PKU-YuanGroup/Video-LLaVA issues and pull requests

#192 - Can you Fix the DEMO. Demo is no longer working

Issue - State: open - Opened by thisurawz1 about 1 month ago

#191 - Can't reproduce results on MSRVTT and MSVD dataset

Issue - State: closed - Opened by 1999Lyd 2 months ago

#190 - Issues with Converting the video-llava Model to ONNX

Issue - State: closed - Opened by Ark1a 3 months ago

#189 - Pretrain and Finetune template versions

Issue - State: open - Opened by xin-li-67 3 months ago - 1 comment

#188 - When I evaluated the ‘TGIF_Zero_Shot_QA’ dataset, the accuracy was only 13%. Should I train first to achieve the 70% accuracy in the paper?

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

#187 - May I ask what is the api_base used for evaluation?

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

#186 - missing file: preprocessor_config.json

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

#185 - Can this model apply a few-shot when inference?

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

#184 - ImportError: cannot import name '_expand_mask' from 'transformers.models.clip.modeling_clip'

Issue - State: open - Opened by qiuchen001 4 months ago - 4 comments

#183 - Update llava_arch.py for device synchronization of mm_projector outputs

Pull Request - State: open - Opened by FangXinyu-0913 4 months ago

#182 - Valley video not found during pretraining.

Issue - State: open - Opened by Aakriti05 4 months ago

#181 - Api is not running properly getting errors In each endpoint

Issue - State: open - Opened by RAJA102002 4 months ago

#180 - Questions about LanguageBind Usage

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

#179 - Multi-GPU inference problem.

Issue - State: open - Opened by jiazheng-xing 4 months ago

#178 - Request for Inference Parameters on VideoLLava

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

#177 - Is it possible to train with languages other than English, and are the 8 frames sampled uniformly across different video lengths?

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

#176 - size mismatch

Issue - State: open - Opened by cs19469 4 months ago

#175 - error:RuntimeError: Error(s) in loading state_dict for CLIPVisionModel: size mismatch for vision_model.embeddings.class_embedding: copying a param with shape torch.Size([1024]) from checkpoint, the shape in current model is torch.Size([768]).

Issue - State: open - Opened by zapqqqwe 5 months ago - 3 comments

#174 - total_frames zero error

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

#173 - pretrained checkpoint

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

#172 - How to increase the sample frames amount

Issue - State: open - Opened by sherlock666 5 months ago

#171 - Issues with finetune_lora.sh

Issue - State: open - Opened by shag1802 5 months ago - 2 comments

#170 - Error with Gradio Client: Please Upgrade Gradio to 4.x and Redeploying HuggingFace Space

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

#169 - Question Regarding Video Frame Processing

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

#168 - How to install videollava together with xformer?

Issue - State: open - Opened by zengbohan0217 5 months ago

#167 - extremely slow with transformers

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

#166 - Training help

Issue - State: open - Opened by felmoreno1726 5 months ago

#165 - About class embedding

Issue - State: open - Opened by feiyu12138 5 months ago

#164 - Video-LLava Upgradation

Issue - State: open - Opened by Tortoise17 6 months ago - 1 comment

#163 - ERROR opening+moov atom not found+mmco: unref short failure

Issue - State: open - Opened by Frank-Dg 6 months ago - 2 comments

#162 - Can the confidence coefficient of an answer be obtained?

Issue - State: closed - Opened by IsabelJimenez99 6 months ago

#161 - Inference model path unclear

Issue - State: open - Opened by Ali2500 6 months ago

#160 - Fix typo and additional instruction in readme

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

#159 - Please specifiy library versions

Issue - State: open - Opened by nahidalam 6 months ago

#158 - Update README.md

Pull Request - State: open - Opened by OliverGrace 6 months ago

#157 - Uri validation issue on Replicate

Issue - State: closed - Opened by Gab1988 6 months ago - 1 comment

#156 - Video-LLaVa now available in the Transformers library!

Issue - State: open - Opened by zucchini-nlp 6 months ago - 52 comments

#155 - chore: Update processing_video.py

Pull Request - State: open - Opened by githubartema 6 months ago

#154 - The problem about the environment

Issue - State: closed - Opened by swiftCC 6 months ago

#153 - Some weights of the model checkpoint at "./Video-LLaVA-7B" were not used when initializing LlavaLlamaForCausalLM:

Issue - State: open - Opened by ssuncheol 6 months ago

#152 - Size mismatch error when running locally.

Issue - State: closed - Opened by ssuncheol 6 months ago - 3 comments

#151 - Problem about pretrain parameter dim size is differen to the model dim size?

Issue - State: open - Opened by NEC09818 6 months ago - 1 comment

#150 - how to load pretrained weight on local (offline)?

Issue - State: open - Opened by jusepv 6 months ago

#149 - Warnings about weights, temperature, top_p, and embedding layer, but it still works. Should I worry about them?

