmup_ratio=0.03, save_steps=save_step, save_total_limit=5, gradient_accumulation_steps=gradient_accumulation, max_steps=max_step, disable_tqdm=False, data_seed=42 ) trainer = Trainer( model=model, args=training_args, train_dataset=new_new_dataset, eval_dataset=None, tokenizer=tokenizer, # Data collator will default to DataCollatorWithPadding, so we change it. data_collator=DataCollatorForLanguageModeling(tokenizer,mlm=False), #compute_metrics=compute_metrics if training_args.do_eval and not is_torch_tpu_available() else None, #preprocess_logits_for_metrics=preprocess_logits_for_metrics #if training_args.do_eval and not is_torch_tpu_available() #else None, ) trainer.train(resume_from_checkpoint=True) ### Expected behavior use the train code uppper my dataset ./gpt_data_v1 have 1000 files, each file size is 120mb start cmd is : python -m torch.distributed.launch --nproc_per_node=8 my_train.py here is result: ![image](https://user-images.githubusercontent.com/15223544/223025177-6f2ec55f-5b2f-42bf-b6cc-dd5f8a8232fe.png) here is memory usage monitor in 12 hours ![image](https://user-images.githubusercontent.com/15223544/223027076-14e32e8b-9608-4282-9a80-f15d0277026d.png) every dataloader work allocate over 24gb cpu memory according to memory usage monitor in 12 hours,sometime small memory releases, but total memory usage is increase. i think datasets streaming mode should not used so much memery,so maybe somewhere has memory leak.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21962/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21962/timeline

completed

null

https://api.github.com/repos/huggingface/transformers/issues/21961

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21961/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21961/comments

https://api.github.com/repos/huggingface/transformers/issues/21961/events

https://github.com/huggingface/transformers/issues/21961

1,610,509,465

I_kwDOCUB6oc5f_myZ

21,961

Support customized vocabulary for decoding (in model.generate)

{ "login": "yuchenlin", "id": 10104354, "node_id": "MDQ6VXNlcjEwMTA0MzU0", "avatar_url": "https://avatars.githubusercontent.com/u/10104354?v=4", "gravatar_id": "", "url": "https://api.github.com/users/yuchenlin", "html_url": "https://github.com/yuchenlin", "followers_url": "https://api.github.com/users/yuchenlin/followers", "following_url": "https://api.github.com/users/yuchenlin/following{/other_user}", "gists_url": "https://api.github.com/users/yuchenlin/gists{/gist_id}", "starred_url": "https://api.github.com/users/yuchenlin/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/yuchenlin/subscriptions", "organizations_url": "https://api.github.com/users/yuchenlin/orgs", "repos_url": "https://api.github.com/users/yuchenlin/repos", "events_url": "https://api.github.com/users/yuchenlin/events{/privacy}", "received_events_url": "https://api.github.com/users/yuchenlin/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "I have read this post: https://huggingface.co/blog/constrained-beam-search\r\n\r\nBut it seems that such Constraints can only support constraints of ensuring some tokens are part of the sentences but cannot prevent other tokens to be selected during decoding. ", "Found this post to use `bad_word_list` as the whole vocab - customized vocab as the input: \r\n\r\nhttps://stackoverflow.com/questions/63920887/whitelist-tokens-for-text-generation-xlnet-gpt-2-in-huggingface-transformers\r\n\r\nWill have a try but sounds like a bit awkward to use. ", "cc @gante ", "Hey @yuchenlin 👋 \r\n\r\nMy first approach would be to use `bad_word_list`, passing to it all but the tokens you want to use. It's a no-code approach, but perhaps not the most efficient computationally.\r\n\r\nAlternatively, you can write your own processor class that sets to `-inf` the logits of all but the tokens you want to consider. To do it, you would have to:\r\n1. Write your own class that implements the logic. You can see plenty of examples in [this file](https://github.com/huggingface/transformers/blob/main/src/transformers/generation/logits_process.py)\r\n2. Use your class at generation time, e.g. \r\n```py\r\ntokens_to_keep = tokenizer(xxx) # xxx = list with your valid words\r\nmy_processor = MyLogitsProcessorClass(tokens_to_keep=tokens_to_keep)\r\nmodel.generate(inputs, ..., logits_processor=LogitsProcessorList([my_processor]))\r\n```\r\n\r\nI hope this short guide helps 🤗 ", "Hi @gante ,\r\n\r\nThanks a lot! Yeah I have tried with the `bad_wordLlist` (see example below) and I found that the generated outputs are much worse than before although they are indeed constrained to the given vocabulary. I was using beam search and I'm not sure if it is because that the vocab is so small that the normalization or other process becomes unstable. \r\n\r\nI will try the logit processor idea as well. Thank you! :D \r\n\r\n```python\r\nfrom transformers import AutoTokenizer, AutoModelForSeq2SeqLM\r\n\r\ntokenizer = AutoTokenizer.from_pretrained(\"google/flan-t5-small\")\r\n\r\nmodel = AutoModelForSeq2SeqLM.from_pretrained(\"google/flan-t5-small\")\r\n\r\n \r\nwhitelist = [\"move\", \"it\", \"pick\", \"up\", \"focus\", \"on\"]\r\nwhitelist_ids = [tokenizer.encode(word)[0] for word in whitelist]\r\nbad_words_ids=[[id] for id in range(tokenizer.vocab_size) if id not in whitelist_ids]\r\n\r\n\r\nencoder_input_str = \"Explain this concept to me: machine learning\"\r\ninput_ids = tokenizer(encoder_input_str, return_tensors=\"pt\").input_ids\r\n\r\noutputs = model.generate(\r\n input_ids,\r\n num_beams=10,\r\n do_sample=False,\r\n num_return_sequences=1,\r\n no_repeat_ngram_size=1,\r\n remove_invalid_values=True,\r\n bad_words_ids = bad_words_ids,\r\n)\r\nprint(tokenizer.decode(outputs[0], skip_special_tokens=True))\r\n```", "@yuchenlin haha yes, the quality of the output will likely decrease significantly, that is to be expected! \r\n\r\nInstead of whitelisting words, consider the \"soft-whitelisting\" alternative: increase the odds of picking a token from the whitelist. You can easily implement this by changing the repetition penalty logits processor to boost the odds of certain tokens :)", "Thanks a lot for the advice! I currently used a simpler method --- adding some random tokens (say 30% of the whole vocab) to the whitelist and it seems to help. \r\n\r\nWill also try your idea soon! Thanks again! :D ", "Just in case you are interested in more diversity of these constraints, I wrote a whole package and paper about this idea: https://github.com/Hellisotherpeople/Constrained-Text-Generation-Studio", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md) are likely to be ignored." ]

1,678

1,681

CONTRIBUTOR

null

### Feature request Use case: Given a small list of tokens that is a subset of the whole vocabulary of the tokenizers for T5. For example, ["put", "move", "pick", "up", "on", "in", "apple", "bag", ....] And when we decode by using `model.generate()`, we want the model only output sentences that consist of words in the above list (i.e., limited vocabulary for beam searching or sampling). Maybe it is already supported in some way? ### Motivation For some applications, we only want to decode sentences with a limited vocabulary instead of allowing open-ended generation. ### Your contribution I'm not sure what is the best way to add this feature, if it is easy to limit the vocab for generate functions, then I can help add this PR.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21961/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21961/timeline

completed

null

https://api.github.com/repos/huggingface/transformers/issues/21960

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21960/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21960/comments

https://api.github.com/repos/huggingface/transformers/issues/21960/events

https://github.com/huggingface/transformers/pull/21960

1,610,465,758

PR_kwDOCUB6oc5LUXzN

21,960

Add missing parameter definition in layoutlm config

{ "login": "Atomnp", "id": 45496355, "node_id": "MDQ6VXNlcjQ1NDk2MzU1", "avatar_url": "https://avatars.githubusercontent.com/u/45496355?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Atomnp", "html_url": "https://github.com/Atomnp", "followers_url": "https://api.github.com/users/Atomnp/followers", "following_url": "https://api.github.com/users/Atomnp/following{/other_user}", "gists_url": "https://api.github.com/users/Atomnp/gists{/gist_id}", "starred_url": "https://api.github.com/users/Atomnp/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Atomnp/subscriptions", "organizations_url": "https://api.github.com/users/Atomnp/orgs", "repos_url": "https://api.github.com/users/Atomnp/repos", "events_url": "https://api.github.com/users/Atomnp/events{/privacy}", "received_events_url": "https://api.github.com/users/Atomnp/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "cc @amyeroberts " ]

1,678

CONTRIBUTOR

null

Four parameters in `LayoutLM` config were missing definitions, Added their definition (copied from BertConfig). # What does this PR do? Fix docs, add parameter definition copying them from BertConfig ## Before submitting - [X] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [ ] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [ ] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [ ] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [ ] Did you write any new necessary tests? ## Who can review? Anyone in the community is free to review the PR once the tests have passed. Feel free to tag members/contributors who may be interested in your PR.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21960/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/21960/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21960", "html_url": "https://github.com/huggingface/transformers/pull/21960", "diff_url": "https://github.com/huggingface/transformers/pull/21960.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21960.patch", "merged_at": 1678116011000 }

https://api.github.com/repos/huggingface/transformers/issues/21959

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21959/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21959/comments

https://api.github.com/repos/huggingface/transformers/issues/21959/events

https://github.com/huggingface/transformers/pull/21959

1,610,282,038

PR_kwDOCUB6oc5LTyLR

21,959

Fix MinNewTokensLengthLogitsProcessor when used with a list of eos tokens

{ "login": "eladsegal", "id": 13485709, "node_id": "MDQ6VXNlcjEzNDg1NzA5", "avatar_url": "https://avatars.githubusercontent.com/u/13485709?v=4", "gravatar_id": "", "url": "https://api.github.com/users/eladsegal", "html_url": "https://github.com/eladsegal", "followers_url": "https://api.github.com/users/eladsegal/followers", "following_url": "https://api.github.com/users/eladsegal/following{/other_user}", "gists_url": "https://api.github.com/users/eladsegal/gists{/gist_id}", "starred_url": "https://api.github.com/users/eladsegal/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/eladsegal/subscriptions", "organizations_url": "https://api.github.com/users/eladsegal/orgs", "repos_url": "https://api.github.com/users/eladsegal/repos", "events_url": "https://api.github.com/users/eladsegal/events{/privacy}", "received_events_url": "https://api.github.com/users/eladsegal/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,678

CONTRIBUTOR

null

# What does this PR do? MinNewTokensLengthLogitsProcessor is missing support for a list of eos token ids. This PR adds the missing support, in the same way it was added in MinLengthLogitsProcessor. ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [x] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [ ] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [ ] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [x] Did you write any new necessary tests? ## Who can review? Anyone in the community is free to review the PR once the tests have passed. Feel free to tag members/contributors who may be interested in your PR. @gante

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21959/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21959/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21959", "html_url": "https://github.com/huggingface/transformers/pull/21959", "diff_url": "https://github.com/huggingface/transformers/pull/21959.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21959.patch", "merged_at": 1678190363000 }

https://api.github.com/repos/huggingface/transformers/issues/21958

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21958/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21958/comments

https://api.github.com/repos/huggingface/transformers/issues/21958/events

https://github.com/huggingface/transformers/issues/21958

1,610,101,047

I_kwDOCUB6oc5f-DE3

21,958

Cannot get the model weight of T5 INT8 model with Transformers 4.26.1

{ "login": "XuhuiRen", "id": 44249229, "node_id": "MDQ6VXNlcjQ0MjQ5MjI5", "avatar_url": "https://avatars.githubusercontent.com/u/44249229?v=4", "gravatar_id": "", "url": "https://api.github.com/users/XuhuiRen", "html_url": "https://github.com/XuhuiRen", "followers_url": "https://api.github.com/users/XuhuiRen/followers", "following_url": "https://api.github.com/users/XuhuiRen/following{/other_user}", "gists_url": "https://api.github.com/users/XuhuiRen/gists{/gist_id}", "starred_url": "https://api.github.com/users/XuhuiRen/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/XuhuiRen/subscriptions", "organizations_url": "https://api.github.com/users/XuhuiRen/orgs", "repos_url": "https://api.github.com/users/XuhuiRen/repos", "events_url": "https://api.github.com/users/XuhuiRen/events{/privacy}", "received_events_url": "https://api.github.com/users/XuhuiRen/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "Cool that this seems to have been fixed for you! Could you tell us what the solution was? (for futur reference and other users who might stumble upon the smae problem)", "@ArthurZucker Hi, Arthur, we still did not fix this issue. Could you please have a check for this issue? We just modify the previous description to make the problem more clear for you to solve. ", "Hey! Is there a reason why you are not using `load_in_8bit = True`? If you install the `bits_and_bytes` library, getting the 8bit quantized version of the model is as easy as the following:\r\n```python \r\n import transformers\r\n from datasets import load_dataset\r\n model_name = 't5-small'\r\n model = transformers.AutoModelForSeq2SeqLM.from_pretrained(model_name,load_in_8bit = True, device_map=\"auto\")\r\n```", "Hello @XuhuiRen ,\r\n\r\nYour script worked fine on the `main` branch of `transformers`\r\n\r\n```python\r\nimport torch\r\nimport transformers\r\nfrom datasets import load_dataset\r\nmodel_name = 't5-small'\r\nmodel_fp32 = transformers.AutoModelForSeq2SeqLM.from_pretrained(\r\n model_name,\r\n)\r\nmodel_int8 = torch.ao.quantization.quantize_dynamic(\r\n model_fp32,\r\n {torch.nn.Linear},\r\n dtype=torch.qint8\r\n)\r\n\r\noutput = model_int8.generate(torch.LongTensor([[0, 1, 2, 3]]))\r\nprint(output)\r\n```\r\n\r\nCan you try it with:\r\n```\r\npip install git+https://github.com/huggingface/transformers.git\r\n```\r\n\r\nFor more context, the PR: https://github.com/huggingface/transformers/pull/21843 solved your issue", "Hi, @ArthurZucker and @younesbelkada really thanks for your reply. Your solution is works for my issue. " ]

1,678

NONE

null

### System Info - `transformers` version: 4.26.1 - Platform: Linux-3.10.0-862.el7.x86_64-x86_64-with-glibc2.17 - Python version: 3.8.13 - Huggingface_hub version: 0.12.1 - PyTorch version (GPU?): 1.13.0+cpu - Tensorflow version (GPU?): not installed (No) - Flax version (CPU?/GPU?/TPU?): not installed (NA) - Jax version: not installed - JaxLib version: not installed - Using GPU in script?: <No> ### Who can help? @ArthurZucker @younesbelkada @sgu ### Information - [ ] The official example scripts - [X] My own modified scripts ### Tasks - [X] An officially supported task in the `examples` folder (such as GLUE/SQuAD, ...) - [ ] My own task or dataset (give details below) ### Reproduction The code is shown as follows: ```python import torch import transformers from datasets import load_dataset model_name = 't5-small' model_fp32 = transformers.AutoModelForSeq2SeqLM.from_pretrained( model_name, ) model_int8 = torch.ao.quantization.quantize_dynamic( model_fp32, {torch.nn.Linear}, dtype=torch.qint8) def get_example_inputs(model_name, dataset_name='sst2'): tokenizer = transformers.AutoTokenizer.from_pretrained(model_name) dataset = load_dataset(dataset_name, split='validation') text = dataset[0]['text'] if dataset_name=='lambada' else dataset[0]['sentence'] input = tokenizer(text, padding='max_length', max_length=195, return_tensors='pt') example_inputs = input['input_ids'][0].to('cpu').unsqueeze(0) return example_inputs example_inputs = get_example_inputs(model_name, dataset_name='lambada') output = model_int8.generate(example_inputs) print(output) ``` The error message is shown as follows, [Issue Report.txt](https://github.com/huggingface/transformers/files/10905558/Issue.Report.txt) ### Expected behavior This example's expected behavior is quantizing the FP32 T5 model into INT8 format using INT8 model to generate the output. This code could make a success with the previous version Transformers 4.26.0. After the recent updates, This code cannot run normally anymore. We found the error is result from the fix of another issue: https://github.com/huggingface/transformers/issues/20287. That is probably because that you have used weight from "self.wo". But that weight becomes a function in the INT8 module. It should be used by "self.wo.weight()". Please reconsider the previous fix for that issue to make it compatible.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21958/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21958/timeline

completed

null

https://api.github.com/repos/huggingface/transformers/issues/21957

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21957/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21957/comments

https://api.github.com/repos/huggingface/transformers/issues/21957/events

https://github.com/huggingface/transformers/pull/21957

1,610,062,223

PR_kwDOCUB6oc5LTG1G

21,957

Update expected values in `XLMProphetNetModelIntegrationTest`

{ "login": "ydshieh", "id": 2521628, "node_id": "MDQ6VXNlcjI1MjE2Mjg=", "avatar_url": "https://avatars.githubusercontent.com/u/2521628?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ydshieh", "html_url": "https://github.com/ydshieh", "followers_url": "https://api.github.com/users/ydshieh/followers", "following_url": "https://api.github.com/users/ydshieh/following{/other_user}", "gists_url": "https://api.github.com/users/ydshieh/gists{/gist_id}", "starred_url": "https://api.github.com/users/ydshieh/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ydshieh/subscriptions", "organizations_url": "https://api.github.com/users/ydshieh/orgs", "repos_url": "https://api.github.com/users/ydshieh/repos", "events_url": "https://api.github.com/users/ydshieh/events{/privacy}", "received_events_url": "https://api.github.com/users/ydshieh/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,677

1,678

COLLABORATOR

null

# What does this PR do? After #21870, we also need to update some expected values in `XLMProphetNetModelIntegrationTest` (as has been done for `ProphetNetModelIntegrationTest` in that PR)

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21957/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21957/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21957", "html_url": "https://github.com/huggingface/transformers/pull/21957", "diff_url": "https://github.com/huggingface/transformers/pull/21957.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21957.patch", "merged_at": 1678090545000 }

https://api.github.com/repos/huggingface/transformers/issues/21956

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21956/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21956/comments

https://api.github.com/repos/huggingface/transformers/issues/21956/events

https://github.com/huggingface/transformers/pull/21956

1,610,029,650

PR_kwDOCUB6oc5LTAp0

21,956

[Generate] Fix gradient_checkpointing and use_cache bug for BLOOM

{ "login": "asrimanth", "id": 30816357, "node_id": "MDQ6VXNlcjMwODE2MzU3", "avatar_url": "https://avatars.githubusercontent.com/u/30816357?v=4", "gravatar_id": "", "url": "https://api.github.com/users/asrimanth", "html_url": "https://github.com/asrimanth", "followers_url": "https://api.github.com/users/asrimanth/followers", "following_url": "https://api.github.com/users/asrimanth/following{/other_user}", "gists_url": "https://api.github.com/users/asrimanth/gists{/gist_id}", "starred_url": "https://api.github.com/users/asrimanth/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/asrimanth/subscriptions", "organizations_url": "https://api.github.com/users/asrimanth/orgs", "repos_url": "https://api.github.com/users/asrimanth/repos", "events_url": "https://api.github.com/users/asrimanth/events{/privacy}", "received_events_url": "https://api.github.com/users/asrimanth/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,677

1,678

CONTRIBUTOR

null

Fixes #21737 for Bloom. ## Before submitting - [x] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [x] Was this discussed/approved via a GitHub issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [ ] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). ## Who can review? cc @younesbelkada @gante

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21956/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21956/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21956", "html_url": "https://github.com/huggingface/transformers/pull/21956", "diff_url": "https://github.com/huggingface/transformers/pull/21956.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21956.patch", "merged_at": 1678114601000 }

https://api.github.com/repos/huggingface/transformers/issues/21955

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21955/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21955/comments

https://api.github.com/repos/huggingface/transformers/issues/21955/events

https://github.com/huggingface/transformers/pull/21955

1,609,993,956

PR_kwDOCUB6oc5LS6Cq

21,955

LLaMA Implementation

{ "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/followers", "following_url": "https://api.github.com/users/zphang/following{/other_user}", "gists_url": "https://api.github.com/users/zphang/gists{/gist_id}", "starred_url": "https://api.github.com/users/zphang/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/zphang/subscriptions", "organizations_url": "https://api.github.com/users/zphang/orgs", "repos_url": "https://api.github.com/users/zphang/repos", "events_url": "https://api.github.com/users/zphang/events{/privacy}", "received_events_url": "https://api.github.com/users/zphang/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "does this work with int8?", "> does this work with int8?\r\n\r\nNo idea! I haven't messed with int8 too much myself. It ought to be compatible with whatever is already supported in the HF models.", "nice work! thanks for the upload and I hope it gets pulled", "_The documentation is not available anymore as the PR was closed or merged._", "It looks like the tests which are currently failing are unrelated to the LLaMA code, so this should be good to review/use.\r\n\r\nIf folks can try it out (particularly with the larger, sharded models) and see if there are any issues, that will be helpful!", "> It looks like the tests which are currently failing are unrelated to the LLaMA code, so this should be good to review/use.\r\n> \r\n> If folks can try it out (particularly with the larger, sharded models) and see if there are any issues, that will be helpful!\r\n\r\nAt lest the convert script seems to work fine. I was able to convert 7B to 30B. I do not have enough ram to convert 65B.", "Great work. thanks for putting this together", "After replacing transformers from Kobold with this PR I am able to load the shards as expected. Just I cant generate anything because Kobold still needs some changes.\r\n![image](https://user-images.githubusercontent.com/7269941/222938895-2e8b9d71-6a88-417d-b7ed-14d8216d2ef4.png)\r\n", "> > does this work with int8?\r\n> \r\n> No idea! I haven't messed with int8 too much myself. It ought to be compatible with whatever is already supported in the HF models.\r\n\r\nInt8 seems not working but float16 is fine, in my hasty put-together test at https://github.com/zsc/llama_infer . Please throw a comment in case you find something!", "@zphang I'm not able to get something like `tokenizer = AutoTokenizer.from_pretrained(\"/data/llama/hf/7b/tokenizer/\")` to work. Is this intentional or just leaving AutoTokenizer for future work?", "> @zphang I'm not able to get something like `tokenizer = AutoTokenizer.from_pretrained(\"/data/llama/hf/7b/tokenizer/\")` to work. Is this intentional or just leaving AutoTokenizer for future work?\r\n\r\nWhat issue are you having / what is the error?", "I have tested the code and these are my findings:\r\n\r\n1. The conversion script works.\r\n2. Loading the model works.\r\n3. Loading the tokenizer with `transformers.LLaMATokenizer.from_pretrained` works.\r\n4. Loading the tokenizer with `AutoTokenizer.from_pretrained` does not work and generates this error:\r\n\r\n```\r\nOSError: /tmp/converted/tokenizer/ does not appear to have a file named config.json. Checkout\r\n'https://huggingface.co//tmp/converted/tokenizer//None' for available files.\r\n\r\n```\r\n\r\n5. The generated text seems to be incoherent. If I try these default values for the generation parameters:\r\n\r\n```\r\nmodel.generate(input_ids, eos_token_id=2, do_sample=True, temperature=1, top_p=1, typical_p=1, repetition_penalty=1, top_k=50, min_length=0, no_repeat_ngram_size=0, num_beams=1, penalty_alpha=0, length_penalty=1, early_stopping=False, max_new_tokens=200).cuda()\r\n```\r\n\r\nwith this prompt:\r\n\r\n```\r\nCommon sense questions and answers\r\n\r\nQuestion: What color is the sky?\r\nFactual answer:\r\n```\r\n\r\nI get\r\n\r\n```\r\nCommon sense questions and answers\r\n\r\nQuestion: What color is the sky?\r\nFactual answer: Tags: python, django, django-models\r\n\r\nQuestion: Using Django with multiple databases\r\n\r\nI am attempting to use django with multiple databases, and I have the following code:\r\n\r\n\\begin{code}\r\nDATABASES = {\r\n 'default': {\r\n 'ENGINE': 'django.db.backends.sqlite3',\r\n 'NAME': ':memory:',\r\n },\r\n 'db_one': {\r\n 'ENGINE': 'django.db.backends.sqlite3',\r\n 'NAME': 'db_one',\r\n },\r\n 'db_two': {\r\n 'ENGINE': 'django.db.backends.sqlite3',\r\n 'NAME': 'db_two',\r\n },\r\n}\r\n```\r\n\r\nIt seems to me that prompts are being completely ignored.\r\n\r\n6. Loading in 8-bit mode with `load_in_8bit=True` works.", "This is OK: `tokenizer = transformers.LLaMATokenizer.from_pretrained(\"/data/llama/hf/7b/tokenizer/\")`\r\n\r\nIf using `tokenizer = AutoTokenizer.from_pretrained(\"/data/llama/hf/7b/tokenizer/\"` then it will complain no \"config.json\". \r\n```\r\nOSError: /data/llama/hf/7b/tokenizer/ does not appear to have a file named config.json. Checkout \r\n'https://huggingface.co//data/llama/hf/7b/tokenizer//None' for available files.\r\n```\r\n\r\nI then hacked by softlinking `/data/llama/hf/7b/tokenizer/special_tokens_map.json` to `/data/llama/hf/7b/tokenizer/config.json` and it works. So maybe just rename?\r\n\r\nAnyway, can now happily play with LLaMA in Hugging Face world and thanks for the great work!", "Thanks for the comments. Looks like the saved tokenizer doesn't work for `AutoTokenizer` but works if you directly instantiate from `LLaMATokenizer`. Maybe one of the HF folks can chime in on the best way to address that.\r\n\r\n> The generated text seems to be incoherent. If I try these default values for the generation parameters:\r\n\r\nCan you check the input_ids you're using to generate? The tokenizer currently adds both BOS and EOS tokens by default, and an EOS might cause the model to ignore your prompt.\r\n\r\nPerhaps I can set EOS to not be added by default so it operates closer to expected behavior.", "For this prompt:\r\n\r\n```\r\n'Common sense questions and answers\\n\\nQuestion: What color is the sky?\\nFactual answer:'\r\n```\r\n\r\nthese are the input_ids:\r\n\r\n```\r\ntensor([[ 1, 13103, 4060, 5155, 322, 6089, 13, 13, 16492, 29901,\r\n 1724, 2927, 338, 278, 14744, 29973, 13, 29943, 19304, 1234,\r\n 29901, 2]], device='cuda:0')\r\n```\r\n\r\nI do not know how to interpret these numbers, but if there is an EOS token in that tensor and that token is causing the text generation to derail, changing that default would be valuable.", "1 is BOS and 2 is EOS. Can you try without the last input id?\r\n\r\nI also added an example in my PR message.", "I confirm that doing this \r\n```\r\n input_ids = input_ids[:, :-1]\r\n\r\n```\r\n\r\nto remove the last input id before calling `model.generate(...)` causes the text generation to become coherent:\r\n\r\n```\r\nCommon sense questions and answers\r\n\r\nQuestion: What color is the sky?\r\nFactual answer: The sky is blue. The sky is blue, and it is a fact that it is blue. The sky is indisputably blue.\r\n\r\n```", "Added a commit that should fix the tokenizer issues, and not add BOS and EOS by default.", "Awesome, I confirm that the text generation is coherent by default now.\r\n\r\nI still cannot load the tokenizer with `AutoTokenizer.from_pretrained`. The error has now changed to this:\r\n\r\n```\r\n File \"/tmp/transformers/src/transformers/models/auto/tokenization_auto.py\", line 694, in from_pretrained\r\n tokenizer_class_py, tokenizer_class_fast = TOKENIZER_MAPPING[type(config)]\r\n File \"/tmp/transformers/src/transformers/models/auto/auto_factory.py\", line 610, in __getitem__\r\n raise KeyError(key)\r\nKeyError: <class 'transformers.models.llama.configuration_llama.LLaMAConfig'>\r\n\r\n```", "> > does this work with int8?\r\n> \r\n> No idea! I haven't messed with int8 too much myself. It ought to be compatible with whatever is already supported in the HF models.\r\n\r\nAfter the fix with EOS, int8 (bitsandbytes) looks decent. Example in https://github.com/zsc/llama_infer/blob/main/README.md", "After https://github.com/huggingface/transformers/pull/21955/commits/459e2ac9f551650ced58deb1c65f06c3d483d606, `AutoTokenizer.from_pretrained` now works as expected.\r\n", "KoboldAI now works", "I'd like to see a more memory-efficient conversion script, the current version loads everything into system memory which makes converting the 30B and 65B variants challenging on some systems", "Yes, this is a quick and dirty version that loads everything into memory.\r\nOne issue is that the way the weights are sharded (for tensor parallelism) is orthogonal to the way that HF shards the weights (by layer). So either we have to load everything in at once, or we have to load/write multiple times. The latter would be slower but useful for folks with less memory.", "Has anyone tested loading 65B with `accelerate` to load on multiple GPUs?", "I can't load the 7B model to cuda with one A4000\r\nshould I just change the gpu?\r\n", "I'm observing some strange behavior with the tokenizer when encoding sequences beginning with a newline:\r\n```\r\n>>> t = AutoTokenizer.from_pretrained(\"llama_hf/tokenizer\")\r\n>>> res = t.encode(\"\\nYou:\")\r\n>>> res\r\n[29871, 13, 3492, 29901]\r\n>>> t.decode(res)\r\n'You:'\r\n```\r\n\r\nThe newline seems to get lost somewhere along the way.\r\n\r\nEDIT: Looking into this, it seems it might be the expected behavior of `sentencepiece`.", "> Has anyone tested loading 65B with `accelerate` to load on multiple GPUs?\r\n\r\n||fp16|int8(bitsandbytes)|\r\n|--|--|--|\r\n|V100|OK, 5xV100|Bad results, short generated sequences|\r\n|A100|OK, 6xA100 when using \"auto\"|OK, 3xA100|\r\n\r\nYes, I currently have a 65B fp16 model running on 6xV100 now (5X should be enough). My working code is at https://github.com/zsc/llama_infer/ . If there are CUDA OOM due to bad distribution of weights among cards, one thing worth trying is tweaking the device_map (`accelerate` seems to only counts weights when enforcing the memory cap in device_map, so there is an art for setting custom cap a little lower for every card, especially card 0).\r\n\r\nStrangely, int8 (LLM.int8 to be specific) for 65B model works like a charm on A100, but leads to bad results on V100 with abnormally short generated sequences.", "> Strangely, int8 (LLM.int8 to be specific) for 65B model works like a charm on A100, but leads to bad results on V100 with abnormally short generated sequences.\r\n\r\nI will have a look at this later next week. The V100 takes a different code path than the A100 because the V100 does not support Int8 tensor cores. I think that is the issue here. We will soon publish FP4 inference which should be more universal and easier to use.", "Jumping on @thomasw21 comment, we sadly cannot accept any code licensed GPLv3 as it would taint the whole Transformers library under that license. This means that the modeling code should be copied from GPT-NeoX whenever possible (with Copied from statements) since I believe that this model is very close to it and that you should be super familiar with it @zphang ;-) , and that no parts of the modeling code should be copy-pasted from the original Llama code.\r\n\r\nWe also cannot attribute Copyright to Meta-AI /Meta in all those files, as attributing that copyright would admit the code in the PR is based on theirs and thus get us back to the license problem." ]

