TideDra VL-RLHF issues - Githubissues

TideDra / VL-RLHF

A RLHF Infrastructure for Vision-Language Models

Apache License 2.0

100 stars 6 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Does this repo support Qwen2VL?

#17 w-zhih opened 1 week ago
0
DPO训练

#16 XxxZzD closed 1 month ago
0
每个模型的学习率设定有什么经验吗？

#15 shipengai closed 2 months ago
3
QwenVL使用了多少卡？global batch size是多少？

#14 shipengai closed 2 months ago
1
[BUG] Value Error noqa:E501

#13 hxhcreate opened 3 months ago
0
支持cogvlm2模型的强化学习训练吗

#12 kaka-Cao opened 3 months ago
0
Any suggestion on how to modify code to train single textual modality

#11 hxhcreate closed 3 months ago
0
微调qwen爆内存

#10 delian11 opened 4 months ago
3
Reproduction of InternLM-XComposer2

#9 ikodoh opened 5 months ago
1
不使用lora报错

#8 yuzeng0-0 opened 5 months ago
1
微调internXC2报错

#7 yuzeng0-0 opened 5 months ago
9
微调LLaVA报错

#6 njucckevin opened 5 months ago
4
Support for InstructBlipPPOTrainer

#5 NoyHanan opened 5 months ago
0
Support for SFT InternLM-XComposer2?

#4 ZhihaoAIRobotic closed 5 months ago
1
请问支持对internvl-1-5的微调吗？如果可以的话显存应该预留多少

#3 sunzx8 opened 5 months ago
2
strip may break chat template

#2 TideDra closed 7 months ago
0
use mmGD as detector

#1 TideDra closed 8 months ago
0