OptimalScale/LMFlow issues and pull requests

#931 - [usability] support qwen2.5 and deepseek

Pull Request - State: open - Opened by wheresmyhair 4 days ago

#930 - [usability] Change dataset check method

Pull Request - State: closed - Opened by wheresmyhair 13 days ago

#929 - How to train a multimodal model with LMFlow?

Issue - State: open - Opened by ZHILIANGZHANG about 1 month ago

#928 - update lora_target_modules in training scripts

Pull Request - State: closed - Opened by Dominic789654 about 1 month ago

#927 - [doc] add FAQ placeholder, add note on arguments for training

Pull Request - State: closed - Opened by lxaw about 1 month ago

#926 - How to go on fine-tuning with a lora checkpoint?

Issue - State: closed - Opened by ORGRUI about 1 month ago - 2 comments

#925 - [usability] Add hymba lora target (#924)

Pull Request - State: closed - Opened by shizhediao about 2 months ago

#924 - [usability] Add hymba lora target

Pull Request - State: closed - Opened by wheresmyhair about 2 months ago

#923 - What is the difference between LMFlow and LLaMA Factory

Issue - State: open - Opened by Artessay about 2 months ago

#922 - Hymba support announcement

Pull Request - State: closed - Opened by shizhediao 2 months ago - 1 comment

#921 - announce support of hymba

Pull Request - State: closed - Opened by shizhediao 2 months ago

#920 - [BUG] Do not use generation of hf model in inferencers

Issue - State: open - Opened by Kamichanw 2 months ago
Labels: pending

#919 - Add Hymba and DoRA support

Pull Request - State: closed - Opened by Dominic789654 2 months ago - 2 comments

#918 - [BUG] Huggingface transformers library has deprecated directly import from transformers.deepspeed

Issue - State: closed - Opened by Kamichanw 2 months ago - 2 comments

#917 - Conversation_template

Issue - State: open - Opened by showgood880702 3 months ago - 4 comments
Labels: enhancement

#916 - Can LMFlow benchmark evaluate my finetuned model on user-defined dataset？

Issue - State: open - Opened by karry5921 3 months ago

#915 - [doc] update default branch

Pull Request - State: closed - Opened by wheresmyhair 3 months ago

#914 - [temp] temporarily restrict transformers version

Pull Request - State: closed - Opened by wheresmyhair 3 months ago

#913 - [BUG] Cannot run with the latest transformer release.

Issue - State: closed - Opened by buptsad 3 months ago - 3 comments

#912 - [doc] Add wandb setup guide

Pull Request - State: closed - Opened by wheresmyhair 4 months ago

#911 - [usability] temporarily change default version to 0.0.8

Pull Request - State: closed - Opened by wheresmyhair 4 months ago

#910 - Pre-training

Issue - State: open - Opened by Yummy416 4 months ago

#909 - [fix] merge lora fix

Pull Request - State: closed - Opened by wheresmyhair 4 months ago

#908 - [BUG] Fail to merge Lora. 'HFDecoderModel' object has no attribute 'backend_model'

Issue - State: closed - Opened by kwunhang 4 months ago

#907 - Does it support qwen2.5?

Issue - State: open - Opened by kwunhang 4 months ago

#906 - How to implement weight decay towards the pre-trained model?

Issue - State: open - Opened by sedol1339 4 months ago

#905 - [usability] deps streamlining

Pull Request - State: closed - Opened by wheresmyhair 4 months ago

#904 - Update & Supplement with new custom optimizer

Pull Request - State: open - Opened by tianshijing 4 months ago

#903 - [BUG]TypeError: Field.init() missing 1 required positional argument: 'kw_only'

Issue - State: closed - Opened by ORGRUI 4 months ago - 3 comments
Labels: good first issue

#902 - Fix load from LoRA weight & empty tools and system

Pull Request - State: closed - Opened by HALIS-sh 4 months ago

#901 - [usability] remove numpy version requirement

Pull Request - State: closed - Opened by wheresmyhair 4 months ago

#900 - [usability] remove numpy version requirement

Pull Request - State: closed - Opened by wheresmyhair 4 months ago

#899 - [Feature] LISA multi GPU support

Pull Request - State: closed - Opened by wheresmyhair 5 months ago

#898 - Can't reproduce AGIEval result of LISA

Issue - State: closed - Opened by BaohaoLiao 5 months ago - 1 comment

#897 - Some question on code of LISA

Issue - State: open - Opened by milong26 5 months ago

#896 - Does it support qwen2?

