14 Pull requests merged by 11 people

• [docs] Redesign

  #31757 merged Mar 3, 2025

• Remove unused code

  #36459 merged Mar 3, 2025

• [Style] fix E721 warnings

  #36474 merged Mar 3, 2025

• Fix edge case for continue_final_message

  #36404 merged Mar 3, 2025

• Fix pipeline+peft interaction

  #36480 merged Mar 3, 2025

• chore: fix message descriptions in arguments and comments

  #36504 merged Mar 3, 2025

• Fix some typos in docs

  #36502 merged Mar 3, 2025

• fix torch_dtype, contiguous, and load_state_dict regression

  #36512 merged Mar 3, 2025

• Fix kwargs UserWarning in SamImageProcessor

  #36479 merged Mar 3, 2025

• Check TRUST_REMOTE_CODE for RealmRetriever for security

  #36511 merged Mar 3, 2025

• Fix loading zero3 weights

  #36455 merged Mar 3, 2025

• Fix _load_state_dict_into_meta_model with device_map=None

  #36488 merged Mar 2, 2025

• Fix couples of issues from #36335

  #36453 merged Mar 1, 2025

• Add Got-OCR 2 Fast image processor and refactor slow one

  #36185 merged Mar 1, 2025

17 Pull requests opened by 15 people

• avoid errors when the size of `input_ids` passed to `PrefixConstrainedLogitsProcessor` is zero

  #36489 opened Mar 1, 2025

• Allow OOV Image Token for LLaVa Next Variants

  #36491 opened Mar 2, 2025

• Make SamVisionEncoder public for better accessibility

  #36493 opened Mar 2, 2025

• Add an event related to forward in the TrainerCallback

  #36496 opened Mar 2, 2025

• Add support for seed in `DataCollatorForLanguageModeling`

  #36497 opened Mar 2, 2025

• [WIP] Add EVEv2 model

  #36498 opened Mar 2, 2025

• Export base streamer.

  #36500 opened Mar 2, 2025

• Refactor object-detection models

  #36514 opened Mar 3, 2025

• Update recommended reviewers

  #36515 opened Mar 3, 2025

• Make ViTPooler configurable

  #36517 opened Mar 3, 2025

• enable/disable compile for quants methods

  #36519 opened Mar 3, 2025

• guard torch version for uint16

  #36520 opened Mar 3, 2025

• Add aya

  #36521 opened Mar 3, 2025

• [docs] Serving LLMs

  #36522 opened Mar 3, 2025

• [docs] Fix quantization overview

  #36523 opened Mar 3, 2025

• chore: Fix typos in docs and examples

  #36524 opened Mar 4, 2025

• chore: enhance messages in docstrings

  #36525 opened Mar 4, 2025

18 Issues closed by 8 people

• tokenizers.apply_chat_template with continue_final_message=True with </think> token

  #36440 closed Mar 3, 2025

• tokenizers.apply_chat_template with `continue_final_message=True` with trailing spaces in input

  #35433 closed Mar 3, 2025

• Confusing behavior when loading PEFT models with pipeline

  #36473 closed Mar 3, 2025

• `_load_state_dict_into_meta_model` - `'NoneType' object has no attribute 'load_state_dict'`

  #36495 closed Mar 3, 2025

• Failed to import transformers.trainer because of the following error (look up to see its traceback): cannot import name 'add_model_info_to_custom_pipelines' from 'transformers.utils'

  #36503 closed Mar 3, 2025

• GRPO Reward Weight Scheduler

  #36490 closed Mar 3, 2025

• please support register_full_backward_pre_hook and register_full_backward_hook

  #36507 closed Mar 3, 2025

• Some Whisper beam search output (sequences_scores, etc.) is lost in _stack_split_outputs

  #32373 closed Mar 3, 2025

• [DEV Testing] Issues with `test_modeling_common`

  #35857 closed Mar 3, 2025

• [BUG]npu zero3 训练自定义模型时，报错Function SumBackward0 returned an invalid gradient at index 0

  #36387 closed Mar 3, 2025

• Load siglip2 error

  #36475 closed Mar 3, 2025

• `padding_side` is of type `bool` when it should be `Literal['right', 'left']`

  #36252 closed Mar 3, 2025

• Add Wan model into Transformers

  #36494 closed Mar 2, 2025

• Bug introduced in `_load_state_dict_into_meta_model` and `to` `v4.49.0`..`v4.50.0.dev`

  #36441 closed Mar 1, 2025

• `model.config.to_diff_dict()` delivers different result to `model.save_pretrained()`

  #35426 closed Mar 1, 2025

• AttributeError: 'Config' object has no attribute '_get_non_default_generation_parameters'

