摘要: orthogonalization/ one metric train、dev/test 划分 开发集和测试集一定来自同一分布 onthe same distribution Human level performance & bayes error human level performance 阅读全文
posted @ 2018-10-23 00:31 smartweed 阅读(354) 评论(0) 推荐(0) 编辑
摘要: 公开课笔记 Bias & variance bias: 1. more epoch 2. deeper network 3.hyperparameters variance : larger dataset 2. regularization regularization L2 norm: weig 阅读全文
posted @ 2018-10-23 00:11 smartweed 阅读(396) 评论(0) 推荐(0) 编辑