Issue - State: open - Opened by yangxue-1 6 months ago - 2 comments

#895 - Specific methods about lisa

Issue - State: open - Opened by orderer0001 6 months ago - 1 comment

#894 - Can lisa training support MiniCPM-V?

Issue - State: open - Opened by orderer0001 6 months ago - 1 comment

#893 - How can I use other optimizers in LISA?

Issue - State: open - Opened by greatzane 6 months ago - 1 comment

#892 - Modify the formatter of function and observation

Pull Request - State: closed - Opened by HALIS-sh 6 months ago

#891 - Why is Lisa’s training loss always 0?

Issue - State: closed - Opened by orderer0001 6 months ago - 1 comment

#890 - Is there any Lisa that supports llama3.1 yet?

Issue - State: open - Opened by orderer0001 6 months ago - 1 comment

#889 - [usability] make dpo eval dataset optional

Pull Request - State: closed - Opened by wheresmyhair 6 months ago

#888 - Problems encountered during speculative decoding execution

Issue - State: open - Opened by PoHaoYen 6 months ago - 1 comment

#887 - Enable RAG inference and evaluation

Pull Request - State: open - Opened by cheng-wang2002 7 months ago

#886 - [BUG]使用远程服务器运行chatbot_gradio.py时意外中断连接

Issue - State: open - Opened by 11whitewater 7 months ago - 2 comments

#885 - [BUG] EvaluatorArguments,

Issue - State: open - Opened by Harshit-tech9 7 months ago - 1 comment

#884 - function-call-finetune

Pull Request - State: closed - Opened by HALIS-sh 7 months ago - 1 comment

#883 - [Feature] Iterative DPO

Pull Request - State: closed - Opened by wheresmyhair 7 months ago - 1 comment

#882 - [Feature] Iterative DPO

Pull Request - State: closed - Opened by wheresmyhair 7 months ago

#881 - update naacl best paper award

Pull Request - State: closed - Opened by shizhediao 7 months ago

#880 - Merge contrib text2image to src

Pull Request - State: open - Opened by Aziily 7 months ago

#879 - [New Feature] Is Mixtral supported?

Issue - State: open - Opened by markusdr 7 months ago - 1 comment

#878 - [Bug Fix] update load_dataset to support long data

Pull Request - State: open - Opened by shizhediao 7 months ago

#877 - [ADD] T2I finetuning with SD1 and SD2 to contrib

Pull Request - State: closed - Opened by Aziily 7 months ago - 2 comments

#876 - Added more custom optimizers

Pull Request - State: closed - Opened by tianshijing 7 months ago - 1 comment

#875 - Added more custom optimizations

Pull Request - State: closed - Opened by tianshijing 7 months ago

#874 - Added more custom optimizations

Pull Request - State: closed - Opened by tianshijing 7 months ago

#873 - Added more custom optimizations

Pull Request - State: closed - Opened by tianshijing 7 months ago

#872 - Add more custom optimizations

Pull Request - State: closed - Opened by tianshijing 7 months ago

#871 - Expanding Optimization Options

Pull Request - State: closed - Opened by tianshijing 7 months ago - 1 comment

#870 - DPO+ZeRO train error

Issue - State: open - Opened by tankeui 8 months ago - 2 comments

#869 - The tool-learning domain fientune script developed based on ToolBench

Pull Request - State: closed - Opened by HALIS-sh 8 months ago

#868 - RAFT Issues

Issue - State: open - Opened by jujipotle 8 months ago - 2 comments

#867 - [Feature] reward model inferencer and dpov2 aligner

Pull Request - State: closed - Opened by wheresmyhair 8 months ago - 2 comments

#866 - [Feature] Reward model inferencer support

Pull Request - State: closed - Opened by wheresmyhair 8 months ago

#865 - Add customized optimizer support

Pull Request - State: closed - Opened by research4pan 8 months ago - 1 comment

#864 - Full parameter fine-tuning bugs

Issue - State: open - Opened by tankeui 8 months ago - 3 comments

#863 - [Feature] Add vllm inference example

Pull Request - State: closed - Opened by wheresmyhair 8 months ago - 1 comment

#862 - [Roadmap] LMFlow Roadmap

Issue - State: open - Opened by wheresmyhair 8 months ago - 2 comments

#861 - [BUG] The text cannot be generated successfully during the Raft step

Issue - State: open - Opened by biaoliu-kiritsugu 8 months ago - 1 comment
Labels: pending

