-
Notifications
You must be signed in to change notification settings - Fork 88
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
PASCAL 数据集训练 #14
Comments
我也遇到了同样的情况 |
请问你在voc上的mAP是多少呀 |
请问你们是如何修改数据写入的 |
请问一下你在VOC训练一个epoch大概要多久? |
11.19更新:
训练集:VOC07+12,测试集:VOC07-test,
使用SGDR学习率调整策略,初始学习率为0.01,学习率调整周期为100个epoches,总共训练610个epoches,
预训练模型:mb3-ssd-lite-Epoch-149-Loss-5.782852862012213.pth,
batch_size:32,
能得到的最好的Loss是4.07,请问怎样进一步调整学习策略?
原提问:
你好,当我使用您提供的PASCAL的预训练模型(mb3-ssd-lite-Epoch-149-Loss-5.782852862012213.pth)对几张图片进行检测时,每张图片都检测不到任何一个物体。
然后,我在上述检查点的基础上进行finetune,使用VOC2007+VOC2012数据集训练了100个epoch,batch_size=32,lr=0.0005+multistep([40,80,]),使用新训练的模型再次检测,部分图片能检测到物体,但是准确度还是比较低。
对此,我有两个问题想请教您:
The text was updated successfully, but these errors were encountered: