ZhengyuZhang
ZhengyuZhang
Hi, author. I really appreciate your interesting work! But I would like to know which dataset is used to train the pretrained models that you provide? or you train the...
我是直接注释掉那句导入预训练模型的代码,然后重新训练,loss一直不降保持在10-20之间,而且经常出现loss为nan的情况,学习率、批次数目和优化器都调过,好像没有变化,请问作者是如何将loss训练到0.几的?这个问题困扰了我很久了。。。是有什么特别的训练策略吗?
Hi author, thanks for your released code first! But I found that the Ax2 pre-trained model is not applicable for the HDDRNet_Ax2 network. looking forward to your reply soon! Thanks...