1,677

1,691

1,678

CONTRIBUTOR

null

# What does this PR do? Implementation of LLaMA models (https://arxiv.org/abs/2302.13971). Model weights can be requested [here](https://docs.google.com/forms/d/e/1FAIpQLSfqNECQnMkycAp2jP4Z9TFX0cGR4uf7b_fBxjY_OjhJILlKGA/viewform). Weight conversion script is included. Weights conversion can be run via: ```bash python src/transformers/models/llama/convert_llama_weights_to_hf.py \ --input_dir /path/to/downloaded/llama/weights \ --model_size 7B \ --output_dir /output/path ``` Models can then be loaded via: ```python tokenizer = transformers.LLaMATokenizer.from_pretrained("/output/path/tokenizer/") model = transformers.LLaMAForCausalLM.from_pretrained("/output/path/llama-7b/") ``` Example: ```bash batch = tokenizer( "The primary use of LLaMA is research on large language models, including", return_tensors="pt", add_special_tokens=False ) batch = {k: v.cuda() for k, v in batch.items()} generated = model.generate(batch["input_ids"], max_length=100) print(tokenizer.decode(generated[0])) ```   Fixes https://github.com/huggingface/transformers/issues/21796 ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [x] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [x] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [x] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [x] Did you write any new necessary tests? ## Who can review? Anyone in the community is free to review the PR once the tests have passed. Feel free to tag members/contributors who may be interested in your PR. @ArthurZucker @younesbelkada

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21955/reactions", "total_count": 200, "+1": 14, "-1": 0, "laugh": 4, "hooray": 4, "confused": 0, "heart": 88, "rocket": 85, "eyes": 5 }

https://api.github.com/repos/huggingface/transformers/issues/21955/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21955", "html_url": "https://github.com/huggingface/transformers/pull/21955", "diff_url": "https://github.com/huggingface/transformers/pull/21955.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21955.patch", "merged_at": null }

https://api.github.com/repos/huggingface/transformers/issues/21954

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21954/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21954/comments

https://api.github.com/repos/huggingface/transformers/issues/21954/events

https://github.com/huggingface/transformers/issues/21954

1,609,969,774

I_kwDOCUB6oc5f9jBu

21,954

`from_pretrained` broken 4.26.1

{ "login": "NicholasKao1029", "id": 45542006, "node_id": "MDQ6VXNlcjQ1NTQyMDA2", "avatar_url": "https://avatars.githubusercontent.com/u/45542006?v=4", "gravatar_id": "", "url": "https://api.github.com/users/NicholasKao1029", "html_url": "https://github.com/NicholasKao1029", "followers_url": "https://api.github.com/users/NicholasKao1029/followers", "following_url": "https://api.github.com/users/NicholasKao1029/following{/other_user}", "gists_url": "https://api.github.com/users/NicholasKao1029/gists{/gist_id}", "starred_url": "https://api.github.com/users/NicholasKao1029/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/NicholasKao1029/subscriptions", "organizations_url": "https://api.github.com/users/NicholasKao1029/orgs", "repos_url": "https://api.github.com/users/NicholasKao1029/repos", "events_url": "https://api.github.com/users/NicholasKao1029/events{/privacy}", "received_events_url": "https://api.github.com/users/NicholasKao1029/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "Hey! If you go to `hf.co/runwayml/stable-diffusion-v1-5` you will se that there are no `tokenizer.json` or any files related to tokenization. However there is a `tokenizer` folder. The new version of transformers supports `subfolder`. \r\nThe model on the hub was modified, thus the command that you are looking for is probably: \r\n```python \r\n>>> from transformers import AutoTokenizer\r\n>>> AutoTokenizer.from_pretrained(\"runwayml/stable-diffusion-v1-5\",subfolder= \"tokenizer\")\r\n```\r\nwhich worked for me 😉 ", "oh yes this is great thank you" ]

1,677

1,678

NONE

null

### System Info - `transformers` version: 4.26.1 - Platform: Linux-5.15.0-53-generic-x86_64-with-glibc2.10 - Python version: 3.8.13 - Huggingface_hub version: 0.12.1 - PyTorch version (GPU?): 1.13.0a0+d0d6b1f (True) - Tensorflow version (GPU?): not installed (NA) - Flax version (CPU?/GPU?/TPU?): not installed (NA) - Jax version: not installed - JaxLib version: not installed - Using GPU in script?: yes - Using distributed or parallel set-up in script?: no ### Who can help? @amyeroberts @ArthurZucker and @younesbelkada ### Information - [ ] The official example scripts - [ ] My own modified scripts ### Tasks - [ ] An officially supported task in the `examples` folder (such as GLUE/SQuAD, ...) - [ ] My own task or dataset (give details below) ### Reproduction code block ``` from transformers import CLIPTokenizer, CLIPTextModel import torch tokenizer=CLIPTokenizer.from_pretrained("runwayml/stable-diffusion-v1-5"), text_encoder=CLIPTextModel.from_pretrained("runwayml/stable-diffusion-v1-5").to("cuda") ``` output ``` Traceback (most recent call last): File "diffusers-oneflow/examples/poc.py", line 17, in <module> tokenizer=CLIPTokenizer.from_pretrained("runwayml/stable-diffusion-v1-5"), File "/opt/conda/lib/python3.8/site-packages/transformers/tokenization_utils_base.py", line 1788, in from_pretrained raise EnvironmentError( OSError: Can't load tokenizer for 'runwayml/stable-diffusion-v1-5'. If you were trying to load it from 'https://huggingface.co/models', make sure you don't have a local directory with the same name. Otherwise, make sure 'runwayml/stable-diffusion-v1-5' is the correct path to a directory containing all relevant files for a CLIPTokenizer tokenizer. ``` ### Expected behavior Expect to be able to consume and not throw

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21954/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21954/timeline

completed

null

https://api.github.com/repos/huggingface/transformers/issues/21953

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21953/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21953/comments

https://api.github.com/repos/huggingface/transformers/issues/21953/events

https://github.com/huggingface/transformers/pull/21953

1,609,908,395

PR_kwDOCUB6oc5LSqJD

21,953

Disable DDP for neuron

{ "login": "sangeethabal", "id": 83724701, "node_id": "MDQ6VXNlcjgzNzI0NzAx", "avatar_url": "https://avatars.githubusercontent.com/u/83724701?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sangeethabal", "html_url": "https://github.com/sangeethabal", "followers_url": "https://api.github.com/users/sangeethabal/followers", "following_url": "https://api.github.com/users/sangeethabal/following{/other_user}", "gists_url": "https://api.github.com/users/sangeethabal/gists{/gist_id}", "starred_url": "https://api.github.com/users/sangeethabal/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sangeethabal/subscriptions", "organizations_url": "https://api.github.com/users/sangeethabal/orgs", "repos_url": "https://api.github.com/users/sangeethabal/repos", "events_url": "https://api.github.com/users/sangeethabal/events{/privacy}", "received_events_url": "https://api.github.com/users/sangeethabal/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,677

1,678

CONTRIBUTOR

null

# What does this PR do? This PR is to disable DDP when using neuron. Currently, we are overwriting the _wrap_model function in trainer.py to disable DDP. We want to avoid overwriting by disabling DDP when using neuron.   ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [x] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [ ] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [ ] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [ ] Did you write any new necessary tests? ## Who can review? @sgugger

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21953/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21953/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21953", "html_url": "https://github.com/huggingface/transformers/pull/21953", "diff_url": "https://github.com/huggingface/transformers/pull/21953.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21953.patch", "merged_at": 1678113225000 }

https://api.github.com/repos/huggingface/transformers/issues/21952

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21952/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21952/comments

https://api.github.com/repos/huggingface/transformers/issues/21952/events

https://github.com/huggingface/transformers/pull/21952

1,609,867,086

PR_kwDOCUB6oc5LSjHd

21,952

docs: improve clarity for language modeling

{ "login": "pdhall99", "id": 20580126, "node_id": "MDQ6VXNlcjIwNTgwMTI2", "avatar_url": "https://avatars.githubusercontent.com/u/20580126?v=4", "gravatar_id": "", "url": "https://api.github.com/users/pdhall99", "html_url": "https://github.com/pdhall99", "followers_url": "https://api.github.com/users/pdhall99/followers", "following_url": "https://api.github.com/users/pdhall99/following{/other_user}", "gists_url": "https://api.github.com/users/pdhall99/gists{/gist_id}", "starred_url": "https://api.github.com/users/pdhall99/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/pdhall99/subscriptions", "organizations_url": "https://api.github.com/users/pdhall99/orgs", "repos_url": "https://api.github.com/users/pdhall99/repos", "events_url": "https://api.github.com/users/pdhall99/events{/privacy}", "received_events_url": "https://api.github.com/users/pdhall99/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,677

1,678

CONTRIBUTOR

null

# What does this PR do? - Improve clarity of tasks/language_modeling and tasks/masked_language_modeling docs. - In example preprocessing, remove `truncation=True` parameter from `tokenizer` so texts aren't truncated before being concatenated and chunked. ## Before submitting - [x] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [x] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [ ] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [ ] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [ ] Did you write any new necessary tests? ## Who can review? @sgugger, @stevhliu

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21952/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21952/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21952", "html_url": "https://github.com/huggingface/transformers/pull/21952", "diff_url": "https://github.com/huggingface/transformers/pull/21952.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21952.patch", "merged_at": 1678126423000 }

https://api.github.com/repos/huggingface/transformers/issues/21951

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21951/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21951/comments

https://api.github.com/repos/huggingface/transformers/issues/21951/events

https://github.com/huggingface/transformers/issues/21951

1,609,682,819

I_kwDOCUB6oc5f8c-D

21,951

TimeSeriesTransformerModel - 'features' Is 'NoneType'

{ "login": "LtlSh", "id": 109275417, "node_id": "U_kgDOBoNpGQ", "avatar_url": "https://avatars.githubusercontent.com/u/109275417?v=4", "gravatar_id": "", "url": "https://api.github.com/users/LtlSh", "html_url": "https://github.com/LtlSh", "followers_url": "https://api.github.com/users/LtlSh/followers", "following_url": "https://api.github.com/users/LtlSh/following{/other_user}", "gists_url": "https://api.github.com/users/LtlSh/gists{/gist_id}", "starred_url": "https://api.github.com/users/LtlSh/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/LtlSh/subscriptions", "organizations_url": "https://api.github.com/users/LtlSh/orgs", "repos_url": "https://api.github.com/users/LtlSh/repos", "events_url": "https://api.github.com/users/LtlSh/events{/privacy}", "received_events_url": "https://api.github.com/users/LtlSh/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "@LtlSh can you kindly try with the main branch of transformers?", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md) are likely to be ignored." ]

1,677

1,681

NONE

null

### System Info Sat Mar 4 10:44:51 2023 +-----------------------------------------------------------------------------+ | NVIDIA-SMI 527.41 Driver Version: 527.41 CUDA Version: 12.0 | |-------------------------------+----------------------+----------------------+ | GPU Name TCC/WDDM | Bus-Id Disp.A | Volatile Uncorr. ECC | | Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. | | | | MIG M. | |===============================+======================+======================| | 0 NVIDIA GeForce ... WDDM | 00000000:01:00.0 On | N/A | | 30% 37C P8 18W / 350W | 550MiB / 24576MiB | 27% Default | | | | N/A | +-------------------------------+----------------------+----------------------+ +-----------------------------------------------------------------------------+ | Processes: | | GPU GI CI PID Type Process name GPU Memory | | ID ID Usage | |=============================================================================| | 0 N/A N/A 1580 C+G ...y\ShellExperienceHost.exe N/A | | 0 N/A N/A 5332 C+G ...e\PhoneExperienceHost.exe N/A | | 0 N/A N/A 7560 C+G ...logioptionsplus_agent.exe N/A | | 0 N/A N/A 7672 C+G ...5n1h2txyewy\SearchApp.exe N/A | | 0 N/A N/A 8268 C+G C:\Windows\explorer.exe N/A | | 0 N/A N/A 11324 C+G ...cw5n1h2txyewy\LockApp.exe N/A | | 0 N/A N/A 12316 C+G ...(x86)\AnyDesk\AnyDesk.exe N/A | | 0 N/A N/A 12908 C+G ...ge\Application\msedge.exe N/A | | 0 N/A N/A 13064 C+G ...2txyewy\TextInputHost.exe N/A | | 0 N/A N/A 13764 C+G ...3d8bbwe\CalculatorApp.exe N/A | | 0 N/A N/A 14880 C+G ...lPanel\SystemSettings.exe N/A | +-----------------------------------------------------------------------------+ ### Who can help? _No response_ ### Information - [ ] The official example scripts - [X] My own modified scripts ### Tasks - [ ] An officially supported task in the `examples` folder (such as GLUE/SQuAD, ...) - [X] My own task or dataset (give details below) ### Reproduction Hi, We're trying to train TimeSeriesTransformerModel, but it seems that there's a variable called 'features' that's not getting assigned. However, we can't figure out why is this happening. Our code: ` # Initializing a default Time Series Transformer configuration configuration = TimeSeriesTransformerConfig(prediction_length = 327, lags_sequence = [0, 0, 0]) # Randomly initializing a model (with random weights) from the configuration model = TimeSeriesTransformerForPrediction(configuration) # Accessing the model configuration configuration = model.config #we dont know if passing the data as a dataframe instead if a tesndor would work #currently model.train() is throwing an error, maybe we need to use a gpu? TODO # Setting the model to training mode model.train() # Defining the loss function and optimizer loss_fn = torch.nn.MSELoss() optimizer = torch.optim.Adam(model.parameters(), lr=1e-3) # Training loop for epoch in range(100): for batch in dataloader: # Forward pass outputs = model( past_values=batch["past_values"], past_time_features=batch["past_time_features"], past_observed_mask=None, static_categorical_features=None, static_real_features=None, future_values=batch["future_values"], future_time_features=batch["future_time_features"], ) loss = loss_fn(outputs, batch) # Backward pass and optimization optimizer.zero_grad() loss.backward() optimizer.step() # Printing the training loss if (epoch + 1) % 10 == 0: print(f"Epoch [{epoch + 1}/100], Loss: {loss.item()}") ` We're getting this error: `Traceback (most recent call last): File "D:\Final Project\fMRI_Ariel_Lital\train.py", line 58, in <module> outputs = model( File "C:\Users\Cognition\anaconda3\envs\ArielLital\lib\site-packages\torch\nn\modules\module.py", line 1130, in _call_impl return forward_call(*input, **kwargs) File "C:\Users\Cognition\anaconda3\envs\ArielLital\lib\site-packages\transformers\models\time_series_transformer\modeling_time_series_transformer.py", line 1813, in forward outputs = self.model( File "C:\Users\Cognition\anaconda3\envs\ArielLital\lib\site-packages\torch\nn\modules\module.py", line 1130, in _call_impl return forward_call(*input, **kwargs) File "C:\Users\Cognition\anaconda3\envs\ArielLital\lib\site-packages\transformers\models\time_series_transformer\modeling_time_series_transformer.py", line 1626, in forward transformer_inputs, scale, static_feat = self.create_network_inputs( File "C:\Users\Cognition\anaconda3\envs\ArielLital\lib\site-packages\transformers\models\time_series_transformer\modeling_time_series_transformer.py", line 1532, in create_network_inputs embedded_cat = self.embedder(static_categorical_features) File "C:\Users\Cognition\anaconda3\envs\ArielLital\lib\site-packages\torch\nn\modules\module.py", line 1130, in _call_impl return forward_call(*input, **kwargs) File "C:\Users\Cognition\anaconda3\envs\ArielLital\lib\site-packages\transformers\models\time_series_transformer\modeling_time_series_transformer.py", line 250, in forward [ File "C:\Users\Cognition\anaconda3\envs\ArielLital\lib\site-packages\transformers\models\time_series_transformer\modeling_time_series_transformer.py", line 251, in <listcomp> embed(cat_feature_slice.squeeze(-1)) AttributeError: 'NoneType' object has no attribute 'squeeze'` Here is the source code where the error occurs - we noticed 'features' is NoneType but don't know why: `class FeatureEmbedder(nn.Module): def __init__(self, cardinalities: List[int], embedding_dims: List[int]) -> None: super().__init__() self.num_features = len(cardinalities) self.embedders = nn.ModuleList([nn.Embedding(c, d) for c, d in zip(cardinalities, embedding_dims)]) def forward(self, features: torch.Tensor) -> torch.Tensor: if self.num_features > 1: # we slice the last dimension, giving an array of length # self.num_features with shape (N,T) or (N) cat_feature_slices = torch.chunk(features, self.num_features, dim=-1) else: cat_feature_slices = [**features**] return torch.cat( [ embed(cat_feature_slice.squeeze(-1)) for embed, cat_feature_slice in zip(self.embedders, cat_feature_slices) ], dim=-1, ) ` Would appreciate your help. @ArthurZucker and @younesbelkadav [train.txt](https://github.com/huggingface/transformers/files/10887890/train.txt) ### Expected behavior We would like to train a TimeSeriesTransformerModelfor for forcasting on tabular data.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21951/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21951/timeline

completed

null

https://api.github.com/repos/huggingface/transformers/issues/21950

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21950/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21950/comments

https://api.github.com/repos/huggingface/transformers/issues/21950/events

https://github.com/huggingface/transformers/issues/21950

1,609,682,793

I_kwDOCUB6oc5f8c9p

21,950

auto_find_batch_size should say what batch size it is using

{ "login": "p-christ", "id": 26346243, "node_id": "MDQ6VXNlcjI2MzQ2MjQz", "avatar_url": "https://avatars.githubusercontent.com/u/26346243?v=4", "gravatar_id": "", "url": "https://api.github.com/users/p-christ", "html_url": "https://github.com/p-christ", "followers_url": "https://api.github.com/users/p-christ/followers", "following_url": "https://api.github.com/users/p-christ/following{/other_user}", "gists_url": "https://api.github.com/users/p-christ/gists{/gist_id}", "starred_url": "https://api.github.com/users/p-christ/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/p-christ/subscriptions", "organizations_url": "https://api.github.com/users/p-christ/orgs", "repos_url": "https://api.github.com/users/p-christ/repos", "events_url": "https://api.github.com/users/p-christ/events{/privacy}", "received_events_url": "https://api.github.com/users/p-christ/received_events", "type": "User", "site_admin": false }

[]

closed

false

{ "login": "muellerzr", "id": 7831895, "node_id": "MDQ6VXNlcjc4MzE4OTU=", "avatar_url": "https://avatars.githubusercontent.com/u/7831895?v=4", "gravatar_id": "", "url": "https://api.github.com/users/muellerzr", "html_url": "https://github.com/muellerzr", "followers_url": "https://api.github.com/users/muellerzr/followers", "following_url": "https://api.github.com/users/muellerzr/following{/other_user}", "gists_url": "https://api.github.com/users/muellerzr/gists{/gist_id}", "starred_url": "https://api.github.com/users/muellerzr/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/muellerzr/subscriptions", "organizations_url": "https://api.github.com/users/muellerzr/orgs", "repos_url": "https://api.github.com/users/muellerzr/repos", "events_url": "https://api.github.com/users/muellerzr/events{/privacy}", "received_events_url": "https://api.github.com/users/muellerzr/received_events", "type": "User", "site_admin": false }

[ { "login": "muellerzr", "id": 7831895, "node_id": "MDQ6VXNlcjc4MzE4OTU=", "avatar_url": "https://avatars.githubusercontent.com/u/7831895?v=4", "gravatar_id": "", "url": "https://api.github.com/users/muellerzr", "html_url": "https://github.com/muellerzr", "followers_url": "https://api.github.com/users/muellerzr/followers", "following_url": "https://api.github.com/users/muellerzr/following{/other_user}", "gists_url": "https://api.github.com/users/muellerzr/gists{/gist_id}", "starred_url": "https://api.github.com/users/muellerzr/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/muellerzr/subscriptions", "organizations_url": "https://api.github.com/users/muellerzr/orgs", "repos_url": "https://api.github.com/users/muellerzr/repos", "events_url": "https://api.github.com/users/muellerzr/events{/privacy}", "received_events_url": "https://api.github.com/users/muellerzr/received_events", "type": "User", "site_admin": false } ]

[ "cc @muellerzr ", "Also, it would be good if auto find batch size also worked for the eval batch size? Otherwise for the eval batch size I still have to guess a few times and hope I don't run out of memory?", "Thanks! Solved with https://github.com/huggingface/transformers/pull/23800", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md) are likely to be ignored." ]

1,677

1,687

NONE

null

### Feature request When using `auto_find_batch_size=True` in the trainer I believe it identifies the right batch size but then it doesn't log it to the console anywhere? It would be good if it could log what batch size it is using? ### Motivation I'd like to know what batch size it is using because then I will know roughly how big a batch can fit in memory - this info would be useful elsewhere. ### Your contribution N/A

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21950/reactions", "total_count": 11, "+1": 11, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21950/timeline

completed

null

https://api.github.com/repos/huggingface/transformers/issues/21949

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21949/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21949/comments

https://api.github.com/repos/huggingface/transformers/issues/21949/events

https://github.com/huggingface/transformers/pull/21949

1,609,614,458

PR_kwDOCUB6oc5LRxF4

21,949

just testing

{ "login": "Raman-Kumar", "id": 32980600, "node_id": "MDQ6VXNlcjMyOTgwNjAw", "avatar_url": "https://avatars.githubusercontent.com/u/32980600?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Raman-Kumar", "html_url": "https://github.com/Raman-Kumar", "followers_url": "https://api.github.com/users/Raman-Kumar/followers", "following_url": "https://api.github.com/users/Raman-Kumar/following{/other_user}", "gists_url": "https://api.github.com/users/Raman-Kumar/gists{/gist_id}", "starred_url": "https://api.github.com/users/Raman-Kumar/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Raman-Kumar/subscriptions", "organizations_url": "https://api.github.com/users/Raman-Kumar/orgs", "repos_url": "https://api.github.com/users/Raman-Kumar/repos", "events_url": "https://api.github.com/users/Raman-Kumar/events{/privacy}", "received_events_url": "https://api.github.com/users/Raman-Kumar/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/transformers/pr_21949). All of your documentation changes will be reflected on that endpoint.", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md) are likely to be ignored." ]

1,677

1,681

NONE

null

# What does this PR do?   Fixes # (issue) ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [ ] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [ ] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [ ] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [ ] Did you write any new necessary tests? ## Who can review? Anyone in the community is free to review the PR once the tests have passed. Feel free to tag members/contributors who may be interested in your PR.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21949/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21949/timeline

null

true

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21949", "html_url": "https://github.com/huggingface/transformers/pull/21949", "diff_url": "https://github.com/huggingface/transformers/pull/21949.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21949.patch", "merged_at": null }

https://api.github.com/repos/huggingface/transformers/issues/21948

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21948/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21948/comments

https://api.github.com/repos/huggingface/transformers/issues/21948/events

https://github.com/huggingface/transformers/pull/21948

1,609,610,111

PR_kwDOCUB6oc5LRwKE

21,948

test pull request for tokengt

{ "login": "Raman-Kumar", "id": 32980600, "node_id": "MDQ6VXNlcjMyOTgwNjAw", "avatar_url": "https://avatars.githubusercontent.com/u/32980600?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Raman-Kumar", "html_url": "https://github.com/Raman-Kumar", "followers_url": "https://api.github.com/users/Raman-Kumar/followers", "following_url": "https://api.github.com/users/Raman-Kumar/following{/other_user}", "gists_url": "https://api.github.com/users/Raman-Kumar/gists{/gist_id}", "starred_url": "https://api.github.com/users/Raman-Kumar/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Raman-Kumar/subscriptions", "organizations_url": "https://api.github.com/users/Raman-Kumar/orgs", "repos_url": "https://api.github.com/users/Raman-Kumar/repos", "events_url": "https://api.github.com/users/Raman-Kumar/events{/privacy}", "received_events_url": "https://api.github.com/users/Raman-Kumar/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/transformers/pr_21948). All of your documentation changes will be reflected on that endpoint.", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md) are likely to be ignored." ]

1,677

1,681

NONE

null

# What does this PR do?   Fixes # (issue) ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [ ] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [ ] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [ ] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [ ] Did you write any new necessary tests? ## Who can review? Anyone in the community is free to review the PR once the tests have passed. Feel free to tag members/contributors who may be interested in your PR.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21948/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21948/timeline

null

true

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21948", "html_url": "https://github.com/huggingface/transformers/pull/21948", "diff_url": "https://github.com/huggingface/transformers/pull/21948.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21948.patch", "merged_at": null }

https://api.github.com/repos/huggingface/transformers/issues/21947

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21947/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21947/comments

https://api.github.com/repos/huggingface/transformers/issues/21947/events

https://github.com/huggingface/transformers/issues/21947

1,609,541,724

I_kwDOCUB6oc5f76hc

21,947

Pre-training language model with Translation Language Modelling (TLM) objective

{ "login": "aloka-fernando", "id": 36888985, "node_id": "MDQ6VXNlcjM2ODg4OTg1", "avatar_url": "https://avatars.githubusercontent.com/u/36888985?v=4", "gravatar_id": "", "url": "https://api.github.com/users/aloka-fernando", "html_url": "https://github.com/aloka-fernando", "followers_url": "https://api.github.com/users/aloka-fernando/followers", "following_url": "https://api.github.com/users/aloka-fernando/following{/other_user}", "gists_url": "https://api.github.com/users/aloka-fernando/gists{/gist_id}", "starred_url": "https://api.github.com/users/aloka-fernando/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/aloka-fernando/subscriptions", "organizations_url": "https://api.github.com/users/aloka-fernando/orgs", "repos_url": "https://api.github.com/users/aloka-fernando/repos", "events_url": "https://api.github.com/users/aloka-fernando/events{/privacy}", "received_events_url": "https://api.github.com/users/aloka-fernando/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md) are likely to be ignored.", "If this is supported in the codebase please provide the steps or refer to an available resource.", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md) are likely to be ignored." ]

