02 2018 档案

caffe output argmax layer
摘要:http://blog.csdn.net/ws_20100/article/details/49184037 http://blog.csdn.net/langb2014/article/details/50988275 (good !) https://www.cnblogs.com/yinhey 阅读全文

posted @ 2018-02-26 11:17 cdekelon 阅读(1169) 评论(0) 推荐(0)

caffe python API
摘要:http://wentaoma.com/2016/08/10/caffe-python-common-api-reference/ 阅读全文

posted @ 2018-02-23 16:08 cdekelon 阅读(144) 评论(0) 推荐(0)

File Descriptor Leak
摘要:https://martin-ueding.de/articles/c-file-descriptors/index.html http://blog.sina.com.cn/s/blog_605f5b4f01013lgv.html (good) https://docs.oracle.com/cd 阅读全文

posted @ 2018-02-13 17:16 cdekelon 阅读(231) 评论(0) 推荐(0)

Xinit and StartX
摘要:http://blog.csdn.net/sjin_1314/article/details/40148853 阅读全文

posted @ 2018-02-08 14:42 cdekelon 阅读(88) 评论(0) 推荐(0)

center loss
摘要:http://blog.csdn.net/u014380165/article/details/76946339 阅读全文

posted @ 2018-02-05 11:25 cdekelon 阅读(86) 评论(0) 推荐(0)

caffe solver
摘要:From http://blog.csdn.net/czp0322/article/details/52161759 solver.prototxt sovler的主要作用就是交替调用前向传导和反向传导 (forward & backward) 来更新神经网络的连接权值,从而达到最小化loss,实际 阅读全文

posted @ 2018-02-03 00:58 cdekelon 阅读(181) 评论(0) 推荐(0)

导航