Skip to content
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

About the datasets of .zip files #17

Open
gina7152316 opened this issue May 11, 2020 · 6 comments
Open

About the datasets of .zip files #17

gina7152316 opened this issue May 11, 2020 · 6 comments

Comments

@gina7152316
Copy link

Hello, thanks for sharing the project.
I tried training it and testing it have a question about datasets:
In configs/celebahq has "zip_root": "../datazip" , I don't know where to download datazip or how to construct contents in datazip by myself.

Thanks!

@ZHANG-hengzhi
Copy link

你好,请问你现在会构造自己的数据集了吗,可以分享一下吗

@gina7152316
Copy link
Author

作者沒有給予回覆,我這邊還是沒有甚麼進展呢

@zhangweiweicpp
Copy link

我也遇到了同样的问题,解决的还望分享一下谢谢

@simplebaseline
Copy link

遇到了同样的问题

@micklexqg
Copy link

the same problem, has anyone solved it?

@FukiBabasaki
Copy link

I have recently understood how the configuration and file structure work.

  • data_loader
    • name: name for the image
    • zip_root: Root directory containing the zip folder with name
    • flist_root: Root directory containing several sub-directories with name which contain the corresponding train.flist, ...

For example,

"data_loader": {
   "name": "faces",
   "zip_root": "dataset",
   "flist_root": "flist",
   ...
}

Then you must have:

  • path_to_project/dataset/faces/faces.zip where faces.zip contains jpg/png input files.
  • path_to_project/flist/faces/train.flist or other flist files which contain paths to images relative to zip folder. In this case, e.g. faces/face1.png`

The author doesn't seem to upload example images but that's fine since the training script automatically produces masks on the go. Hope this helps!

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

6 participants