Skip to content

对于nus-wide数据集 #25

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.

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

Closed
xiemitiao opened this issue Sep 29, 2024 · 5 comments
Closed

对于nus-wide数据集 #25

xiemitiao opened this issue Sep 29, 2024 · 5 comments

Comments

@xiemitiao
Copy link

您好,请问nus-wide数据集是怎么清洗的?筛查后的结果不能符合论文中186,557大小数据集的要求。

@penghu-cs
Copy link
Owner

您好,数据集是用的:Deep cross-modal hashing 这篇文章的,供您参考。
链接:https://github.com/jiangqy/DCMH-CVPR2017/tree/master/DCMH_matlab/DCMH_matlab

@xiemitiao
Copy link
Author

您好,数据集是用的:Deep cross-modal hashing 这篇文章的,供您参考。 链接:https://github.com/jiangqy/DCMH-CVPR2017/tree/master/DCMH_matlab/DCMH_matlab

作者您好,请问nus-wide的参数中诸如margin、shift等参数要如何配置呢?

@penghu-cs
Copy link
Owner

如下参数供您参考:
python UCCH.py --data_name nus_wide_tc10_fea --num_hiden_layers 3 2 --max_epochs 20 --train_batch_size 256 --lr 0.0001 --optimizer Adam --shift 0.1 --alpha 0.7 --margin 0.2 --bit 16

@xiemitiao
Copy link
Author

您好,国庆快乐,请问对于nus-wide在raw data上该如何配置呢?

@penghu-cs
Copy link
Owner

可以将_fea去掉即可,如下代码供您参考:

python UCCH.py --data_name nus_wide_tc10 --num_hiden_layers 3 2 --max_epochs 20 --train_batch_size 256 --lr 0.0001 --optimizer Adam --shift 0.1 --alpha 0.7 --margin 0.2 --bit 16 --warmup_epoch 5 --pretrain -a vgg11

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants