Ecosyste.ms: Issues

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

GitHub / dvlab-research/MGM issues and pull requests

#35 - Can provide laion-gpt4v dataset images zip?

Issue - State: open - Opened by lucasjinreal 7 months ago - 4 comments

#33 - 关于token数量的疑问

Issue - State: closed - Opened by DimplesL 7 months ago - 5 comments

#32 - batch inference

Issue - State: closed - Opened by mukundkhanna123 7 months ago - 2 comments

#31 - Mini-Gemini-2B evaluation error

Issue - State: closed - Opened by laserwave 7 months ago - 2 comments

#30 - size mismatch about convnext_large_d_320

Issue - State: closed - Opened by cyy-1234 7 months ago - 3 comments

#29 - Could you offer LAION GPT4V images

Issue - State: open - Opened by daicver 7 months ago - 4 comments

#28 - Training on customized LLM loss got sort of high

Issue - State: closed - Opened by lucasjinreal 7 months ago - 6 comments

#27 - Demo page is not working

Issue - State: closed - Opened by 09u2h4n 7 months ago

#26 - No image generated while run python -m minigemini.serve.cli --gen

Issue - State: closed - Opened by Xiaolong-RRL 7 months ago - 2 comments

#25 - Fixing the pretrain and sft stage ALLaVa images issue.

Issue - State: closed - Opened by lucasjinreal 8 months ago - 2 comments

#24 - How to train 34B model?

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

#23 - datasets preparation

Issue - State: closed - Opened by KevinPanJun 8 months ago - 2 comments

#22 - evaluate MMMU_val of 2B error

Issue - State: closed - Opened by lalulxm 8 months ago - 2 comments

#21 - Not find vision tower: model_zoo/OpenAI/clip-vit-large-patch14-336

Issue - State: closed - Opened by wangzijian1010 8 months ago - 3 comments

#20 - Pretrain data not found in AllaVA

Issue - State: closed - Opened by lucasjinreal 8 months ago - 16 comments

#19 - Where is the data of Generation-related Instructions in Section 3.3?

Issue - State: closed - Opened by jiashenggu 8 months ago - 5 comments

#18 - performing finetune on top of mini-gemini-8x7b-HD

Issue - State: closed - Opened by lightmatmul 8 months ago - 8 comments

#16 - Update README.md

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

#15 - Similarities to LLaVA-HR

Issue - State: closed - Opened by luogen1996 8 months ago - 2 comments

#14 - issues running cli inference example

Issue - State: closed - Opened by lightmatmul 8 months ago - 2 comments

#13 - 3 gradio issues

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

#12 - Minor issue of the HR feature map size

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

#11 - CLI of 2B does not work

Issue - State: closed - Opened by yiranyyu 8 months ago - 3 comments

#10 - Limitations on current image generating method

Issue - State: closed - Opened by lucasjinreal 8 months ago - 8 comments

#9 - 4 bit loading fails

Issue - State: closed - Opened by lucyknada 8 months ago - 2 comments

#8 - Begin training loss a little high

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

#7 - AutoTokenizer resolve error

Issue - State: closed - Opened by yiranyyu 8 months ago - 4 comments

#6 - Training model got some error

Issue - State: closed - Opened by lucasjinreal 8 months ago - 2 comments

#5 - Possible positional embedding in Patch Info Mining process

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

#4 - images_aux encode error

Issue - State: closed - Opened by duisuangsheng 8 months ago - 3 comments

#3 - Demo inference wrong?

Issue - State: closed - Opened by lucasjinreal 8 months ago - 3 comments

#2 - Why does any stage didn't optimize vision encoder?

Issue - State: closed - Opened by lucasjinreal 8 months ago - 5 comments

#1 - any idea coping with long video input?

Issue - State: closed - Opened by dragen1860 8 months ago - 1 comment