Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / haotian-liu/LLaVA issues and pull requests
#1705 - [Question] 想学习【MLLM多模态】快加学习交流群
Issue -
State: open - Opened by km1994 3 days ago
#1704 - [Question] What is the optimizer of LLaVA?
Issue -
State: open - Opened by 20191864218 4 days ago
#1703 - How to judge whether we train the vision tower during our lora-fine tuning
Issue -
State: open - Opened by lmingze 7 days ago
#1702 - what is the difference between liuhaotian/llava-v1.5-7b and (vicuna-7b-v1.5 + vision tower + mm_projector)
Issue -
State: open - Opened by lmingze 7 days ago
#1701 - [Question] Unable to generate pydantic-core schema for <class 'starlette.requests.Request'>. Set `arbitrary_types_allowed=True`
Issue -
State: open - Opened by sxchen123 8 days ago
- 1 comment
#1700 - [Question] May I ask if there is a 7B-sized file of Llama_2_7b_chat?
Issue -
State: open - Opened by xlnn 8 days ago
#1699 - [Question] if not vision_tower.is_loaded: AttributeError: 'NoneType' object has no attribute 'is_loaded'
Issue -
State: open - Opened by shen1005 10 days ago
#1698 - [Usage] The difference between finetune_lora.sh and finetune_task_lora.sh
Issue -
State: open - Opened by PixelChen24 10 days ago
#1697 - [Question] Why the link of LAION/CC/SBU BLIP-Caption Concept-balanced 558K Meta Data(meta.json) is empty?
Issue -
State: open - Opened by Liuqibaa 10 days ago
#1696 - [Question] What base python version should `LLaVA` use ? like `3.10` ,`3.11` ?
Issue -
State: closed - Opened by q2333gh 11 days ago
- 2 comments
#1695 - [Question] Why is the output always numerical when using model inference, like this
Issue -
State: open - Opened by yuese1234 12 days ago
#1694 - Can I load the parameters of llava1.5 with full parameter fine-tuning
Issue -
State: open - Opened by zhangzef 12 days ago
#1693 - [Usage] can I load
Issue -
State: closed - Opened by zhangzef 12 days ago
#1692 - Error while fine-tuning llava-v1.6 using finetune_task_lora.sh:DeepSpeed Zero-3 is not compatible with `low_cpu_mem_usage=True` or with passing a `device_map`.
Issue -
State: open - Opened by LuYinMiao 13 days ago
#1691 - [Question] How can I this truble when i use LLaVa_Med?
Issue -
State: open - Opened by TimoFan1998 15 days ago
#1690 - [Question] A series of questions about fine tuning , I want to learn this stuff
Issue -
State: open - Opened by bdv29 15 days ago
#1688 - Enhance error handling and optimize file handling
Pull Request -
State: open - Opened by mandlinsarah 16 days ago
#1687 - OSError: Error no file named pytorch_model.bin, tf_model.h5, model.ckpt.index or flax_model.msgpack found in directory ./checkpoints/vicuna-7b-v1.5.
Issue -
State: open - Opened by Leo-Lsc 16 days ago
#1686 - [Question] Why is the accuracy low when I evaluate llava-v1.5-7b-lora on VQAv2 ?
Issue -
State: open - Opened by tanghao2118 18 days ago
- 1 comment
#1685 - [Question] Any ideas on how to design a customized loss function for LLaVA and integrate several new layers that would be compatible with it?
Issue -
State: open - Opened by powenlo 19 days ago
#1683 - [Question] how much vram for batch 1 full fine tune?
Issue -
State: open - Opened by nub2927 20 days ago
#1682 - [Question] 歪曲了历史,数据源精美精西
Issue -
State: open - Opened by wushh 21 days ago
- 3 comments
#1681 - [Usage] Adding very few parameters when using LoRA to finetune LLaVA 1.5
Issue -
State: open - Opened by XiaoruiMaLU 21 days ago
#1680 - [Question] Custom Conversations
Issue -
State: open - Opened by KansaiUser 22 days ago
#1679 - [Usage] Some weights of LlavaLlamaForCausalLM were not initialized from the model checkpoint
Issue -
State: open - Opened by RobitsG 22 days ago
- 2 comments
#1678 - [Usage] Can you release the code for concept balancing filtering in pretrain data?
Issue -
State: open - Opened by williamium3000 22 days ago
#1677 - [Question] why model generate takes more time the second time?
Issue -
State: open - Opened by KansaiTraining 23 days ago
#1676 - [Question] Is there a parameter to limit the amount of text output, or is the output more streamlined
Issue -
State: open - Opened by zou-yawen 23 days ago
#1675 - [Question] How can I use CLIPVisionTowerS2?
Issue -
State: closed - Opened by dongbinShin96 24 days ago
#1674 - [Question] How to get output embeddings or last hidden states?
Issue -
State: open - Opened by flab305 26 days ago
#1674 - [Question] How to get output embeddings or last hidden states?
Issue -
State: open - Opened by flab305 26 days ago
#1673 - [Question] Understanding the Licensing and Commercial Use of LLaVa-NeXT
Issue -
State: open - Opened by sayanbiswas59 27 days ago
#1673 - [Question] Understanding the Licensing and Commercial Use of LLaVa-NeXT
Issue -
State: open - Opened by sayanbiswas59 27 days ago
#1672 - [Question] The difference between llava ckpt and llava_hf ckpt?
Issue -
State: open - Opened by hxhcreate 28 days ago
#1670 - [Usage] pretrain about "gradient_accumulation_steps" para use
Issue -
State: closed - Opened by wanlipeng 29 days ago
#1669 - [Feature request] Add better support for Brazilian Portuguese
Issue -
State: open - Opened by insinfo 29 days ago
- 1 comment
#1668 - [Usage] about fine tuning
Issue -
State: open - Opened by liucheny 30 days ago
#1667 - [Usage] From Hugging face llava gives no response
Issue -
State: open - Opened by MonoMarkor about 1 month ago
#1666 - [Usage] Error - `ollama._types.ResponseError: error parsing llm response stream: error: {"content":"internal_error"}`
Issue -
State: open - Opened by laksh-2193 about 1 month ago
- 1 comment
#1665 - [Question] How to correctly write a prompt for this model?
Issue -
State: open - Opened by KansaiTraining about 1 month ago
#1665 - [Question] How to correctly write a prompt for this model?
Issue -
State: open - Opened by KansaiTraining about 1 month ago
#1664 - [Question] evaluation problem in textVQA
Issue -
State: open - Opened by jiinhui about 1 month ago
#1664 - [Question] evaluation problem in textVQA
Issue -
State: open - Opened by jiinhui about 1 month ago
#1663 - [Usage] How do i get a LLaVA model converted to contain vision tower related config items?
Issue -
State: closed - Opened by BountyMage about 1 month ago
- 2 comments
#1662 - [Question] Support for Multi-Image Input in One-Shot/Few-Shot Learning Scenarios
Issue -
State: open - Opened by vedernikovphoto about 1 month ago
- 1 comment
#1661 - [Question] 单卡训练正常,多卡训练loss == 0
Issue -
State: open - Opened by Camellia-hz about 1 month ago
- 2 comments
#1660 - Dependency errors
Issue -
State: open - Opened by Chanete about 1 month ago
- 1 comment
#1659 - [Question] The pure text process in function "prepare_inputs_labels_for_multimodal"
Issue -
State: open - Opened by zhanglixuan0720 about 1 month ago
- 1 comment
#1658 - [Question] When finetuning with my custom dataset, do I need to mix the custom datasets and original LLaVA instruction datasets?