1,677

1,683

NONE

null

### Feature request For large-scale pre-training MLM objective is heavily used. eg: BERT (Devlin et al., 2018), XLM-R (Connaue et al.,2019). The resources available from huggingface is to train a language model (LM) using MLM or CLM objectives. (ie. https://github.com/huggingface/notebooks/blob/main/examples/language_modeling.ipynb) To recreate XLM (Lample and Connaue, 2019) I wish to pre-train my own language model using both MLM and TLM objectives. Please advice on how to do this using the huggingface transformers. Thank you. ### Motivation MLM + TLM is a common pre-training objective during language modelling training. Therefore for further improvement to these models, we need to first train using MLM + TLM objectives. In addition to this, if I need to customize the masking to be done for noun terms only, or verb terms only, please let me know whether this is allowed. ### Your contribution I would appreciate if the pre-training on TLM objective is also provided.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21947/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21947/timeline

completed

null

https://api.github.com/repos/huggingface/transformers/issues/21946

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21946/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21946/comments

https://api.github.com/repos/huggingface/transformers/issues/21946/events

https://github.com/huggingface/transformers/pull/21946

1,609,196,426

PR_kwDOCUB6oc5LQV3s

21,946

Fix gradient checkpointing bug in Roformer

{ "login": "KMFODA", "id": 35491698, "node_id": "MDQ6VXNlcjM1NDkxNjk4", "avatar_url": "https://avatars.githubusercontent.com/u/35491698?v=4", "gravatar_id": "", "url": "https://api.github.com/users/KMFODA", "html_url": "https://github.com/KMFODA", "followers_url": "https://api.github.com/users/KMFODA/followers", "following_url": "https://api.github.com/users/KMFODA/following{/other_user}", "gists_url": "https://api.github.com/users/KMFODA/gists{/gist_id}", "starred_url": "https://api.github.com/users/KMFODA/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/KMFODA/subscriptions", "organizations_url": "https://api.github.com/users/KMFODA/orgs", "repos_url": "https://api.github.com/users/KMFODA/repos", "events_url": "https://api.github.com/users/KMFODA/events{/privacy}", "received_events_url": "https://api.github.com/users/KMFODA/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,677

CONTRIBUTOR

null

This PR fixes a bug that a user can encounter while using generate and models that use gradient_checkpointing. Fixes Issue https://github.com/huggingface/transformers/issues/21737 cc @younesbelkada or @gante

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21946/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21946/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21946", "html_url": "https://github.com/huggingface/transformers/pull/21946", "diff_url": "https://github.com/huggingface/transformers/pull/21946.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21946.patch", "merged_at": 1677944674000 }

https://api.github.com/repos/huggingface/transformers/issues/21945

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21945/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21945/comments

https://api.github.com/repos/huggingface/transformers/issues/21945/events

https://github.com/huggingface/transformers/pull/21945

1,609,193,713

PR_kwDOCUB6oc5LQVSv

21,945

Fix gradient checkpointing bug in Rembert

{ "login": "KMFODA", "id": 35491698, "node_id": "MDQ6VXNlcjM1NDkxNjk4", "avatar_url": "https://avatars.githubusercontent.com/u/35491698?v=4", "gravatar_id": "", "url": "https://api.github.com/users/KMFODA", "html_url": "https://github.com/KMFODA", "followers_url": "https://api.github.com/users/KMFODA/followers", "following_url": "https://api.github.com/users/KMFODA/following{/other_user}", "gists_url": "https://api.github.com/users/KMFODA/gists{/gist_id}", "starred_url": "https://api.github.com/users/KMFODA/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/KMFODA/subscriptions", "organizations_url": "https://api.github.com/users/KMFODA/orgs", "repos_url": "https://api.github.com/users/KMFODA/repos", "events_url": "https://api.github.com/users/KMFODA/events{/privacy}", "received_events_url": "https://api.github.com/users/KMFODA/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,677

CONTRIBUTOR

null

This PR fixes a bug that a user can encounter while using generate and models that use gradient_checkpointing. Fixes Issue https://github.com/huggingface/transformers/issues/21737 cc @younesbelkada or @gante

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21945/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21945/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21945", "html_url": "https://github.com/huggingface/transformers/pull/21945", "diff_url": "https://github.com/huggingface/transformers/pull/21945.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21945.patch", "merged_at": 1677944647000 }

https://api.github.com/repos/huggingface/transformers/issues/21944

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21944/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21944/comments

https://api.github.com/repos/huggingface/transformers/issues/21944/events

https://github.com/huggingface/transformers/pull/21944

1,609,191,459

PR_kwDOCUB6oc5LQUz1

21,944

Fix gradient checkpointing bug in Pegasus

{ "login": "KMFODA", "id": 35491698, "node_id": "MDQ6VXNlcjM1NDkxNjk4", "avatar_url": "https://avatars.githubusercontent.com/u/35491698?v=4", "gravatar_id": "", "url": "https://api.github.com/users/KMFODA", "html_url": "https://github.com/KMFODA", "followers_url": "https://api.github.com/users/KMFODA/followers", "following_url": "https://api.github.com/users/KMFODA/following{/other_user}", "gists_url": "https://api.github.com/users/KMFODA/gists{/gist_id}", "starred_url": "https://api.github.com/users/KMFODA/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/KMFODA/subscriptions", "organizations_url": "https://api.github.com/users/KMFODA/orgs", "repos_url": "https://api.github.com/users/KMFODA/repos", "events_url": "https://api.github.com/users/KMFODA/events{/privacy}", "received_events_url": "https://api.github.com/users/KMFODA/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,677

CONTRIBUTOR

null

This PR fixes a bug that a user can encounter while using generate and models that use gradient_checkpointing. Fixes Issue https://github.com/huggingface/transformers/issues/21737 cc @younesbelkada or @gante

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21944/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21944/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21944", "html_url": "https://github.com/huggingface/transformers/pull/21944", "diff_url": "https://github.com/huggingface/transformers/pull/21944.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21944.patch", "merged_at": 1677944613000 }

https://api.github.com/repos/huggingface/transformers/issues/21943

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21943/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21943/comments

https://api.github.com/repos/huggingface/transformers/issues/21943/events

https://github.com/huggingface/transformers/pull/21943

1,609,188,166

PR_kwDOCUB6oc5LQUFV

21,943

Fix gradient checkpointing bug in OPT

{ "login": "KMFODA", "id": 35491698, "node_id": "MDQ6VXNlcjM1NDkxNjk4", "avatar_url": "https://avatars.githubusercontent.com/u/35491698?v=4", "gravatar_id": "", "url": "https://api.github.com/users/KMFODA", "html_url": "https://github.com/KMFODA", "followers_url": "https://api.github.com/users/KMFODA/followers", "following_url": "https://api.github.com/users/KMFODA/following{/other_user}", "gists_url": "https://api.github.com/users/KMFODA/gists{/gist_id}", "starred_url": "https://api.github.com/users/KMFODA/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/KMFODA/subscriptions", "organizations_url": "https://api.github.com/users/KMFODA/orgs", "repos_url": "https://api.github.com/users/KMFODA/repos", "events_url": "https://api.github.com/users/KMFODA/events{/privacy}", "received_events_url": "https://api.github.com/users/KMFODA/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,677

CONTRIBUTOR

null

This PR fixes a bug that a user can encounter while using generate and models that use gradient_checkpointing. Fixes Issue https://github.com/huggingface/transformers/issues/21737 cc @younesbelkada or @gante

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21943/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21943/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21943", "html_url": "https://github.com/huggingface/transformers/pull/21943", "diff_url": "https://github.com/huggingface/transformers/pull/21943.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21943.patch", "merged_at": 1677944578000 }

https://api.github.com/repos/huggingface/transformers/issues/21942

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21942/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21942/comments

https://api.github.com/repos/huggingface/transformers/issues/21942/events

https://github.com/huggingface/transformers/pull/21942

1,609,187,121

PR_kwDOCUB6oc5LQT3K

21,942

[examples/speech-recognition] Add SpecAugment to run_speech_recognition_seq2seq.py

{ "login": "bofenghuang", "id": 38185248, "node_id": "MDQ6VXNlcjM4MTg1MjQ4", "avatar_url": "https://avatars.githubusercontent.com/u/38185248?v=4", "gravatar_id": "", "url": "https://api.github.com/users/bofenghuang", "html_url": "https://github.com/bofenghuang", "followers_url": "https://api.github.com/users/bofenghuang/followers", "following_url": "https://api.github.com/users/bofenghuang/following{/other_user}", "gists_url": "https://api.github.com/users/bofenghuang/gists{/gist_id}", "starred_url": "https://api.github.com/users/bofenghuang/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/bofenghuang/subscriptions", "organizations_url": "https://api.github.com/users/bofenghuang/orgs", "repos_url": "https://api.github.com/users/bofenghuang/repos", "events_url": "https://api.github.com/users/bofenghuang/events{/privacy}", "received_events_url": "https://api.github.com/users/bofenghuang/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "Hi @sgugger,\r\n\r\nThanks for the review ! I've removed the SpecAugment arguments from `ModelArguments` and left some default values as asked. Also secured the attribute access in the line 447 to define `return_attention_mask` :)", "I think it's okay to have the whisper-specific changes as they are in this PR. We have similar things in `run_translation` for some specific models as well.", "Thanks for the clarification @sgugger and @bofenghuang for the contribution! Merging in that case 🤗" ]

1,677

1,678

CONTRIBUTOR

null

# What does this PR do?   Hello 👋, In this PR, I tried to add SpecAugment to run_speech_recognition_seq2seq.py training example since it's asked in https://github.com/huggingface/transformers/pull/21298. I tried to not impact the training of other seq2seq models using this script. But I might still have missed something. PS: Also removed useless argument `text_column` :) Thanks in advance! ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [x] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [x] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [ ] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [ ] Did you write any new necessary tests? ## Who can review? cc @ArthurZucker @sanchit-gandhi @sgugger Anyone in the community is free to review the PR once the tests have passed. Feel free to tag members/contributors who may be interested in your PR.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21942/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21942/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21942", "html_url": "https://github.com/huggingface/transformers/pull/21942", "diff_url": "https://github.com/huggingface/transformers/pull/21942.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21942.patch", "merged_at": 1678294772000 }

https://api.github.com/repos/huggingface/transformers/issues/21941

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21941/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21941/comments

https://api.github.com/repos/huggingface/transformers/issues/21941/events

https://github.com/huggingface/transformers/pull/21941

1,609,137,345

PR_kwDOCUB6oc5LQJVe

21,941

Adding Type Hints to TF_Pegasus model

{ "login": "mollerup23", "id": 69806327, "node_id": "MDQ6VXNlcjY5ODA2MzI3", "avatar_url": "https://avatars.githubusercontent.com/u/69806327?v=4", "gravatar_id": "", "url": "https://api.github.com/users/mollerup23", "html_url": "https://github.com/mollerup23", "followers_url": "https://api.github.com/users/mollerup23/followers", "following_url": "https://api.github.com/users/mollerup23/following{/other_user}", "gists_url": "https://api.github.com/users/mollerup23/gists{/gist_id}", "starred_url": "https://api.github.com/users/mollerup23/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/mollerup23/subscriptions", "organizations_url": "https://api.github.com/users/mollerup23/orgs", "repos_url": "https://api.github.com/users/mollerup23/repos", "events_url": "https://api.github.com/users/mollerup23/events{/privacy}", "received_events_url": "https://api.github.com/users/mollerup23/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "Hi @mollerup23, mostly looks good! One thing to watch out for is that in some cases the default value of the argument has been changed. It's easy to see if you look in the GitHub \"Files changed\" tab (see the image - `return_dict` had its default argument changed)\r\n![image](https://user-images.githubusercontent.com/12866554/223161978-7abf2912-865d-4975-80bb-f38899440350.png)\r\n\r\nIf you fix the instances where that happened and double-check that it's all okay in the Files Changed tab, we should be good to go!", "Hi @Rocketknight1, I updated and committed again. Hopefully these fixes help, let me know if there is anything else I should do!" ]

1,677

1,678

CONTRIBUTOR

null

# What does this PR do? Added type hints for remaining `call()` functions for the Pegasus model (changes made only in the models/pegasus/modeling_tf_pegasus.py file). Fixes # [(16059)](https://github.com/huggingface/transformers/issues/16059) ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [x] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [x] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. [approval](https://github.com/huggingface/transformers/issues/16059#issuecomment-1441895896) - [ ] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [ ] Did you write any new necessary tests? ## Who can review? Anyone in the community is free to review the PR once the tests have passed. Feel free to tag members/contributors who may be interested in your PR. @Rocketknight1

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21941/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21941/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21941", "html_url": "https://github.com/huggingface/transformers/pull/21941", "diff_url": "https://github.com/huggingface/transformers/pull/21941.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21941.patch", "merged_at": 1678723110000 }

https://api.github.com/repos/huggingface/transformers/issues/21940

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21940/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21940/comments

https://api.github.com/repos/huggingface/transformers/issues/21940/events

https://github.com/huggingface/transformers/pull/21940

1,609,100,685

PR_kwDOCUB6oc5LQBk2

21,940

[CI] Fix ci

{ "login": "ArthurZucker", "id": 48595927, "node_id": "MDQ6VXNlcjQ4NTk1OTI3", "avatar_url": "https://avatars.githubusercontent.com/u/48595927?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ArthurZucker", "html_url": "https://github.com/ArthurZucker", "followers_url": "https://api.github.com/users/ArthurZucker/followers", "following_url": "https://api.github.com/users/ArthurZucker/following{/other_user}", "gists_url": "https://api.github.com/users/ArthurZucker/gists{/gist_id}", "starred_url": "https://api.github.com/users/ArthurZucker/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ArthurZucker/subscriptions", "organizations_url": "https://api.github.com/users/ArthurZucker/orgs", "repos_url": "https://api.github.com/users/ArthurZucker/repos", "events_url": "https://api.github.com/users/ArthurZucker/events{/privacy}", "received_events_url": "https://api.github.com/users/ArthurZucker/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "> 2 other slow tests are failing, but they were also failing prior to #20211 .\r\n\r\nCould you share the names of these 2 tests.\r\nI can check on CI runners. I did it once on Friday for one test you mentioned, but let's make sure.\r\n", "There is :\r\n- `tests/models/deformable_detr/test_modeling_deformable_detr.py::DeformableDetrModelIntegrationTests::test_inference_object_detection_head_with_box_refine_two_stage`\r\n- `tests/models/deformable_detr/test_modeling_deformable_detr.py::DeformableDetrModelIntegrationTests::test_inference_object_detection_head_equivalence_cpu_gpu`\r\n- `tests/models/deformable_detr/test_modeling_deformable_detr.py::DeformableDetrModelIntegrationTests::test_inference_object_detection_head`\r\nRan them on the CI runner an they are all green so LGTM. " ]

1,677

1,678

COLLABORATOR

null

# What does this PR do? A typo made during the full cleanup of the code was making 2 test fail. 2 other slow tests are failing, but they were also failing prior to #20211 .

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21940/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21940/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21940", "html_url": "https://github.com/huggingface/transformers/pull/21940", "diff_url": "https://github.com/huggingface/transformers/pull/21940.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21940.patch", "merged_at": 1678112548000 }

https://api.github.com/repos/huggingface/transformers/issues/21939

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21939/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21939/comments

https://api.github.com/repos/huggingface/transformers/issues/21939/events

https://github.com/huggingface/transformers/pull/21939

1,609,080,631

PR_kwDOCUB6oc5LP9Xe

21,939

Add TF contrastive image text finetuning example

{ "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.github.com/users/Rocketknight1/followers", "following_url": "https://api.github.com/users/Rocketknight1/following{/other_user}", "gists_url": "https://api.github.com/users/Rocketknight1/gists{/gist_id}", "starred_url": "https://api.github.com/users/Rocketknight1/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Rocketknight1/subscriptions", "organizations_url": "https://api.github.com/users/Rocketknight1/orgs", "repos_url": "https://api.github.com/users/Rocketknight1/repos", "events_url": "https://api.github.com/users/Rocketknight1/events{/privacy}", "received_events_url": "https://api.github.com/users/Rocketknight1/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "Update: Woah! The argument names to `PushToHubCallback` across our examples became outdated and tests weren't picking that up. I'm fixing it in this PR too." ]

1,677

1,678

MEMBER

null

This PR adds a TF port of the PyTorch example for finetuning the `TFVisionTextDualEncoderModel` class. Functionality is largely the same, but I used a `tf.data` pipeline to efficiently stream images instead of `torchvision`. I also added the ability to specify separate image/text models with arguments, whereas in the PyTorch example you have to create the dual encoder with a separate script. I also caught a small bug in the original model code while writing this - loss is a scalar rather than having shape `(1,)`. That's fixed in here too!

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21939/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21939/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21939", "html_url": "https://github.com/huggingface/transformers/pull/21939", "diff_url": "https://github.com/huggingface/transformers/pull/21939.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21939.patch", "merged_at": 1678121861000 }

https://api.github.com/repos/huggingface/transformers/issues/21938

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21938/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21938/comments

https://api.github.com/repos/huggingface/transformers/issues/21938/events

https://github.com/huggingface/transformers/pull/21938

1,609,039,943

PR_kwDOCUB6oc5LP0ra

21,938

[Whisper] Fix feature normalization in `WhisperFeatureExtractor`

{ "login": "bofenghuang", "id": 38185248, "node_id": "MDQ6VXNlcjM4MTg1MjQ4", "avatar_url": "https://avatars.githubusercontent.com/u/38185248?v=4", "gravatar_id": "", "url": "https://api.github.com/users/bofenghuang", "html_url": "https://github.com/bofenghuang", "followers_url": "https://api.github.com/users/bofenghuang/followers", "following_url": "https://api.github.com/users/bofenghuang/following{/other_user}", "gists_url": "https://api.github.com/users/bofenghuang/gists{/gist_id}", "starred_url": "https://api.github.com/users/bofenghuang/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/bofenghuang/subscriptions", "organizations_url": "https://api.github.com/users/bofenghuang/orgs", "repos_url": "https://api.github.com/users/bofenghuang/repos", "events_url": "https://api.github.com/users/bofenghuang/events{/privacy}", "received_events_url": "https://api.github.com/users/bofenghuang/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,677

CONTRIBUTOR

null

# What does this PR do?   Hello @ArthurZucker @sanchit-gandhi , In this PR, I tried to fix the feature normalization in `WhisperFeatureExtractor` which is actually not used. The line 354 takes the `input_features` from the line 340. In addition, the `zero_mean_unit_var_norm` expects a `padded_inputs["attention_mask"]` also in the sample level (48000) as `padded_inputs["input_features"]`. IMO it should not be rescaled in the line 344 (it's also repeated in the line 363). Please let me know what you think :) ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [ ] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [ ] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [ ] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [ ] Did you write any new necessary tests? ## Who can review? Anyone in the community is free to review the PR once the tests have passed. Feel free to tag members/contributors who may be interested in your PR.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21938/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21938/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21938", "html_url": "https://github.com/huggingface/transformers/pull/21938", "diff_url": "https://github.com/huggingface/transformers/pull/21938.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21938.patch", "merged_at": 1677871274000 }

https://api.github.com/repos/huggingface/transformers/issues/21937

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21937/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21937/comments

https://api.github.com/repos/huggingface/transformers/issues/21937/events

https://github.com/huggingface/transformers/issues/21937

1,608,970,591

I_kwDOCUB6oc5f5vFf

21,937

Whisper does not respect `config.forced_decoder_ids`

{ "login": "sanchit-gandhi", "id": 93869735, "node_id": "U_kgDOBZhWpw", "avatar_url": "https://avatars.githubusercontent.com/u/93869735?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sanchit-gandhi", "html_url": "https://github.com/sanchit-gandhi", "followers_url": "https://api.github.com/users/sanchit-gandhi/followers", "following_url": "https://api.github.com/users/sanchit-gandhi/following{/other_user}", "gists_url": "https://api.github.com/users/sanchit-gandhi/gists{/gist_id}", "starred_url": "https://api.github.com/users/sanchit-gandhi/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sanchit-gandhi/subscriptions", "organizations_url": "https://api.github.com/users/sanchit-gandhi/orgs", "repos_url": "https://api.github.com/users/sanchit-gandhi/repos", "events_url": "https://api.github.com/users/sanchit-gandhi/events{/privacy}", "received_events_url": "https://api.github.com/users/sanchit-gandhi/received_events", "type": "User", "site_admin": false }

[]

closed

false

{ "login": "sanchit-gandhi", "id": 93869735, "node_id": "U_kgDOBZhWpw", "avatar_url": "https://avatars.githubusercontent.com/u/93869735?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sanchit-gandhi", "html_url": "https://github.com/sanchit-gandhi", "followers_url": "https://api.github.com/users/sanchit-gandhi/followers", "following_url": "https://api.github.com/users/sanchit-gandhi/following{/other_user}", "gists_url": "https://api.github.com/users/sanchit-gandhi/gists{/gist_id}", "starred_url": "https://api.github.com/users/sanchit-gandhi/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sanchit-gandhi/subscriptions", "organizations_url": "https://api.github.com/users/sanchit-gandhi/orgs", "repos_url": "https://api.github.com/users/sanchit-gandhi/repos", "events_url": "https://api.github.com/users/sanchit-gandhi/events{/privacy}", "received_events_url": "https://api.github.com/users/sanchit-gandhi/received_events", "type": "User", "site_admin": false }

[ { "login": "sanchit-gandhi", "id": 93869735, "node_id": "U_kgDOBZhWpw", "avatar_url": "https://avatars.githubusercontent.com/u/93869735?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sanchit-gandhi", "html_url": "https://github.com/sanchit-gandhi", "followers_url": "https://api.github.com/users/sanchit-gandhi/followers", "following_url": "https://api.github.com/users/sanchit-gandhi/following{/other_user}", "gists_url": "https://api.github.com/users/sanchit-gandhi/gists{/gist_id}", "starred_url": "https://api.github.com/users/sanchit-gandhi/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sanchit-gandhi/subscriptions", "organizations_url": "https://api.github.com/users/sanchit-gandhi/orgs", "repos_url": "https://api.github.com/users/sanchit-gandhi/repos", "events_url": "https://api.github.com/users/sanchit-gandhi/events{/privacy}", "received_events_url": "https://api.github.com/users/sanchit-gandhi/received_events", "type": "User", "site_admin": false } ]

[ "This is based on the behavior that we enforce through the generate function. If there is a `generation_config`, it will be used and has priority over the `config`. \r\nThere should be a warning deprecating the control of generation through the `config`. See [here](https://github.com/ArthurZucker/transformers/blob/main/src/transformers/generation/utils.py#LL538C17-L538C17).\r\nThis is a minor breaking change but should also be adresses with deprecation cycle " ]

1,677

1,678

CONTRIBUTOR

null

### System Info - `transformers` version: 4.27.0.dev0 - Platform: macOS-13.1-arm64-arm-64bit - Python version: 3.9.13 - Huggingface_hub version: 0.11.1 - PyTorch version (GPU?): 1.13.1 (False) - Tensorflow version (GPU?): not installed (NA) - Flax version (CPU?/GPU?/TPU?): 0.6.3 (cpu) - Jax version: 0.4.1 - JaxLib version: 0.4.1 ### Who can help? @sanchit-gandhi @ArthurZucker ### Reproduction Previously, the recommended method for setting the language and transcription for Whisper inference was by setting the `config.forced_decoder_ids`. However, this method no longer works on the main branch, for example if we force the model to generate in French: ```python from transformers import WhisperProcessor, WhisperForConditionalGeneration from datasets import load_dataset # load model and processor processor = WhisperProcessor.from_pretrained("openai/whisper-tiny") model = WhisperForConditionalGeneration.from_pretrained("openai/whisper-tiny") ds = load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation") sample = ds[0]["audio"] input_features = processor(sample["array"], sampling_rate=sample["sampling_rate"], return_tensors="pt").input_features # set the forced ids model.config.forced_decoder_ids = processor.get_decoder_prompt_ids(language="french", task="transcribe") # generate token ids predicted_ids = model.generate(input_features) # decode token ids to text transcription = processor.batch_decode(predicted_ids, skip_special_tokens=False) print(transcription) ``` **Print Output**: ``` ['<|startoftranscript|><|en|><|transcribe|><|notimestamps|> Mr. Quilter is the apostle of the middle classes and we are glad to welcome his gospel.<|endoftext|>'] ``` ### Expected behavior Language token should correspond to the forced decoder token id (e.g. `"<|fr|>"` in this example). This was the case in previous transformer versions, so there has been a breaking change that we need to remedy.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21937/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21937/timeline

completed

null

https://api.github.com/repos/huggingface/transformers/issues/21936

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21936/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21936/comments

https://api.github.com/repos/huggingface/transformers/issues/21936/events

https://github.com/huggingface/transformers/issues/21936

1,608,938,023

I_kwDOCUB6oc5f5nIn

21,936

Whisper breaks on poor quality speech audio

{ "login": "frankiedrake", "id": 12988773, "node_id": "MDQ6VXNlcjEyOTg4Nzcz", "avatar_url": "https://avatars.githubusercontent.com/u/12988773?v=4", "gravatar_id": "", "url": "https://api.github.com/users/frankiedrake", "html_url": "https://github.com/frankiedrake", "followers_url": "https://api.github.com/users/frankiedrake/followers", "following_url": "https://api.github.com/users/frankiedrake/following{/other_user}", "gists_url": "https://api.github.com/users/frankiedrake/gists{/gist_id}", "starred_url": "https://api.github.com/users/frankiedrake/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/frankiedrake/subscriptions", "organizations_url": "https://api.github.com/users/frankiedrake/orgs", "repos_url": "https://api.github.com/users/frankiedrake/repos", "events_url": "https://api.github.com/users/frankiedrake/events{/privacy}", "received_events_url": "https://api.github.com/users/frankiedrake/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "cc @ArthurZucker and @sanchit-gandhi ", "Hey! Did you get similar results using the original `Whisper` model? Most probably this is just the model not being very good. ", "> Hey! Did you get similar results using the original `Whisper` model? Most probably this is just the model not being very good.\r\n\r\nNo, I don't face such problem with original `Whisper`", "OK! There can be a few things to try, but if you can share an audio file with me it would help a lot! ", "@frankiedrake Have you tried using `return_timestamps` ? When experimenting I found that the default for openai is using timestamps (even if not shown) and that the model seems to perform better than without.", "> \r\n\r\nSorry for a noob question, where should I put this parameter? I tried to add `no_timestamps=False` to `processor.get_decoder_prompt_ids` call and also to the three other methods call, but result didn't change.\r\n\r\nBtw it works well with `medium` model, but it's too heavy for my setup", "> OK! There can be a few things to try, but if you can share an audio file with me it would help a lot!\r\n\r\nUnfortunately, I cannot share it because of NDA, but I'll try to record one myself", "`generate(..., return_timestamps=True)` to get the proper generation mode.\r\nThat's correct right @ArthurZucker ", "> return_timestamps\r\n\r\nI tried this, but got an error, that model's `forward` method isn't aware of this parameter\r\n\r\n```ValueError: The following `model_kwargs` are not used by the model: ['return_timestamps'] (note: typos in the generate arguments will also show up in this list)```", "`pipe = pipeline(..., return_timestamps=True)` should work though.", "> @frankiedrake Have you tried using `return_timestamps` ? When experimenting I found that the default for openai is using timestamps (even if not shown) and that the model seems to perform better than without.\r\n\r\nI tested all variants, result is the same", "Without being able to reproduce it's really hard. Could you dive to the level of logits and figure out any potential differences ?\r\n\r\nI'm pretty sure it should come down to a configuration difference in the end, but if *we* can't reproduce, it'd be hard to understand. ", "> Without being able to reproduce it's really hard. Could you dive to the level of logits and figure out any potential differences ?\r\n> \r\n> I'm pretty sure it should come down to a configuration difference in the end, but if _we_ can't reproduce, it'd be hard to understand.\r\n\r\nOkay, thank you for suggestions, I'll try to look at it, and also will try to find an audio I could share with you", "You are probably not using `main`! The timestamps were not part of the initial release 😉 ", "@ArthurZucker Correct! I installed the library from main branch and the transcription is now better with one file, but remains the same with another, probably I can tune the parameters further to make it work even better", "Glad that I could help 😉 ", "> OK! There can be a few things to try, but if you can share an audio file with me it would help a lot!\r\n\r\nGuys, I was able to record an [audio](https://github.com/frankiedrake/demo/blob/master/whisper_test.wav) that reproduces a problem, but it wasn't so easy, so the text I recorded doesn't actually exists 😁, anyway I'd expect the model gives some text instead of repeating a single word multiple times.\r\nAlso, I noticed that if I don't specify language manually the model seems not to break, in my case it detected Chinese language and emitted some readable text.\r\nOutput of Ukrainian:\r\n`Вереси, єдине м'якості, єдине ісцемо, і два, і бере, і бере, і бере, і бере, і бере, і бере, і бере, і бере, і бере, і бере.`\r\n\r\nThe original whisper produces:\r\n`Веріості, є рівні, є мільйорівні, є сьогодні, є два, є бере, є бере, чіжено, чіжено, нервно, нервно.`\r\n ", "> f I don't specify language manually the model seems not to break\r\n\r\nCan you share how you specify language in both `openai` and `transformers` ? The difference is likely coming from there.", "\r\n\r\n\r\n> > f I don't specify language manually the model seems not to break\r\n> \r\n> Can you share how you specify language in both `openai` and `transformers` ? The difference is likely coming from there.\r\n\r\nI really doubt that it's related because it reproduces even if I don't specify the language in other audios", "Seems like it's a known [issue](https://github.com/microsoft/DialoGPT/issues/45#issuecomment-680087019) with transformers models, and tuning `temperature` and `repetition_penalty` params of `generate` method helps, I was able to stop the model from repeating a single word, text now looks much much better, but seems I noticed a small cut at the end of the audio, will investigate a bit deeper. I read about `repetition_penalty` before but I didn't succeed, maybe the conjunction with the `temperature` is significant. \r\nMaybe you can shed some light on why these parameters are so important?", "> Maybe you can shed some light on why these parameters are so important?\r\n\r\nLMs are know to hallucinate by repeating tokens, or several tokens. I don't think there's a good consensus on why it's the case, but it's a very well know issue on LMs. Adding penalties does help, but it's a clutch in my very personal opinion.\r\n\r\nMaybe the openai defaults are still different the ones we have\r\n", "Do we agree the openai test you're doing is simply \r\n\r\n```\r\nwhisper whisper_test.wav --model small\r\n```\r\n\r\nAnd get this\r\n\r\n```\r\nDetecting language using up to the first 30 seconds. Use `--language` to specify the language\r\nDetected language: Chinese\r\n[00:00.420 --> 00:02.000] 我問他們\r\n[00:02.000 --> 00:03.320] 他們在那邊\r\n[00:03.320 --> 00:05.280] 順便他們等著過去\r\n[00:05.280 --> 00:06.980] 還要去\r\n[00:06.980 --> 00:08.260] 剛剛說他們背後\r\n```\r\n \r\nright ?\r\n", "I have the following result\r\n```\r\nDetecting language using up to the first 30 seconds. Use `--language` to specify the language\r\nDetected language: Chinese\r\n[00:00.000 --> 00:01.700] 要生一年年老了\r\n[00:01.700 --> 00:03.280] 一生不一年老\r\n[00:03.480 --> 00:05.960] 練一陣子\r\n[00:05.960 --> 00:06.700] 熱情不二\r\n[00:06.700 --> 00:07.620] 謝謝大家\r\n```\r\n", "Can you share anything reproducible ? Right now it's a back and forth and we can't reproduce anything on our end.\r\n\r\nPlease share a clear (small) script, that I can copy past that should reproduce the issue on `transformers@main` and `whisper@main` otherwise, it's going to be too tedious for us to investigate.", "> Can you share anything reproducible ? Right now it's a back and forth and we can't reproduce anything on our end.\r\n> \r\n> Please share a clear (small) script, that I can copy past that should reproduce the issue on `transformers@main` and `whisper@main` otherwise, it's going to be too tedious for us to investigate.\r\n\r\nBut I shared a file with you, what exactly you can't reproduce?", "I don't have the same output for neither `openai/whisper` nor `transformers` on your file.\r\n", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md) are likely to be ignored." ]

