Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / huggingface/peft issues and pull requests

#815 - Tst extend unit tests 8bit quantization

Pull Request - State: closed - Opened by BenjaminBossan about 1 year ago - 3 comments

#814 - Fixed test issue for SD model with disabled LoRA adapter

Pull Request - State: closed - Opened by kovalexal about 1 year ago - 10 comments

#814 - Fixed test issue for SD model with disabled LoRA adapter

Pull Request - State: open - Opened by kovalexal about 1 year ago - 9 comments

#813 - add docs chatbot v0

Pull Request - State: open - Opened by pacman100 about 1 year ago - 5 comments

#813 - add docs chatbot v0

Pull Request - State: closed - Opened by pacman100 about 1 year ago - 6 comments

#812 - Question -- is `TaskType` used for anything?

Issue - State: closed - Opened by radekosmulski about 1 year ago - 4 comments

#812 - Question -- is `TaskType` used for anything?

Issue - State: closed - Opened by radekosmulski about 1 year ago - 4 comments

#811 - Prompt tuning: Tensors must have same number of dimensions: got 2 and 1

Issue - State: closed - Opened by andysingal about 1 year ago - 1 comment

#810 - Why can't we evaluate wer/cer while training using PEFT?

Issue - State: closed - Opened by MightyStud about 1 year ago - 2 comments

#810 - Why can't we evaluate wer/cer while training using PEFT?

Issue - State: open - Opened by MightyStud about 1 year ago - 1 comment

#809 - Fix seq2seq prompt tuning (#439)

Pull Request - State: open - Opened by glerzing about 1 year ago - 2 comments

#809 - Fix seq2seq prompt tuning (#439)

Pull Request - State: closed - Opened by glerzing about 1 year ago - 2 comments

#808 - What is the correct way to apply LoRA on a custom model (not models on HuggingFace)?

Issue - State: closed - Opened by DtYXs about 1 year ago - 16 comments

#807 - [skip ci] [WIP] Move tuners to subpackages

Pull Request - State: open - Opened by BenjaminBossan about 1 year ago - 1 comment

#807 - MNT: Move tuners to subpackages

Pull Request - State: closed - Opened by BenjaminBossan about 1 year ago - 4 comments

#806 - HFValidationError on local PeftModel.from_pretrained

Issue - State: closed - Opened by LazerJesus about 1 year ago - 2 comments

#804 - Multi adapter weight conflicts when services are concurrent

Issue - State: closed - Opened by Nipi64310 about 1 year ago - 10 comments

#804 - Multi adapter weight conflicts when services are concurrent

Issue - State: closed - Opened by Nipi64310 about 1 year ago - 22 comments

#803 - Error when sharding model within deepspeed.zero.Init context

Issue - State: closed - Opened by softsweetengineer about 1 year ago - 1 comment

#802 - Improve config handling

Issue - State: closed - Opened by BenjaminBossan about 1 year ago - 1 comment

#801 - BERT and LoRA BERT Number of Parameters Mismatch

Issue - State: closed - Opened by uygarkurt about 1 year ago - 1 comment

#796 - Add support for Pix2Struct

Issue - State: open - Opened by NielsRogge about 1 year ago - 2 comments

#796 - Add support for Pix2Struct

Issue - State: closed - Opened by NielsRogge about 1 year ago - 3 comments

#795 - Loading the base model and the PEFT model at the same time

Issue - State: closed - Opened by CandyPanda-LS about 1 year ago - 2 comments

#794 - Fix unbound error in ia3.py

Pull Request - State: closed - Opened by His-Wardship about 1 year ago - 14 comments

#794 - Fix unbound error in ia3.py

Pull Request - State: closed - Opened by His-Wardship about 1 year ago - 14 comments

#793 - Lora model (after fine tuning) working exactly the same as base model

Issue - State: closed - Opened by Luke-4 about 1 year ago - 25 comments

#793 - Lora model (after fine tuning) working exactly the same as base model

Issue - State: closed - Opened by Luke-4 about 1 year ago - 19 comments

#792 - multiple adapter load_model fails when adding bias

Issue - State: closed - Opened by DavidPeleg6 about 1 year ago - 4 comments

#790 - PR #389 breaks Flash Attention 2 with peft

Issue - State: open - Opened by rationalism about 1 year ago - 10 comments

#790 - PR #389 breaks Flash Attention 2 with peft

Issue - State: closed - Opened by rationalism about 1 year ago - 11 comments

#785 - GPU memory consumption increases when using a qunatized model with the PEFT