  #35543 closed Mar 1, 2025

• Prompt_ids feature causing repetitions and hallucinations

  #35603 closed Mar 1, 2025

• convert_llama_weight_to_hf.py

  #35820 closed Mar 1, 2025

11 Issues opened by 11 people

• GraniteMoe’s implementation is not compatible with HF’s peft

  #36518 opened Mar 3, 2025

• Object detection tutorial uses buggy dataset, may lead to crash during training

  #36516 opened Mar 3, 2025

• Allow parameters of the ViTPooler to be configurable, with the default values set to the current hardcoded values

  #36513 opened Mar 3, 2025

• model.generate function is not compatible with custom position_ids

  #36510 opened Mar 3, 2025

• Can not use prompt tuning inference

  #36509 opened Mar 3, 2025

• haggingface model unsupported torch backward hook

  #36508 opened Mar 3, 2025

• model from_pretrained bug in 4.50.dev0 in these days

  #36506 opened Mar 3, 2025

• add a param to control cache in streamer when return output

  #36505 opened Mar 3, 2025

• TypeError: object of type 'IterableDataset' has no len()

  #36501 opened Mar 3, 2025

• Support Distill Depth Anything

  #36499 opened Mar 2, 2025

• Error at scatter num_items_in_batch in ddp/dp

  #36492 opened Mar 2, 2025

69 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• `GPT2Model` StaticCache support

  #35761 commented on Mar 3, 2025 • 6 new comments

• Fix sdpa in sam and refactor relative position embeddings

  #36422 commented on Mar 4, 2025 • 5 new comments

• Integrate SwanLab for offline/online experiment tracking and local visualization

  #36433 commented on Mar 3, 2025 • 3 new comments

• Fix Batch Size Mismatch When Using `crops_n_layers` in `mask-generation` Pipeline #35530

  #35627 commented on Mar 3, 2025 • 3 new comments

• Customize docstrings fast image processor

  #36466 commented on Mar 3, 2025 • 2 new comments

• Add FAST

  #35476 commented on Mar 2, 2025 • 1 new comment

• 🚧 [WiP] Add Janus model

  #36053 commented on Mar 4, 2025 • 1 new comment

• Support QuestionAnswering Module for ModernBert based models.

  #35566 commented on Mar 3, 2025 • 1 new comment

• Fix model saving bug post training with tensor parallel in Accelerate

  #36434 commented on Mar 3, 2025 • 1 new comment

• [ModernBERT] Add CausalLM functionality to ModernBERT

  #35946 commented on Mar 3, 2025 • 0 new comments

• Add generation config validation using Pydantic

  #35910 commented on Mar 3, 2025 • 0 new comments

• Introduce modular files for speech models

  #35902 commented on Mar 3, 2025 • 0 new comments

• Add Doge model

  #35891 commented on Mar 2, 2025 • 0 new comments

• Adds GGUF support for Gemma models

  #35887 commented on Mar 3, 2025 • 0 new comments

• Add padding-free to bamba

  #35861 commented on Mar 3, 2025 • 0 new comments

• Github action for auto-assigning reviewers

  #35846 commented on Mar 3, 2025 • 0 new comments

• Add MultipleChoice & QuestionAnswering heads to ModernBERT

  #35825 commented on Mar 3, 2025 • 0 new comments

• Add StyleTTS 2

  #35790 commented on Mar 3, 2025 • 0 new comments

• Add AIMv2 to Transformers

  #35550 commented on Mar 3, 2025 • 0 new comments

• 🔴 Video processors as a separate class

  #35206 commented on Mar 3, 2025 • 0 new comments

• Add Phi-3.5-vision

  #36036 commented on Mar 3, 2025 • 0 new comments

• Proper performant flex attention implementation

  #36103 commented on Mar 3, 2025 • 0 new comments

• add DeepSpeed tensor parallel initialization.

  #36114 commented on Mar 3, 2025 • 0 new comments

• Try working around the processor registration bugs

  #36184 commented on Mar 3, 2025 • 0 new comments

• Add evolla rebase main

  #36232 commented on Mar 4, 2025 • 0 new comments

• Add support for DeepseekAI's DeepseekVL

  #36248 commented on Mar 2, 2025 • 0 new comments

• [docs] Update README

  #36265 commented on Mar 3, 2025 • 0 new comments

• Remove remote code warning

  #36285 commented on Mar 3, 2025 • 0 new comments

• Fix ONNX export for sequence classification head

  #36332 commented on Mar 3, 2025 • 0 new comments

• Handle DAC conversion when using weight_norm with newer PyTorch versions

  #36393 commented on Mar 2, 2025 • 0 new comments

• HPU support

  #36424 commented on Mar 3, 2025 • 0 new comments

• Avoid error when using `torch.compile` and `DataCollatorWithFlattening`

  #36450 commented on Mar 2, 2025 • 0 new comments

• Fix fp16 ONNX export for RT-DETR and RT-DETRv2

  #36460 commented on Mar 3, 2025 • 0 new comments

• Fix incorrect attention mask truncate in WhisperFlashAttention2

  #36477 commented on Mar 3, 2025 • 0 new comments

• Sanitize Model Module Names to Follow Python Conventions

  #36478 commented on Mar 3, 2025 • 0 new comments

• Mask2FormerImageProcessor support overlapping features

  #35536 commented on Mar 3, 2025 • 0 new comments

• model.parameters() return [Parameter containing: tensor([], device='cuda:0', dtype=torch.bfloat16, requires_grad=True)] when using zero3

