MultiPoseNet.pytorch

MultiPoseNet.pytorch copied to clipboard

Hello, I was wondering what's the difference between your released baseline model and the the models you used in your paper. Because I found there is a noticeable gap, which...

ok-sweat

detection subnet training

9

你好，检测子网络我训练了80epoch，val_loss只降到0.48，跟baseline模型相差有点大，请问当时训练baseline的时候就是现在的设置了吗？ 按照你的预设的max_epoch=50，训练到完成loss曲线仍一直下降，学习率(1e-5)还未发生过改变。 ### my model(80epoch): Validation loss: mean: 0.48589630182399307, std: 0.056519281222650285 ### baseline model: Validation loss: mean: 0.3989471616440041, std: 0.06292749785476406

accountcwd

different input size of the two subnet when training

10

你好，我注意到在training中，keypoint_subnet 的输入是480x480，而detection_subnet的是608x608，请问是出于什么考虑才设置不同的分辨率去训练这两个网络呢？ 另外，我训练的keypoint_subnet经过48个epoch，val_loss为0.00312（上一个best为35epoch时的0.00313），这结果可以吧？

accountcwd