1,677

1,681

NONE

null

### System Info - `transformers` version: 4.26.1 - Platform: Linux-5.19.0-32-generic-x86_64-with-glibc2.17 - Python version: 3.8.16 - Huggingface_hub version: 0.11.1 - PyTorch version (GPU?): 1.13.1+cu117 (False) - Tensorflow version (GPU?): not installed (NA) - Flax version (CPU?/GPU?/TPU?): not installed (NA) - Jax version: not installed - JaxLib version: not installed - Using GPU in script?: no - Using distributed or parallel set-up in script?: no ### Who can help? _No response_ ### Information - [ ] The official example scripts - [X] My own modified scripts ### Tasks - [X] An officially supported task in the `examples` folder (such as GLUE/SQuAD, ...) - [X] My own task or dataset (give details below) ### Reproduction I don't know if it's a bug, but it's definitely not an expected behaviour for me. Also I saw a thread with such behaviour where @Narsil said that "the model is repeating itself", but I can't find it right now, I'll update the issue when I do. To recognize audio file I'm using script that I found in one of the threads here on github [link](https://colab.research.google.com/drive/1Qz9hUL3Z3SxHLUt7f4vuzZG7KEgV0ofk?usp=share_link) ``` processor = WhisperProcessor.from_pretrained("openai/whisper-small") model = WhisperForConditionalGeneration.from_pretrained("openai/whisper-small") model.config.forced_decoder_ids = processor.get_decoder_prompt_ids(task="transcribe") input_speech, sr = audio2numpy.audio_from_file(file) input_features = processor(input_speech, return_tensors="pt", sampling_rate=16000).input_features predicted_ids = model.generate(input_features, max_length=model.config.max_length, repetition_penalty=1) transcription = processor.batch_decode(predicted_ids, skip_special_tokens=True) ``` And it works okay if the speech is clear and the utterance is detected good, but when speaker talks fast or not legibly enough, or if there's little silence in audio, then the transcription becomes ugly. It looks like: "Привіт, хороша погода але але але але але але але але але але але але але але але але" Currently I'm using only ukrainian files so I'm not aware if it happens in other languages. ### Expected behavior The text is recognized along the whole audio file without breaking

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21936/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21936/timeline

completed

null

https://api.github.com/repos/huggingface/transformers/issues/21935

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21935/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21935/comments

https://api.github.com/repos/huggingface/transformers/issues/21935/events

https://github.com/huggingface/transformers/pull/21935

1,608,927,712

PR_kwDOCUB6oc5LPc07

21,935

Update feature selection in to_tf_dataset

{ "login": "amyeroberts", "id": 22614925, "node_id": "MDQ6VXNlcjIyNjE0OTI1", "avatar_url": "https://avatars.githubusercontent.com/u/22614925?v=4", "gravatar_id": "", "url": "https://api.github.com/users/amyeroberts", "html_url": "https://github.com/amyeroberts", "followers_url": "https://api.github.com/users/amyeroberts/followers", "following_url": "https://api.github.com/users/amyeroberts/following{/other_user}", "gists_url": "https://api.github.com/users/amyeroberts/gists{/gist_id}", "starred_url": "https://api.github.com/users/amyeroberts/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/amyeroberts/subscriptions", "organizations_url": "https://api.github.com/users/amyeroberts/orgs", "repos_url": "https://api.github.com/users/amyeroberts/repos", "events_url": "https://api.github.com/users/amyeroberts/events{/privacy}", "received_events_url": "https://api.github.com/users/amyeroberts/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "cc @MKhalusova Reference on updated API for selecting features from TF datasets", "Thanks for updating the docs too! Looks neat :)", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md) are likely to be ignored." ]

1,677

1,682

COLLABORATOR

null

# What does this PR do? Updates feature selection to ensure returned dataset structure is consistent after merging of datasets PR: https://github.com/huggingface/datasets/pull/5602. The PR makes it possible to return a TF dataset with a dict structure even if only a single feature is selected. Compatibility with this version of datasets was run with [this commit](https://github.com/huggingface/transformers/pull/21935/commits/b64204bc1093edd7e3666ad76354fa09405cf4ec) and had a [successful run](https://github.com/huggingface/transformers/actions/runs/4555221658/jobs/8034018899). Note: In all the cases here, the examples were tested with and without these updates. The models would successfully train with both the new and old dataset structures. ## Before submitting - [x] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [x] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [ ] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [ ] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [ ] Did you write any new necessary tests?

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21935/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21935/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21935", "html_url": "https://github.com/huggingface/transformers/pull/21935", "diff_url": "https://github.com/huggingface/transformers/pull/21935.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21935.patch", "merged_at": 1682354071000 }

https://api.github.com/repos/huggingface/transformers/issues/21934

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21934/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21934/comments

https://api.github.com/repos/huggingface/transformers/issues/21934/events

https://github.com/huggingface/transformers/issues/21934

1,608,907,562

I_kwDOCUB6oc5f5fsq

21,934

Faster `Skipping the first batches` in Trainer

{ "login": "SamuelLarkin", "id": 7314973, "node_id": "MDQ6VXNlcjczMTQ5NzM=", "avatar_url": "https://avatars.githubusercontent.com/u/7314973?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SamuelLarkin", "html_url": "https://github.com/SamuelLarkin", "followers_url": "https://api.github.com/users/SamuelLarkin/followers", "following_url": "https://api.github.com/users/SamuelLarkin/following{/other_user}", "gists_url": "https://api.github.com/users/SamuelLarkin/gists{/gist_id}", "starred_url": "https://api.github.com/users/SamuelLarkin/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SamuelLarkin/subscriptions", "organizations_url": "https://api.github.com/users/SamuelLarkin/orgs", "repos_url": "https://api.github.com/users/SamuelLarkin/repos", "events_url": "https://api.github.com/users/SamuelLarkin/events{/privacy}", "received_events_url": "https://api.github.com/users/SamuelLarkin/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "This is already done in the main branch. You just need to have `Accelerate` installed as an extra dependency.", "Thanks for the info but I do have `accelerate==0.16.0` installed as reported in my logs but I still get extremely slow `Skipping the first batches`\r\n\r\n```yaml\r\nCONDA: transformers-4.20.1\r\nname: transformers-4.20.1\r\nchannels:\r\n - pytorch\r\n - huggingface\r\n - anaconda\r\n - conda-forge\r\n - defaults\r\ndependencies:\r\n - _libgcc_mutex=0.1=conda_forge\r\n - _openmp_mutex=4.5=2_kmp_llvm\r\n - abseil-cpp=20211102.0=hd4dd3e8_0\r\n - aiohttp=3.8.1=py39h7f8727e_1\r\n - aiosignal=1.2.0=pyhd3eb1b0_0\r\n - arrow-cpp=8.0.0=py39h60b952e_0\r\n - async-timeout=4.0.1=pyhd3eb1b0_0\r\n - attrs=21.4.0=pyhd3eb1b0_0\r\n - aws-c-common=0.4.57=he6710b0_1\r\n - aws-c-event-stream=0.1.6=h2531618_5\r\n - aws-checksums=0.1.9=he6710b0_0\r\n - aws-sdk-cpp=1.8.185=hce553d0_0\r\n - blas=1.0=mkl\r\n - boost-cpp=1.73.0=h7f8727e_12\r\n - bottleneck=1.3.4=py39hce1f21e_0\r\n - brotli=1.0.9=he6710b0_2\r\n - brotlipy=0.7.0=py39h27cfd23_1003\r\n - bzip2=1.0.8=h7b6447c_0\r\n - c-ares=1.18.1=h7f8727e_0\r\n - ca-certificates=2023.01.10=h06a4308_0\r\n - certifi=2022.12.7=py39h06a4308_0\r\n - cffi=1.15.0=py39hd667e15_1\r\n - charset-normalizer=2.0.4=pyhd3eb1b0_0\r\n - click=8.0.4=py39h06a4308_0\r\n - cryptography=37.0.1=py39h9ce1e76_0\r\n - cudatoolkit=11.6.0=hecad31d_10\r\n - cycler=0.11.0=pyhd3eb1b0_0\r\n - dataclasses=0.8=pyh6d0b6a4_7\r\n - datasets=2.3.2=py_0\r\n - dbus=1.13.18=hb2f20db_0\r\n - dill=0.3.4=pyhd3eb1b0_0\r\n - expat=2.4.4=h295c915_0\r\n - filelock=3.6.0=pyhd3eb1b0_0\r\n - fontconfig=2.13.1=h6c09931_0\r\n - fonttools=4.25.0=pyhd3eb1b0_0\r\n - freetype=2.11.0=h70c0345_0\r\n - frozenlist=1.2.0=py39h7f8727e_0\r\n - fsspec=2022.3.0=py39h06a4308_0\r\n - gflags=2.2.2=he6710b0_0\r\n - giflib=5.2.1=h7b6447c_0\r\n - glib=2.69.1=h4ff587b_1\r\n - glog=0.5.0=h2531618_0\r\n - grpc-cpp=1.46.1=h33aed49_0\r\n - gst-plugins-base=1.14.0=h8213a91_2\r\n - gstreamer=1.14.0=h28cd5cc_2\r\n - h5py=3.7.0=py39h737f45e_0\r\n - hdf5=1.10.6=hb1b8bf9_0\r\n - huggingface_hub=0.8.1=py_0\r\n - icu=58.2=he6710b0_3\r\n - idna=3.3=pyhd3eb1b0_0\r\n - importlib-metadata=4.11.3=py39h06a4308_0\r\n - importlib_metadata=4.11.3=hd3eb1b0_0\r\n - intel-openmp=2021.4.0=h06a4308_3561\r\n - joblib=1.1.0=pyhd3eb1b0_0\r\n - jpeg=9e=h7f8727e_0\r\n - kiwisolver=1.4.2=py39h295c915_0\r\n - krb5=1.19.2=hac12032_0\r\n - lcms2=2.12=h3be6417_0\r\n - ld_impl_linux-64=2.38=h1181459_1\r\n - lerc=3.0=h295c915_0\r\n - libboost=1.73.0=h28710b8_12\r\n - libbrotlicommon=1.0.9=h166bdaf_7\r\n - libbrotlidec=1.0.9=h166bdaf_7\r\n - libbrotlienc=1.0.9=h166bdaf_7\r\n - libclang=10.0.1=default_hb85057a_2\r\n - libcurl=7.82.0=h0b77cf5_0\r\n - libdeflate=1.8=h7f8727e_5\r\n - libedit=3.1.20210910=h7f8727e_0\r\n - libev=4.33=h7f8727e_1\r\n - libevent=2.1.12=h8f2d780_0\r\n - libffi=3.3=he6710b0_2\r\n - libgcc=7.2.0=h69d50b8_2\r\n - libgcc-ng=12.1.0=h8d9b700_16\r\n - libgfortran-ng=7.5.0=ha8ba4b0_17\r\n - libgfortran4=7.5.0=ha8ba4b0_17\r\n - libllvm10=10.0.1=hbcb73fb_5\r\n - libnghttp2=1.46.0=hce63b2e_0\r\n - libpng=1.6.37=hbc83047_0\r\n - libpq=12.9=h16c4e8d_3\r\n - libprotobuf=3.20.1=h4ff587b_0\r\n - libssh2=1.10.0=h8f2d780_0\r\n - libstdcxx-ng=11.2.0=h1234567_1\r\n - libthrift=0.15.0=hcc01f38_0\r\n - libtiff=4.4.0=hecacb30_0\r\n - libutf8proc=2.6.1=h27cfd23_0\r\n - libuuid=1.0.3=h7f8727e_2\r\n - libwebp=1.2.2=h55f646e_0\r\n - libwebp-base=1.2.2=h7f8727e_0\r\n - libxcb=1.15=h7f8727e_0\r\n - libxkbcommon=1.0.1=hfa300c1_0\r\n - libxml2=2.9.14=h74e7548_0\r\n - libxslt=1.1.35=h4e12654_0\r\n - libzlib=1.2.12=h166bdaf_1\r\n - llvm-openmp=14.0.4=he0ac6c6_0\r\n - lz4-c=1.9.3=h295c915_1\r\n - matplotlib=3.5.1=py39h06a4308_1\r\n - matplotlib-base=3.5.1=py39ha18d171_1\r\n - mkl=2021.4.0=h06a4308_640\r\n - mkl-service=2.4.0=py39h7f8727e_0\r\n - mkl_fft=1.3.1=py39hd3c417c_0\r\n - mkl_random=1.2.2=py39h51133e4_0\r\n - multidict=5.2.0=py39h7f8727e_2\r\n - multiprocess=0.70.12.2=py39h7f8727e_0\r\n - munkres=1.1.4=py_0\r\n - ncurses=6.3=h7f8727e_2\r\n - nodejs=6.11.2=h3db8ef7_0\r\n - nspr=4.33=h295c915_0\r\n - nss=3.74=h0370c37_0\r\n - numexpr=2.8.1=py39h807cd23_2\r\n - numpy=1.22.3=py39he7a7128_0\r\n - numpy-base=1.22.3=py39hf524024_0\r\n - openssl=1.1.1t=h7f8727e_0\r\n - orc=1.7.4=h07ed6aa_0\r\n - packaging=21.3=pyhd3eb1b0_0\r\n - pandas=1.4.2=py39h295c915_0\r\n - parquet-cpp=1.5.1=h34088ae_4\r\n - pcre=8.45=h295c915_0\r\n - pillow=9.2.0=py39hace64e9_1\r\n - pip=21.2.4=py39h06a4308_0\r\n - ply=3.11=py39h06a4308_0\r\n - protobuf=3.20.1=py39h295c915_0\r\n - pyarrow=8.0.0=py39h992f0b0_0\r\n - pycparser=2.21=pyhd3eb1b0_0\r\n - pyopenssl=22.0.0=pyhd3eb1b0_0\r\n - pyparsing=3.0.4=pyhd3eb1b0_0\r\n - pyqt=5.15.7=py39h6a678d5_1\r\n - pyqt5-sip=12.11.0=py39h6a678d5_1\r\n - pysocks=1.7.1=py39h06a4308_0\r\n - python=3.9.12=h12debd9_1\r\n - python-dateutil=2.8.2=pyhd3eb1b0_0\r\n - python-xxhash=2.0.2=py39h7f8727e_0\r\n - python_abi=3.9=1_cp39\r\n - pytorch=1.12.0=py3.9_cuda11.6_cudnn8.3.2_0\r\n - pytorch-mutex=1.0=cuda\r\n - pytz=2022.1=py39h06a4308_0\r\n - pyyaml=6.0=py39h7f8727e_1\r\n - qt-main=5.15.2=h327a75a_6\r\n - qt-webengine=5.15.9=hd2b0992_4\r\n - qtwebkit=5.212=h4eab89a_4\r\n - re2=2022.04.01=h295c915_0\r\n - readline=8.1.2=h7f8727e_1\r\n - regex=2022.3.15=py39h7f8727e_0\r\n - requests=2.27.1=pyhd3eb1b0_0\r\n - s2n=1.3.0=h9b69904_0\r\n - sacremoses=master=py_0\r\n - scikit-learn=1.0.2=py39h51133e4_1\r\n - scipy=1.7.3=py39hc147768_0\r\n - setuptools=61.2.0=py39h06a4308_0\r\n - sip=6.6.2=py39h6a678d5_0\r\n - six=1.16.0=pyhd3eb1b0_1\r\n - snappy=1.1.9=h295c915_0\r\n - sqlite=3.38.5=hc218d9a_0\r\n - tensorboardx=2.2=pyhd3eb1b0_0\r\n - threadpoolctl=2.2.0=pyh0d69192_0\r\n - tk=8.6.12=h1ccaba5_0\r\n - tokenizers=0.12.1=py39_0\r\n - toml=0.10.2=pyhd3eb1b0_0\r\n - tornado=6.1=py39h27cfd23_0\r\n - tqdm=4.64.0=py39h06a4308_0\r\n - transformers=4.20.1=pyhd8ed1ab_0\r\n - typing-extensions=4.1.1=hd3eb1b0_0\r\n - typing_extensions=4.1.1=pyh06a4308_0\r\n - tzdata=2022a=hda174b7_0\r\n - urllib3=1.26.9=py39h06a4308_0\r\n - utf8proc=2.6.1=h27cfd23_0\r\n - wheel=0.37.1=pyhd3eb1b0_0\r\n - xxhash=0.8.0=h7f8727e_3\r\n - xz=5.2.5=h7f8727e_1\r\n - yaml=0.2.5=h7b6447c_0\r\n - yarl=1.6.3=py39h27cfd23_0\r\n - zipp=3.8.0=py39h06a4308_0\r\n - zlib=1.2.12=h166bdaf_1\r\n - zstd=1.5.2=ha4553b6_0\r\n - pip:\r\n - absl-py==1.3.0\r\n - accelerate==0.16.0\r\n - asttokens==2.2.1\r\n - backcall==0.2.0\r\n - cachetools==5.2.1\r\n - codetiming==1.4.0\r\n - decorator==5.1.1\r\n - executing==1.2.0\r\n - google-auth==2.15.0\r\n - google-auth-oauthlib==0.4.6\r\n - grpcio==1.51.1\r\n - ipython==8.7.0\r\n - jedi==0.18.2\r\n - levenshtein==0.20.9\r\n - markdown==3.4.1\r\n - markupsafe==2.1.1\r\n - matplotlib-inline==0.1.6\r\n - more-itertools==9.0.0\r\n - oauthlib==3.2.2\r\n - parso==0.8.3\r\n - pexpect==4.8.0\r\n - pickleshare==0.7.5\r\n - prompt-toolkit==3.0.36\r\n - psutil==5.9.4\r\n - ptyprocess==0.7.0\r\n - pudb==2022.1.3\r\n - pure-eval==0.2.2\r\n - py-spy==0.3.3+computecanada\r\n - pyasn1==0.4.8\r\n - pyasn1-modules==0.2.8\r\n - pygments==2.13.0\r\n - rapidfuzz==2.13.7\r\n - requests-oauthlib==1.3.1\r\n - rsa==4.9\r\n - stack-data==0.6.2\r\n - tensorboard==2.11.0\r\n - tensorboard-data-server==0.6.1\r\n - tensorboard-plugin-wit==1.8.1\r\n - traitlets==5.7.1\r\n - urwid==2.1.2\r\n - urwid-readline==0.13\r\n - wcwidth==0.2.5\r\n - werkzeug==2.2.2\r\nprefix: /gpfs/projects/DT/mtp/WMT20/opt/miniconda3/envs/transformers-4.20.1\r\n```", "I should point out that I'm running `examples/pytorch/language-modeling/run_mlm.py`.", "I said on the the main branch. You have Transformers 4.20 installed, you need a source install.", "Thank you for the much needed speed improvement.\r\nI'm now using `accelerate==0.16.0` with `transformers==4.27.0.dev0` and it no longer takes 11h to skip the first batches." ]

1,677

1,679

CONTRIBUTOR

null

### Feature request Improve speed when skipping the first batches in `trainer.py`. ### Motivation Skipping batches should be fast. ``` Skipping the first batches: 76%|███████▌ | 93017/122500 [6:35:47<2:26:12, 3.36it/s] ``` In this example, the 'Trainer` already spent ~6h30 simply skipping batches and it estimates another ~2h30 to complete the task. This should not take that long to accomplish as those batches are not used and simply discarded. Early investigation show that the `dataloader` is invoke which implies that it samples, fetches and collates the data where collating can be expensive and useless. Wouldn't simply looping X times the `dataloader.sampler` be sufficient to resume the state of training? This would be a light process that could be done prior to the training loop. Perhaps, as an alternate solution, we could temporarily attach a noop collator to `train_dataloader` while skipping the first batches. ### Your contribution I would be glad to provide a PR. I can investigate the issue further but I would need advices on the matter as I don't have a setup to test all possible combinations of `dataloader`, distributed, `deepspeed` and so forth.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21934/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21934/timeline

completed

null

https://api.github.com/repos/huggingface/transformers/issues/21933

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21933/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21933/comments

https://api.github.com/repos/huggingface/transformers/issues/21933/events

https://github.com/huggingface/transformers/pull/21933

1,608,855,040

PR_kwDOCUB6oc5LPNP8

21,933

Update README logo

{ "login": "gary149", "id": 3841370, "node_id": "MDQ6VXNlcjM4NDEzNzA=", "avatar_url": "https://avatars.githubusercontent.com/u/3841370?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gary149", "html_url": "https://github.com/gary149", "followers_url": "https://api.github.com/users/gary149/followers", "following_url": "https://api.github.com/users/gary149/following{/other_user}", "gists_url": "https://api.github.com/users/gary149/gists{/gist_id}", "starred_url": "https://api.github.com/users/gary149/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gary149/subscriptions", "organizations_url": "https://api.github.com/users/gary149/orgs", "repos_url": "https://api.github.com/users/gary149/repos", "events_url": "https://api.github.com/users/gary149/events{/privacy}", "received_events_url": "https://api.github.com/users/gary149/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "nice!" ]

1,677

1,678

1,677

CONTRIBUTOR

null

# What does this PR do? Update the README logo, mainly to have it visible in dark-mode (instead of black on black).

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21933/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21933/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21933", "html_url": "https://github.com/huggingface/transformers/pull/21933", "diff_url": "https://github.com/huggingface/transformers/pull/21933.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21933.patch", "merged_at": 1677862660000 }

https://api.github.com/repos/huggingface/transformers/issues/21932

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21932/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21932/comments

https://api.github.com/repos/huggingface/transformers/issues/21932/events

https://github.com/huggingface/transformers/issues/21932

1,608,828,095

I_kwDOCUB6oc5f5MS_

21,932

Some text in the international README files are in the wrong language

{ "login": "shermansiu", "id": 12627125, "node_id": "MDQ6VXNlcjEyNjI3MTI1", "avatar_url": "https://avatars.githubusercontent.com/u/12627125?v=4", "gravatar_id": "", "url": "https://api.github.com/users/shermansiu", "html_url": "https://github.com/shermansiu", "followers_url": "https://api.github.com/users/shermansiu/followers", "following_url": "https://api.github.com/users/shermansiu/following{/other_user}", "gists_url": "https://api.github.com/users/shermansiu/gists{/gist_id}", "starred_url": "https://api.github.com/users/shermansiu/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/shermansiu/subscriptions", "organizations_url": "https://api.github.com/users/shermansiu/orgs", "repos_url": "https://api.github.com/users/shermansiu/repos", "events_url": "https://api.github.com/users/shermansiu/events{/privacy}", "received_events_url": "https://api.github.com/users/shermansiu/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md) are likely to be ignored." ]

1,677

1,681

CONTRIBUTOR

null

### System Info Some of the text in the international README files, such as the model descriptions, are in the wrong language. ### Who can help? _No response_ ### Reproduction N/A ### Expected behavior The text in each README file should be in the correct language. A good starting point is looking at the model list for each language.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21932/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21932/timeline

completed

null

https://api.github.com/repos/huggingface/transformers/issues/21931

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21931/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21931/comments

https://api.github.com/repos/huggingface/transformers/issues/21931/events

https://github.com/huggingface/transformers/pull/21931

1,608,770,641

PR_kwDOCUB6oc5LO7GF

21,931

[CLAP] Support batched inputs for CLAP. Fixes pipeline issues

{ "login": "ArthurZucker", "id": 48595927, "node_id": "MDQ6VXNlcjQ4NTk1OTI3", "avatar_url": "https://avatars.githubusercontent.com/u/48595927?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ArthurZucker", "html_url": "https://github.com/ArthurZucker", "followers_url": "https://api.github.com/users/ArthurZucker/followers", "following_url": "https://api.github.com/users/ArthurZucker/following{/other_user}", "gists_url": "https://api.github.com/users/ArthurZucker/gists{/gist_id}", "starred_url": "https://api.github.com/users/ArthurZucker/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ArthurZucker/subscriptions", "organizations_url": "https://api.github.com/users/ArthurZucker/orgs", "repos_url": "https://api.github.com/users/ArthurZucker/repos", "events_url": "https://api.github.com/users/ArthurZucker/events{/privacy}", "received_events_url": "https://api.github.com/users/ArthurZucker/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "Just need to update the expected values for the doctest and will merge", "Pipeline, CI and doctests are all green 😉 " ]

1,677

COLLABORATOR

null

# What does this PR do? Support batching the `is_longer` part of the input for `zero_shot_audio_classification`. The model was not able to run batched inputs. Added a test to support this.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21931/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21931/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21931", "html_url": "https://github.com/huggingface/transformers/pull/21931", "diff_url": "https://github.com/huggingface/transformers/pull/21931.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21931.patch", "merged_at": 1677865339000 }

https://api.github.com/repos/huggingface/transformers/issues/21930

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21930/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21930/comments

https://api.github.com/repos/huggingface/transformers/issues/21930/events

https://github.com/huggingface/transformers/pull/21930

1,608,700,096

PR_kwDOCUB6oc5LOsFP

21,930

Avoid failure in `check_repo.py` due to missing backends

{ "login": "ydshieh", "id": 2521628, "node_id": "MDQ6VXNlcjI1MjE2Mjg=", "avatar_url": "https://avatars.githubusercontent.com/u/2521628?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ydshieh", "html_url": "https://github.com/ydshieh", "followers_url": "https://api.github.com/users/ydshieh/followers", "following_url": "https://api.github.com/users/ydshieh/following{/other_user}", "gists_url": "https://api.github.com/users/ydshieh/gists{/gist_id}", "starred_url": "https://api.github.com/users/ydshieh/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ydshieh/subscriptions", "organizations_url": "https://api.github.com/users/ydshieh/orgs", "repos_url": "https://api.github.com/users/ydshieh/repos", "events_url": "https://api.github.com/users/ydshieh/events{/privacy}", "received_events_url": "https://api.github.com/users/ydshieh/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,677

