snow-wind-001
snow-wind-001
试了一些参数,mAP达到0.88 
> 你的loss降到多少了? > 你的loss降到多少了? test loss = 7左右
> > 时代 > > 你的30个epoch是什么意思?在作者代码中是如何设置30个epoch的?谢谢 在config.py改的
加载了预权重,调整了epoch大小,训练分多段进行,观察最终test的损失,实时进行调整。
@chang48 @dapladoc @churnikov @alexmelekhin
@chang48 @dapladoc @churnikov @alexmelekhin
Can I understand that to put new photos into a list of one photo per category, I calculate the distances through pairwise_dist, or use pairwiseImgesPostprocessor to process the queue, and...
@chang48 @dapladoc @churnikov @alexmelekhin
My level is very limited, and the problem may also be due to my lack of understanding of this project, but then I ran the postprocessor.jpynb file and made modifications,...
https://colab.research.google.com/drive/1o5CQbbafdKyiooewOHfAqmxJ1b--6UJ8?usp=drive_link