rnn-from-scratch-master

代码说明：

  RNN神经网络的应用和概念，RNN源代码和使用方法(You can find that the parameters `(W, U, V)` are shared in different time steps. And the output in each time step can be**softmax**. So you can use**cross entropy** loss as an error function and use some optimizing method (e.g. gradient descent) to calculate the optimized parameters `(W, U, V)`. Let recap the equations of our RNN: )

文件列表：

rnn-from-scratch-master
.......................\activation.py,409,2016-04-18
.......................\data
.......................\....\reddit-comments-2015-08.csv,7610868,2016-04-18
.......................\figures
.......................\.......\gradient.png
.......................\.......\init.png



.......................\.......\rnn-compuattion-graph_2.png,22968,2016-04-18
.......................\.......\rnn.jpg,44320,2016-04-18
.......................\.......\rnn_equation.png,1213,2016-04-18
.......................\.......\rnn_eval.png,19356,2016-04-18
.......................\.......\rnn_loss.png,1681,2016-04-18
.......................\.......\rnn_loss_2.png,976,2016-04-18
.......................\gate.py,475,2016-04-18
.......................\layer.py,869,2016-04-18
.......................\output.py,334,2016-04-18
.......................\preprocessing.py,2770,2016-04-18
.......................\README.md,15570,2016-04-18
.......................\rnn.py,4373,2016-04-18
.......................\rnnlm.py,357,2016-04-18
.......................\__pycache__
.......................\...........\activation.cpython-34.pyc,1094,2016-04-18
.......................\...........\gate.cpython-34.pyc,1190,2016-04-18
.......................\...........\layer.cpython-34.pyc,1196,2016-04-18
.......................\...........\output.cpython-34.pyc,819,2016-04-18
.......................\...........\preprocessing.cpython-34.pyc,3712,2016-04-18
.......................\...........\rnn.cpython-34.pyc,4332,2016-04-18