AILab-CVC SEED issues - Githubissues

AILab-CVC / SEED

Official implementation of SEED-LLaMA (ICLR 2024).

https://ailab-cvc.github.io/seed

Other

576 stars 31 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Request for Instruction Tunning

#53 URRealHero opened 2 months ago
0
预训练数据

#52 cxjtju opened 3 months ago
0
How to run seed-2 image reconstruction inference?

#51 zhangqingwu opened 3 months ago
0
Inquiry Regarding the Evaluation Details of SEED-LLaMA

#50 nth2000 opened 3 months ago
0
hydra.errors.InstantiationException: Error locating target './models.seed_llama_tokenizer.SeedLlamaTokenizer.from_pretrained', set env var HYDRA_FULL_ERROR=1 to see chained exception

#49 cxjtju closed 3 months ago
1
instruction tuning data example

#48 cxjtju opened 3 months ago
0
中文效果

#47 cxjtju opened 3 months ago
0
Will consider try with new Magvit2 image tokenizer?

#46 lucasjinreal opened 5 months ago
0
maximum number of tokens

#45 Caixy1113 opened 5 months ago
0
Clear Training Instructions

#44 eslambakr opened 5 months ago
0
Evaluating the quality of codebooks using image-text/text-image retrieval as proxy.

#43 SxJyJay opened 5 months ago
0
Where are the core code lines that transform image to discrete token?

#42 guotong1988 closed 6 months ago
1
How to train the model on my own dataset?

#41 zhi-xuan-chen opened 6 months ago
0
Question about T2I recall1 of COCO

#40 leexinhao opened 6 months ago
0
Instruction Tuning hyperparameters for SEED-LLaMA

#39 dhdbsrlw opened 7 months ago
0
Does model has Chinese OCR ability?

#38 luohao123 opened 7 months ago
0
sh train_scripts/causal_qformer.sh is not working!

#37 leedaehan-kev opened 7 months ago
1
How long does it take to train the SEED tokenizer?

#36 ys-zong opened 7 months ago
0
Codebook Training Epochs

#35 Revliter opened 7 months ago
1
Whether can SEED process more than 2 images once?

#34 qwqwq1445 opened 7 months ago
0
Reproduce SEED LLaMA evaluation

#33 hyomin14 opened 7 months ago
0
Missing Multimodel Pretraining step

#32 shubhamgarg21 opened 8 months ago
1
Fixes for issues in generating pre-training data by converting images into discrete tokens

#31 shubhamgarg21 opened 8 months ago
0
train_scripts/causal_qformer.sh not working

#30 shubhamgarg21 opened 8 months ago
1
install.sh not working

#29 shubhamgarg21 closed 8 months ago
1
Difference in 'blocks' and 'blocks_for_image'

#28 zheedong closed 2 months ago
2
Training Data of Tokenizer

#27 zheedong opened 8 months ago
2
如何获取训练数据？

#26 APiaoG opened 8 months ago
1
Train data

#25 APiaoG opened 8 months ago
1
Pretrained LLM version

#24 hyomin14 closed 8 months ago
1
Stage I Contrastive Learning : What is 'final' causal embedding?

#23 zheedong closed 8 months ago
1
Training code

#22 koda-11 opened 9 months ago
1
About add the quantized image tokens to pretrained language tokenizer.

#21 Jiushanhuadao opened 9 months ago
1
Demo Not Working

#20 yzeng58 opened 10 months ago
1
Question on how task27 generates images

#19 JunZhan2000 closed 10 months ago
0
Questions about EVA-CLIP-G used in SEED.

#18 Haochen-Wang409 closed 8 months ago
2
How to force model to generate image?

#17 haochuan-li opened 11 months ago
2
Hyperparameter for training SEED Tokenizer

#16 Cheolhyun-Mun opened 11 months ago
1
Evaluation Results

#15 chancharikmitra opened 11 months ago
0
Training Code

#14 ChangeNext opened 11 months ago
1
What is frozen text/image encoder?

#13 zheedong closed 9 months ago
3
Training code for SEED-LLaMA

#12 shubhamgarg21 opened 11 months ago
1
More explanation about pilot experiments

#11 zheedong closed 11 months ago
1
Fail on Inference

#10 SihengLi99 closed 11 months ago
2
CIDEr reproduce hyperparameter

#9 zheedong closed 11 months ago
2
About Gradio Version

#8 MajorDavidZhang closed 11 months ago
1
What is the difference between SEED-2-1 and SEED-2?

#7 JunZhan2000 closed 11 months ago
1
新的seed-llama-v2-1的模型，请问会开源吗？

#6 yonghenglh6 closed 11 months ago
1
Training Code ?

#5 achen46 closed 1 year ago
1
Update README.md

#4 computerscienceiscool opened 1 year ago
0