demo.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "diverse-senegal",
   "metadata": {},
   "outputs": [],
   "source": [
    "from picograd.engine import Var\n",
    "from picograd.graph_viz import ForwardGraphViz"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "infectious-marsh",
   "metadata": {},
   "source": [
    "## Reverse mode auto-differentiation aka Backpropagation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "available-progressive",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "24.7041\n",
      "138.8338\n",
      "645.5773\n"
     ]
    }
   ],
   "source": [
    "a = Var(-4.0, label='a')\n",
    "b = Var(2.0, label='b')\n",
    "c = a + b; c.label='c'\n",
    "d = a * b + b**3; d.label='d'\n",
    "c += c + 1\n",
    "c += 1 + c + (-a)\n",
    "d += d * 2 + (b + a).relu()\n",
    "d += 3 * d + (b - a).relu()\n",
    "e = c - d; e.label='e'\n",
    "f = e**2; f.label='f'\n",
    "g = f / 2.0; g.label='g'\n",
    "g += 10.0 / f\n",
    "print(f'{g.data:.4f}') # prints 24.7041, the outcome of this forward pass\n",
    "g.backward()\n",
    "print(f'{a.grad:.4f}') # prints 138.8338, i.e. the numerical value of dg/da\n",
    "print(f'{b.grad:.4f}') # prints 645.5773, i.e. the numerical value of dg/db"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "taken-copyright",
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_computational_graph(root, rankdir='LR'):\n",
    "    graph_builder = ForwardGraphViz()\n",
    "    return graph_builder.create_graph(root, rankdir=rankdir)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "adjusted-malpractice",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/svg+xml": [
       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
       "<!-- Generated by graphviz version 2.40.1 (20161225.0304)\n",
       " -->\n",
       "<!-- Title: %3 Pages: 1 -->\n",
       "<svg width=\"741pt\" height=\"210pt\"\n",
       " viewBox=\"0.00 0.00 740.99 210.00\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 206)\">\n",
       "<title>%3</title>\n",
       "<polygon fill=\"#ffffff\" stroke=\"transparent\" points=\"-4,4 -4,-206 736.9942,-206 736.9942,4 -4,4\"/>\n",
       "<!-- 140161765950688 -->\n",
       "<g id=\"node1\" class=\"node\">\n",
       "<title>140161765950688</title>\n",
       "<polygon fill=\"none\" stroke=\"#000000\" points=\"212,-132.5 212,-201.5 298,-201.5 298,-132.5 212,-132.5\"/>\n",
       "<text text-anchor=\"middle\" x=\"255\" y=\"-186.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\"> </text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"212,-178.5 298,-178.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"255\" y=\"-163.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">data: 1.0000</text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"212,-155.5 298,-155.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"255\" y=\"-140.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">grad: 1.0000</text>\n",
       "</g>\n",
       "<!-- 140161765951472+ -->\n",
       "<g id=\"node9\" class=\"node\">\n",
       "<title>140161765951472+</title>\n",
       "<ellipse fill=\"none\" stroke=\"#000000\" cx=\"361\" cy=\"-123\" rx=\"27\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"361\" y=\"-119.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">+</text>\n",
       "</g>\n",
       "<!-- 140161765950688&#45;&gt;140161765951472+ -->\n",
       "<g id=\"edge4\" class=\"edge\">\n",
       "<title>140161765950688&#45;&gt;140161765951472+</title>\n",
       "<path fill=\"none\" stroke=\"#000000\" d=\"M298.1143,-149.1035C308.2389,-144.9008 318.8625,-140.491 328.4358,-136.5172\"/>\n",
       "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"330.0728,-139.6273 337.9669,-132.5609 327.3891,-133.1622 330.0728,-139.6273\"/>\n",
       "</g>\n",
       "<!-- 140161764847512 -->\n",
       "<g id=\"node2\" class=\"node\">\n",
       "<title>140161764847512</title>\n",
       "<polygon fill=\"none\" stroke=\"#000000\" points=\"0,-88.5 0,-157.5 86,-157.5 86,-88.5 0,-88.5\"/>\n",
       "<text text-anchor=\"middle\" x=\"43\" y=\"-142.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\"> </text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"0,-134.5 86,-134.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"43\" y=\"-119.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">data: 2.0000</text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"0,-111.5 86,-111.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"43\" y=\"-96.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">grad: 1.0000</text>\n",
       "</g>\n",
       "<!-- 140161765952928* -->\n",
       "<g id=\"node4\" class=\"node\">\n",
       "<title>140161765952928*</title>\n",