Issue -
State: open - Opened by zy1996829 about 1 month ago
#1657 - [Usage] Out of bounds error
Issue -
State: open - Opened by happywinder about 1 month ago
#1656 - comment out torch requirement
Pull Request -
State: closed - Opened by nguyendinhson-kaist about 1 month ago
#1656 - comment out torch requirement
Pull Request -
State: closed - Opened by nguyendinhson-kaist about 1 month ago
#1655 - [Question]detailed information about tokenizer and encoder
Issue -
State: open - Opened by dnkscu about 1 month ago
#1654 - [Usage] why the demo of LLaVA-1.5 can not use now?
Issue -
State: open - Opened by moonnnpie about 1 month ago
#1654 - [Usage] why the demo of LLaVA-1.5 can not use now?
Issue -
State: open - Opened by moonnnpie about 1 month ago
#1653 - [Question] Could you tell me any difference between llava-v1.5-13b.jsonl and old llava-v1.5-7b.jsonl?
Issue -
State: open - Opened by yukio0321 about 1 month ago
#1651 - [Usage] LLaVA-v1.6-vicuna-7b generates incomplete sentence
Issue -
State: closed - Opened by gymbeijing about 1 month ago
- 1 comment
#1650 - [Question] Learning is completed, but only the weights of the projector are output.
Issue -
State: closed - Opened by kouyakamada about 1 month ago
#1649 - Why are the first image and the remaining images processed separately? in prepare_inputs_labels_for_multimodal()
Issue -
State: open - Opened by shure-dev about 1 month ago
#1648 - llava 1.2.2.post1 requires torch==2.1.2, but you have torch 2.0.1 which is incompatible.
Issue -
State: open - Opened by Mike-ihr about 1 month ago
- 3 comments
#1647 - [Usage] Freezing Vision Encoder During LoRA Training
Issue -
State: open - Opened by xie-qiang about 1 month ago
#1646 - You are using a model of type llava to instantiate a model of type llava_llama. This is not supported for all configurations of models and can yield errors.
Issue -
State: open - Opened by XCF-Mike about 1 month ago
- 1 comment
#1645 - [Question] Inference Code
Issue -
State: closed - Opened by bryanwong17 about 2 months ago
#1644 - [Question] Misaligned image_grid_pinpoints
Issue -
State: open - Opened by Forence1999 about 2 months ago
#1643 - about multi-image input
Issue -
State: open - Opened by eternal8080 about 2 months ago
#1642 - [Question] Image and Text Embeddings For Downstream Tasks
Issue -
State: open - Opened by dipikakhullar about 2 months ago
- 1 comment
#1641 - load_dataset('liuhaotian/LLaVA-Instruct-150K') ERROR
Issue -
State: open - Opened by YerongLi about 2 months ago
#1639 - [Usage] KeyError: 'LlavaMistralConfig' when I start a model-worker with a fine-tuned Llava-v1.6-mistral-7B model.
Issue -
State: open - Opened by yunsaijc about 2 months ago
#1638 - Error when load model in 4bit
Issue -
State: open - Opened by rin2401 about 2 months ago
- 1 comment
#1637 - [Question] Fine-tuned model ignore some of the captions
Issue -
State: open - Opened by AbdulrahmanSoliman1 about 2 months ago
#1636 - How can I custom loss function during fine-tuning?[Question]
Issue -
State: open - Opened by qingyunyanran about 2 months ago
- 4 comments
#1635 - Error when saving model: Invalid generation config due to conflicting parameters
Issue -
State: closed - Opened by ohhan777 about 2 months ago
- 5 comments
#1634 - NETWORK ERROR DUE TO HIGH TRAFFIC. PLEASE REGENERATE OR REFRESH THIS PAGE.[Usage]
Issue -
State: open - Opened by sivang about 2 months ago
#1633 - Can this model be adapted to torch2.3?
Issue -
State: open - Opened by zxy1728 about 2 months ago
#1632 - [Question] About how to deal with multiple-answer question in vqav2
Issue -
State: open - Opened by BroJunn about 2 months ago
#1632 - [Question] About how to deal with multiple-answer question in vqav2
Issue -
State: open - Opened by BroJunn about 2 months ago
#1630 - [Usage] Visual instruction tuning for LLaVa 1.6
Issue -
State: open - Opened by mattia-re-learn about 2 months ago
- 6 comments
#1628 - 拉了一个多模态大模型技术交流群,大家可以加入进来进行技术交流
Issue -
State: open - Opened by feihuamantian about 2 months ago
- 3 comments
#1625 - [Usage] There is no output from the ASSISTANT when using CLI Inference????