COLLABORATOR

null

# What does this PR do? My 2 PRs #21903 and #21909 forgot to check missing backends just like `check_all_models_are_auto_configured`, and it could fail on users env where some backends are missinig. This PR fixes this by checking missing backends. Sorry about that. **Remark** ``` from transformers.models.auto.modeling_flax_auto import FLAX_MODEL_MAPPING_NAMES ``` this worked even without backend jax/flax. That's why I missed.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21930/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21930/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21930", "html_url": "https://github.com/huggingface/transformers/pull/21930", "diff_url": "https://github.com/huggingface/transformers/pull/21930.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21930.patch", "merged_at": 1677854061000 }

https://api.github.com/repos/huggingface/transformers/issues/21929

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21929/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21929/comments

https://api.github.com/repos/huggingface/transformers/issues/21929/events

https://github.com/huggingface/transformers/pull/21929

1,608,604,265

PR_kwDOCUB6oc5LOWqy

21,929

[Flan-UL2] Add-flan-ul2

{ "login": "ArthurZucker", "id": 48595927, "node_id": "MDQ6VXNlcjQ4NTk1OTI3", "avatar_url": "https://avatars.githubusercontent.com/u/48595927?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ArthurZucker", "html_url": "https://github.com/ArthurZucker", "followers_url": "https://api.github.com/users/ArthurZucker/followers", "following_url": "https://api.github.com/users/ArthurZucker/following{/other_user}", "gists_url": "https://api.github.com/users/ArthurZucker/gists{/gist_id}", "starred_url": "https://api.github.com/users/ArthurZucker/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ArthurZucker/subscriptions", "organizations_url": "https://api.github.com/users/ArthurZucker/orgs", "repos_url": "https://api.github.com/users/ArthurZucker/repos", "events_url": "https://api.github.com/users/ArthurZucker/events{/privacy}", "received_events_url": "https://api.github.com/users/ArthurZucker/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "Also, why are the international README entries in the wrong language?", "This should not be the case! Will fix this. ", "@sgugger it seems that the README translations are not correct for several languages (korean, jp, etc) due to the fact that they do not follow the correct structure (\"released with the repository\", \"released with the blogpost\", ..), As this requires a bit of work I think that we can address a solution in a follow-up PR! Wdyt?", "Btw, I opened a new issue for the README translations to keep things nice and organized (#21932).", "_The documentation is not available anymore as the PR was closed or merged._", "Thanks a lot! " ]

1,677

COLLABORATOR

null

# What does this PR do? Adds the documentation for the FLan-UL2 model cc @younesbelkada Fixes #21917

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21929/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21929/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21929", "html_url": "https://github.com/huggingface/transformers/pull/21929", "diff_url": "https://github.com/huggingface/transformers/pull/21929.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21929.patch", "merged_at": 1677862645000 }

https://api.github.com/repos/huggingface/transformers/issues/21928

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21928/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21928/comments

https://api.github.com/repos/huggingface/transformers/issues/21928/events

https://github.com/huggingface/transformers/pull/21928

1,608,584,437

PR_kwDOCUB6oc5LOSVw

21,928

Use large VM for `repo_utils_job`

{ "login": "ydshieh", "id": 2521628, "node_id": "MDQ6VXNlcjI1MjE2Mjg=", "avatar_url": "https://avatars.githubusercontent.com/u/2521628?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ydshieh", "html_url": "https://github.com/ydshieh", "followers_url": "https://api.github.com/users/ydshieh/followers", "following_url": "https://api.github.com/users/ydshieh/following{/other_user}", "gists_url": "https://api.github.com/users/ydshieh/gists{/gist_id}", "starred_url": "https://api.github.com/users/ydshieh/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ydshieh/subscriptions", "organizations_url": "https://api.github.com/users/ydshieh/orgs", "repos_url": "https://api.github.com/users/ydshieh/repos", "events_url": "https://api.github.com/users/ydshieh/events{/privacy}", "received_events_url": "https://api.github.com/users/ydshieh/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "Can you just try to do a fake code modification in one of the repo utils so that the tests are run and we can check it works?", "Should be fine with 2 runs\r\n\r\nhttps://app.circleci.com/pipelines/github/huggingface/transformers/58978/workflows/d2090147-1c3f-4b21-8840-36409ab3424b\r\n\r\nhttps://app.circleci.com/pipelines/github/huggingface/transformers/58978/workflows/09bc67f7-a079-4865-9ec2-9094820afa0e/jobs/719891\r\n\r\nand it shows docker/large", "_The documentation is not available anymore as the PR was closed or merged._" ]

1,677

COLLABORATOR

null

# What does this PR do? Use `large` VM for `repo_utils_job`, as #21856 add `torch` for that job which requires more memory.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21928/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21928/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21928", "html_url": "https://github.com/huggingface/transformers/pull/21928", "diff_url": "https://github.com/huggingface/transformers/pull/21928.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21928.patch", "merged_at": 1677850983000 }

https://api.github.com/repos/huggingface/transformers/issues/21927

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21927/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21927/comments

https://api.github.com/repos/huggingface/transformers/issues/21927/events

https://github.com/huggingface/transformers/pull/21927

1,608,580,260

PR_kwDOCUB6oc5LORiv

21,927

Fix `ZeroShotAudioClassificationPipeline` doctest

{ "login": "ydshieh", "id": 2521628, "node_id": "MDQ6VXNlcjI1MjE2Mjg=", "avatar_url": "https://avatars.githubusercontent.com/u/2521628?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ydshieh", "html_url": "https://github.com/ydshieh", "followers_url": "https://api.github.com/users/ydshieh/followers", "following_url": "https://api.github.com/users/ydshieh/following{/other_user}", "gists_url": "https://api.github.com/users/ydshieh/gists{/gist_id}", "starred_url": "https://api.github.com/users/ydshieh/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ydshieh/subscriptions", "organizations_url": "https://api.github.com/users/ydshieh/orgs", "repos_url": "https://api.github.com/users/ydshieh/repos", "events_url": "https://api.github.com/users/ydshieh/events{/privacy}", "received_events_url": "https://api.github.com/users/ydshieh/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,677

1,679

1,677

COLLABORATOR

null

# What does this PR do? Fix `ZeroShotAudioClassificationPipeline` doctest

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21927/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21927/timeline

null

true

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21927", "html_url": "https://github.com/huggingface/transformers/pull/21927", "diff_url": "https://github.com/huggingface/transformers/pull/21927.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21927.patch", "merged_at": null }

https://api.github.com/repos/huggingface/transformers/issues/21926

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21926/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21926/comments

https://api.github.com/repos/huggingface/transformers/issues/21926/events

https://github.com/huggingface/transformers/pull/21926

1,608,421,217

PR_kwDOCUB6oc5LNvRq

21,926

Sync preprocesses before loading the processor at run_speech_recognition_ctc.py

{ "login": "mpenagar", "id": 1698682, "node_id": "MDQ6VXNlcjE2OTg2ODI=", "avatar_url": "https://avatars.githubusercontent.com/u/1698682?v=4", "gravatar_id": "", "url": "https://api.github.com/users/mpenagar", "html_url": "https://github.com/mpenagar", "followers_url": "https://api.github.com/users/mpenagar/followers", "following_url": "https://api.github.com/users/mpenagar/following{/other_user}", "gists_url": "https://api.github.com/users/mpenagar/gists{/gist_id}", "starred_url": "https://api.github.com/users/mpenagar/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/mpenagar/subscriptions", "organizations_url": "https://api.github.com/users/mpenagar/orgs", "repos_url": "https://api.github.com/users/mpenagar/repos", "events_url": "https://api.github.com/users/mpenagar/events{/privacy}", "received_events_url": "https://api.github.com/users/mpenagar/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "cc @sanchit-gandhi ", "Updated seq2seq ASR fine-tuning script. I'm not very good with github, I guess there is no need to do a new PR.", "Wait... there is something I don't get correctly.\r\n\r\nAs far as I understand from the [(documentation )](https://huggingface.co/transformers/v4.11.3/main_classes/trainer.html#transformers.TrainingArguments.main_process_first) , any code inside a block `with training_args.main_process_first():` should be executed only by the main process:\r\n\r\n```\r\nA context manager for torch distributed environment where\r\non needs to do something on the main process, while blocking\r\nreplicas, and when it’s finished releasing the replicas.\r\n\r\nOne such use is for datasets’s map feature which to be efficient\r\nshould be run once on the main process, which upon completion\r\nsaves a cached version of results and which then automatically\r\ngets loaded by the replicas.\r\n```\r\n\r\nBut in my experience, the code is executed by all the processes, not just the main one. Take this minimal `example.py':\r\n\r\n```python\r\nfrom transformers import TrainingArguments,HfArgumentParser\r\nfrom transformers.trainer_utils import is_main_process\r\n\r\ndef main():\r\n parser = HfArgumentParser((TrainingArguments,))\r\n training_args, = parser.parse_args_into_dataclasses()\r\n rank = training_args.local_rank\r\n main_process = is_main_process(rank)\r\n print(f'\\nBEFORE WITH - local_rank={rank} is_main_process={main_process}')\r\n with training_args.main_process_first():\r\n print(f'\\nINSIDE WITH - local_rank={rank}')\r\n\r\nif __name__ == \"__main__\":\r\n main()\r\n```\r\n\r\nIf I execute it in a 4 GPU node:\r\n\r\n`OMP_NUM_THREADS=1 python3 -m torch.distributed.launch --nproc_per_node 4 example.py --output_dir none`\r\n\r\nThe synching is working, but all processes execute the \"INSIDE\" `print`\r\n\r\nExecuting with newer `torchrun` does the same:\r\n\r\n`OMP_NUM_THREADS=1 torchrun --standalone --nnodes=1 --nproc_per_node=4 example.py --output_dir none`\r\n\r\nWhat I am geting wrong?", "No, as the name indicates, it executes the code in the context manager on the main process, and then on all the others. The code is indeed executed in all processes, just in a certain order.\r\n\r\nSince with Datasets, everything is cached, executing the preprocessing inside that contextmanager means that process 0 will do the preprocessing, and then all the other will load the result from the cache without needing to do the preprocessing.", "Ok, then the PR is not correct, since all the processes will try to write the json files. I removed the original:\r\n\r\n`if is_main_process(training_args.local_rank):`\r\n\r\nthat should be there inside the `with` block...", "Indeed, your changes are perfect. Is this ready to be merged now?", "It is working in my end without any problem", "Is this good for merge @mpenagar? Changes LGTM!", "Yes, it is ready. Anyway, I don't know how github works. Should I close the PR (there is a \"Close\" button there)?", "Awesome, thanks for confirming @mpenagar and for your contribution 🤗" ]

1,677

1,680

CONTRIBUTOR

null

# What does this PR do? Make sure all processes wait until data is saved before loading the processor from the `output_dir` in the `pytorch/speech-recognition/run_speech_recognition_ctc.py` example. Issue: * Non-main proccesses might try to load the processor from the `output_dir` before it is saved. ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [x] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [ ] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [ ] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [ ] Did you write any new necessary tests? ## Who can review? Anyone in the community is free to review the PR once the tests have passed. Feel free to tag members/contributors who may be interested in your PR.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21926/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21926/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21926", "html_url": "https://github.com/huggingface/transformers/pull/21926", "diff_url": "https://github.com/huggingface/transformers/pull/21926.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21926.patch", "merged_at": 1680701765000 }

https://api.github.com/repos/huggingface/transformers/issues/21925

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21925/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21925/comments

https://api.github.com/repos/huggingface/transformers/issues/21925/events

https://github.com/huggingface/transformers/issues/21925

1,608,400,837

I_kwDOCUB6oc5f3j_F

21,925

[Whisper] Index error with model weights like base, large-v2

{ "login": "kurianbenoy-sentient", "id": 101088788, "node_id": "U_kgDOBgZ-FA", "avatar_url": "https://avatars.githubusercontent.com/u/101088788?v=4", "gravatar_id": "", "url": "https://api.github.com/users/kurianbenoy-sentient", "html_url": "https://github.com/kurianbenoy-sentient", "followers_url": "https://api.github.com/users/kurianbenoy-sentient/followers", "following_url": "https://api.github.com/users/kurianbenoy-sentient/following{/other_user}", "gists_url": "https://api.github.com/users/kurianbenoy-sentient/gists{/gist_id}", "starred_url": "https://api.github.com/users/kurianbenoy-sentient/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/kurianbenoy-sentient/subscriptions", "organizations_url": "https://api.github.com/users/kurianbenoy-sentient/orgs", "repos_url": "https://api.github.com/users/kurianbenoy-sentient/repos", "events_url": "https://api.github.com/users/kurianbenoy-sentient/events{/privacy}", "received_events_url": "https://api.github.com/users/kurianbenoy-sentient/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "Do you have the sample audio that triggers the bug ?\r\nI couldn't reproduce locally.\r\n\r\nAslo, is there any particular reason for having such a complex script ? I can see several things that could slow down this code more than it should.\r\n\r\n`pipe = pipeline(task=\"automatic-speech-recognition\", model=model_dir, device=device)` should work out of the box.\r\n\r\nHere for instance the device is set during prediction, which is fcalled repetively could slow down things a bit (once the device is set it shouldn't move afterwards. Even the pipeline creation, is not big, but still could be done ahead of time.\r\n\r\n\r\nCheers ! ", "@Narsil my apologies. I got this error with specific weights in transformers library v4.26.1. \r\n\r\nYou can reproduce this error with the file in the below link.\r\n\r\nhttps://drive.google.com/file/d/1ubw9PLlo5NwB7xwTwfVnuRWKH8L5XPIx/view?usp=sharing\r\n\r\nI am using whisper models for offline inference. So simply passing the pipeline directly won't work:\r\n\r\n```\r\npipe = pipeline(task=\"automatic-speech-recognition\", model=model_dir, device=device)\r\n```\r\n\r\nThat's why I have to manually pass the pipeline:\r\n\r\n```\r\npipe = pipeline(\r\n task=\"automatic-speech-recognition\",\r\n model=self.model,\r\n tokenizer=self.tokenizer,\r\n feature_extractor=self.feature_extractor,\r\n framework=\"pt\",\r\n chunk_length_s=30,\r\n generate_kwargs={\"max_new_tokens\": 1024},\r\n device=dno,\r\n return_timestamps=timestamp\r\n )\r\n```\r\n\r\nThanks for pointing the issue in device_no, I will set it in such a way it won't always be called each time when method is called. ", "While in the latest version in master I am getting the error: v4.27.0-dev for all model weights, when I use the above code:\r\n\r\n```\r\nTraceback (most recent call last): File \"/root/whisper_project/microservice-ai-whisper/deploy/Whisper.py\", line 99, in <module> print(model.predict_raw(payload)) \r\nile \"/root/whisper_project/microservice-ai-whisper/deploy/Whisper.py\", line 72, in predict_raw\r\n results = pipe(a_file.name) File \"/root/mambaforge/lib/python3.10/site-packages/transformers/pipelines/automatic_speech_recognition.py\", line 272, in __call__ return super().__call__(inputs, **kwargs) File \"/root/mambaforge/lib/python3.10/site-packages/transformers/pipelines/base.py\", line 1101, in __call__ return next( File \"/root/mambaforge/lib/python3.10/site-packages/transformers/pipelines/pt_utils.py\", line 124, in __next__ item = next(self.iterator) File \"/root/mambaforge/lib/python3.10/site-packages/transformers/pipelines/pt_utils.py\", line 266, in __next__ processed = self.infer(next(self.iterator), **self.params) File \"/root/mambaforge/lib/python3.10/site-packages/transformers/pipelines/base.py\", line 1015, in forward model_outputs = self._forward(model_inputs, **forward_params) File \"/root/mambaforge/lib/python3.10/site-packages/transformers/pipelines/automatic_speech_recognition.py\", line 445, in _forward tokens = self.model.generate( File \"/root/mambaforge/lib/python3.10/site-packages/transformers/models/whisper/modeling_whisper.py\", line 1534, in generate logits_processor = [WhisperTimeStampLogitsProcessor(generation_config)] File \"/root/mambaforge/lib/python3.10/site-packages/transformers/generation/logits_process.py\", line 935, in __init__ self.no_timestamps_token_id = generate_config.no_timestamps_token_id AttributeError: 'GenerationConfig' object has no attribute 'no_timestamps_token_id' \r\n```\r\n\r\n![image](https://user-images.githubusercontent.com/101088788/223036652-fa889119-fbe1-41ed-87b1-91db3cfd29ee.png)\r\n", "Ok this works:\r\n\r\n```python\r\nfrom transformers import pipeline\r\n\r\npipe = pipeline(\r\n task=\"automatic-speech-recognition\",\r\n model=\"openai/whisper-large-v2\",\r\n chunk_length_s=30,\r\n device=0,\r\n return_timestamps=True,\r\n)\r\n\r\nout = pipe(\"sample.wav\")\r\nprint(out)\r\n```\r\n\r\nThen you can definitely use the same simplicity with local values. Just save the pipeline\r\n\r\n`pipe.save_pretrained(\"whisper-local\")`\r\n\r\nAnd then \r\n```python\r\nfrom transformers import pipeline\r\n\r\npipe = pipeline(\r\n task=\"automatic-speech-recognition\",\r\n model=\"./whisper_local\",\r\n chunk_length_s=30,\r\n device=0,\r\n return_timestamps=True,\r\n)\r\n```\r\n\r\nShould work.\r\n\r\nNo for the latest error, I'm guessing this has to do with recent changes in the configuration of whisper.\r\nIs that possible @ArthurZucker ?", "Thank you @Narsil for techniques to reduce complexity. \r\n\r\nI didn't we can save pipeline also locally.", "Yes, if you are using main, and want to use timestamps, you should make sure that the model has the `no_timestamp_id`. \r\nThis is because this is a new feature, and for proper generation, we recommend setting a `generation_config`. \r\nAlso, if you check both models, they have a different `configuration.forced_decoder_ids`. This is probably what was causing the difference in behaviour between the `large` and `base` (for example).\r\n\r\n\r\n", "@ArthurZucker how can I make sure that model has the `no_timestamp_id`?", "You should save/push the update `generation_config`. \r\nThe simplest you can do is the following: \r\n```python \r\nfrom transformers import GenerationConfig, WhisperForConditionalGeneration\r\nmodel = WhisperForConditionalGeneration.from_pretrained(\"your_pretrained_checkpoint\")\r\ngeneration_config = GenerationConfig.from_pretrained(\"openai/whisper-base\") # if you are using a multilingual model\r\nmodel.generation_config = generation_config\r\nmodel.push_to_hub(\"your_pretrained_checkpoint\", use_auth_token = \"your_token_if_not_logged_in\", create_pr = True)\r\n```", "Thanks for your answer. Hope you will resolve this issue before taking it in next stable version. I am closing this issue for now.\r\n" ]

1,677

1,678

NONE

null

### System Info - `transformers` version: 4.27.0.dev0 - Platform: Linux-5.4.0-66-generic-x86_64-with-glibc2.31 - Python version: 3.10.9 - Huggingface_hub version: 0.12.1 - PyTorch version (GPU?): 1.13.1 (True) - Tensorflow version (GPU?): not installed (NA) - Flax version (CPU?/GPU?/TPU?): not installed (NA) - Jax version: not installed - JaxLib version: not installed - Using GPU in script?: Yes - Using distributed or parallel set-up in script?: No ### Who can help? @sanchit-gandhi @Narsil ### Information - [ ] The official example scripts - [X] My own modified scripts ### Tasks - [X] An officially supported task in the `examples` folder (such as GLUE/SQuAD, ...) - [ ] My own task or dataset (give details below) ### Reproduction 1. I have been using Whisper from transformers library with my own custom script. The full code is attached below for reference: ``` import argparse import base64 import os import tempfile import torch from transformers import ( AutoFeatureExtractor, AutoTokenizer, WhisperForConditionalGeneration, WhisperProcessor, pipeline, ) from utils.b64 import convert_to_base64 model_dir = "openai/whisper-base" class Whisper: def __init__(self): self.tokenizer = AutoTokenizer.from_pretrained(model_dir) self.feature_extractor = AutoFeatureExtractor.from_pretrained(model_dir) self.processor = WhisperProcessor.from_pretrained(model_dir) self.device = "cuda" if torch.cuda.is_available() else "cpu" # print([x for x in Path(model_dir).iterdir()]) self.model = WhisperForConditionalGeneration.from_pretrained(model_dir).to(self.device) self.is_timestamp = False def predict_raw(self, payload): if payload is None: return {"inputerror": "JSON expected"} if "wav_base64" not in payload: return {"inputerror": "Missing key wav_base64 in payload."} afs = payload["wav_base64"] if not isinstance(afs, str): return {"inputerror": "Audio file to passed as input in base64 format"} if "timestamps" not in payload: timestamp = self.is_timestamp elif "timestamps" in payload and type(payload["timestamps"]) != str: return {"inputerror": "timestamps should be of string datatype"} elif "timestamps" in payload and payload ["timestamps"] != "true": return {"inputerror": "timestamps payload should be of Value True"} else: timestamp = True lang = payload.get("language") print(lang) afs = base64.b64decode(afs) dno = torch.cuda.current_device() if self.device == "cuda" else -1 with tempfile.NamedTemporaryFile() as a_file: a_file.write(afs) pipe = pipeline( task="automatic-speech-recognition", model=self.model, tokenizer=self.tokenizer, feature_extractor=self.feature_extractor, framework="pt", chunk_length_s=30, generate_kwargs={"max_new_tokens": 1024}, device=dno, return_timestamps=timestamp ) if lang: self.model.config.forced_decoder_ids = self.processor.get_decoder_prompt_ids( task="transcribe", language=lang ) if timestamp: results = pipe(a_file.name) timestamp_info = [{"text": x["text"], "start": x["timestamp"][0],"end": x["timestamp"][1]} for x in results["chunks"]] return {"text": results["text"], "timestamps": timestamp_info} else: return pipe(a_file.name) if __name__ == "__main__": parser = argparse.ArgumentParser() parser.add_argument( "audio_file", type=str, help="Input the path to audio file you want to transcribe", ) parser.add_argument( "--language", type=str, help="Input the language", ) args = parser.parse_args() kwargs = vars(args) b64 = convert_to_base64(kwargs["audio_file"]) if kwargs["language"]: payload = {"wav_base64": b64, "language": kwargs["language"]} payload = {"wav_base64": b64, "timestamps": "true"} model = Whisper() print(model.predict_raw(payload)) payload = {"wav_base64": b64} print(model.predict_raw(payload)) ``` 2. When I am using model weights of `base` and `large-v2`. The code is getting an error as below: ``` Traceback (most recent call last): File "/root/whisper_project/microservice-ai-whisper/deploy/debug.py", line 102, in <module> print(model.predict_raw(payload)) File "/root/whisper_project/microservice-ai-whisper/deploy/debug.py", line 75, in predict_raw results = pipe(a_file.name) File "/root/mambaforge/lib/python3.10/site-packages/transformers/pipelines/automatic_speech_recognition.py", line 272, in __call__ return super().__call__(inputs, **kwargs) File "/root/mambaforge/lib/python3.10/site-packages/transformers/pipelines/base.py", line 1101, in __call__ return next( File "/root/mambaforge/lib/python3.10/site-packages/transformers/pipelines/pt_utils.py", line 124, in __next__ item = next(self.iterator) File "/root/mambaforge/lib/python3.10/site-packages/transformers/pipelines/pt_utils.py", line 266, in __next__ processed = self.infer(next(self.iterator), **self.params) File "/root/mambaforge/lib/python3.10/site-packages/transformers/pipelines/base.py", line 1015, in forward model_outputs = self._forward(model_inputs, **forward_params) File "/root/mambaforge/lib/python3.10/site-packages/transformers/pipelines/automatic_speech_recognition.py", line 445, in _forward tokens = self.model.generate( File "/root/mambaforge/lib/python3.10/site-packages/transformers/models/whisper/modeling_whisper.py", line 1543, in generate return super().generate( File "/root/mambaforge/lib/python3.10/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context return func(*args, **kwargs) File "/root/mambaforge/lib/python3.10/site-packages/transformers/generation/utils.py", line 1406, in generate return self.greedy_search( File "/root/mambaforge/lib/python3.10/site-packages/transformers/generation/utils.py", line 2211, in greedy_search next_token_logits = outputs.logits[:, -1, :] IndexError: index -1 is out of bounds for dimension 1 with size 0 ``` 3. I noticed this issue only with some model weights. The exact same code works with medium and large-v1 model weights ### Expected behavior Get the transcribed output along with corresponding timestamps

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21925/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21925/timeline

completed

null

https://api.github.com/repos/huggingface/transformers/issues/21924

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21924/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21924/comments

https://api.github.com/repos/huggingface/transformers/issues/21924/events

https://github.com/huggingface/transformers/pull/21924

1,608,382,882

PR_kwDOCUB6oc5LNnAM

21,924

[WIP] [Flax] Improving Docs

{ "login": "Shubhamai", "id": 51819922, "node_id": "MDQ6VXNlcjUxODE5OTIy", "avatar_url": "https://avatars.githubusercontent.com/u/51819922?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Shubhamai", "html_url": "https://github.com/Shubhamai", "followers_url": "https://api.github.com/users/Shubhamai/followers", "following_url": "https://api.github.com/users/Shubhamai/following{/other_user}", "gists_url": "https://api.github.com/users/Shubhamai/gists{/gist_id}", "starred_url": "https://api.github.com/users/Shubhamai/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Shubhamai/subscriptions", "organizations_url": "https://api.github.com/users/Shubhamai/orgs", "repos_url": "https://api.github.com/users/Shubhamai/repos", "events_url": "https://api.github.com/users/Shubhamai/events{/privacy}", "received_events_url": "https://api.github.com/users/Shubhamai/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/transformers/pr_21924). All of your documentation changes will be reflected on that endpoint.", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md) are likely to be ignored." ]

1,677

1,681

CONTRIBUTOR

null

# What does this PR do? Adding relevant Jax/Flax code in `<frameworkcontent>` in Transformers Docs. Issues: - Since Flax has no `Trainer` class, no code exist for `Train` section in the tasks guide ([example](https://huggingface.co/docs/transformers/tasks/token_classification#train)). ## Before submitting - [x] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [x] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [ ] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [ ] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [ ] Did you write any new necessary tests? ## Who can review? Anyone in the community is free to review the PR once the tests have passed. Feel free to tag members/contributors who may be interested in your PR. *Not mentioning due to currently in WIP* - Documentation: sgugger, stevhliu and MKhalusova - Flax: sanchit-gandhi

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21924/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21924/timeline

null

true

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21924", "html_url": "https://github.com/huggingface/transformers/pull/21924", "diff_url": "https://github.com/huggingface/transformers/pull/21924.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21924.patch", "merged_at": null }

https://api.github.com/repos/huggingface/transformers/issues/21923

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21923/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21923/comments

https://api.github.com/repos/huggingface/transformers/issues/21923/events

https://github.com/huggingface/transformers/pull/21923

1,608,317,817

PR_kwDOCUB6oc5LNYjx

21,923

Fix `AlignModelTest` tests

{ "login": "ydshieh", "id": 2521628, "node_id": "MDQ6VXNlcjI1MjE2Mjg=", "avatar_url": "https://avatars.githubusercontent.com/u/2521628?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ydshieh", "html_url": "https://github.com/ydshieh", "followers_url": "https://api.github.com/users/ydshieh/followers", "following_url": "https://api.github.com/users/ydshieh/following{/other_user}", "gists_url": "https://api.github.com/users/ydshieh/gists{/gist_id}", "starred_url": "https://api.github.com/users/ydshieh/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ydshieh/subscriptions", "organizations_url": "https://api.github.com/users/ydshieh/orgs", "repos_url": "https://api.github.com/users/ydshieh/repos", "events_url": "https://api.github.com/users/ydshieh/events{/privacy}", "received_events_url": "https://api.github.com/users/ydshieh/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,677

