Added RNN

code/rnn.py

@@ -1,4 +1,6 @@
+import torch
 import torch.nn as nn
+#from torch.autograd import Variable
 
 class RNN(nn.Module):
     def __init__(self, n_classes):
@@ -17,16 +19,16 @@ def forward(self, inputs, hidden):
 
 def train(model, inputs, targets):
     for i in range(len(inputs)):
-        target = Variable([targets[i]])
+        target = Variable(targets[i])
         name = inputs[i]
         hidden = Variable(torch.zeros(1,128))
         model.zero_grad()
 
         for char in name:
             input_ = Variable(torch.FloatTensor(char))
-            out, hidden = model(input_, hidden)
+            pred, hidden = model(input_, hidden)
 
-        loss = criterion(out, target)
+        loss = criterion(pred, target)
         loss.backward()
 
         for p in model.parameters():

docs/deep_learning.rst

@@ -6,32 +6,35 @@ Deep Learning
 
 .. contents:: :local:
 
-
-General
-=======
+Autoencoder
+===========
 
 Be the first to `contribute! <https://github.com/bfortuner/ml-cheatsheet>`__
 
-Autoencoder
-===========
+- `Deep Learning Book <http://www.deeplearningbook.org/contents/autoencoders.html>`_
+
 
 CNN
 ===
 
 Be the first to `contribute! <https://github.com/bfortuner/ml-cheatsheet>`__
 
-`cnn reference <http://www.deeplearningbook.org/contents/convnets.html>`_
+- `Deep Learning Book <http://www.deeplearningbook.org/contents/convnets.html>`_
+
+
+GAN
+===
+
+- `Deep Learning Book <http://www.deeplearningbook.org/contents/generative_models.html>`_
 
 
 RNN
 ===
 
 Description of RNN use case and basic architecture.
 
-.. image:: images/relu.png
-      :align: center      
-      :width: 256 px      
-      :height: 256 px     
+.. image:: images/rnn.png
+      :align: center
 
 .. rubric:: Model
 
@@ -40,20 +43,24 @@ Description of RNN use case and basic architecture.
 
 .. rubric:: Training
 
-Explanation of training loop
+In this example, our input is a list of last names, where each name is 
+a variable length array of one-hot encoded characters. Our target is is a list of
+indices representing the class (language) of the name.
 
-.. literalinclude:: ../code/rnn.py   
+1. For each input name..
+2. Initialize the hidden vector
+3. Loop through the characters and predict the class
+4. Pass the final character's prediction to the loss function
+5. Backprop and update the weights
+
+.. literalinclude:: ../code/rnn.py
       :pyobject: train
 
 .. rubric:: Further reading
 
+- `Jupyter notebook <https://github.com/bfortuner/ml-cheatsheet/notebooks/rnn.ipynb>`_
 - `Deep Learning Book <http://www.deeplearningbook.org/contents/rnn.html>`_
 
-GAN
-===
-
-- ` Deep Learning Book <http://www.deeplearningbook.org/contents/generative_models.html>`_
-
 
 VAE
 ===

docs/index.rst

@@ -45,14 +45,14 @@ Brief visual explanations of machine learning concepts with diagrams, code examp
 
 .. toctree::
     :maxdepth: 1
-    :caption: Algorithms (empty)
+    :caption: Algorithms
 
     Applications <applications>
     Classification <classification_algos>
     Clustering <clustering_algos>
     Deep Learning <deep_learning>
-    Generative <generative_algos>
     Regression <regression_algos>
+    Reinforcement Learning <reinforcement_learning>
 
 .. toctree::
     :maxdepth: 1

docs/reinforcement_learning.rst

@@ -0,0 +1,15 @@
+.. _reinforcement_learning:
+
+======================
+Reinforcement Learning
+======================
+
+Be the first to `contribute! <https://github.com/bfortuner/ml-cheatsheet>`__
+
+
+.. rubric:: References
+
+.. [1] Example Reference
+
+
+