#860 - [Feature] vllm inferencer and memory safe vllm inferencer

Pull Request - State: closed - Opened by wheresmyhair 8 months ago - 2 comments

#859 - [Feature] Iterative DPO support

Pull Request - State: closed - Opened by wheresmyhair 8 months ago

#858 - demo of DPO with QLoRA (w Llama3 70B Instruct)

Pull Request - State: open - Opened by anchen1011 8 months ago - 1 comment

#857 - Usability update

Pull Request - State: closed - Opened by wheresmyhair 8 months ago

#856 - json load dataset takes for a long time

Issue - State: closed - Opened by biaoliu-kiritsugu 8 months ago - 1 comment

#855 - multi-gpu full para train error

Issue - State: open - Opened by tankeui 8 months ago - 2 comments

#854 - [Feature] Add PPO support

Pull Request - State: closed - Opened by wheresmyhair 8 months ago

#853 - Add multi node README

Pull Request - State: closed - Opened by research4pan 8 months ago

#852 - [Model] hf model modification and inheritance change

Pull Request - State: closed - Opened by wheresmyhair 8 months ago

#851 - [Feature] PPO Support

Pull Request - State: closed - Opened by wheresmyhair 8 months ago

#850 - [Usability] Add preset lora target modules

Pull Request - State: closed - Opened by wheresmyhair 8 months ago

#849 - [Model Support] Qwen2 update

Pull Request - State: closed - Opened by wheresmyhair 8 months ago

#848 - Add langchain chatbot

Pull Request - State: closed - Opened by YanxinLu 8 months ago

#847 - [Bug fix] Fix tokenizer multiprocessing in reward model

Pull Request - State: closed - Opened by wheresmyhair 8 months ago

#846 - [Bug fix] Blocking function args missing fix

Pull Request - State: closed - Opened by wheresmyhair 8 months ago

#845 - [Bug fix] Tokenization multiprocessing fix

Pull Request - State: closed - Opened by wheresmyhair 8 months ago

#844 - Long context summarize demo

Pull Request - State: closed - Opened by HALIS-sh 9 months ago - 1 comment

#843 - Readme update

Pull Request - State: closed - Opened by wheresmyhair 9 months ago

#842 - Full parameter fine-tuning cannot be trained

Issue - State: open - Opened by orderer0001 9 months ago - 1 comment

#841 - Training was successful on a single card 4090GPU, but an error was reported on a 3*4090GPU. why

Issue - State: open - Opened by orderer0001 9 months ago - 1 comment

#840 - [BUG]when map the dataset, i set the num_proc = 2 or 4, it will make mistakes.

Issue - State: open - Opened by nicosouth 9 months ago - 8 comments

#839 - Add supported models table

Pull Request - State: closed - Opened by wheresmyhair 9 months ago

#838 - Add paired conversation dataset description

Pull Request - State: closed - Opened by wheresmyhair 9 months ago

#837 - Add github stale bot

Pull Request - State: closed - Opened by wheresmyhair 9 months ago

#836 - Reward modeling support

Pull Request - State: closed - Opened by wheresmyhair 9 months ago - 1 comment

#835 - Doc dataset page update

Pull Request - State: closed - Opened by wheresmyhair 9 months ago

#834 - Reward modeling support

Pull Request - State: closed - Opened by wheresmyhair 9 months ago

#833 - Discussion about LISA

Issue - State: open - Opened by caoshuai03 9 months ago - 2 comments

#832 - Add finetuning doc

Pull Request - State: closed - Opened by wheresmyhair 9 months ago

GitHub / OptimalScale/LMFlow issues and pull requests