intial commit

MNIST.ipynb

@@ -0,0 +1,54 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# MNIST Digit Classification (from scratch!)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Background: MNIST is a dataset of images of single digits. \n",
+    "\n",
+    "Here, you will find the code for digit classification of the MNIST dataset (digits 1-9) **from scratch**.\n",
+    "\n",
+    "<br>My goal is to investigate the differences and similarity between **multilayer perceptons (a type of artificial neural network) are to the human brain (biological neural networks)**."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Key Notes\n",
+    "\n",
+    "- Every progressive layer in the network corresponds to increasingly larger and larger edges. Etc. The second layer corresponds to small edges, and the next layer corresponds to overall shapes (ex. a loop) from those small edges\n",
+    "    - pixels -> edges -> patterns -> digits\n",
+    "- "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "language_info": {
+   "name": "python"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}

Neural Networks/notes.ipynb

@@ -0,0 +1,47 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Neural Networks"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Different types\n",
+    "\n",
+    "- Multilayer perceptron (plain vanilla neural network)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Key Vocabulary\n",
+    "\n",
+    "- **Neuron**: thing that holds a number\n",
+    "    - The number held is called the neuron's **activation**\n",
+    "\n",
+    "- **Hidden Layers**: layers between the input & output layers\n",
+    "    - the activations from one layer determine the activations of the next layer"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "language_info": {
+   "name": "python"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}