draft 3

min-char-rnn-nb.ipynb

@@ -1,5 +1,21 @@
 {
  "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## RNN Language Model\n",
+    "\n",
+    "Below is a diagram of the RNN computation that we will implement below. We're plugging characters into the RNN with a 1-hot encoding and expecting it to predict the next character. In this example the training data is the string \"hello\", so there are 4 letters in the vocabulary: [h,e,l,o]."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<img src=\"rnnlm.jpeg\">"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": 1,