2016年07月_蜗牛一步一步往上爬

12月 10月 09月 08月 07月 06月 05月 04月 03月 02月 01月

原创 CS231n Class Notes- lecture11 ConvNets in Practice

这节课主要讲了一些在实际使用 CNN 时的技巧，可以加速训练网络。很多内容需要图解，所以我就直接截图了。所有图都是从 CS231n lecture 11的课件中截取的。The power of small filtersThree 3 x 3 conv gives similar representational power as a single 7 x 7 convolution 这是在 V

2016-07-30 20:59:35 1722

原创 Understanding LSTM Networks

学习RNN，发现一个很好的 LSTM 讲解。

2016-07-29 11:04:11 453

原创 Leetcode168. Excel Sheet Column Title

Probelm DescriptionGiven a positive integer, return its corresponding column title as appear in an Excel sheet. For example:1 -> A2 -> B3 -> C...26 -> Z27 -> AA28 -> AB Thinking:这个问题是 Easy 型的，也

2016-07-20 19:42:14 488

原创 cs231n - assignment1 - neural net 梯度推导

Implementing a Neural Network In this exercise we will develop a neural network with fully-connected layers to perform classification, and test it out on the CIFAR-10 dataset.这里开始采用矩阵的形式来推导梯度，而且将

2016-07-18 19:19:10 8541 3

原创 cs231n - assignment1 - softmax 梯度推导

Softmax exercise Complete and hand in this completed worksheet (including its outputs and any supporting code outside of the worksheet) with your assignment submission. For more details see the assig

2016-07-16 15:19:07 16497 23

原创 cs231n - assignment1 - linear-svm 梯度推导

Multiclass Support Vector Machine exercise Complete and hand in this completed worksheet (including its outputs and any supporting code outside of the worksheet) with your assignment submission. F

2016-07-15 22:59:27 8929 6

原创 cs231n - assignment1- k-Nearest Neighbor Classifier 梯度推导

assignment web pagek-Nearest Neighbor (kNN) exercise Complete and hand in this completed worksheet (including its outputs and any supporting code outside of the worksheet) with your assignment submis

2016-07-14 20:07:23 1253

原创 Google gflags usage on windows

gflags 是 google 的开源项目，用来解析程序的命令行输入参数。传统的方法是手工解析argv参数，或者使用getopt函数。两种方法都比较费劲。使用Google gflags可以大大简化命令行参数处理。下面主要记录如何在windows下利用CMake编译gflags，并用CMakeLists来使用gflags,以及基本的使用方法。Complie gflags By CMake在github

2016-07-01 14:06:14 3927