InternVL

InternVL copied to clipboard

Questions about your model in video-mme

1

I noticed your latest good results on video-mme (https://video-mme.github.io/home_page.html#leaderboard), ranking 9th, the parameter size is 20B, the number of image frames is 10 frames, you also announced this good result...

zmj1203

Thank you for your excellent job! Another question: since the InternVL2 blog does not mention any updates to the ViT models (InternViT-300M-448px / InternViT-6B-448px-V1-5), I wonder if you have made...

kekethu

DummyScheduler issue while fine-tuning with LoRA

3

我按照以下的文档进行 fine-tune the LoRA ，遇到DummyScheduler的错误，请问应该怎么fix？ https://github.com/OpenGVLab/InternVL/blob/main/document/how_to_finetune_internvl_chat_v1_2_on_a_custom_dataset.md 运行的命令： CUDA_VISIBLE_DEVICES=0,1 sh shell/hermes2_yi34b/internvl_chat_v1_2_hermes2_yi34b_448_res_finetune_continue_lora.sh 遇到的问题： File "/root/miniconda3/lib/python3.9/site-packages/accelerate/accelerator.py", line 1468, in _prepare_deepspeed raise ValueError( ValueError: You cannot create a `DummyScheduler` without specifying a scheduler in...

ff151

InternVL2-26B模型微调后，报UnboundLocalError: local variable 'head_num' referenced before assignment错误

9

 InternVL2-26B模型微调后，报UnboundLocalError: local variable 'head_num' referenced before assignment错误

feihuamantian

safetensors_rust.SafetensorError: Error while deserializing header: HeaderTooLarge

2

/root/miniconda3/envs/internvl/lib/python3.9/site-packages/transformers/generation/configuration_utils.py:397: UserWarning: `do_sample` is set to `False`. However, `top_p` is set to `None` -- this flag is only used in sample-based generation modes. You should set `do_sample=True` or unset `top_p`....

yangtao-git

[Bug] load video frames bug.

1

### Checklist - [X] 1. I have searched related issues but cannot get the expected help. - [X] 2. The bug has not been fixed in the latest version. -...

xmy0916

[Bug] fused_adam问题

3

### Checklist - [ ] 1. I have searched related issues but cannot get the expected help. - [ ] 2. The bug has not been fixed in the latest...

1518630367

Device errors running the multi-gpu demo on huggingface for InternVL2-76B

3

### Checklist - [X] 1. I have searched related issues but cannot get the expected help. - [X] 2. The bug has not been fixed in the latest version. -...

ThibaultGROUEIX

Model returns nonsensical output with AWQ format

2

### Checklist - [X] 1. I have searched related issues but cannot get the expected help. - [X] 2. The bug has not been fixed in the latest version. -...

pykeras

识别图片出现 输出结果 乱码问题

4

https://huggingface.co/OpenGVLab/InternVL-Chat-V1-5 下载上述文件，按照要求，下载transformers==4.37.2的版本，运行Model Usage中的代码，输出结果乱码

xiaotaozi121096