       "<ellipse fill=\"none\" stroke=\"#000000\" cx=\"149\" cy=\"-79\" rx=\"27\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"149\" y=\"-75.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">*</text>\n",
       "</g>\n",
       "<!-- 140161764847512&#45;&gt;140161765952928* -->\n",
       "<g id=\"edge7\" class=\"edge\">\n",
       "<title>140161764847512&#45;&gt;140161765952928*</title>\n",
       "<path fill=\"none\" stroke=\"#000000\" d=\"M86.1143,-105.1035C96.2389,-100.9008 106.8625,-96.491 116.4358,-92.5172\"/>\n",
       "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"118.0728,-95.6273 125.9669,-88.5609 115.3891,-89.1622 118.0728,-95.6273\"/>\n",
       "</g>\n",
       "<!-- 140161765952928 -->\n",
       "<g id=\"node3\" class=\"node\">\n",
       "<title>140161765952928</title>\n",
       "<polygon fill=\"none\" stroke=\"#000000\" points=\"212,-44.5 212,-113.5 298,-113.5 298,-44.5 212,-44.5\"/>\n",
       "<text text-anchor=\"middle\" x=\"255\" y=\"-98.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">*_res</text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"212,-90.5 298,-90.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"255\" y=\"-75.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">data: 2.0000</text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"212,-67.5 298,-67.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"255\" y=\"-52.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">grad: 1.0000</text>\n",
       "</g>\n",
       "<!-- 140161765952928&#45;&gt;140161765951472+ -->\n",
       "<g id=\"edge5\" class=\"edge\">\n",
       "<title>140161765952928&#45;&gt;140161765951472+</title>\n",
       "<path fill=\"none\" stroke=\"#000000\" d=\"M298.1143,-96.8965C308.2389,-101.0992 318.8625,-105.509 328.4358,-109.4828\"/>\n",
       "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"327.3891,-112.8378 337.9669,-113.4391 330.0728,-106.3727 327.3891,-112.8378\"/>\n",
       "</g>\n",
       "<!-- 140161765952928*&#45;&gt;140161765952928 -->\n",
       "<g id=\"edge1\" class=\"edge\">\n",
       "<title>140161765952928*&#45;&gt;140161765952928</title>\n",
       "<path fill=\"none\" stroke=\"#000000\" d=\"M176.0225,-79C183.9553,-79 192.9196,-79 201.8881,-79\"/>\n",
       "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"201.961,-82.5001 211.961,-79 201.9609,-75.5001 201.961,-82.5001\"/>\n",
       "</g>\n",
       "<!-- 140161765951920 -->\n",
       "<g id=\"node5\" class=\"node\">\n",
       "<title>140161765951920</title>\n",
       "<polygon fill=\"none\" stroke=\"#000000\" points=\"646.9942,-88.5 646.9942,-157.5 732.9942,-157.5 732.9942,-88.5 646.9942,-88.5\"/>\n",
       "<text text-anchor=\"middle\" x=\"689.9942\" y=\"-142.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">y</text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"646.9942,-134.5 732.9942,-134.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"689.9942\" y=\"-119.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">data: 3.0000</text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"646.9942,-111.5 732.9942,-111.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"689.9942\" y=\"-96.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">grad: 1.0000</text>\n",
       "</g>\n",
       "<!-- 140161765951920ReLU -->\n",
       "<g id=\"node6\" class=\"node\">\n",
       "<title>140161765951920ReLU</title>\n",
       "<ellipse fill=\"none\" stroke=\"#000000\" cx=\"578.4971\" cy=\"-123\" rx=\"32.4942\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"578.4971\" y=\"-119.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">ReLU</text>\n",
       "</g>\n",
       "<!-- 140161765951920ReLU&#45;&gt;140161765951920 -->\n",
       "<g id=\"edge2\" class=\"edge\">\n",
       "<title>140161765951920ReLU&#45;&gt;140161765951920</title>\n",
       "<path fill=\"none\" stroke=\"#000000\" d=\"M611.0464,-123C619.0851,-123 627.907,-123 636.65,-123\"/>\n",
       "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"636.8255,-126.5001 646.8255,-123 636.8254,-119.5001 636.8255,-126.5001\"/>\n",
       "</g>\n",
       "<!-- 140161764847568 -->\n",
       "<g id=\"node7\" class=\"node\">\n",
       "<title>140161764847568</title>\n",
       "<polygon fill=\"none\" stroke=\"#000000\" points=\"0,-.5 0,-69.5 86,-69.5 86,-.5 0,-.5\"/>\n",
       "<text text-anchor=\"middle\" x=\"43\" y=\"-54.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">x</text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"0,-46.5 86,-46.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"43\" y=\"-31.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">data: 1.0000</text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"0,-23.5 86,-23.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"43\" y=\"-8.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">grad: 2.0000</text>\n",
