-
Notifications
You must be signed in to change notification settings - Fork 5.6k
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
[PaddlePaddle Hackathon] add ResNeXt #36070
Conversation
Thanks for your contribution! |
@LielinJiang 我试了所有我所迁移的模型,直接删除 name 可以直接加载预训练权重(无任何warning),那是不是就可以不用修改clas权重的名字了呀? 另外我和队友还做了一下基准测试,请问像这样是可以的嘛(见上面)? |
那麻烦在统一程序中,实例化两次看看是否会报错,如果不会报错,那就没问题了:
|
好哒好哒~刚刚在基准测试所用 Notebook 里增加了多次实例化的测试,是可以直接跑通的 更新后的 notebook:https://aistudio.baidu.com/aistudio/projectdetail/2422677?contributionType=1 请问这样是不是就可以进行提交了呀? |
可以的辛苦了 |
好的感谢~~~ |
""" | ||
|
||
def __init__(self, | ||
layers=50, |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
已修改~
@LielinJiang 十分抱歉,刚刚在修改其他模型时候发现 ResNeXt 出了点问题,之后又重新 commit 修复了这个问题,这导致之前的 review 被视为 dismiss 了( •̥́ ˍ •̀ू ),可以重新 review 一下嘛? 此外,以下三个模型也均已完成修改,均增加了 |
f148941
to
ae5f8ac
Compare
* add resnext model * add zh docs * add unittest * test performance Co-authored-by: Ainavo <ainavo@163.com> Co-authored-by: pithygit <pyg20200403@163.com>
已经解决冲突,CI 也已通过(除去需要 Approve 的和非 Required 的),可以麻烦再次 Review 一下嘛? |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
PR types
New features
PR changes
APIs
Describe
status: Pending Review
Performance
AiStudio 测试详情:https://aistudio.baidu.com/aistudio/projectdetail/2422677?contributionType=1
基准参考:https://github.com/PaddlePaddle/PaddleClas/blob/release/2.2/docs/en/ImageNet_models_en.md
括号中为以 PaddleClas 性能基准为参考的偏差值