Skip to content

Issues: TinyLLaVA/TinyLLaVA_Factory

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Clear current search query, filters, and sorts
38 Open 43 Closed
38 Open 43 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

Checkpoints using base method and LLaVA dataset
#11 opened Feb 29, 2024 by G-JWLee updated Mar 1, 2024
1
what the difference between v1 and v1.1?
#19 opened Mar 11, 2024 by Eric-is-good updated Mar 11, 2024
1
Use merge_lora_weights.py Error:AttributeError: 'TinyLlavaPhiConfig' object has no attribute 'attention_bias'
#24 opened Mar 15, 2024 by mengjiexu updated Mar 15, 2024
3
请问phi-2是像llava-phi那样预先在文本指令上微调过的吗，还是直接拿基座模型连接视觉编码器进行预训练呢？
#33 opened Mar 20, 2024 by Yang-bug-star updated Mar 20, 2024
1
Did you train the whole LLM in the pretraining stage of share recipe？
#34 opened Mar 20, 2024 by Yang-bug-star updated Mar 20, 2024
3
Is it possible to pretrain tinyllama-3b on 2 V100s ?
#37 opened Mar 22, 2024 by Yang-bug-star updated Mar 22, 2024
1
Need training script
#16 opened Mar 7, 2024 by dddraxxx updated Mar 23, 2024
4
eval problem
#39 opened Mar 23, 2024 by UestcJay updated Mar 23, 2024
1
SigLIP image processing flow
#41 opened Mar 28, 2024 by eternalding updated Mar 28, 2024
Data download
#43 opened Apr 2, 2024 by wangfengjuan updated Apr 3, 2024
3
请问本代码能否在单卡4090上做微调呢？最少显存占用多少
#46 opened Apr 8, 2024 by hby1997 updated Apr 8, 2024
1
Able to merge 1.5B model, but unable to run eval
#50 opened Apr 17, 2024 by tanveer-sayyed updated Apr 19, 2024
6
请问单卡4090能对TinyLLaVA-1.5B进行增量预训练吗？
#51 opened Apr 22, 2024 by Lyzz123 updated Apr 26, 2024
4
Any plans on training a new model based on Phi-3?
#52 opened Apr 26, 2024 by TheBobbyliu updated Apr 28, 2024
1
Loss
#53 opened May 3, 2024 by Yu-xm updated May 7, 2024
2
请问qwen应该将--version 设为mpt吗还是plain？
#54 opened May 8, 2024 by Yang-bug-star updated May 11, 2024
1
Can you share the Loss Curve?
#47 opened Apr 8, 2024 by necrophagists updated May 15, 2024
1
Missing llavabench.sh
#57 opened May 16, 2024 by cooleel updated May 16, 2024
question about custom fine-tuning
#58 opened May 16, 2024 by josephtey updated May 16, 2024
add phi3
#59 opened May 17, 2024 by Jayantverma2 updated May 17, 2024
Can you share the finetune.sh and pretrain.sh to train TinyLLaVA-1.5B?
#42 opened Apr 2, 2024 by fyting updated May 23, 2024
2
Is share_textvqa is the same as textvqa in llava dataset?
#65 opened May 24, 2024 by cooleel updated May 25, 2024
1
MM-Vet download link broken
#66 opened May 26, 2024 by ryan-caesar-ramos updated May 26, 2024
2
the loss is nan when pre-training tinyllama using share recipe
#13 opened Mar 4, 2024 by xushilin1 updated May 29, 2024
1
What does "share" mean in the last line of the model performance table.
#69 opened May 29, 2024 by lijiannuist updated May 29, 2024
1
ProTip! Add no:assignee to see everything that’s not assigned.