Issue - State: open - Opened by shamanez about 1 year ago - 1 comment

#785 - GPU memory consumption increases when using a qunatized model with the PEFT

Issue - State: closed - Opened by shamanez about 1 year ago - 2 comments

#784 - Peft model signature

Pull Request - State: closed - Opened by kiansierra about 1 year ago - 20 comments

#784 - Peft model signature

Pull Request - State: closed - Opened by kiansierra about 1 year ago - 20 comments

#783 - Automatic Model Signature for PeftModel

Issue - State: closed - Opened by kiansierra about 1 year ago - 1 comment

#783 - Automatic Model Signature for PeftModel

Issue - State: closed - Opened by kiansierra about 1 year ago - 1 comment

#782 - It just doesn't calculate the metrics

Issue - State: closed - Opened by xiehuanyi about 1 year ago - 2 comments

#782 - It just doesn't calculate the metrics

Issue - State: open - Opened by xiehuanyi about 1 year ago - 2 comments

#781 - Can I apply peft to my own model?

Issue - State: closed - Opened by moonriver0922 about 1 year ago - 4 comments

#780 - Add new method: GLoRA

Issue - State: closed - Opened by HsunGong about 1 year ago - 3 comments

#780 - Add new method: GLoRA

Issue - State: open - Opened by HsunGong about 1 year ago - 1 comment

#777 - Is it possible to have a peft model on a pefted model?

Issue - State: open - Opened by 2catycm about 1 year ago - 3 comments

#777 - Is it possible to have a peft model on a pefted model?

Issue - State: closed - Opened by 2catycm about 1 year ago - 3 comments

#774 - Peft Offload_dir should be offload_folder

Issue - State: open - Opened by larawehbe about 1 year ago - 1 comment

#774 - Peft Offload_dir should be offload_folder

Issue - State: closed - Opened by larawehbe about 1 year ago - 1 comment

#771 - GPTQ Integration

Pull Request - State: closed - Opened by SunMarc about 1 year ago - 2 comments

#771 - GPTQ Integration

Pull Request - State: closed - Opened by SunMarc about 1 year ago - 2 comments

#768 - Improve Training Loss

Issue - State: closed - Opened by ahmadmustafaanis about 1 year ago - 2 comments

#766 - LoRA `Linear4bit` is unmergeable

Issue - State: closed - Opened by SergeyTsimfer about 1 year ago - 2 comments

#766 - LoRA `Linear4bit` is unmergeable

Issue - State: closed - Opened by SergeyTsimfer about 1 year ago - 5 comments

#765 - Performance degradation with HF Trainer when training LayoutLMv3

Issue - State: open - Opened by anthony2261 about 1 year ago - 1 comment

#765 - Performance degradation with HF Trainer when training LayoutLMv3

Issue - State: closed - Opened by anthony2261 about 1 year ago - 4 comments

#764 - Peft for SEQ_CLS not compatible with BloomForSequenceClassification

Issue - State: closed - Opened by poedator about 1 year ago - 1 comment

#763 - Extend AdaptionPrompt and Add Multi-Modal AdaptionPromptV2

Pull Request - State: closed - Opened by PanQiWei about 1 year ago - 7 comments

#763 - Extend AdaptionPrompt and Add Multi-Modal AdaptionPromptV2

Pull Request - State: open - Opened by PanQiWei about 1 year ago - 6 comments

#762 - Size Mismatch Error When Loading Pretrained Model with Expanded Embedding Layer

Issue - State: open - Opened by hzphzp about 1 year ago - 3 comments

#762 - Size Mismatch Error When Loading Pretrained Model with Expanded Embedding Layer

Issue - State: closed - Opened by hzphzp about 1 year ago - 6 comments

#761 - fine-tuning OpenClip with Hugingface's PEFT (such as LoRA)

Issue - State: open - Opened by KyanChen about 1 year ago - 48 comments

#761 - fine-tuning OpenClip with Hugingface's PEFT (such as LoRA)

Issue - State: open - Opened by KyanChen about 1 year ago - 46 comments

#759 - Mixed-task batching

Issue - State: open - Opened by einarbmag about 1 year ago - 4 comments

#759 - Mixed-task batching

Issue - State: closed - Opened by einarbmag about 1 year ago - 5 comments

#756 - PEFT for Multiple Choice?

