![](https://img-blog.csdnimg.cn/20201014180756724.png?x-oss-process=image/resize,m_fixed,h_64,w_64)
deep learning
docdok
诚信做人,踏实做事,坚持,创新,悟性,效率
展开
-
caffe中的iteration,batch_size, epochs理解
epoch: Forward and Backward pass of all training examples ( not used in Caffe)batch: how many images in one passiterations: how many batches1. Batch SizeBatch size in mainly depended to yo原创 2016-03-29 17:21:14 · 6298 阅读 · 0 评论 -
caffe训练时总是会占用0号GPU卡
Sometimes, we will change the default GPU 0 to others when we are training deep learning models using Caffe, however, we also find that gpu 0 is always used (~100 MBs) .To solver this, we can simp原创 2017-06-21 11:49:30 · 1514 阅读 · 0 评论