Issue -
State: open - Opened by xlxcomputer about 2 months ago
- 1 comment
#1624 - [Usage] from llava.mm_utils import ( ImportError: cannot import name 'load_pretrained_model' from 'llava.mm_utils'
Issue -
State: closed - Opened by seven112233 about 2 months ago
- 1 comment
#1622 - [Usage] You are using a model of type llava to instantiate a model of type llava_llama. This is not supported for all configurations of models and can yield errors.
Issue -
State: open - Opened by matengxiaotiancai 2 months ago
- 2 comments
#1619 - [Question] "Size mismatch" Error when finetuning from a projector
Issue -
State: open - Opened by hvgupta 2 months ago
- 3 comments
#1618 - Some images of ocr vqa data in llava_v1_5_mix665k.json do not exist!
Issue -
State: open - Opened by Vicent0205 2 months ago
- 5 comments
#1613 - [Questio why 'mm_vision_select_layer' == -2 in config ? n]
Issue -
State: open - Opened by fmy7834 2 months ago
- 2 comments
#1610 - [Question] 想换掉Vicuna 模型
Issue -
State: open - Opened by wanglongpeng1 2 months ago
- 1 comment
#1610 - [Question] 想换掉Vicuna 模型
Issue -
State: open - Opened by wanglongpeng1 2 months ago
- 1 comment
#1603 - Pretraining LLaVA with SigLIP
Pull Request -
State: open - Opened by nahidalam 2 months ago
- 2 comments
#1600 - Need either a `state_dict` or a `save_folder` containing offloaded weights.
Issue -
State: open - Opened by zhaodaojie 2 months ago
- 1 comment
#1597 - [Usage] how to load the trained lora?
Issue -
State: closed - Opened by feiyangsuo 2 months ago
- 4 comments
#1582 - device mis-match error on pre-training
Issue -
State: open - Opened by oroojlooy 3 months ago
- 3 comments
#1582 - device mis-match error on pre-training
Issue -
State: open - Opened by oroojlooy 3 months ago
- 1 comment
#1580 - [Question] Mac run but get importlib.metadata.PackageNotFoundError: No package metadata was found for bitsandbytes
Issue -
State: open - Opened by Timothychen00 3 months ago
- 4 comments
#1578 - [Question] Why LLaVA-1.5-13b keep repeatting answer in the first round conversation?
Issue -
State: open - Opened by PoopBear1 3 months ago
- 3 comments
#1573 - add qwen2 support for pretraining and finetuning
Pull Request -
State: open - Opened by TobyYang7 3 months ago
- 5 comments
#1568 - "Assertion `srcIndex < srcSelectDimSize` failed" in Docker on some systems
Issue -
State: closed - Opened by Careiner 3 months ago
- 3 comments
#1568 - "Assertion `srcIndex < srcSelectDimSize` failed" in Docker on some systems
Issue -
State: closed - Opened by Careiner 3 months ago
- 3 comments
#1567 - Having issues while merging LoRA attention weights
Issue -
State: closed - Opened by anas-zafar 3 months ago
- 4 comments
#1554 - [Usage] Not able to fine tune the LLaVA model with llava-v1.5-7b.
Issue -
State: open - Opened by ayushgupta9198 3 months ago
- 6 comments
#1552 - [Usage] Merging LoRa weights into llava-13b fails with bizarre error
Issue -
State: open - Opened by maxall41 3 months ago
- 3 comments
#1551 - [Discussion] Request for Guidance on Stage Two: Converting LLaVA-V1.6 into 4-bit GGUF Format (PAPER)
Issue -
State: open - Opened by rohithbojja 3 months ago
- 2 comments