Meituan-AutoML MobileVLM issues

Meituan-AutoML / MobileVLM

Strong and Open Vision Language Assistant for Mobile Devices

Apache License 2.0

1.04k stars 66 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

[Question]: Any Online Resources on how to fine tune MobileLLaMA-1.4B-Chat on custom dataset with LoRA ?

#62 mtami-zoor opened 4 days ago
0
[Question] What is the minimum size of an image that can be classified?

#61 ChangGiMoon opened 1 month ago
0
how to train on multi nodes with deepspeed (take "pre_train" for example)

#60 thunder95 closed 2 months ago
0
Model Misclassification (All Images Classified as NSFW)

#59 mehranfvs opened 3 months ago
0
Any demo on how to run it on mobile

#58 surajat17 opened 3 months ago
0
cli_video_demo.py 问题

#57 enkiGithub opened 3 months ago
0
Is this able to provide the position?

#56 husnoo opened 3 months ago
0
inference

#55 Arbor334 opened 3 months ago
2
Not being able to run an inference using llama.cpp

#54 Hardik-Choraria opened 4 months ago
0
What's torch_dtype used during evaluation?

#53 EasonXiao-888 closed 4 months ago
0
what is in the pretrain_data for mobileVLM2? @er-muyue

#52 xiyuanhao opened 6 months ago
2
where can I download mm_projector

#51 wuwu-C opened 6 months ago
0
running gguf Mobile VLM on Mac (metal)

#50 rezacopol opened 6 months ago
1
sft loss curve

#49 shufangxun opened 6 months ago
0
Interest in training a Phi-3 version?

#48 kinchahoy opened 6 months ago
0
pre-train dataset

#47 FabienCode closed 6 months ago
3
I'm curious that why MobileLLaMA 2.7B is faster then OpenLLaMA reported in v1 paper?

#46 XiaohuJoshua opened 7 months ago
2
fix v1&&v2 lora conflict

#45 sxu1997 closed 7 months ago
0
About the supply of mm_projector.bin

#44 Bobby-youngking opened 7 months ago
1
questions about fine tune

#43 Bobby-youngking opened 7 months ago
0
Warning about tokenizer

#42 xushilin1 opened 8 months ago
0
Results of Evaluation with MobileVLM_V2-1.7B on GQA/POPE/MME Dataset is Different from that in Paper

#41 ZiangWu-77 closed 6 months ago
1
FlashAttention1.0.9 error in 2080Ti

#40 miraclewkf opened 8 months ago
1
MobileVLM_V2-3B doesn't work on llama.cpp ?

#39 weizhou1991 opened 8 months ago
1
CUDA out of memory

#38 qazzombie opened 8 months ago
3
Can MobileVLM v2 receive multiple images at once？

#37 wuwu-C opened 8 months ago
2
请问支持量化推理吗？

#36 LianghuiGuo closed 8 months ago
1
A100推理为啥很慢呢？是我Inference姿势不对吗？

#35 LianghuiGuo closed 8 months ago
2
How to use our own dataset to test?

#34 QvQKing closed 8 months ago
2
请问推理支持batch跑吗？

#33 LianghuiGuo closed 8 months ago
1
有中文的readme没有？两张rtx A4000跑不起来，报错

#32 life2048 closed 8 months ago
2
Problem of Finetune

#31 QvQKing closed 8 months ago
2
mobilevlm_v2 with lora

#30 rangmiao closed 8 months ago
2
llama.c转换mobileLVM-V2报错

#29 shinerdeng opened 8 months ago
7
Where can I obtain the data used for MobileVLM v2 training?

#28 xushilin1 closed 8 months ago
2
Repetitive response problem with VLM

#27 Muhammad4hmed closed 8 months ago
11
Llama.cpp usage does not work with the latest v2-1.7B model

#26 l3utterfly closed 8 months ago
3
Update README.md

#25 eltociear closed 8 months ago
0
Hugging Face demo

#24 merveenoyan closed 8 months ago
2
clip部分 ldp 跟直接编码mlp的时间对比

#23 JennieGao-njust closed 8 months ago
3
any lora for stage 2?

#22 shawnricecake closed 8 months ago
2
Update README.md

#21 YangYang-DLUT closed 9 months ago
0
请问支持中文推理/中文数据微调吗？

#20 LianghuiGuo opened 9 months ago
1
Is it possible to provide pretrained mm_projector.bin

#19 KP-Zhang closed 8 months ago
2
Significant Performance Drop in Model Replication

#18 rangmiao closed 10 months ago
6
funtune with lora error

#17 BadUncleBoy opened 10 months ago
8
Does it support Chinese instruction for mobileVLM?

#16 ghoshaw closed 10 months ago
1
Some question on selection device and card

#15 alone4u1314 closed 9 months ago
3
'get_tensor' error occurs during inference on mobile devices

#14 wwqq closed 10 months ago
2
关于手机上运行时间的求教

#13 JennieGao-njust closed 8 months ago
12