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

How to run using SSD model #33

Closed
Liang-ZX opened this issue Aug 2, 2021 · 4 comments
Closed

How to run using SSD model #33

Liang-ZX opened this issue Aug 2, 2021 · 4 comments
Labels
enhancement New feature or request

Comments

@Liang-ZX
Copy link

Liang-ZX commented Aug 2, 2021

请问是否有提供使用SSD网络的代码,以及对应的预训练模型是什么,文章中好像没有谈

@yuantn
Copy link
Owner

yuantn commented Aug 3, 2021

使用 SSD 网络的代码正在整理中,对应的预训练模型为 open-mmlab 提供的 vgg16_caffe 模型。


The code for using the SSD network is being arranged, and the corresponding pre-trained model is the vgg16_caffe model provided by open-mmlab.

@yuantn
Copy link
Owner

yuantn commented Aug 9, 2021

使用 SSD 检测器的代码已经整理完毕,你可以更新这个目录(重新下载或重新克隆),并参考 这里 关于使用 SSD 检测器的说明来实现在 SSD 检测器基础上的训练。


The code for using the SSD detector is ready, you can update this directory (re-download or re-clone), and refer to here for the instruction of using the SSD detector to implement the training based on the SSD detector.

@yuantn yuantn added the enhancement New feature or request label Aug 9, 2021
@Liang-ZX
Copy link
Author

Liang-ZX commented Aug 10, 2021 via email

@yuantn
Copy link
Owner

yuantn commented Aug 10, 2021

首先,我的 RetinaNet for COCO 的实验确实是用当前代码库中的代码实现的,但我并不能看到你上传的截图。

另外,我在论文中从未声明 SSD 的 epoch 数为 100,反而是 300(240+60,如论文 4.1 节所述)。对于 random 的结果,我是去掉了所有在未标注集合上训练的过程,并以随机的方式挑选图像得出的结果。


First, the experiment of RetinaNet for COCO is indeed implemented with the code in the repository, but I cannot see the screenshot you uploaded.

In addition, I have not declared in the paper that the epoch number of SSD is 100, but 300 (240+60, as described in section 4.1 of the paper). For the random result, I removed all the training process on the unlabeled set, and selected the image randomly.

@yuantn yuantn closed this as completed Sep 1, 2021
yuantn added a commit that referenced this issue Sep 18, 2021
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
None yet
Development

No branches or pull requests

2 participants