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

用所给数据集训练时,LMDBDataset类中必须使用到backend='caffe',请问是否有以backend='pickle'为基础的数据集? #4

Open
lolccsdz opened this issue May 16, 2024 · 4 comments

Comments

@lolccsdz
Copy link

因为caffe框架所需要的cuda版本过低,现在一些模型可能会出现内部的包与cuda不兼容问题。而所给数据集必须用caffe读取,所以想问问是否有不需要caffe框架就能读取的数据集?非常感谢。([email protected]

@lolccsdz
Copy link
Author

lolccsdz commented May 16, 2024

训练时使用了网盘里的ICVL_31_100.db,会报错:_pickle.UnpicklingError: invalid load key, '\x08'
将LMDBDataset默认的backend='pickle'改为backend='caffe',使用了caffe框架,才能训练。
请问这该如何修改?

@Zeqiang-Lai
Copy link
Owner

Zeqiang-Lai commented May 16, 2024

我之前的转换脚本找不到了,大概就是

需要用 caffe 读取,然后用 lmdb 再存一次,变成 pickle 格式的。可以参考原来的 lmdb build 函数

把这行修改成 pickle 格式序列化
https://github.com/Vandermode/QRNN3D/blob/78e0b5adaeaeb51d8ac7df17526090468a180745/utility/lmdb_data.py#L74

@lolccsdz
Copy link
Author

lolccsdz commented May 17, 2024

非常感谢您的解答。

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