COLLABORATOR

null

# What does this PR do? - Fix `AlignModelTest` torchscript tests. This model has non-persistent buffer, and need a bit more care (same as in common test) ```python self.register_buffer( "token_type_ids", torch.zeros(self.position_ids.size(), dtype=torch.long), persistent=False ) ``` - Fix `AlignModelTest.test_multi_gpu_data_parallel_forward`: same as CLIP, we need a even number for `batch_size`, as this model have `logits_per_image` and `logits_per_text` which is of shape `(batch_size, batch_size)`, and in order to gather across devices, we need the second dim being the same.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21923/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21923/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21923", "html_url": "https://github.com/huggingface/transformers/pull/21923", "diff_url": "https://github.com/huggingface/transformers/pull/21923.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21923.patch", "merged_at": 1677851230000 }

https://api.github.com/repos/huggingface/transformers/issues/21922

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21922/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21922/comments

https://api.github.com/repos/huggingface/transformers/issues/21922/events

https://github.com/huggingface/transformers/pull/21922

1,608,265,173

PR_kwDOCUB6oc5LNM9e

21,922

update model_split_percents

{ "login": "ydshieh", "id": 2521628, "node_id": "MDQ6VXNlcjI1MjE2Mjg=", "avatar_url": "https://avatars.githubusercontent.com/u/2521628?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ydshieh", "html_url": "https://github.com/ydshieh", "followers_url": "https://api.github.com/users/ydshieh/followers", "following_url": "https://api.github.com/users/ydshieh/following{/other_user}", "gists_url": "https://api.github.com/users/ydshieh/gists{/gist_id}", "starred_url": "https://api.github.com/users/ydshieh/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ydshieh/subscriptions", "organizations_url": "https://api.github.com/users/ydshieh/orgs", "repos_url": "https://api.github.com/users/ydshieh/repos", "events_url": "https://api.github.com/users/ydshieh/events{/privacy}", "received_events_url": "https://api.github.com/users/ydshieh/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,677

COLLABORATOR

null

# What does this PR do? #21883 changed `WhisperModelTest`'s `model_split_percents ` from `[0.5, 0.7, 0.9]` to `= [0.8, 0.9]` to make `test_model_parallelism` work. But it fails `test_disk_offload` which uses `self.model_split_percents[0]`. This PR adds back `0.5`. With `model_split_percents = [0.5, 0.8, 0.9]`, the relevant Whisper tests using `model_split_percents` all pass.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21922/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21922/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21922", "html_url": "https://github.com/huggingface/transformers/pull/21922", "diff_url": "https://github.com/huggingface/transformers/pull/21922.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21922.patch", "merged_at": 1677850509000 }

https://api.github.com/repos/huggingface/transformers/issues/21921

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21921/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21921/comments

https://api.github.com/repos/huggingface/transformers/issues/21921/events

https://github.com/huggingface/transformers/pull/21921

1,608,253,076

PR_kwDOCUB6oc5LNKVE

21,921

Fix gradient checkpointing megatron bert

{ "login": "KMFODA", "id": 35491698, "node_id": "MDQ6VXNlcjM1NDkxNjk4", "avatar_url": "https://avatars.githubusercontent.com/u/35491698?v=4", "gravatar_id": "", "url": "https://api.github.com/users/KMFODA", "html_url": "https://github.com/KMFODA", "followers_url": "https://api.github.com/users/KMFODA/followers", "following_url": "https://api.github.com/users/KMFODA/following{/other_user}", "gists_url": "https://api.github.com/users/KMFODA/gists{/gist_id}", "starred_url": "https://api.github.com/users/KMFODA/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/KMFODA/subscriptions", "organizations_url": "https://api.github.com/users/KMFODA/orgs", "repos_url": "https://api.github.com/users/KMFODA/repos", "events_url": "https://api.github.com/users/KMFODA/events{/privacy}", "received_events_url": "https://api.github.com/users/KMFODA/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,677

CONTRIBUTOR

null

This PR fixes a bug that a user can encounter while using generate and models that use gradient_checkpointing. Fixes Issue https://github.com/huggingface/transformers/issues/21737 cc @younesbelkada or @gante

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21921/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21921/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21921", "html_url": "https://github.com/huggingface/transformers/pull/21921", "diff_url": "https://github.com/huggingface/transformers/pull/21921.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21921.patch", "merged_at": 1677844222000 }

https://api.github.com/repos/huggingface/transformers/issues/21920

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21920/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21920/comments

https://api.github.com/repos/huggingface/transformers/issues/21920/events

https://github.com/huggingface/transformers/pull/21920

1,608,252,928

PR_kwDOCUB6oc5LNKTA

21,920

Fix gradient checkpointing bug in mvp

{ "login": "KMFODA", "id": 35491698, "node_id": "MDQ6VXNlcjM1NDkxNjk4", "avatar_url": "https://avatars.githubusercontent.com/u/35491698?v=4", "gravatar_id": "", "url": "https://api.github.com/users/KMFODA", "html_url": "https://github.com/KMFODA", "followers_url": "https://api.github.com/users/KMFODA/followers", "following_url": "https://api.github.com/users/KMFODA/following{/other_user}", "gists_url": "https://api.github.com/users/KMFODA/gists{/gist_id}", "starred_url": "https://api.github.com/users/KMFODA/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/KMFODA/subscriptions", "organizations_url": "https://api.github.com/users/KMFODA/orgs", "repos_url": "https://api.github.com/users/KMFODA/repos", "events_url": "https://api.github.com/users/KMFODA/events{/privacy}", "received_events_url": "https://api.github.com/users/KMFODA/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,677

CONTRIBUTOR

null

This PR fixes a bug that a user can encounter while using generate and models that use gradient_checkpointing. Fixes Issue https://github.com/huggingface/transformers/issues/21737 cc @younesbelkada or @gante

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21920/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21920/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21920", "html_url": "https://github.com/huggingface/transformers/pull/21920", "diff_url": "https://github.com/huggingface/transformers/pull/21920.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21920.patch", "merged_at": 1677844190000 }

https://api.github.com/repos/huggingface/transformers/issues/21919

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21919/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21919/comments

https://api.github.com/repos/huggingface/transformers/issues/21919/events

https://github.com/huggingface/transformers/pull/21919

1,608,077,381

PR_kwDOCUB6oc5LMkEX

21,919

Fix wrong documentation about DataCollator padding defaults

{ "login": "substanc3-dev", "id": 6898693, "node_id": "MDQ6VXNlcjY4OTg2OTM=", "avatar_url": "https://avatars.githubusercontent.com/u/6898693?v=4", "gravatar_id": "", "url": "https://api.github.com/users/substanc3-dev", "html_url": "https://github.com/substanc3-dev", "followers_url": "https://api.github.com/users/substanc3-dev/followers", "following_url": "https://api.github.com/users/substanc3-dev/following{/other_user}", "gists_url": "https://api.github.com/users/substanc3-dev/gists{/gist_id}", "starred_url": "https://api.github.com/users/substanc3-dev/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/substanc3-dev/subscriptions", "organizations_url": "https://api.github.com/users/substanc3-dev/orgs", "repos_url": "https://api.github.com/users/substanc3-dev/repos", "events_url": "https://api.github.com/users/substanc3-dev/events{/privacy}", "received_events_url": "https://api.github.com/users/substanc3-dev/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,677

CONTRIBUTOR

null

# What does this PR do? Fixes some documation for DataCollatorForTokenClassification and DataCollatorForSeq2Seq, which previously mentioned no padding is the default, which is false. ## Before submitting - [x] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [x] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [ ] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [ ] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [ ] Did you write any new necessary tests? ## Who can review? Anyone in the community is free to review the PR once the tests have passed. Feel free to tag members/contributors who may be interested in your PR (@sgugger).

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21919/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21919/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21919", "html_url": "https://github.com/huggingface/transformers/pull/21919", "diff_url": "https://github.com/huggingface/transformers/pull/21919.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21919.patch", "merged_at": 1677862314000 }

https://api.github.com/repos/huggingface/transformers/issues/21918

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21918/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21918/comments

https://api.github.com/repos/huggingface/transformers/issues/21918/events

https://github.com/huggingface/transformers/pull/21918

1,607,975,616

PR_kwDOCUB6oc5LMOJz

21,918

Fix gradient checkpointing bug in MBart

{ "login": "KMFODA", "id": 35491698, "node_id": "MDQ6VXNlcjM1NDkxNjk4", "avatar_url": "https://avatars.githubusercontent.com/u/35491698?v=4", "gravatar_id": "", "url": "https://api.github.com/users/KMFODA", "html_url": "https://github.com/KMFODA", "followers_url": "https://api.github.com/users/KMFODA/followers", "following_url": "https://api.github.com/users/KMFODA/following{/other_user}", "gists_url": "https://api.github.com/users/KMFODA/gists{/gist_id}", "starred_url": "https://api.github.com/users/KMFODA/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/KMFODA/subscriptions", "organizations_url": "https://api.github.com/users/KMFODA/orgs", "repos_url": "https://api.github.com/users/KMFODA/repos", "events_url": "https://api.github.com/users/KMFODA/events{/privacy}", "received_events_url": "https://api.github.com/users/KMFODA/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,677

CONTRIBUTOR

null

This PR fixes a bug that a user can encounter while using generate and models that use gradient_checkpointing. Fixes Issue https://github.com/huggingface/transformers/issues/21737 cc @younesbelkada or @gante

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21918/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21918/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21918", "html_url": "https://github.com/huggingface/transformers/pull/21918", "diff_url": "https://github.com/huggingface/transformers/pull/21918.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21918.patch", "merged_at": 1677844168000 }

https://api.github.com/repos/huggingface/transformers/issues/21917

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21917/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21917/comments

https://api.github.com/repos/huggingface/transformers/issues/21917/events

https://github.com/huggingface/transformers/issues/21917

1,607,951,523

I_kwDOCUB6oc5f12Sj

21,917

Add FLAN-UL2

{ "login": "shermansiu", "id": 12627125, "node_id": "MDQ6VXNlcjEyNjI3MTI1", "avatar_url": "https://avatars.githubusercontent.com/u/12627125?v=4", "gravatar_id": "", "url": "https://api.github.com/users/shermansiu", "html_url": "https://github.com/shermansiu", "followers_url": "https://api.github.com/users/shermansiu/followers", "following_url": "https://api.github.com/users/shermansiu/following{/other_user}", "gists_url": "https://api.github.com/users/shermansiu/gists{/gist_id}", "starred_url": "https://api.github.com/users/shermansiu/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/shermansiu/subscriptions", "organizations_url": "https://api.github.com/users/shermansiu/orgs", "repos_url": "https://api.github.com/users/shermansiu/repos", "events_url": "https://api.github.com/users/shermansiu/events{/privacy}", "received_events_url": "https://api.github.com/users/shermansiu/received_events", "type": "User", "site_admin": false }

[ { "id": 1843244711, "node_id": "MDU6TGFiZWwxODQzMjQ0NzEx", "url": "https://api.github.com/repos/huggingface/transformers/labels/New%20model", "name": "New model", "color": "fbca04", "default": false, "description": "" } ]

closed

false

null

[]

[ "@DanielHesslow (since you ported the original UL2 weights). I would like to contribute, but I'm not too sure how to convert the weights from JAX to PyTorch.", "I had a dirty dirty script which unfortunately lives on my old dev machine that I don't have with me at the moment 😅 \r\n\r\nI basically just loaded the t5 weights and went through and renamed every thing to match the HF format. ", "Hey! Thanks for opening, they will be available on the hub soon! We are converting them with @younesbelkada ", "The model is already out! (https://huggingface.co/google/flan-ul2)\r\n@younesbelkada has a space comparing Flan-T5-XXL and Flan-UL2 here: https://huggingface.co/spaces/ybelkada/i-like-flan-ul2" ]

1,677

CONTRIBUTOR

null

### Model description UL2 is a unified framework for pretraining models that are universally effective across datasets and setups. UL2 uses Mixture-of-Denoisers (MoD), a pre-training objective that combines diverse pre-training paradigms together. UL2 introduces a notion of mode switching, wherein downstream fine-tuning is associated with specific pre-training schemes. FLAN-UL2 has the same configuration as the original UL2 20B model, except that it has been instruction tuned with Flan. ### Open source status - [X] The model implementation is available - [X] The model weights are available ### Provide useful links for the implementation The model architecture (UL2) is already in Huggingface Transformers. The 20B model weights are here: https://github.com/google-research/google-research/tree/master/ul2#checkpoints

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21917/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21917/timeline

completed

null

https://api.github.com/repos/huggingface/transformers/issues/21916

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21916/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21916/comments

https://api.github.com/repos/huggingface/transformers/issues/21916/events

https://github.com/huggingface/transformers/pull/21916

1,607,853,726

PR_kwDOCUB6oc5LL0sH

21,916

[In progress] Add warning padding attention mask

{ "login": "anruijian", "id": 115125339, "node_id": "U_kgDOBtysWw", "avatar_url": "https://avatars.githubusercontent.com/u/115125339?v=4", "gravatar_id": "", "url": "https://api.github.com/users/anruijian", "html_url": "https://github.com/anruijian", "followers_url": "https://api.github.com/users/anruijian/followers", "following_url": "https://api.github.com/users/anruijian/following{/other_user}", "gists_url": "https://api.github.com/users/anruijian/gists{/gist_id}", "starred_url": "https://api.github.com/users/anruijian/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/anruijian/subscriptions", "organizations_url": "https://api.github.com/users/anruijian/orgs", "repos_url": "https://api.github.com/users/anruijian/repos", "events_url": "https://api.github.com/users/anruijian/events{/privacy}", "received_events_url": "https://api.github.com/users/anruijian/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "cc @gante ", "Thank you for the comment!\r\n\r\nBased on my understanding, [this line of code](https://github.com/huggingface/transformers/pull/21916/files#diff-6b72b98c4c2dcfc6cc606843917733f5d858374fbc22a735ff483bbc0c1e63eaR1054) enables the checking process only once during the forward pass, so it should not significantly impact performance.\r\n\r\nThe current warning method only issue warnings when `attention_mask` is necessary (due to the presence of padding tokens in the input), but no `attention_mask` is provided. In other cases where `attention_mask` is not required, no warning is issued. The additional checking on special tokens allows a more detailed warning message. \r\n\r\nI agree that your suggested method is more concise and efficient, but it may generate warnings when `attention_mask` is not needed. \r\n\r\nSince it's my first time contributing to the community, I don't have a strong opinion towards either solution. The original work is by @ydshieh and @patrickvonplaten. Perhaps they have additional insights and can suggest a more effective solution.", "> Why not a simple logger.warning_once()\r\n\r\nThis is recently introduced :-)", "@anruijian It checks `input_ids` until there is a batch in which a `pad_token_id` exists. If a user is working on a problem where they have no `pad_token_id` on their data and they don't pass the `attention_mask`, there is a check made every forward pass. I'd strongly advocate for a simple warning when the `attention_mask` is not passed 🤗 \r\n\r\nAs a side note, we have related problems at other points in the code base. Getting into the habit of passing the `attention_mask` would really make everyone happier!", "@gante Just to confirm before updating the PR, we are going to remove `warn_if_pad_token_in_input_ids_no_attention_mask` method and use `logger.warning_once` in `forward()`:\r\n```python\r\ndef forward(...):\r\n ...\r\n if not attention_mask:\r\n logger.warning_once(\r\n \"\\nWe strongly recommend passing an `attention_mask` to avoid possibly incorrectly computing the\"\r\n \" attention weights. \"\r\n )\r\n ...\r\n```", "@anruijian correct :) I would add a short example in the warning, such as `(e.g. to correctly mask the pad tokens)`, but I'll leave that up to you!", "@gante \r\n```python\r\ndef forward(...):\r\n ...\r\n if not attention_mask:\r\n logger.warning_once(\r\n \"\\nWe strongly recommend passing an `attention_mask` to avoid possibly incorrectly computing the\"\r\n \" attention weights. Example to correctly mask the pad tokens: model(input_ids, attention_mask=attention_mask).\"\r\n \" See https://huggingface.co/docs/transformers/v4.23.1/en/troubleshooting#incorrect-output-when-padding-tokens-arent-masked for more details.\"\r\n )\r\n ...\r\n```\r\nDoes this example look good to you? I also link the official doc on the issue. Not sure if it's too long. Let me know what you think about this. Thanks!\r\n", "@anruijian sounds good to me! (A minor nit: the link is for v4.23 of the docs, should be `https://huggingface.co/docs/transformers/troubleshooting#incorrect-output-when-padding-tokens-arent-masked` instead)", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md) are likely to be ignored." ]

1,677

1,683

CONTRIBUTOR

null

# What does this PR do?   Fixes #16136 ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [ ] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [ ] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [ ] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [ ] Did you write any new necessary tests? ## Who can review? Anyone in the community is free to review the PR once the tests have passed. Feel free to tag members/contributors who may be interested in your PR.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21916/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21916/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21916", "html_url": "https://github.com/huggingface/transformers/pull/21916", "diff_url": "https://github.com/huggingface/transformers/pull/21916.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21916.patch", "merged_at": null }

https://api.github.com/repos/huggingface/transformers/issues/21915

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21915/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21915/comments

https://api.github.com/repos/huggingface/transformers/issues/21915/events

https://github.com/huggingface/transformers/issues/21915

1,607,847,050

I_kwDOCUB6oc5f1cyK

21,915

Mask2Former ImageProcessor produces different results on Mac vs Windows.

{ "login": "nickponline", "id": 590151, "node_id": "MDQ6VXNlcjU5MDE1MQ==", "avatar_url": "https://avatars.githubusercontent.com/u/590151?v=4", "gravatar_id": "", "url": "https://api.github.com/users/nickponline", "html_url": "https://github.com/nickponline", "followers_url": "https://api.github.com/users/nickponline/followers", "following_url": "https://api.github.com/users/nickponline/following{/other_user}", "gists_url": "https://api.github.com/users/nickponline/gists{/gist_id}", "starred_url": "https://api.github.com/users/nickponline/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/nickponline/subscriptions", "organizations_url": "https://api.github.com/users/nickponline/orgs", "repos_url": "https://api.github.com/users/nickponline/repos", "events_url": "https://api.github.com/users/nickponline/events{/privacy}", "received_events_url": "https://api.github.com/users/nickponline/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "Here is the image I used.\r\n![0023](https://user-images.githubusercontent.com/590151/222617740-0088ded3-cd49-46df-aa23-0c2a30605729.jpg)\r\n", "Also cc @alaradirik ", "Thanks for raising this issue @nickponline and for all the details! \r\n\r\nCould you give details on how you're reading in the image e.g. through torchvision and the format the image is saved in? If I download the image in the comment above I get different results than in the snippet.\r\n\r\n```\r\nimport torchvision\r\n\r\n# Load in downloaded image\r\nimage = torchvision.io.read_image('222617740-0088ded3-cd49-46df-aa23-0c2a30605729.jpg')\r\nimage = image.numpy()\r\nprint(image.dtype, image.shape, image.sum()) # uint8 (3, 1000, 1000) 443861838\r\n```", "@amyeroberts @sgugger \r\n\r\nI'm reading the image with PIL\r\n\r\n```\r\nfrom PIL import Image\r\nimage = Image.open(filename)\r\nimage = image.convert('RGB')\r\nimage = np.array(image)\r\nimage = image.astype(np.float32)\r\nimage = image.transpose(2,0,1)\r\n```\r\n\r\nAt that point I have confirmed the the `image` is identical on both Windows and Mac. Also after inference further in the code the Mac result is the worse than the windows result if that help. But it's the image processor that is generating a different result for identical inputs. ", "@amyeroberts @sgugger the means and stds of the input image are different on Windows and Mac after `ImageProcessor` forward call:\r\n\r\nWindows\r\n```\r\nmean = [-0.4228946 -0.17078026 0.25235963]\r\nstd = [0.81622934 0.699496 0.71027416]\r\n```\r\n\r\nMac\r\n```\r\nmean = [-1.229962 -1.1720737 -0.6407509]\r\nstd = [1.5912648 1.5453817 1.7506045]\r\n```", "@amyeroberts @sgugger I updated the repro snippet above to make it easier to confirm.", "@nickponline - thank you very much for extra details! I'll dig into this and try to figure out what's happening 🕵️‍♀️ ", "@amyeroberts @sgugger I feel the issue is here:\r\n\r\nhttps://github.com/huggingface/transformers/blob/main/src/transformers/image_transforms.py#L159\r\n\r\nThe image is already in the range `[0..255]` and after the rescale and then `image.astype(np.uint8)` the arrays are different on Windows and Mac. ", "Calling in backup here: https://stackoverflow.com/questions/75632469/why-does-np-astypeuint8-give-different-results-on-windows-versus-mac 😀", "Confirming that this works with `Python 3.10.6+ (Mac) Numpy 1.24.2+`. ShruggingFace 🤷‍♂️. It must be a bug or change of behavior in Numpy or Python. Can close. ", "@nickponline Thanks for the updates and all the work digging into this! \r\n\r\nLooking at the line you highlighted and conversation on stackoverflow, it seems there's two things happening, resulting in this issue: \r\n* Rescaling the pixel values by multiplying by 255 if the input image is of type `float32`. Resulting in pixel values between 0 and 65,025. Then casting to `uint8` [here](https://github.com/huggingface/transformers/blob/fcf813417aa34f3a0ea7d283f7d4f6b0834cf098/src/transformers/image_transforms.py#L162)\r\n* Different overflow behaviour in numpy - as highlighted in [the stackoverflow comment](https://stackoverflow.com/a/75632979)\r\n\r\nIn this case, updating numpy will give consistent results between the OS's, however the resulting pixel_values from the image processor may not be sensible or produce good predictions from the model, depending on how the values are cast when overflow occurs. \r\n\r\nThe first issue is tricky to handle - the logic is partly there for backwards compatibility as resizing was handled by the PIL library and, when converting to PIL images, whether to rescale the pixel values was inferred by the type. The assumption is that raw pixel values are of an int type and between 0-255; unnormalized float type pixel values have values between 0-1. \r\n\r\nI think there's two possible things we can do to address these issues in the future: \r\n* Add an additional check on pixel values before rescaling\r\n* Raise a warning when casting to uint8 if overflow is going to occur \r\nI'll open a PR for these. \r\n\r\nAs a side note, you don't need to convert your images to float before feeding into the image processor. You can pass in the PIL images directly. \r\n\r\np.s. thanks for coining 'Shrugging Face' - I shall be using it in the future! \r\n" ]

1,677

1,678

NONE

null

### System Info >>> transformers.__version__ '4.27.0.dev0' >>> Python 3.10.6 Windows vs Mac ### Who can help? @amyeroberts ### Information - [ ] The official example scripts - [ ] My own modified scripts ### Tasks - [ ] An officially supported task in the `examples` folder (such as GLUE/SQuAD, ...) - [X] My own task or dataset (give details below) ### Reproduction ``` import torch from transformers import AutoImageProcessor, Mask2FormerForUniversalSegmentation processor = AutoImageProcessor.from_pretrained("facebook/mask2former-swin-large-cityscapes-instance", reduce_labels=False, ignore_index=255, do_resize=True, size=dict(width=500, height=500), do_normalize=True, image_mean=[0.485, 0.456, 0.406], image_std=[0.229, 0.224, 0.225]) device = torch.device("cpu") image = Image.open(filename1) image = image.convert('RGB') image = np.array(image) image = image.astype(np.float32) image = image.transpose(2,0,1) print(image.dtype, image.shape, image.mean((1, 2))) # float32 (3, 1000, 1000) [156.41327 149.47672 137.97989] ret = processor([image], return_tensors="pt") pixel_values = ret["pixel_values"].to(device) print(pixel_values.dtype, pixel_values.shape, pixel_values[0].mean((1, 2)), pixel_values[0].std((1, 2))) ``` Windows ``` float32 (3, 1000, 1000) [156.41327 149.47672 137.97989] mean = [-0.4228946 -0.17078026 0.25235963] std = [0.81622934 0.699496 0.71027416] ``` Mac ``` float32 (3, 1000, 1000) [156.41327 149.47672 137.97989] mean = [-1.229962 -1.1720737 -0.6407509] std = [1.5912648 1.5453817 1.7506045] ``` ### Expected behavior Same result on Windows and Mac

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21915/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21915/timeline

completed

null

https://api.github.com/repos/huggingface/transformers/issues/21914

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21914/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21914/comments

https://api.github.com/repos/huggingface/transformers/issues/21914/events

https://github.com/huggingface/transformers/pull/21914

1,607,841,971

PR_kwDOCUB6oc5LLyUK

21,914

feat: filter try/except when looking at custom code

{ "login": "zanussbaum", "id": 33707069, "node_id": "MDQ6VXNlcjMzNzA3MDY5", "avatar_url": "https://avatars.githubusercontent.com/u/33707069?v=4", "gravatar_id": "", "url": "https://api.github.com/users/zanussbaum", "html_url": "https://github.com/zanussbaum", "followers_url": "https://api.github.com/users/zanussbaum/followers", "following_url": "https://api.github.com/users/zanussbaum/following{/other_user}", "gists_url": "https://api.github.com/users/zanussbaum/gists{/gist_id}", "starred_url": "https://api.github.com/users/zanussbaum/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/zanussbaum/subscriptions", "organizations_url": "https://api.github.com/users/zanussbaum/orgs", "repos_url": "https://api.github.com/users/zanussbaum/repos", "events_url": "https://api.github.com/users/zanussbaum/events{/privacy}", "received_events_url": "https://api.github.com/users/zanussbaum/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "Thanks again!", "Happy to help! Thanks for the great packages!\n\nOn Fri, Mar 3, 2023 at 8:44 AM Sylvain Gugger ***@***.***>\nwrote:\n\n> Thanks again!\n>\n> —\n> Reply to this email directly, view it on GitHub\n> <https://github.com/huggingface/transformers/pull/21914#issuecomment-1453553255>,\n> or unsubscribe\n> <https://github.com/notifications/unsubscribe-auth/AIBFIPJCABTGQJDB3PCJKPDW2HYTDANCNFSM6AAAAAAVOB6YRQ>\n> .\n> You are receiving this because you authored the thread.Message ID:\n> ***@***.***>\n>\n" ]

1,677

CONTRIBUTOR

null

# What does this PR do?   Fixes #21912 ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [ ] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [x] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [ ] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [ ] Did you write any new necessary tests? ## Who can review? Anyone in the community is free to review the PR once the tests have passed. Feel free to tag members/contributors who may be interested in your PR. @sgugger

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21914/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21914/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21914", "html_url": "https://github.com/huggingface/transformers/pull/21914", "diff_url": "https://github.com/huggingface/transformers/pull/21914.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21914.patch", "merged_at": 1677851040000 }

https://api.github.com/repos/huggingface/transformers/issues/21913

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21913/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21913/comments

https://api.github.com/repos/huggingface/transformers/issues/21913/events

https://github.com/huggingface/transformers/issues/21913

1,607,789,680

I_kwDOCUB6oc5f1Oxw

21,913

[performance] from_pretrained is still much slower than torch.load and seems to be initializing weights

{ "login": "moyix", "id": 34380, "node_id": "MDQ6VXNlcjM0Mzgw", "avatar_url": "https://avatars.githubusercontent.com/u/34380?v=4", "gravatar_id": "", "url": "https://api.github.com/users/moyix", "html_url": "https://github.com/moyix", "followers_url": "https://api.github.com/users/moyix/followers", "following_url": "https://api.github.com/users/moyix/following{/other_user}", "gists_url": "https://api.github.com/users/moyix/gists{/gist_id}", "starred_url": "https://api.github.com/users/moyix/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/moyix/subscriptions", "organizations_url": "https://api.github.com/users/moyix/orgs", "repos_url": "https://api.github.com/users/moyix/repos", "events_url": "https://api.github.com/users/moyix/events{/privacy}", "received_events_url": "https://api.github.com/users/moyix/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "Thank you for trying to analyse this, @moyix and for wanting to make things faster.\r\n\r\nI dug into it and here is what I have to share with you.\r\n\r\n# What's happening for real\r\n\r\nIt's pretty clear from your profiler report that the diff comes from weights init which as you said get overwritten with weights.\r\n\r\nIndeed this is what's happening here. Except you are mixing 2 things.\r\n\r\nAs you discovered lazy model init was implemented here https://github.com/huggingface/transformers/pull/11471 and it later was improved upon in multiple PRs. This was done only for `_init_weights` functions defined in the modeling code of `transformers`.\r\n\r\nNow you're forgetting about calls like \r\n\r\nhttps://github.com/huggingface/transformers/blob/37e0974afcbccdc85da59d51b44e1437b6b3caea/src/transformers/models/codegen/modeling_codegen.py#L117-L119\r\n\r\nwhich of course by default call their init functions:\r\n\r\n```\r\n File \"/mnt/nvme0/code/huggingface/transformers-master/src/transformers/models/codegen/modeling_codegen.py\", line 117, in __init__\r\n self.qkv_proj = nn.Linear(self.embed_dim, self.embed_dim * 3, bias=False)\r\n File \"/home/stas/anaconda3/envs/py38-pt113/lib/python3.8/site-packages/torch/nn/modules/linear.py\", line 101, in __init__\r\n self.reset_parameters()\r\n File \"/home/stas/anaconda3/envs/py38-pt113/lib/python3.8/site-packages/torch/nn/modules/linear.py\", line 107, in reset_parameters\r\n init.kaiming_uniform_(self.weight, a=math.sqrt(5))\r\n File \"/home/stas/anaconda3/envs/py38-pt113/lib/python3.8/site-packages/torch/nn/init.py\", line 396, in kaiming_uniform_\r\n```\r\n\r\nSo that overhead all comes from pytorch `nn.Module` submodules and not `_init_weights` defined in the modeling code of `transformers`.\r\n\r\nYou're wanting to use a huge 14GB model and it surely adds some 30sec to init it.\r\n\r\nThe problem is that you're comparing loading the weights only with instantiating the model plus loading the weights, so of course they aren't the same thing. But we agree that it's a pointless waste of compute and time to init weights that are going to be overwritten moments later.\r\n\r\nTo test I changed pytorch's `kaiming_uniform_` to be:\r\n```\r\ndef kaiming_uniform_(\r\n tensor: Tensor, a: float = 0, mode: str = 'fan_in', nonlinearity: str = 'leaky_relu'\r\n):\r\n return tensor\r\n```\r\nand the same for `uniform_` and `from_pretrained` was as fast as you wanted it to be.\r\n\r\nhint: perhaps you can use it as a hack until a better solution is provided - simply monkey patch the init functions with a no-op (I hope I covered the ones that are used here).\r\n\r\n```\r\nfrom transformers import CodeGenForCausalLM\r\nimport torch.nn.init\r\ntorch.nn.init.kaiming_uniform_ = lambda x, *args, **kwargs: x\r\ntorch.nn.init.uniform_ = lambda x, *args, **kwargs: x\r\nCodeGenForCausalLM.from_pretrained('Salesforce/codegen-6B-mono')\r\n```\r\nof course, I assume you are either doing inference or you have all weights in the distributed file - so no important init is missed.\r\n\r\nthis I think should give you the speed closer to `torch.load`\r\n\r\n# What can be done\r\n\r\nBut why you'd say can't you skip those inits?\r\n\r\nWe actually are able to do so since pytorch-1.10 where special functionality was added.\r\n- https://pytorch.org/docs/stable/generated/torch.nn.utils.skip_init.html#torch.nn.utils.skip_init\r\n- https://pytorch.org/tutorials/prototype/skip_param_init.html\r\n\r\nLooking at the requirements it actually appears to be possible despite needing to support pytorch<1.10 as well.\r\n\r\nThe modules will have to be adapted to meet 2 requirements:\r\nhttps://pytorch.org/tutorials/prototype/skip_param_init.html#updating-modules-to-support-skipping-initialization\r\nI will repaste them here:\r\n\r\n1. The module must accept a device kwarg in its constructor that is passed to any parameters or buffers created during construction.\r\n2. The module must not perform any computation on parameters or buffers in its constructor except initialization (i.e. functions from torch.nn.init).\r\n\r\nThe first one is certainly possible since doing:\r\n\r\n```\r\n- def __init__(self, foo, bar):\r\n+ def __init__(self, foo, bar, device=None):\r\n```\r\nshould be backward compatible.\r\n\r\nI think the 2nd requirement should be somewhat possible, but I can't speak for the multitude of models we have.\r\n\r\nOnce this is done, the rest of the `from_pretrained` will need to be adapted to use the `device` argument as in the example of the tutorial,\r\n```\r\nm = nn.Linear(10, 5, device='meta')\r\n```\r\nbut of course it will be `m = ModelName(..., device='meta')`\r\n\r\nI think this needs to happen sooner than later as it'd greatly simplify the various juggling we have during the loading process (after updating all the models, e.g. like `low_cpu_mem_usage` functionality). But needing to support torch<1.10 might make this somewhat messy. I'm not sure.\r\n\r\nSo now let me bring here @sgugger and @patrickvonplaten to take over as I'm currently working on a different project, and they can decide on whether the project is ready for this major change or not quite yet and then you can use my hack ;)\r\n\r\np.s. BTW, while studying your report I have invalidated your suggestion that there was a general `from_pretrained` regression, but to do that I had to use a different class since `CodeGenForCausalLM` was added only recently. I went all the way back to `transformers==4.14` and `t5-large` loads with the same speed as the latest version.\r\n\r\n**edit** Additional solutions are added in:\r\n- https://github.com/huggingface/transformers/issues/21913#issuecomment-1453482689\r\n- https://github.com/huggingface/transformers/issues/21913#issuecomment-1453858274", "I'm curious, are you doing inference or finetuning? Because for the latter usually the init overhead is usually irrelevant.\r\n\r\nFast loading is also important for debug.\r\n\r\nI think I'm going to propose to pytorch this new feature:\r\n```\r\nwith torch.inference:\r\n m = MyModel(...)\r\n```\r\nand it would just work and be really fast w/o the overhead of init'ing weights which will be overloaded from pretrained weights.\r\n", "Thanks for the very comprehensive answer! That makes perfect sense :) I am indeed doing inference and trying to get the batch size correct – so having to wait a long time for the model load each attempt (only to get a CUDA out of memory error) was a bit painful.\r\n\r\nThat hack helps a lot for now, thanks!", "Using `low_cpu_mem_usage=True` will initialize the model on the meta device (requires Accelerate as an extra dep) and should speed up the initialization as a result. This will become the default mid-term but we need some more preparation work by making the tests more robust for `from_pretrained` to make sure we absolutely don't break anything.", "Some additional solutions coming from pytorch-slack where I asked [this question](https://pytorch.slack.com/archives/C3PDTEV8E/p1677813090248699):\r\n\r\n1. install pytorch-nightly from instructions at https://pytorch.org/get-started/locally/ (or if you read this later when pytorch==2.0 is released any 2.0 and higher version will do).\r\n\r\nnow you can do:\r\n\r\n```\r\n with torch.device(\"cuda\"):\r\n model = CodeGenForCausalLM.from_pretrained('Salesforce/codegen-6B-mono')\r\n```\r\n\r\nso it instantiates the model directly on your gpu and all the inits are run much faster. This solution is just a bit slower than cancelling out the init functions. plus your model will already be on gpu, so no copying overhead from cpu.\r\n\r\nInstead of using the context manager you can just set the default device like so:\r\n\r\n```\r\ntorch.set_default_device('cuda')\r\n```\r\nand you no longer need to indent your existing code.\r\n\r\n1b. Using materialization on the `meta` device will be really fast as it will cancel out the init functions and won't even waste time on allocating memory for the weights:\r\n```\r\n with torch.device(\"meta\"):\r\n model = CodeGenForCausalLM.from_pretrained('Salesforce/codegen-6B-mono')\r\n```\r\nbut the resulting model isn't usable right away and requires additional manipulations to materialize it on the target device with the preloaded weights. This most likely have to be done by `transformers` unless pytorch comes up with a magical method a user could do themselves.\r\n\r\ncredits: @alband and @stephenroller\r\n\r\n2. Another solution comes from https://pytorch.org/torchdistx/latest/deferred_init.html, but it requires tweaking `from_pretrained` to support `from torchdistx.deferred_init import deferred_init, materialize_module` and this experimental package isn't easy to install since it requires CUDA extensions building (though not for this functionality), so we can't make `transformers` depend on it. It will have to be upstreamed into pytorch first.\r\n\r\ncredits: @cbalioglu", "In extension of @stas00 's number one, one might enhance the context manager solution with a diversion of the `init` functions. I wrote up [a bit more detail on my blog](http://lernapparat.de/faster-model-init).\r\n", "@stas00 your solution is great, tested it a bit. Is there any timeline for this feature and could one help with integration? Would be interested to know what are the team's thoughts on integrating this feature within the `Trainer` but also `pipelines`? Happy to help if I can!", "For the timeline questions we need to ask @sgugger ", "The `low_cpu_mem_usage=True` option is already there in Transformers and usable today. Changing the default will take more time to ensure backward compatibility.", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md) are likely to be ignored.", "I know this issue is closed but just some relevant feedback: I'm also facing extremely slow performance with the `from_pretrained` method, this time in a conda environment. I tried the `low_cpu_mem_usage=True` solution, but this requires a more recent version of `transformers` than is available in the conda repos so I can't. Reported already on [Stack Overflow](https://stackoverflow.com/questions/76059654/getting-a-error-when-running-gptneoxforcausallm-from-transformers-library-namee).\r\n\r\nTLDR: for a chunk of users (anyone who has to use a conda environment) the `low_cpu_mem_usage=True` parameter is not available or usable.", "Hey @tomwagstaff-opml, thanks for reporting.\r\n\r\nI believe you're using the `transformers` version from the main channel of anaconda, but we don't (and none of the open-source project maintainers do) maintain this version. This is maintained by the anaconda team.\r\n\r\nIn our README we indicate that you should use the [huggingface channel in order to install the package](https://github.com/huggingface/transformers#with-conda).\r\n\r\nPlease install it as such:\r\n\r\n```\r\nconda install -c huggingface transformers\r\n```\r\n\r\nor, alternatively, use the conda-forge channel which is also the latest version:\r\n\r\n```\r\nconda install -c conda-forge transformers\r\n```", "Thanks for your help @LysandreJik - installing `transformers` from the Hugging Face channel has worked and allowed me to try out the `low_cpu_mem_usage` parameter" ]

1,677

1,683

1,681

NONE

null

### System Info - `transformers` version: 4.26.1 - Platform: Linux-5.15.0-52-generic-x86_64-with-glibc2.31 - Python version: 3.10.9 - Huggingface_hub version: 0.12.1 - PyTorch version (GPU?): 2.0.0.dev20230224+cu118 (True) - Tensorflow version (GPU?): not installed (NA) - Flax version (CPU?/GPU?/TPU?): not installed (NA) - Jax version: not installed - JaxLib version: not installed - Using GPU in script?: No - Using distributed or parallel set-up in script?: No ### Who can help? @stas00, @patrickvonplaten ### Information - [ ] The official example scripts - [ ] My own modified scripts ### Tasks - [ ] An officially supported task in the `examples` folder (such as GLUE/SQuAD, ...) - [ ] My own task or dataset (give details below) ### Reproduction Loading a model with `from_pretrained` takes much longer than the underlying torch.load. For example, for the `Salesforce/codegen-6B-mono` model, `CodeGenForCausalLM.from_pretrained('Salesforce/codegen-6B-mono')` takes ~38 seconds, whereas `torch.load()` on its `pytorch_model.bin` takes just ~5.4 seconds. This is very similar to #9205, but is happening with the latest transformers from pip (4.26.1), so possibly a regression? Short repro: ```python import time import torch from transformers import CodeGenForCausalLM t1 = time.time() CodeGenForCausalLM.from_pretrained('Salesforce/codegen-6B-mono') t2 = time.time() print("Load took", t2-t1, "seconds") ``` Prints **Load took 37.78910255432129 seconds** ```python import time import torch from transformers.utils import cached_file torch.load(cached_file('Salesforce/codegen-6B-mono', 'pytorch_model.bin')) ``` Prints **Load took 5.443041801452637 seconds** Based on profiling the HF from_pretrained script, it seems like ~75% of the time is being spent doing random initialization of weights that are about to be overwritten. This is the same problem that was fixed in PR #11471 so I'm not sure what's going on here. Here's the cProfile output and output from gprof2dot: [loadmodel_profile.txt](https://github.com/huggingface/transformers/files/10877225/loadmodel_profile.txt) [hf_loadmodel_new.pdf](https://github.com/huggingface/transformers/files/10877227/hf_loadmodel_new.pdf) ### Expected behavior `from_pretrained` should skip weight initialization when loading a pretrained model.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21913/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21913/timeline

completed

null

https://api.github.com/repos/huggingface/transformers/issues/21912

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21912/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21912/comments

https://api.github.com/repos/huggingface/transformers/issues/21912/events

https://github.com/huggingface/transformers/issues/21912

1,607,672,293

I_kwDOCUB6oc5f0yHl

21,912

Allow for try/except imports for custom code

{ "login": "zanussbaum", "id": 33707069, "node_id": "MDQ6VXNlcjMzNzA3MDY5", "avatar_url": "https://avatars.githubusercontent.com/u/33707069?v=4", "gravatar_id": "", "url": "https://api.github.com/users/zanussbaum", "html_url": "https://github.com/zanussbaum", "followers_url": "https://api.github.com/users/zanussbaum/followers", "following_url": "https://api.github.com/users/zanussbaum/following{/other_user}", "gists_url": "https://api.github.com/users/zanussbaum/gists{/gist_id}", "starred_url": "https://api.github.com/users/zanussbaum/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/zanussbaum/subscriptions", "organizations_url": "https://api.github.com/users/zanussbaum/orgs", "repos_url": "https://api.github.com/users/zanussbaum/repos", "events_url": "https://api.github.com/users/zanussbaum/events{/privacy}", "received_events_url": "https://api.github.com/users/zanussbaum/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "Indeed that sounds like a nice feature to have. I think the easiest way to deal with it would be to remove all try/except blocks from the content [here](https://github.com/huggingface/transformers/blob/37e0974afcbccdc85da59d51b44e1437b6b3caea/src/transformers/dynamic_module_utils.py#L117) before the tests of the imports. If you want to take a stab at it, happy to review a PR!", "Yes I'll give it a go and send a PR out in a few!" ]

1,677

CONTRIBUTOR

null

### Feature request When uploading a model with custom code, I wanted to try and use Flash Attention in one of the modules. However, to get around a case where people might use it on CPU, I added `try/except` block for the import. However, I get an error when downloading locally like `This modeling file requires the following packages that were not found in your environment` which seems to come from [this file](https://github.com/huggingface/transformers/blob/main/src/transformers/dynamic_module_utils.py#L112). ### Motivation I want to be able to write custom model code that allows for optional imports if possible (like FlashAttention) ### Your contribution I could try a crack at a PR?

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21912/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21912/timeline

completed

null

https://api.github.com/repos/huggingface/transformers/issues/21911

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21911/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21911/comments

https://api.github.com/repos/huggingface/transformers/issues/21911/events

https://github.com/huggingface/transformers/pull/21911

1,607,395,048

PR_kwDOCUB6oc5LKP9k

21,911

Avoid modeling tests run in pipeline CI jobs

{ "login": "ydshieh", "id": 2521628, "node_id": "MDQ6VXNlcjI1MjE2Mjg=", "avatar_url": "https://avatars.githubusercontent.com/u/2521628?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ydshieh", "html_url": "https://github.com/ydshieh", "followers_url": "https://api.github.com/users/ydshieh/followers", "following_url": "https://api.github.com/users/ydshieh/following{/other_user}", "gists_url": "https://api.github.com/users/ydshieh/gists{/gist_id}", "starred_url": "https://api.github.com/users/ydshieh/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ydshieh/subscriptions", "organizations_url": "https://api.github.com/users/ydshieh/orgs", "repos_url": "https://api.github.com/users/ydshieh/repos", "events_url": "https://api.github.com/users/ydshieh/events{/privacy}", "received_events_url": "https://api.github.com/users/ydshieh/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/transformers/pr_21911). All of your documentation changes will be reflected on that endpoint." ]

1,677

COLLABORATOR

null

# What does this PR do? Avoid modeling tests run in pipeline CI jobs. PR #21887 applied ``` @is_pipeline_test class PipelineTesterMixin: ``` Together the changes in #21516 ``` class BertModelTest(ModelTesterMixin, PipelineTesterMixin, unittest.TestCase): ``` these 2 make all modeling test methods become pipeline tests, therefore CircleCI pipeline jobs run all pipeline tests + all model tests. This causes OOM as pipeline jobs are run with `-n 8`. This PR applies `@is_pipeline_test` to each test methods instead of the test class to avoid such issue. ### Effect Run ```python python -m pytest -m is_pipeline_test -v tests/models/bridgetower/test_modeling_bridgetower.py::BridgeTowerModelTest::test_model ``` #### On main test pass #### on this PR test deselected

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21911/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21911/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21911", "html_url": "https://github.com/huggingface/transformers/pull/21911", "diff_url": "https://github.com/huggingface/transformers/pull/21911.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21911.patch", "merged_at": 1677788586000 }

https://api.github.com/repos/huggingface/transformers/issues/21910

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21910/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21910/comments

https://api.github.com/repos/huggingface/transformers/issues/21910/events

https://github.com/huggingface/transformers/pull/21910

1,607,390,271

PR_kwDOCUB6oc5LKO8X

21,910

Fix doctests for TFVisionTextDualEncoder

{ "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.github.com/users/Rocketknight1/followers", "following_url": "https://api.github.com/users/Rocketknight1/following{/other_user}", "gists_url": "https://api.github.com/users/Rocketknight1/gists{/gist_id}", "starred_url": "https://api.github.com/users/Rocketknight1/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Rocketknight1/subscriptions", "organizations_url": "https://api.github.com/users/Rocketknight1/orgs", "repos_url": "https://api.github.com/users/Rocketknight1/repos", "events_url": "https://api.github.com/users/Rocketknight1/events{/privacy}", "received_events_url": "https://api.github.com/users/Rocketknight1/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/transformers/pr_21910). All of your documentation changes will be reflected on that endpoint." ]

1,677

MEMBER

null

So I might have just copy-pasted all the PyTorch doctests into the TF class and made the CI angry. But it's fixed now!

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21910/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21910/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21910", "html_url": "https://github.com/huggingface/transformers/pull/21910", "diff_url": "https://github.com/huggingface/transformers/pull/21910.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21910.patch", "merged_at": 1677802691000 }

https://api.github.com/repos/huggingface/transformers/issues/21909

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21909/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21909/comments

https://api.github.com/repos/huggingface/transformers/issues/21909/events

https://github.com/huggingface/transformers/pull/21909

1,607,375,601

PR_kwDOCUB6oc5LKLzY

21,909

Cleanup more auto mapping names

{ "login": "ydshieh", "id": 2521628, "node_id": "MDQ6VXNlcjI1MjE2Mjg=", "avatar_url": "https://avatars.githubusercontent.com/u/2521628?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ydshieh", "html_url": "https://github.com/ydshieh", "followers_url": "https://api.github.com/users/ydshieh/followers", "following_url": "https://api.github.com/users/ydshieh/following{/other_user}", "gists_url": "https://api.github.com/users/ydshieh/gists{/gist_id}", "starred_url": "https://api.github.com/users/ydshieh/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ydshieh/subscriptions", "organizations_url": "https://api.github.com/users/ydshieh/orgs", "repos_url": "https://api.github.com/users/ydshieh/repos", "events_url": "https://api.github.com/users/ydshieh/events{/privacy}", "received_events_url": "https://api.github.com/users/ydshieh/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,677

COLLABORATOR

null

# What does this PR do? Just a follow-up PR of #21903. Rebase on main once #21911 (or a fix from your side) being merged will be OK.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21909/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21909/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21909", "html_url": "https://github.com/huggingface/transformers/pull/21909", "diff_url": "https://github.com/huggingface/transformers/pull/21909.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21909.patch", "merged_at": 1677851025000 }

https://api.github.com/repos/huggingface/transformers/issues/21908

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21908/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21908/comments

https://api.github.com/repos/huggingface/transformers/issues/21908/events

https://github.com/huggingface/transformers/pull/21908

1,607,260,282

PR_kwDOCUB6oc5LJzAv

21,908

Temporarily skip 3 tests in `BridgeTowerModelTest`

{ "login": "ydshieh", "id": 2521628, "node_id": "MDQ6VXNlcjI1MjE2Mjg=", "avatar_url": "https://avatars.githubusercontent.com/u/2521628?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ydshieh", "html_url": "https://github.com/ydshieh", "followers_url": "https://api.github.com/users/ydshieh/followers", "following_url": "https://api.github.com/users/ydshieh/following{/other_user}", "gists_url": "https://api.github.com/users/ydshieh/gists{/gist_id}", "starred_url": "https://api.github.com/users/ydshieh/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ydshieh/subscriptions", "organizations_url": "https://api.github.com/users/ydshieh/orgs", "repos_url": "https://api.github.com/users/ydshieh/repos", "events_url": "https://api.github.com/users/ydshieh/events{/privacy}", "received_events_url": "https://api.github.com/users/ydshieh/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,677

COLLABORATOR

null

# What does this PR do? skip for now

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21908/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21908/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21908", "html_url": "https://github.com/huggingface/transformers/pull/21908", "diff_url": "https://github.com/huggingface/transformers/pull/21908.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21908.patch", "merged_at": 1677780964000 }

https://api.github.com/repos/huggingface/transformers/issues/21907

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21907/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21907/comments

https://api.github.com/repos/huggingface/transformers/issues/21907/events

https://github.com/huggingface/transformers/pull/21907

1,607,204,738

PR_kwDOCUB6oc5LJnE-

21,907

Update modeling_funnel.py

{ "login": "robinbg", "id": 28918243, "node_id": "MDQ6VXNlcjI4OTE4MjQz", "avatar_url": "https://avatars.githubusercontent.com/u/28918243?v=4", "gravatar_id": "", "url": "https://api.github.com/users/robinbg", "html_url": "https://github.com/robinbg", "followers_url": "https://api.github.com/users/robinbg/followers", "following_url": "https://api.github.com/users/robinbg/following{/other_user}", "gists_url": "https://api.github.com/users/robinbg/gists{/gist_id}", "starred_url": "https://api.github.com/users/robinbg/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/robinbg/subscriptions", "organizations_url": "https://api.github.com/users/robinbg/orgs", "repos_url": "https://api.github.com/users/robinbg/repos", "events_url": "https://api.github.com/users/robinbg/events{/privacy}", "received_events_url": "https://api.github.com/users/robinbg/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,677

NONE

null

Just for check, don't merge # What does this PR do?   Fixes # (issue) ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [ ] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [ ] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [ ] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [ ] Did you write any new necessary tests? ## Who can review? Anyone in the community is free to review the PR once the tests have passed. Feel free to tag members/contributors who may be interested in your PR.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21907/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21907/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21907", "html_url": "https://github.com/huggingface/transformers/pull/21907", "diff_url": "https://github.com/huggingface/transformers/pull/21907.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21907.patch", "merged_at": null }

https://api.github.com/repos/huggingface/transformers/issues/21906

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21906/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21906/comments

https://api.github.com/repos/huggingface/transformers/issues/21906/events

https://github.com/huggingface/transformers/pull/21906

1,607,196,243

PR_kwDOCUB6oc5LJlQW

21,906

faster forward following what is done for images

{ "login": "ArthurZucker", "id": 48595927, "node_id": "MDQ6VXNlcjQ4NTk1OTI3", "avatar_url": "https://avatars.githubusercontent.com/u/48595927?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ArthurZucker", "html_url": "https://github.com/ArthurZucker", "followers_url": "https://api.github.com/users/ArthurZucker/followers", "following_url": "https://api.github.com/users/ArthurZucker/following{/other_user}", "gists_url": "https://api.github.com/users/ArthurZucker/gists{/gist_id}", "starred_url": "https://api.github.com/users/ArthurZucker/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ArthurZucker/subscriptions", "organizations_url": "https://api.github.com/users/ArthurZucker/orgs", "repos_url": "https://api.github.com/users/ArthurZucker/repos", "events_url": "https://api.github.com/users/ArthurZucker/events{/privacy}", "received_events_url": "https://api.github.com/users/ArthurZucker/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "Failure seems unrelated" ]

1,677

COLLABORATOR

null

# What does this PR do? Follow up of #21897 Keeps batch_size active, but restricts it to the imagepart. Calculates candidate_labels features only once (there's no way with this current pipeline to send different candidate_labels so we can take the optimization. Ultra narrowed for CLIP but the pipeline has existed for 4 months now with no new models. Given the popularity of CLIP for diffusion models, I think this is ok to overspecify. We can always relax later when new models come in. This allows to downgrade from ChunkPipeline to regular PIpeline

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21906/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21906/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21906", "html_url": "https://github.com/huggingface/transformers/pull/21906", "diff_url": "https://github.com/huggingface/transformers/pull/21906.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21906.patch", "merged_at": 1677820699000 }

https://api.github.com/repos/huggingface/transformers/issues/21905

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21905/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21905/comments

https://api.github.com/repos/huggingface/transformers/issues/21905/events

https://github.com/huggingface/transformers/pull/21905

1,607,131,884

PR_kwDOCUB6oc5LJXOi

21,905

Adds "causal-lm-with-past" to codegen

{ "login": "corey-nm", "id": 109536191, "node_id": "U_kgDOBodjvw", "avatar_url": "https://avatars.githubusercontent.com/u/109536191?v=4", "gravatar_id": "", "url": "https://api.github.com/users/corey-nm", "html_url": "https://github.com/corey-nm", "followers_url": "https://api.github.com/users/corey-nm/followers", "following_url": "https://api.github.com/users/corey-nm/following{/other_user}", "gists_url": "https://api.github.com/users/corey-nm/gists{/gist_id}", "starred_url": "https://api.github.com/users/corey-nm/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/corey-nm/subscriptions", "organizations_url": "https://api.github.com/users/corey-nm/orgs", "repos_url": "https://api.github.com/users/corey-nm/repos", "events_url": "https://api.github.com/users/corey-nm/events{/privacy}", "received_events_url": "https://api.github.com/users/corey-nm/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "Thanks for your PR, but ONNX support is now in the optimum library and we don't accept new PRs in Transformers.", "Okay, thanks, closing then!" ]

1,677

NONE

null

# What does this PR do?   Fixes # (issue) ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [x] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [ ] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [ ] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [ ] Did you write any new necessary tests? ## Who can review? Anyone in the community is free to review the PR once the tests have passed. Feel free to tag members/contributors who may be interested in your PR.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21905/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21905/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21905", "html_url": "https://github.com/huggingface/transformers/pull/21905", "diff_url": "https://github.com/huggingface/transformers/pull/21905.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21905.patch", "merged_at": null }

https://api.github.com/repos/huggingface/transformers/issues/21904

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21904/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21904/comments

https://api.github.com/repos/huggingface/transformers/issues/21904/events

https://github.com/huggingface/transformers/pull/21904

1,607,049,289

PR_kwDOCUB6oc5LJFeQ

21,904

Add Blip and Blip2 for pipeline tests

{ "login": "ydshieh", "id": 2521628, "node_id": "MDQ6VXNlcjI1MjE2Mjg=", "avatar_url": "https://avatars.githubusercontent.com/u/2521628?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ydshieh", "html_url": "https://github.com/ydshieh", "followers_url": "https://api.github.com/users/ydshieh/followers", "following_url": "https://api.github.com/users/ydshieh/following{/other_user}", "gists_url": "https://api.github.com/users/ydshieh/gists{/gist_id}", "starred_url": "https://api.github.com/users/ydshieh/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ydshieh/subscriptions", "organizations_url": "https://api.github.com/users/ydshieh/orgs", "repos_url": "https://api.github.com/users/ydshieh/repos", "events_url": "https://api.github.com/users/ydshieh/events{/privacy}", "received_events_url": "https://api.github.com/users/ydshieh/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "Thank you ! !!!" ]

1,677

COLLABORATOR

null

# What does this PR do? A continuation of (not merged) #21802. @NielsRogge I will add you as the contributor. @Narsil Just in case you want to take a look too, as you reviewed #21904, and asked for adding the tests.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21904/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21904/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21904", "html_url": "https://github.com/huggingface/transformers/pull/21904", "diff_url": "https://github.com/huggingface/transformers/pull/21904.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21904.patch", "merged_at": 1677777635000 }

https://api.github.com/repos/huggingface/transformers/issues/21903

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21903/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21903/comments

https://api.github.com/repos/huggingface/transformers/issues/21903/events

https://github.com/huggingface/transformers/pull/21903

1,606,812,003

PR_kwDOCUB6oc5LISNI

21,903

Clean up auto mapping names

{ "login": "ydshieh", "id": 2521628, "node_id": "MDQ6VXNlcjI1MjE2Mjg=", "avatar_url": "https://avatars.githubusercontent.com/u/2521628?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ydshieh", "html_url": "https://github.com/ydshieh", "followers_url": "https://api.github.com/users/ydshieh/followers", "following_url": "https://api.github.com/users/ydshieh/following{/other_user}", "gists_url": "https://api.github.com/users/ydshieh/gists{/gist_id}", "starred_url": "https://api.github.com/users/ydshieh/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ydshieh/subscriptions", "organizations_url": "https://api.github.com/users/ydshieh/orgs", "repos_url": "https://api.github.com/users/ydshieh/repos", "events_url": "https://api.github.com/users/ydshieh/events{/privacy}", "received_events_url": "https://api.github.com/users/ydshieh/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "so far on `main`\r\n```bash\r\n`layoutxlm` appears in the mapping `PROCESSOR_MAPPING_NAMES` but it is not defined in the keys of `CONFIG_MAPPING_NAMES`.\r\n`wav2vec2_with_lm` appears in the mapping `PROCESSOR_MAPPING_NAMES` but it is not defined in the keys of `CONFIG_MAPPING_NAMES`.\r\n`blip_2` appears in the mapping `MODEL_MAPPING_NAMES` but it is not defined in the keys of `CONFIG_MAPPING_NAMES`.\r\n`decision_transformer_gpt2` appears in the mapping `MODEL_MAPPING_NAMES` but it is not defined in the keys of `CONFIG_MAPPING_NAMES`.\r\n`nllb` appears in the mapping `MODEL_MAPPING_NAMES` but it is not defined in the keys of `CONFIG_MAPPING_NAMES`.\r\n\r\n```", "_The documentation is not available anymore as the PR was closed or merged._" ]

1,677

COLLABORATOR

null

# What does this PR do? Clean up auto mapping names + add a check 🐕‍🦺

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21903/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21903/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21903", "html_url": "https://github.com/huggingface/transformers/pull/21903", "diff_url": "https://github.com/huggingface/transformers/pull/21903.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21903.patch", "merged_at": 1677773691000 }

https://api.github.com/repos/huggingface/transformers/issues/21902

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21902/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21902/comments

https://api.github.com/repos/huggingface/transformers/issues/21902/events

https://github.com/huggingface/transformers/issues/21902

1,606,792,420

I_kwDOCUB6oc5fxbTk

21,902

how to beam search

{ "login": "Mryangkaitong", "id": 23132307, "node_id": "MDQ6VXNlcjIzMTMyMzA3", "avatar_url": "https://avatars.githubusercontent.com/u/23132307?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Mryangkaitong", "html_url": "https://github.com/Mryangkaitong", "followers_url": "https://api.github.com/users/Mryangkaitong/followers", "following_url": "https://api.github.com/users/Mryangkaitong/following{/other_user}", "gists_url": "https://api.github.com/users/Mryangkaitong/gists{/gist_id}", "starred_url": "https://api.github.com/users/Mryangkaitong/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Mryangkaitong/subscriptions", "organizations_url": "https://api.github.com/users/Mryangkaitong/orgs", "repos_url": "https://api.github.com/users/Mryangkaitong/repos", "events_url": "https://api.github.com/users/Mryangkaitong/events{/privacy}", "received_events_url": "https://api.github.com/users/Mryangkaitong/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "Please use the [forums](https://discuss.huggingface.co/) to ask such questions.", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md) are likely to be ignored." ]

1,677

1,681

NONE

null

The current interface is similar to `output=model.generate(**inputs, num_beams=4, no_repeat_ngram_size=7)` but decoding one by one is too slow. Does beam seach support batch-by-batch decoding? How to write if support。

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21902/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21902/timeline

completed

null

https://api.github.com/repos/huggingface/transformers/issues/21901

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21901/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21901/comments

https://api.github.com/repos/huggingface/transformers/issues/21901/events

https://github.com/huggingface/transformers/pull/21901

1,606,788,221

PR_kwDOCUB6oc5LINCH

21,901

[WIP] Creating automated test with release candidate of safetensors.