       "</g>\n",
       "<!-- 140161764847568&#45;&gt;140161765952928* -->\n",
       "<g id=\"edge8\" class=\"edge\">\n",
       "<title>140161764847568&#45;&gt;140161765952928*</title>\n",
       "<path fill=\"none\" stroke=\"#000000\" d=\"M86.1143,-52.8965C96.2389,-57.0992 106.8625,-61.509 116.4358,-65.4828\"/>\n",
       "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"115.3891,-68.8378 125.9669,-69.4391 118.0728,-62.3727 115.3891,-68.8378\"/>\n",
       "</g>\n",
       "<!-- 140161765951472 -->\n",
       "<g id=\"node8\" class=\"node\">\n",
       "<title>140161765951472</title>\n",
       "<polygon fill=\"none\" stroke=\"#000000\" points=\"424,-88.5 424,-157.5 510,-157.5 510,-88.5 424,-88.5\"/>\n",
       "<text text-anchor=\"middle\" x=\"467\" y=\"-142.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">+_res</text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"424,-134.5 510,-134.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"467\" y=\"-119.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">data: 3.0000</text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"424,-111.5 510,-111.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"467\" y=\"-96.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">grad: 1.0000</text>\n",
       "</g>\n",
       "<!-- 140161765951472&#45;&gt;140161765951920ReLU -->\n",
       "<g id=\"edge6\" class=\"edge\">\n",
       "<title>140161765951472&#45;&gt;140161765951920ReLU</title>\n",
       "<path fill=\"none\" stroke=\"#000000\" d=\"M510.1077,-123C518.5595,-123 527.4205,-123 535.857,-123\"/>\n",
       "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"535.881,-126.5001 545.881,-123 535.8809,-119.5001 535.881,-126.5001\"/>\n",
       "</g>\n",
       "<!-- 140161765951472+&#45;&gt;140161765951472 -->\n",
       "<g id=\"edge3\" class=\"edge\">\n",
       "<title>140161765951472+&#45;&gt;140161765951472</title>\n",
       "<path fill=\"none\" stroke=\"#000000\" d=\"M388.0225,-123C395.9553,-123 404.9196,-123 413.8881,-123\"/>\n",
       "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"413.961,-126.5001 423.961,-123 413.9609,-119.5001 413.961,-126.5001\"/>\n",
       "</g>\n",
       "</g>\n",
       "</svg>\n"
      ],
      "text/plain": [
       "<graphviz.graphs.Digraph at 0x7f79f4454400>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x = Var(1.0, label='x')\n",
    "y = (x * 2 + 1).relu(); y.label='y'\n",
    "y.backward()\n",
    "plot_computational_graph(y)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "valid-percentage",
   "metadata": {},
   "source": [
    "## Single Neuron"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "occupied-serve",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Neuron(2, relu)\n",
      "[Var(data=0.07822845723954064), Var(data=0.06208218763697615), Var(data=0.1581407774947987)]\n"
     ]
    }
   ],
   "source": [
    "import random\n",
    "from picograd.nn import Neuron\n",
    "\n",
    "random.seed(5486)\n",
    "n = Neuron(2, activation='relu')\n",
    "print(n)\n",
    "print(n.parameters())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "expired-intensity",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/svg+xml": [
       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
       "<!-- Generated by graphviz version 2.40.1 (20161225.0304)\n",
       " -->\n",
       "<!-- Title: %3 Pages: 1 -->\n",
       "<svg width=\"960pt\" height=\"342pt\"\n",
       " viewBox=\"0.00 0.00 959.99 342.00\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 338)\">\n",
       "<title>%3</title>\n",
       "<polygon fill=\"#ffffff\" stroke=\"transparent\" points=\"-4,4 -4,-338 955.9942,-338 955.9942,4 -4,4\"/>\n",
       "<!-- 140161764566016 -->\n",
       "<g id=\"node1\" class=\"node\">\n",
       "<title>140161764566016</title>\n",
       "<polygon fill=\"none\" stroke=\"#000000\" points=\"643,-132.5 643,-201.5 729,-201.5 729,-132.5 643,-132.5\"/>\n",
       "<text text-anchor=\"middle\" x=\"686\" y=\"-186.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">+_res</text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"643,-178.5 729,-178.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"686\" y=\"-163.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">data: 0.1122</text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"643,-155.5 729,-155.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"686\" y=\"-140.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">grad: 1.0000</text>\n",
       "</g>\n",
       "<!-- 140161764565680ReLU -->\n",
       "<g id=\"node6\" class=\"node\">\n",