  #35994 commented on Mar 3, 2025 • 0 new comments

• Qwen2VLForConditionalGeneration doesn't work with MPS devices

  #36413 commented on Mar 3, 2025 • 0 new comments

• Support H100 training with FP8 in Trainer and Deepspeed

  #25333 commented on Mar 2, 2025 • 0 new comments

• Support SDPA & Flash Attention 2 for LayoutLMv3

  #35467 commented on Mar 2, 2025 • 0 new comments

• LayerDrop broken in various Flax models (Whisper/BART/more...)

  #35468 commented on Mar 2, 2025 • 0 new comments

• Memory Access out of bounds in mra/cuda_kernel.cu::index_max_cuda_kernel()

  #35507 commented on Mar 2, 2025 • 0 new comments

• Very slow to load deep seekv3 int4 model and device_map="auto" "sequential" bug

  #35522 commented on Mar 2, 2025 • 0 new comments

• adalomo and deepspeed zero3 offload error

  #35977 commented on Mar 2, 2025 • 0 new comments

• llama code break with torch compile

  #36484 commented on Mar 2, 2025 • 0 new comments

• Speed up image processors - cast to array before BatchFeature

  #31205 commented on Mar 2, 2025 • 0 new comments

• A warning message showing that `MultiScaleDeformableAttention.so` is not found in `/root/.cache/torch_extensions` if `ninja` is installed with `transformers`

  #35349 commented on Mar 2, 2025 • 0 new comments

• [Whisper] TypeError: '<=' not supported between instances of 'NoneType' and 'float'

  #33552 commented on Mar 1, 2025 • 0 new comments

• Saving model with shared tensors fails on cpu but succeeds on gpu

  #33688 commented on Mar 1, 2025 • 0 new comments

• Tokenizer does not split text according to newly added input tokens

  #35447 commented on Mar 1, 2025 • 0 new comments

• Can't use Trainer on mps device

  #35954 commented on Mar 1, 2025 • 0 new comments

• DeepSeek V3 Support

  #35425 commented on Mar 1, 2025 • 0 new comments

• Samhq model addition

  #35147 commented on Mar 3, 2025 • 0 new comments

• Add TimesFM Time Series Forecasting Model

  #34082 commented on Mar 3, 2025 • 0 new comments

• TypeError: CustomTrainer.compute_loss() got an unexpected keyword argument 'num_items_in_batch'

  #36331 commented on Mar 4, 2025 • 0 new comments

• Groq inference provider

  #36353 commented on Mar 3, 2025 • 0 new comments

• Dtensor support requires torch>=2.5.1

  #36472 commented on Mar 3, 2025 • 0 new comments

• Gemma2 (quantized) inference is broken - torch._dynamo.exc.UserError: Dynamic control flow is not supported at the moment.

  #36485 commented on Mar 3, 2025 • 0 new comments

• The output tensor's data type is not torch.long when the input text is empty.

  #36277 commented on Mar 3, 2025 • 0 new comments

• Possible bug when using cosine lr scheduler with gradient accumulation

  #35484 commented on Mar 3, 2025 • 0 new comments

• Accidentally allocating 2x memory in new caching_allocator_warmup

  #36483 commented on Mar 3, 2025 • 0 new comments

• Misleading documentation for `is_decoder` configuration parameter

  #36482 commented on Mar 3, 2025 • 0 new comments

• Add type checking to CI

  #36481 commented on Mar 3, 2025 • 0 new comments

• ViTPose tutorial fails

  #36454 commented on Mar 3, 2025 • 0 new comments

• SAM mask-generation - crops_n_layers

  #35530 commented on Mar 3, 2025 • 0 new comments

• Enhance the memory efficiency of loading large models (400B) to prevent out-of-memory errors when using tensor parallelism.

  #36467 commented on Mar 3, 2025 • 0 new comments

• Add EVEv2 : an Encoder-free VLM

  #36379 commented on Mar 3, 2025 • 0 new comments

• Inference with FSDP during training affects checkpoints

  #34530 commented on Mar 3, 2025 • 0 new comments

• Recomputed tensor size does not match when using activation checkpointing when using FSDP and accelerate

  #34928 commented on Mar 3, 2025 • 0 new comments