Issue - State: open - Opened by secretlycarl 6 months ago

#148 - Impossible to install on windows

Issue - State: open - Opened by secretlycarl 7 months ago

#147 - add-llama3

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

#146 - 推理多张图片时报错 IndexError: list index out of range

Issue - State: open - Opened by Qinger27 7 months ago - 1 comment

#145 - Multi-GPU inference enabled following LLaVA repo

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

#144 - 训练时报错AttributeError: 'DeepSpeedCPUAdam' object has no attribute 'ds_opt_adam'

Issue - State: open - Opened by Qinger27 7 months ago - 5 comments

#143 - Unable to install flash attn module

Issue - State: closed - Opened by anantalp 7 months ago - 1 comment

#142 - Seems it has very limited understanding ability..

Issue - State: open - Opened by advenTure423 7 months ago

#141 - 为什么loss一直为0

Issue - State: open - Opened by xienan0326 7 months ago

#140 - About contrastive learning

Issue - State: open - Opened by mjkmain 7 months ago

#139 - About the license

Issue - State: open - Opened by Pickpate 7 months ago

#138 - Similarity index and time - codes

Issue - State: open - Opened by Tortoise17 8 months ago

#137 - Tendency to keep repeating the same sentence

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

#136 - image / video with different MLPs

Issue - State: open - Opened by wangziyizy 8 months ago

#135 - Unable inference video locally.

Issue - State: open - Opened by Tiancheng-htc 8 months ago - 1 comment

#134 - Gradio APP upgrade

Issue - State: open - Opened by Kamakshi8104 8 months ago

#133 - How should I obtain llava_image_tune_.json, videochatgpt_tune_.json, and nlp_tune.json

Issue - State: open - Opened by laurenlong 8 months ago - 5 comments

#132 - Reproducibility. Can't install the project `pip install -e .`

Issue - State: open - Opened by felmoreno1726 8 months ago

#131 - Error when loading released model on huggingface

Issue - State: open - Opened by Leo-Yuyang 8 months ago

#130 - How should I obtain llava_image_tune_.json, videochatgpt_tune_.json, and nlp_tune.json

Issue - State: closed - Opened by Annie1900 8 months ago - 1 comment

#129 - More robust- retries & concurrent inputs

Pull Request - State: open - Opened by bschreck 8 months ago

#128 - ImportError: dlopen: cannot load any more object with static TLS

Issue - State: open - Opened by digdigdigging 8 months ago

#127 - finetune with lora

Issue - State: open - Opened by cm-xcju 8 months ago - 14 comments

#126 - We couldn't connect to 'https://huggingface.co' to load this file

Issue - State: closed - Opened by HolyDragon57 8 months ago - 1 comment

#125 - Tokenizer in different code version

Issue - State: closed - Opened by countytown 8 months ago

#124 - Is there any way to speed up the inference

Issue - State: open - Opened by Coronal-Halo 8 months ago - 4 comments

#123 - How to increase frame from 8 to a bigger number

Issue - State: open - Opened by tarunmis 8 months ago - 10 comments

#122 - How to set the config to use multi node and multi GPUS?

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

#121 - RuntimeError : Error(s) in loading state_dict for LlavaLlamaForCausalLM

Issue - State: open - Opened by zhangye0402 9 months ago - 2 comments

#120 - 解码器设置为opencv时，某些视频在推理时出现异常

Issue - State: open - Opened by Chenjm08 9 months ago

#119 - Transformer version

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

#118 - Huggingface data files - corrupted zip files?

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

#117 - 推理过程中出现错误

Issue - State: closed - Opened by zhangye0402 9 months ago - 3 comments

#116 - Invalid link to MoE-LLaVA

Issue - State: closed - Opened by DDVD233 9 months ago - 2 comments

#115 - Parameter Explanations

Issue - State: open - Opened by Davidyao99 9 months ago - 2 comments

#114 - video eval error

Issue - State: closed - Opened by wsxwd 9 months ago - 4 comments

#112 - Error while loading finetuned model for inferencing.

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

#109 - "RuntimeError: CUDA error: device-side assert triggered"

Issue - State: open - Opened by tarunmis 9 months ago - 6 comments

#108 - Inference on more than 8 frames

Issue - State: open - Opened by Nihel01 9 months ago - 2 comments

#107 - when training new model, I got stuck in the middle of the training

Issue - State: open - Opened by sunwhw 9 months ago - 3 comments

#105 - 推理 GPU bug

Issue - State: open - Opened by 827648313 9 months ago - 2 comments

#104 - 断言错误问题

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

#102 - Distributed Inference Doesn't Work

Issue - State: open - Opened by dfan 9 months ago - 10 comments

#100 - TypeError: 'NoneType' object is not callable

Issue - State: open - Opened by yumianhuli1 9 months ago - 3 comments

#99 - Finetuning with LORA

Issue - State: open - Opened by joycech333 9 months ago - 2 comments

#95 - 推理bug

Issue - State: open - Opened by 827648313 9 months ago - 12 comments

#90 - 有人遇到运行视频推理index out of bounds这个问题吗？

Issue - State: open - Opened by hotbbsun 10 months ago - 3 comments

#82 - Support multiple rounds of video conversations?

Issue - State: open - Opened by JiaweiZhao-git 10 months ago - 4 comments

#79 - Some installation suggestions of [pip install .e] for my homies

Issue - State: open - Opened by LIN-SHANG 10 months ago - 1 comment

GitHub / PKU-YuanGroup/Video-LLaVA issues and pull requests