{ "login": "Narsil", "id": 204321, "node_id": "MDQ6VXNlcjIwNDMyMQ==", "avatar_url": "https://avatars.githubusercontent.com/u/204321?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Narsil", "html_url": "https://github.com/Narsil", "followers_url": "https://api.github.com/users/Narsil/followers", "following_url": "https://api.github.com/users/Narsil/following{/other_user}", "gists_url": "https://api.github.com/users/Narsil/gists{/gist_id}", "starred_url": "https://api.github.com/users/Narsil/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Narsil/subscriptions", "organizations_url": "https://api.github.com/users/Narsil/orgs", "repos_url": "https://api.github.com/users/Narsil/repos", "events_url": "https://api.github.com/users/Narsil/events{/privacy}", "received_events_url": "https://api.github.com/users/Narsil/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/transformers/pr_21901). All of your documentation changes will be reflected on that endpoint.", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md) are likely to be ignored." ]

1,677

1,681

CONTRIBUTOR

null

# What does this PR do?   Fixes # (issue) ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [ ] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [ ] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [ ] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [ ] Did you write any new necessary tests? ## Who can review? Anyone in the community is free to review the PR once the tests have passed. Feel free to tag members/contributors who may be interested in your PR.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21901/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21901/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21901", "html_url": "https://github.com/huggingface/transformers/pull/21901", "diff_url": "https://github.com/huggingface/transformers/pull/21901.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21901.patch", "merged_at": null }

https://api.github.com/repos/huggingface/transformers/issues/21900

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21900/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21900/comments

https://api.github.com/repos/huggingface/transformers/issues/21900/events

https://github.com/huggingface/transformers/pull/21900

1,606,753,450

PR_kwDOCUB6oc5LIFZL

21,900

Make TFForceTokensLogitsProcessor exportable as tf concrete function.

{ "login": "mfuntowicz", "id": 2241520, "node_id": "MDQ6VXNlcjIyNDE1MjA=", "avatar_url": "https://avatars.githubusercontent.com/u/2241520?v=4", "gravatar_id": "", "url": "https://api.github.com/users/mfuntowicz", "html_url": "https://github.com/mfuntowicz", "followers_url": "https://api.github.com/users/mfuntowicz/followers", "following_url": "https://api.github.com/users/mfuntowicz/following{/other_user}", "gists_url": "https://api.github.com/users/mfuntowicz/gists{/gist_id}", "starred_url": "https://api.github.com/users/mfuntowicz/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/mfuntowicz/subscriptions", "organizations_url": "https://api.github.com/users/mfuntowicz/orgs", "repos_url": "https://api.github.com/users/mfuntowicz/repos", "events_url": "https://api.github.com/users/mfuntowicz/events{/privacy}", "received_events_url": "https://api.github.com/users/mfuntowicz/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/transformers/pr_21900). All of your documentation changes will be reflected on that endpoint.", "Also, check CI haha", "Will handle these 👍🏻 ", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md) are likely to be ignored." ]

1,677

1,681

MEMBER

null

Previously we were doing an inner conversion from `[[int]]` to `dict[int, int]` but `dict` is not something concrete functions like. This PR provides a fully compatible concrete function export, removing the inner transformation to `dict` tu use `tf.Tensor` only.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21900/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21900/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21900", "html_url": "https://github.com/huggingface/transformers/pull/21900", "diff_url": "https://github.com/huggingface/transformers/pull/21900.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21900.patch", "merged_at": null }

https://api.github.com/repos/huggingface/transformers/issues/21899

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21899/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21899/comments

https://api.github.com/repos/huggingface/transformers/issues/21899/events

https://github.com/huggingface/transformers/issues/21899

1,606,637,031

I_kwDOCUB6oc5fw1Xn

21,899

Enable to pass two or model pretrained models to the Trainer.

{ "login": "shamanez", "id": 16892570, "node_id": "MDQ6VXNlcjE2ODkyNTcw", "avatar_url": "https://avatars.githubusercontent.com/u/16892570?v=4", "gravatar_id": "", "url": "https://api.github.com/users/shamanez", "html_url": "https://github.com/shamanez", "followers_url": "https://api.github.com/users/shamanez/followers", "following_url": "https://api.github.com/users/shamanez/following{/other_user}", "gists_url": "https://api.github.com/users/shamanez/gists{/gist_id}", "starred_url": "https://api.github.com/users/shamanez/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/shamanez/subscriptions", "organizations_url": "https://api.github.com/users/shamanez/orgs", "repos_url": "https://api.github.com/users/shamanez/repos", "events_url": "https://api.github.com/users/shamanez/events{/privacy}", "received_events_url": "https://api.github.com/users/shamanez/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md) are likely to be ignored." ]

1,677

1,681

CONTRIBUTOR

null

### Feature request I tried a compound architecture that uses two tower models that get initialized with from_pretrained methods. Then I wrapped the Custom model class with the Pretrained model class. It works well in a single GPU. But fails in multiple GPU settings. ### Motivation It would be useful to change models like DPR. ### Your contribution I have tested a few methods and can help to improve.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21899/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21899/timeline

completed

null

https://api.github.com/repos/huggingface/transformers/issues/21898

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21898/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21898/comments

https://api.github.com/repos/huggingface/transformers/issues/21898/events

https://github.com/huggingface/transformers/pull/21898

1,606,611,803

PR_kwDOCUB6oc5LHmXM

21,898

fix typo in Bart's attention

{ "login": "kashif", "id": 8100, "node_id": "MDQ6VXNlcjgxMDA=", "avatar_url": "https://avatars.githubusercontent.com/u/8100?v=4", "gravatar_id": "", "url": "https://api.github.com/users/kashif", "html_url": "https://github.com/kashif", "followers_url": "https://api.github.com/users/kashif/followers", "following_url": "https://api.github.com/users/kashif/following{/other_user}", "gists_url": "https://api.github.com/users/kashif/gists{/gist_id}", "starred_url": "https://api.github.com/users/kashif/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/kashif/subscriptions", "organizations_url": "https://api.github.com/users/kashif/orgs", "repos_url": "https://api.github.com/users/kashif/repos", "events_url": "https://api.github.com/users/kashif/events{/privacy}", "received_events_url": "https://api.github.com/users/kashif/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "thank you!" ]

1,677

CONTRIBUTOR

null

# What does this PR do? Fix a typo in Bart's attention.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21898/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21898/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21898", "html_url": "https://github.com/huggingface/transformers/pull/21898", "diff_url": "https://github.com/huggingface/transformers/pull/21898.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21898.patch", "merged_at": 1677764967000 }

https://api.github.com/repos/huggingface/transformers/issues/21897

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21897/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21897/comments

https://api.github.com/repos/huggingface/transformers/issues/21897/events

https://github.com/huggingface/transformers/pull/21897

1,606,533,351

PR_kwDOCUB6oc5LHVWT

21,897

Faster zero shot image

{ "login": "Narsil", "id": 204321, "node_id": "MDQ6VXNlcjIwNDMyMQ==", "avatar_url": "https://avatars.githubusercontent.com/u/204321?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Narsil", "html_url": "https://github.com/Narsil", "followers_url": "https://api.github.com/users/Narsil/followers", "following_url": "https://api.github.com/users/Narsil/following{/other_user}", "gists_url": "https://api.github.com/users/Narsil/gists{/gist_id}", "starred_url": "https://api.github.com/users/Narsil/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Narsil/subscriptions", "organizations_url": "https://api.github.com/users/Narsil/orgs", "repos_url": "https://api.github.com/users/Narsil/repos", "events_url": "https://api.github.com/users/Narsil/events{/privacy}", "received_events_url": "https://api.github.com/users/Narsil/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "@yessenzhar", "_The documentation is not available anymore as the PR was closed or merged._", "Does it also do a cross product ? Then possibly yes.", "I mostly copy pasted the whole sequence. Will open a PR ", "> I mostly copy pasted the whole sequence. Will open a PR\r\n\r\nThe cross product is in the model itself (image_batch_size x text_batch_size).", "CLAP has the same implementation for loss and etc. So we should see a similar behaviour \r\n", "@Narsil Thank you for sorting this out this quick." ]

1,677

CONTRIBUTOR

null

# What does this PR do? Should superseed: https://github.com/huggingface/transformers/pull/21861#event-8644623666 - Keeps `batch_size` active, but restricts it to the `image`part. - Calculates `candidate_labels` features only once (there's no way with this current pipeline to send different candidate_labels so we can take the optimization. - Ultra narrowed for `CLIP` but the pipeline has existed for 4 months now with no new models. Given the popularity of CLIP for diffusion models, I think this is ok to overspecify. We can always relax later when new models come in. This allows to downgrade from ChunkPipeline to regular PIpeline   Fixes # (issue) ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [ ] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [ ] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [ ] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [ ] Did you write any new necessary tests? ## Who can review? Anyone in the community is free to review the PR once the tests have passed. Feel free to tag members/contributors who may be interested in your PR.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21897/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21897/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21897", "html_url": "https://github.com/huggingface/transformers/pull/21897", "diff_url": "https://github.com/huggingface/transformers/pull/21897.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21897.patch", "merged_at": 1677782782000 }

https://api.github.com/repos/huggingface/transformers/issues/21896

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21896/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21896/comments

https://api.github.com/repos/huggingface/transformers/issues/21896/events

https://github.com/huggingface/transformers/pull/21896

1,606,412,521

PR_kwDOCUB6oc5LG69O

21,896

[T5 doc] Fix confusing documentation about `d_kv`

{ "login": "ArthurZucker", "id": 48595927, "node_id": "MDQ6VXNlcjQ4NTk1OTI3", "avatar_url": "https://avatars.githubusercontent.com/u/48595927?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ArthurZucker", "html_url": "https://github.com/ArthurZucker", "followers_url": "https://api.github.com/users/ArthurZucker/followers", "following_url": "https://api.github.com/users/ArthurZucker/following{/other_user}", "gists_url": "https://api.github.com/users/ArthurZucker/gists{/gist_id}", "starred_url": "https://api.github.com/users/ArthurZucker/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ArthurZucker/subscriptions", "organizations_url": "https://api.github.com/users/ArthurZucker/orgs", "repos_url": "https://api.github.com/users/ArthurZucker/repos", "events_url": "https://api.github.com/users/ArthurZucker/events{/privacy}", "received_events_url": "https://api.github.com/users/ArthurZucker/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,677

COLLABORATOR

null

# What does this PR do? Fixes #21641. The documentation mentioned that `d_kv` must be equal to `d_model//num_heads` while this does not hold in the code. The code states that `d_kv = inner_dim//num_heads`, but inner dim can be different from the d_model.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21896/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21896/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21896", "html_url": "https://github.com/huggingface/transformers/pull/21896", "diff_url": "https://github.com/huggingface/transformers/pull/21896.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21896.patch", "merged_at": 1677762446000 }

https://api.github.com/repos/huggingface/transformers/issues/21895

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21895/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21895/comments

https://api.github.com/repos/huggingface/transformers/issues/21895/events

https://github.com/huggingface/transformers/pull/21895

1,606,312,157

PR_kwDOCUB6oc5LGk5r

21,895

Make error message more informative

{ "login": "Atomnp", "id": 45496355, "node_id": "MDQ6VXNlcjQ1NDk2MzU1", "avatar_url": "https://avatars.githubusercontent.com/u/45496355?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Atomnp", "html_url": "https://github.com/Atomnp", "followers_url": "https://api.github.com/users/Atomnp/followers", "following_url": "https://api.github.com/users/Atomnp/following{/other_user}", "gists_url": "https://api.github.com/users/Atomnp/gists{/gist_id}", "starred_url": "https://api.github.com/users/Atomnp/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Atomnp/subscriptions", "organizations_url": "https://api.github.com/users/Atomnp/orgs", "repos_url": "https://api.github.com/users/Atomnp/repos", "events_url": "https://api.github.com/users/Atomnp/events{/privacy}", "received_events_url": "https://api.github.com/users/Atomnp/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._" ]

1,677

CONTRIBUTOR

null

# What does this PR do? Make error message more descriptive by adding one word, changes the message `The hidden size ({config.hidden_size}) is not a multiple of the number of attention` to `The hidden size ({config.hidden_size}) is not a multiple of the number of attention heads` in the previous error message, the word "head" was missing  ## Before submitting - [X] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [ ] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [ ] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [ ] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [ ] Did you write any new necessary tests? ## Who can review? Anyone in the community is free to review the PR once the tests have passed. Feel free to tag members/contributors who may be interested in your PR.

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21895/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21895/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21895", "html_url": "https://github.com/huggingface/transformers/pull/21895", "diff_url": "https://github.com/huggingface/transformers/pull/21895.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21895.patch", "merged_at": null }

https://api.github.com/repos/huggingface/transformers/issues/21894

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21894/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21894/comments

https://api.github.com/repos/huggingface/transformers/issues/21894/events

https://github.com/huggingface/transformers/pull/21894

1,606,281,683

PR_kwDOCUB6oc5LGeUw

21,894

Add FlaxWhisperForAudioClassification model

{ "login": "raghavanone", "id": 115454562, "node_id": "U_kgDOBuGyYg", "avatar_url": "https://avatars.githubusercontent.com/u/115454562?v=4", "gravatar_id": "", "url": "https://api.github.com/users/raghavanone", "html_url": "https://github.com/raghavanone", "followers_url": "https://api.github.com/users/raghavanone/followers", "following_url": "https://api.github.com/users/raghavanone/following{/other_user}", "gists_url": "https://api.github.com/users/raghavanone/gists{/gist_id}", "starred_url": "https://api.github.com/users/raghavanone/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/raghavanone/subscriptions", "organizations_url": "https://api.github.com/users/raghavanone/orgs", "repos_url": "https://api.github.com/users/raghavanone/repos", "events_url": "https://api.github.com/users/raghavanone/events{/privacy}", "received_events_url": "https://api.github.com/users/raghavanone/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/transformers/pr_21894). All of your documentation changes will be reflected on that endpoint.", "> Modelling code looks good @raghavanone! Nice one on getting this working so quickly 🙌 Do you want to have a go at adding the encoder-only tests? See the PyTorch WhisperForAudioClassficiation PR for details, think you can also add these quite quickly :)\r\n\r\nI have added the Encoder tests, But some test are failing, The FlaxWhisperForAudioClassification class extends FlaxWhisperPreTrainedModel . Due to this inheritance, the call method expects decoder related params. \r\n\r\nShould the FlaxWhisperForAudioClassification not extend FlaxWhisperPreTrainedModel instead create a new pretrainedclass ? ", "Hey @raghavanone! The PyTorch model has just been merged (https://github.com/huggingface/transformers/pull/21754), so you can rebase onto main to get the required config changes:\r\n```\r\ngit fetch upstream\r\ngit rebase upstream/main\r\n```\r\nThis will fix the failing Flax tests we're getting here: https://app.circleci.com/pipelines/github/huggingface/transformers/58972/workflows/2388bd70-553e-412f-9ee7-0599cace5639/jobs/719829\r\n\r\nThe only thing to make sure is that the first time you push after rebasing, you **force push** to origin:\r\n```\r\ngit add .\r\ngit commit -m \"Some new changes after rebase\"\r\ngit push -f origin fix_issue_21779\r\n```\r\n\r\nYou only have to force push once, the next time you can just regular push:\r\n```\r\ngit add .\r\ngit commit -m \"Some more changes\"\r\ngit push -u origin fix_issue_21779\r\n```", "@sanchit-gandhi There are 2 test failing here, I am unable to get the same failure locally in my machine. Any pointers on how to replicate failing test and fix it ? ", "Hey @raghavanone! Would you mind going through the previous review comments and marking them as resolved where you've addressed them? I'll then get you a final review asap! Thanks!", "Hey @raghavanone - I think the commit history has been corrupted for this PR? Gentle reminder that one must force push after rebasing: https://github.com/huggingface/transformers/pull/21894#issuecomment-1458359220 Think this is probably the culprit for the 250 extra commits!\r\n\r\nIn this instance, it's probably best to close this PR in favour of a new one that only contains the new changes you with to merge. Sorry about that!", "Closing in favour of #22883" ]

1,677

1,683

CONTRIBUTOR

null

# What does this PR do? Fix : #21779 Please review and let me know changes @sanchit-gandhi

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21894/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21894/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21894", "html_url": "https://github.com/huggingface/transformers/pull/21894", "diff_url": "https://github.com/huggingface/transformers/pull/21894.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21894.patch", "merged_at": null }

https://api.github.com/repos/huggingface/transformers/issues/21893

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21893/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21893/comments

https://api.github.com/repos/huggingface/transformers/issues/21893/events

https://github.com/huggingface/transformers/pull/21893

1,606,256,828

PR_kwDOCUB6oc5LGY_G

21,893

[ZAC] fix ci daily

{ "login": "ArthurZucker", "id": 48595927, "node_id": "MDQ6VXNlcjQ4NTk1OTI3", "avatar_url": "https://avatars.githubusercontent.com/u/48595927?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ArthurZucker", "html_url": "https://github.com/ArthurZucker", "followers_url": "https://api.github.com/users/ArthurZucker/followers", "following_url": "https://api.github.com/users/ArthurZucker/following{/other_user}", "gists_url": "https://api.github.com/users/ArthurZucker/gists{/gist_id}", "starred_url": "https://api.github.com/users/ArthurZucker/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ArthurZucker/subscriptions", "organizations_url": "https://api.github.com/users/ArthurZucker/orgs", "repos_url": "https://api.github.com/users/ArthurZucker/repos", "events_url": "https://api.github.com/users/ArthurZucker/events{/privacy}", "received_events_url": "https://api.github.com/users/ArthurZucker/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "_The documentation is not available anymore as the PR was closed or merged._", "Failing test is unrelated" ]

1,677

COLLABORATOR

null

# What does this PR do? Fixes the loading of zero shot audio classification pipeline by providing a correct revision. The previous one was destroyed when checkpoints were overwritten at some point. (that's also when the model card dissapeared).

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21893/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21893/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21893", "html_url": "https://github.com/huggingface/transformers/pull/21893", "diff_url": "https://github.com/huggingface/transformers/pull/21893.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21893.patch", "merged_at": 1677750364000 }

https://api.github.com/repos/huggingface/transformers/issues/21892

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21892/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21892/comments

https://api.github.com/repos/huggingface/transformers/issues/21892/events

https://github.com/huggingface/transformers/issues/21892

1,606,239,921

I_kwDOCUB6oc5fvUax

21,892

[i18n-<languageCode>] Translating docs to <languageName>

{ "login": "sachiweb", "id": 102918669, "node_id": "U_kgDOBiJqDQ", "avatar_url": "https://avatars.githubusercontent.com/u/102918669?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachiweb", "html_url": "https://github.com/sachiweb", "followers_url": "https://api.github.com/users/sachiweb/followers", "following_url": "https://api.github.com/users/sachiweb/following{/other_user}", "gists_url": "https://api.github.com/users/sachiweb/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachiweb/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachiweb/subscriptions", "organizations_url": "https://api.github.com/users/sachiweb/orgs", "repos_url": "https://api.github.com/users/sachiweb/repos", "events_url": "https://api.github.com/users/sachiweb/events{/privacy}", "received_events_url": "https://api.github.com/users/sachiweb/received_events", "type": "User", "site_admin": false }

[ { "id": 2796628563, "node_id": "MDU6TGFiZWwyNzk2NjI4NTYz", "url": "https://api.github.com/repos/huggingface/transformers/labels/WIP", "name": "WIP", "color": "234C99", "default": false, "description": "Label your PR/Issue with WIP for some long outstanding Issues/PRs that are work in progress" } ]

closed

false

null

[]

1,677

NONE

null

Hi! Let's bring the documentation to all the <languageName>-speaking community 🌐 (currently 0 out of 267 complete) Who would want to translate? Please follow the 🤗 [TRANSLATING guide](https://github.com/huggingface/transformers/blob/main/docs/TRANSLATING.md). Here is a list of the files ready for translation. Let us know in this issue if you'd like to translate any, and we'll add your name to the list. Some notes: * Please translate using an informal tone (imagine you are talking with a friend about transformers 🤗). * Please translate in a gender-neutral way. * Add your translations to the folder called `<languageCode>` inside the [source folder](https://github.com/huggingface/transformers/tree/main/docs/source). * Register your translation in `<languageCode>/_toctree.yml`; please follow the order of the [English version](https://github.com/huggingface/transformers/blob/main/docs/source/en/_toctree.yml). * Once you're finished, open a pull request and tag this issue by including #issue-number in the description, where issue-number is the number of this issue. Please ping @ArthurZucker, @sgugger for review. * 🙋 If you'd like others to help you with the translation, you can also post in the 🤗 [forums](https://discuss.huggingface.co/). ## Get Started section - [ ] [index.mdx](https://github.com/huggingface/transformers/blob/main/docs/source/en/index.mdx) https://github.com/huggingface/transformers/pull/20180 - [ ] [quicktour.mdx](https://github.com/huggingface/transformers/blob/main/docs/source/en/quicktour.mdx) (waiting for initial PR to go through) - [ ] [installation.mdx](https://github.com/huggingface/transformers/blob/main/docs/source/en/installation.mdx). ## Tutorial section - [ ] [pipeline_tutorial.mdx](https://github.com/huggingface/transformers/blob/main/docs/source/en/pipeline_tutorial.mdx) - [ ] [autoclass_tutorial.mdx](https://github.com/huggingface/transformers/blob/master/docs/source/autoclass_tutorial.mdx) - [ ] [preprocessing.mdx](https://github.com/huggingface/transformers/blob/main/docs/source/en/preprocessing.mdx) - [ ] [training.mdx](https://github.com/huggingface/transformers/blob/main/docs/source/en/training.mdx) - [ ] [accelerate.mdx](https://github.com/huggingface/transformers/blob/main/docs/source/en/accelerate.mdx) - [ ] [model_sharing.mdx](https://github.com/huggingface/transformers/blob/main/docs/source/en/model_sharing.mdx) - [ ] [multilingual.mdx](https://github.com/huggingface/transformers/blob/main/docs/source/en/multilingual.mdx)

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21892/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21892/timeline

completed

null

https://api.github.com/repos/huggingface/transformers/issues/21891

https://api.github.com/repos/huggingface/transformers

https://api.github.com/repos/huggingface/transformers/issues/21891/labels{/name}

https://api.github.com/repos/huggingface/transformers/issues/21891/comments

https://api.github.com/repos/huggingface/transformers/issues/21891/events

https://github.com/huggingface/transformers/pull/21891

1,606,038,926

PR_kwDOCUB6oc5LFqHM

21,891

[Time-Series] Autoformer model

{ "login": "elisim", "id": 17675462, "node_id": "MDQ6VXNlcjE3Njc1NDYy", "avatar_url": "https://avatars.githubusercontent.com/u/17675462?v=4", "gravatar_id": "", "url": "https://api.github.com/users/elisim", "html_url": "https://github.com/elisim", "followers_url": "https://api.github.com/users/elisim/followers", "following_url": "https://api.github.com/users/elisim/following{/other_user}", "gists_url": "https://api.github.com/users/elisim/gists{/gist_id}", "starred_url": "https://api.github.com/users/elisim/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/elisim/subscriptions", "organizations_url": "https://api.github.com/users/elisim/orgs", "repos_url": "https://api.github.com/users/elisim/repos", "events_url": "https://api.github.com/users/elisim/events{/privacy}", "received_events_url": "https://api.github.com/users/elisim/received_events", "type": "User", "site_admin": false }

[]

closed

false

null

[]

[ "One small open issue left, is adding the series decomposition to the decoder with the trend input. Will do after the initial review :)", "_The documentation is not available anymore as the PR was closed or merged._", "some of the TF tests are failing and I believe they are unrelated", "PR is green ", "thank you @amyeroberts will get it fixed!", "@amyeroberts, thank you for the comprehensive CR! I sincerely appreciate the effort and time you dedicated to thoroughly assessing this pull request. \r\n\r\nWill be fixed!", "CR changes I did:\r\n\r\r\n* Added `layer_norm_eps` \r\n* Model layers are now take the config, except the `AutoformerAttention` which I wasn't sure about\n* Better variables names \n* Addressed questions I could answer\n\n`fix-copies` is falling because diffs with the time-series-transformer. It's about to decide if to change time-series-transformer here, or to remove \"copied from...\"\r\n\r\n@kashif @amyeroberts \n\n" ]

1,677

1,685

CONTRIBUTOR

null

# What does this PR do? Adding Time Series Autoformer model https://arxiv.org/abs/2106.13008 Related issue: #21890 @kashif :) ## Differences between the vanilla transformer <img width="484" alt="image" src="https://user-images.githubusercontent.com/17675462/229438415-6a5bba78-c3bf-47b1-966a-f3664a0921e0.png"> * Introduced Series Decomposition in encoder & decoder --- done, waiting for review * Replaced canonical self-attention with autocorrelation block --- done, waiting for review * Added seasonal and trend inputs for the decoder --- added todo places in the code * trend seasonal pseudo code: ``` mean_data = mean(enc_input) zeros = zeros() # size: x_dec.size(0), prediction_length, x_dec.size(2) seasonal_init, trend_init = decomp_layer(enc_input) trend_init = concat(trend_init, mean_data) seasonal_init = concat(seasonal_init, zeros) ```

{ "url": "https://api.github.com/repos/huggingface/transformers/issues/21891/reactions", "total_count": 1, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 1, "rocket": 0, "eyes": 0 }

https://api.github.com/repos/huggingface/transformers/issues/21891/timeline

null

false

{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/21891", "html_url": "https://github.com/huggingface/transformers/pull/21891", "diff_url": "https://github.com/huggingface/transformers/pull/21891.diff", "patch_url": "https://github.com/huggingface/transformers/pull/21891.patch", "merged_at": 1685435013000 }