Issue - State: closed - Opened by jacob-morrison about 1 year ago - 7 comments

#754 - Merging LORA weights and saving safetensors to launch a TGI server

Issue - State: open - Opened by Mohamedhabi about 1 year ago - 2 comments

#754 - Merging LORA weights and saving safetensors to launch a TGI server

Issue - State: closed - Opened by Mohamedhabi about 1 year ago - 2 comments

#752 - specify rank per layer or per tensor type

Issue - State: closed - Opened by MichelNivard about 1 year ago - 2 comments

#749 - [`core`] PEFT refactor + introducing `inject_adapter_in_model` public method

Pull Request - State: closed - Opened by younesbelkada about 1 year ago - 2 comments

#749 - [`core`] PEFT refactor + introducing `inject_adapter_in_model` public method

Pull Request - State: closed - Opened by younesbelkada about 1 year ago - 2 comments

#748 - Feature for using onnx model as base model when using lora weight

Issue - State: closed - Opened by bohyunshin about 1 year ago - 5 comments

#748 - Feature for using onnx model as base model when using lora weight

Issue - State: open - Opened by bohyunshin about 1 year ago - 4 comments

#746 - Please fix Lora model resume in transformers when using DeepSpeed

Issue - State: closed - Opened by lucasjinreal over 1 year ago - 9 comments

#746 - Please fix Lora model resume in transformers when using DeepSpeed

Issue - State: open - Opened by lucasjinreal over 1 year ago - 6 comments

#739 - CUDA out of memory. Is it a matter of data size? Always stop at the same step!

Issue - State: closed - Opened by dsj96 about 1 year ago - 2 comments

#738 - Method to unload an adapter, to allow the memory to be freed

Issue - State: closed - Opened by uyhcire about 1 year ago - 3 comments
Labels: solved

#738 - Method to unload an adapter, to allow the memory to be freed

Issue - State: closed - Opened by uyhcire about 1 year ago - 2 comments
Labels: solved

#735 - Add dense layers to TRANSFORMERS_MODELS_TO_LORA_TARGET_MODULES_MAPPING

Issue - State: closed - Opened by BramVanroy about 1 year ago - 2 comments

#735 - Add dense layers to TRANSFORMERS_MODELS_TO_LORA_TARGET_MODULES_MAPPING

Issue - State: closed - Opened by BramVanroy about 1 year ago - 2 comments

#732 - [minor bugfix] prevent lora Linear4bit from sneakily changing compute_dtype to float32

Pull Request - State: closed - Opened by justheuristic about 1 year ago - 6 comments

#725 - Use device_map to load adapters weight onto specified device

Pull Request - State: open - Opened by zhangyilun about 1 year ago - 5 comments

#725 - Use device_map to load adapters weight onto specified device

Pull Request - State: closed - Opened by zhangyilun about 1 year ago - 7 comments

#723 - quantize open-Flamingo error : init reverted

Issue - State: closed - Opened by YerongLi about 1 year ago - 2 comments

#723 - quantize open-Flamingo error : init reverted

Issue - State: closed - Opened by YerongLi about 1 year ago - 2 comments

#720 - Use LoRA and prompt tuning at the same time?

Issue - State: closed - Opened by chengyin38 about 1 year ago - 3 comments

#720 - Use LoRA and prompt tuning at the same time?

Issue - State: closed - Opened by chengyin38 about 1 year ago - 2 comments

#719 - [WIP] Add functionality to support AdaMix

Pull Request - State: closed - Opened by rishabbala about 1 year ago - 21 comments

#717 - Release version 0.5.0.dev0

Pull Request - State: closed - Opened by pacman100 about 1 year ago - 1 comment

#715 - FIX: Removes warnings about unknown pytest marker

Pull Request - State: closed - Opened by BenjaminBossan about 1 year ago - 1 comment

#714 - `AutoTokenizer` bug when using `LlamaTokenizer`

Issue - State: closed - Opened by sidnb13 about 1 year ago - 1 comment

#714 - `AutoTokenizer` bug when using `LlamaTokenizer`

Issue - State: open - Opened by sidnb13 about 1 year ago - 1 comment

#713 - Can prefix tuning be used for multi-query model like bigcode/starcoder?

Issue - State: closed - Opened by ainilian about 1 year ago - 4 comments

#713 - Can prefix tuning be used for multi-query model like bigcode/starcoder?

Issue - State: closed - Opened by ainilian about 1 year ago - 4 comments

#711 - How to change the location of soft tokens in prompt tuning

Issue - State: closed - Opened by XueTianci about 1 year ago - 4 comments

#711 - How to change the location of soft tokens in prompt tuning

Issue - State: open - Opened by XueTianci about 1 year ago - 3 comments