DragonEmperorG / QDeepOdo

Reproduce the paper “A Novel Deep Odometry Network for VehiclePositioning Based on Smartphone” method
2 stars 1 forks source link

【讨论】文章中对于DeepOdo网络训练中的时间窗口参数N的选取 #2

Open DragonEmperorG opened 10 months ago

DragonEmperorG commented 10 months ago

在Google Smartphone Decimeter Challenge 2023数据集上将一条轨迹整个作为一段时间序列数据送入网络进行训练,验证集Loss不下降的问题,这样窗口参数的选择可能是问题的原因。整个网络的参数不足以支持超长时间序列的预测,减小窗口参数的大小,产生更多的数据段使得数据规模和网络参数能够匹配