Added Readme.md files for Char-RNN and MNIST

text/char-rnn/Readme.md

@@ -0,0 +1,4 @@
+# Character Level Language Model using RNNs
+This is an implementation of RNN based Character Level Language Model. In this model,RNN model(based on LSTMs) is given a chunk of huge text on which it is trained to  model the probability distribution of the next character in the sequence given a sequence of previous characters. This will then allow the model to generate new text one character at a time
+
+To train this Model, Shakespeare.txt was used. However, any text can also be used to train the model.

vision/mnist/Readme.md

@@ -0,0 +1,9 @@
+# Simple NN models for Computer Vision on MNIST Dataset
+This folder contains following models implemented on MNIST Dataset for Hand-Written Digit Recognition:
+1. Simple Multi-layered Perceptron model(<i>mlp.jl</i>)
+2. Simple Convolutional Neural Networks(<i>conv.jl</i>)
+3. Simple Autoencoders(<i>autoencoder.jl</i>)
+4. Variational Autoencoders(<i>vae</i>)
+
+## About MNIST Dataset
+MNIST Dataset is a large database of Handwritten Digits that is commonly used for training various Image Processing Systems