http://ufldl.stanford.edu/tutorial/supervised/SoftmaxRegression/
http://ufldl.stanford.edu/tutorial/supervised/ExerciseSupervisedNeuralNetwork/
http://ufldl.stanford.edu/tutorial/supervised/MultiLayerNeuralNetworks/
https://www.zhihu.com/question/23765351
http://ufldl.stanford.edu/tutorial/supervised/LogisticRegression/
http://ufldl.stanford.edu/tutorial/supervised/MultiLayerNeuralNetworks/
http://www.pyimagesearch.com/2016/09/12/softmax-classifiers-explained/
https://github.com/Kulbear/deep-learning-nano-foundation/wiki/ReLU-and-Softmax-Activation-Functions
https://en.wikipedia.org/wiki/Softmax_function
https://zh.wikipedia.org/wiki/Softmax%E5%87%BD%E6%95%B0
http://dataaspirant.com/2017/03/07/difference-between-softmax-function-and-sigmoid-function/
http://cs231n.github.io/linear-classify/
http://knet.readthedocs.io/en/latest/softmax.html
http://ufldl.stanford.edu/wiki/index.php/Softmax_Regression