luogen1996 LaVIN issues

luogen1996 / LaVIN

[NeurIPS 2023] Official implementations of "Cheap and Quick: Efficient Vision-Language Instruction Tuning for Large Language Models"

502 stars 36 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

COCO Caption

#43 liangyi-qianwan opened 4 months ago
0
Any suggestions to reduce the training epochs? like tuning lr or something else?

#42 baiyuting opened 5 months ago
0
Where to get 15-eph-pretrain.pth to run the demo?

#41 YatingPan opened 8 months ago
1
lavin-7B结果达不到，两次结果差别很大，一样的代码

#40 yanghu819 opened 8 months ago
8
License

#39 RobertBiehl opened 9 months ago
0
提供一个用于 mme 评估的 shell 脚本？

#38 baiyuting opened 9 months ago
0
3090 24G

#37 hsb1995 opened 9 months ago
0
Regarding the tokenizer.model in /data/weights dir

#36 wanboyang opened 10 months ago
1
I would like to execute "gradio_client" reference

#35 choibigo opened 10 months ago
0
why starting the "input" in base_prompt.py with two "\n"

#34 moured opened 10 months ago
1
when eval the same checkpoint, different batch size get different result, how to handle this problem?

#33 baiyuting opened 10 months ago
0
Gradient Is Not Accumulated?

#32 HenryPengZou opened 11 months ago
1
vicuna 7b 版本怎么选择？

#31 baiyuting opened 11 months ago
0
Do you support llama2 now？

#30 hangzeli05 opened 11 months ago
2
About the 4bit quantification

#29 Pefect96 opened 1 year ago
1
关于repAdapter_Router 代码的疑问

#28 verigle opened 1 year ago
2
Beginner cuestion

#27 mauridev777 opened 1 year ago
1
目前7B单卡精度略低于论文值，请问可能的原因

#25 verigle opened 1 year ago
2
update requirements.txt and add docker file, format code

#24 yzbx opened 1 year ago
0
A sample docker and update for requirments.txt

#23 yzbx opened 1 year ago
0
How to continue fine-tuning training on the basis of your pre training fine-tuning nodes【LaVIN-7B】or 【LaVIN-13B】

#22 hangzeli08 opened 1 year ago
2
Questions about training related to NCCL

#21 air-tea opened 1 year ago
3
Questions about environment variables

#20 anshkumar closed 1 year ago
0
Figure 2 questions about Adapter

#19 HYUNMIN-HWANG opened 1 year ago
0
[BUG] when NaN loss encountered->zeroDivisionError: float division by zero

#18 verigle opened 1 year ago
1
Incorrect path used to load tokenizer in eval.py

#17 angelala3252 opened 1 year ago
0
[data]请问在MME benchmark上进行评测的模型是只使用了instrcution数据（约210k）训练的嘛？

#16 aopolin-lv closed 1 year ago
2
7B lite 单卡没有达到报的点数，请教可能的原因。

#15 yanghu819 closed 1 year ago
4
Demo可以在24G的显卡上运行吗，爆显存了

#14 wooozihui opened 1 year ago
1
Could you please share the code to generate the instruct data?

#13 aixiaodewugege opened 1 year ago
0
4-bit multi-gpu training

#12 Yingyue-L opened 1 year ago
1
data 文件夹下 captions.json 文件如何获取？

#11 verigle closed 1 year ago
1
关于Repadapter的位置问题

#10 YuyingShang opened 1 year ago
1
请问官方的 sqa-llama-7b-lite.pth 用的训练参数是什么呀？我这边复现的时候差距还挺大的。是卡不够多还是？

#9 CheungZeeCn closed 1 year ago
4
What does the parameter "max_seq_len" do exactly?

#8 WizardMx closed 1 year ago
1
请问有考虑用 lora来替换部分adapter吗？

#7 CheungZeeCn closed 1 year ago
2
GPT-4 evaluation

#6 WizardMx closed 1 year ago
1
LaVin-7B model on task Multimodal ChatBot

#5 Dicomsky closed 1 year ago
3
A little confusion of Figure 2

#4 WizardMx closed 1 year ago
4
typo in your citation~

#3 WizardMx closed 1 year ago
3
torch.distributed.elastic.multiprocessing.api: failed exitcode 1 local rand 1

#2 Yingyue-L closed 1 year ago
12
Computing output likelihoods with the model

#1 vishaal27 opened 1 year ago
1