dvlab-research MGM issues

dvlab-research / MGM

Official repo for "Mini-Gemini: Mining the Potential of Multi-modality Vision Language Models"

Apache License 2.0

3.19k stars 278 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

article error

#136 TuuSiwei opened 1 week ago
0
Unable to Merge LoRA Weights with Base Model: ValueError: Can't find 'adapter_config.json' at ...

#135 PARSA-MHMDI opened 3 weeks ago
0
CHAIR Evaluation

#134 itsqyh opened 2 months ago
0
TypeError: MGMllamaForCausalLM.forward() got an unexpected keyword argument 'cache_position'

#133 czm708033 opened 2 months ago
2
ImportError: cannot import name 'packaging' from 'pkg_resources'

#132 LiuRicky closed 2 months ago
1
when use stable-diffusion,AttributeError: 'NoneType' object has no attribute 'tokenize'

#131 ALR-alr opened 3 months ago
1
llama3 result is repeated many times

#130 pennypengpm opened 3 months ago
1
Does MGM support in-context(few-shot) inference?

#129 waltonfuture opened 3 months ago
0
Will there be support for Qwen2?

#128 huxian0402 opened 3 months ago
0
How to access hidden states?

#127 Divyanshsingh1910 opened 3 months ago
1
I get this error: WARNING: tokenization mismatch: 156 vs. 161. (ignored) when I finetune llama3

#126 shidingz opened 3 months ago
1
使用多gpu启动worker，对话时报错

#125 kimi360 closed 3 months ago
1
Do you meet the error "MGMConfig"?

#124 strawberryrs620 opened 4 months ago
0
loss 0 and grad nan

#123 TuuSiwei closed 4 months ago
3
error in loading

#122 TuuSiwei closed 4 months ago
0
Requirement for pretraining weights of LLaMa-3-8B-Instruct

#121 shiwk23 opened 4 months ago
0
Can provide laion-gpt4v dataset images zip?

#120 TuuSiwei opened 4 months ago
0
mgm-34b-hd, should have a 'model_type' key in its config.json

#119 chrisx599 opened 4 months ago
2
Inference problem about the demo.

#118 ApolloRay opened 4 months ago
1
The data for alignment and finetuning contains duplicates. Can you please explain why this is happening?

#117 KANGRuipeng opened 4 months ago
1
dataset miss problem

#116 TuuSiwei closed 4 months ago
1
| EORROR | stderr | RecursionError: Maximun recursion depth exceeded in comparison

#115 linyf38 opened 4 months ago
1
May I ask if the current inference code does not support multi images input

#114 Angelalilyer opened 4 months ago
1
多轮对话修改图像输入后报错

#113 pennypengpm opened 4 months ago
0
Generation-related Instructions dataset link

#112 berry-ding opened 4 months ago
1
关于多机多卡效果不如单机多卡好的问题

#111 DePengW opened 4 months ago
1
Loss does not decrease

#110 yfthu closed 4 months ago
0
可以放一下生成generation_pure_text数据的代码吗

#109 pennypengpm closed 4 months ago
4
LLama 70B support

#108 PrateekPal641 opened 5 months ago
0
Inference speed

#107 PrateekPal641 opened 5 months ago
0
lora initialisation missing from builder.py

#106 adrielkuek opened 5 months ago
1
Error while loading model with transformers library

#105 PrateekPal641 closed 5 months ago
3
Congratulations for the best LLaVA derived models !

#104 deepbeepmeep opened 5 months ago
1
Some weights of the model checkpoint were not used when initializing MGMLlamaForCausalLM

#103 charlesCXK opened 5 months ago
2
how to use stage2 ckpt fine-tuning stage3？

#102 linqinguang opened 5 months ago
1
计划加入DPO训练来缓解模型幻觉问题吗

#101 jiezhangGt opened 5 months ago
0
Take input image as condition.

#100 Adenialzz closed 5 months ago
2
How to fix [NETWORK ERROR DUE TO HIGH TRAFFIC. ] on MacOS ?

#99 seasoncool opened 5 months ago
2
stage2 loss is 0

#98 jiezhangGt closed 5 months ago
1
Excessive Length of Responses from Mini Gemini

#97 Dopplenum closed 5 months ago
1
AttributeError: 'OpenCLIPVisionTower' object has no attribute 'device'

#96 l1019008146 opened 5 months ago
2
Use of ocr in Evaluation

#95 bruceisme opened 5 months ago
1
'LlamaForCausalLM' object has no attribute 'get_vision_tower'

#94 HongLouyemeng closed 5 months ago
1
使用cli调用自定义微调模型，出现'OpenCLIPVisionTower' object has no attribute 'device'

#93 HongLouyemeng closed 5 months ago
16
pretrain error: lack of preprocessor_config.json

#92 jiezhangGt closed 5 months ago
1
Which deepspeed version is it

#91 Kareneveve opened 5 months ago
2
当我使用推理命令的时候出现网络错误，无法构建推理的接口

#90 HongLouyemeng closed 5 months ago
2
请问为什么在训练llama的脚本中，预训练和微调所使用的conv不一样

#89 shidingz opened 5 months ago
1
model asks self questions and answers

#88 Bowei-Li opened 5 months ago
2
为什么输出结果为nan呢

#87 freja-zy closed 5 months ago
1