卷积神经网络
关注
分享
扫一扫
Co丿Hx
西南石油大学CS研究生。发表ccfc类论文两篇，专利一项，获国家奖学金等多项奖学金，现就职于招银网络科技。
展开
-
Keras Tutorial - The Happy House 本地跑通版
这是整个项目目录testKeras.pyimport numpy as npfrom keras import layersfrom keras.layers import Input, Dense, Activation, ZeroPadding2D, BatchNormalization, Flatten, Conv2Dfrom keras.layers import AveragePo...原创 2018-06-27 19:47:38 · 477 阅读 · 0 评论 -
FlowNet: Learning Optical Flow with Convolutional Networks
FlowNet: Learning Optical Flow with Convolutional Networks2015 IEEE International Conference on Computer Visionoutline•Introduction•RelatedWork•Dataset Schedules•Stacking Networks•Small ...原创 2018-09-17 18:13:18 · 450 阅读 · 0 评论 -
A `Concatenate` layer requires inputs with matching shapes except for the concat axis.
使用keras跑别人代码时报错A `Concatenate` layer requires inputs with matching shapes except for the concat axis.发现维度不匹配，通过网上找资料发现是channel_last与channel_first问题。在$HOME/.keras/keras.json目录下将channels_last改成channel...原创 2018-09-25 10:43:22 · 5403 阅读 · 0 评论 -
cuda实现二维卷积共享内存
本次cuda课作业为实现cuda卷积。核心代码为1.cpu卷积代码void Conv2(float** filter, float** arr, float** res, int filter_size, int arr_size) { int temp; for (int i = 0; i<arr_size; i++) { for (int j = 0; j<ar...原创 2019-06-06 16:03:00 · 5972 阅读 · 21 评论