       "<title>140161764565680ReLU</title>\n",
       "<ellipse fill=\"none\" stroke=\"#000000\" cx=\"797.4971\" cy=\"-167\" rx=\"32.4942\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"797.4971\" y=\"-163.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">ReLU</text>\n",
       "</g>\n",
       "<!-- 140161764566016&#45;&gt;140161764565680ReLU -->\n",
       "<g id=\"edge11\" class=\"edge\">\n",
       "<title>140161764566016&#45;&gt;140161764565680ReLU</title>\n",
       "<path fill=\"none\" stroke=\"#000000\" d=\"M729.1077,-167C737.5595,-167 746.4205,-167 754.857,-167\"/>\n",
       "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"754.881,-170.5001 764.881,-167 754.8809,-163.5001 754.881,-170.5001\"/>\n",
       "</g>\n",
       "<!-- 140161764566016+ -->\n",
       "<g id=\"node2\" class=\"node\">\n",
       "<title>140161764566016+</title>\n",
       "<ellipse fill=\"none\" stroke=\"#000000\" cx=\"580\" cy=\"-167\" rx=\"27\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"580\" y=\"-163.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">+</text>\n",
       "</g>\n",
       "<!-- 140161764566016+&#45;&gt;140161764566016 -->\n",
       "<g id=\"edge1\" class=\"edge\">\n",
       "<title>140161764566016+&#45;&gt;140161764566016</title>\n",
       "<path fill=\"none\" stroke=\"#000000\" d=\"M607.0225,-167C614.9553,-167 623.9196,-167 632.8881,-167\"/>\n",
       "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"632.961,-170.5001 642.961,-167 632.9609,-163.5001 632.961,-170.5001\"/>\n",
       "</g>\n",
       "<!-- 140161764897288 -->\n",
       "<g id=\"node3\" class=\"node\">\n",
       "<title>140161764897288</title>\n",
       "<polygon fill=\"none\" stroke=\"#000000\" points=\"212,-88.5 212,-157.5 302,-157.5 302,-88.5 212,-88.5\"/>\n",
       "<text text-anchor=\"middle\" x=\"257\" y=\"-142.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\"> </text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"212,-134.5 302,-134.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"257\" y=\"-119.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">data: 0.0621</text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"212,-111.5 302,-111.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"257\" y=\"-96.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">grad: &#45;2.0000</text>\n",
       "</g>\n",
       "<!-- 140161764565960* -->\n",
       "<g id=\"node15\" class=\"node\">\n",
       "<title>140161764565960*</title>\n",
       "<ellipse fill=\"none\" stroke=\"#000000\" cx=\"365\" cy=\"-123\" rx=\"27\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"365\" y=\"-119.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">*</text>\n",
       "</g>\n",
       "<!-- 140161764897288&#45;&gt;140161764565960* -->\n",
       "<g id=\"edge14\" class=\"edge\">\n",
       "<title>140161764897288&#45;&gt;140161764565960*</title>\n",
       "<path fill=\"none\" stroke=\"#000000\" d=\"M302.176,-123C310.7108,-123 319.548,-123 327.8032,-123\"/>\n",
       "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"327.8821,-126.5001 337.8821,-123 327.8821,-119.5001 327.8821,-126.5001\"/>\n",
       "</g>\n",
       "<!-- 140161764565624 -->\n",
       "<g id=\"node4\" class=\"node\">\n",
       "<title>140161764565624</title>\n",
       "<polygon fill=\"none\" stroke=\"#000000\" points=\"0,-220.5 0,-289.5 86,-289.5 86,-220.5 0,-220.5\"/>\n",
       "<text text-anchor=\"middle\" x=\"43\" y=\"-274.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\"> </text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"0,-266.5 86,-266.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"43\" y=\"-251.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">data: 1.0000</text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"0,-243.5 86,-243.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"43\" y=\"-228.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">grad: 0.0782</text>\n",
       "</g>\n",
       "<!-- 140161764565904* -->\n",
       "<g id=\"node13\" class=\"node\">\n",
       "<title>140161764565904*</title>\n",
       "<ellipse fill=\"none\" stroke=\"#000000\" cx=\"149\" cy=\"-211\" rx=\"27\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"149\" y=\"-207.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">*</text>\n",
       "</g>\n",
       "<!-- 140161764565624&#45;&gt;140161764565904* -->\n",
       "<g id=\"edge10\" class=\"edge\">\n",
       "<title>140161764565624&#45;&gt;140161764565904*</title>\n",
       "<path fill=\"none\" stroke=\"#000000\" d=\"M86.1143,-237.1035C96.2389,-232.9008 106.8625,-228.491 116.4358,-224.5172\"/>\n",
       "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"118.0728,-227.6273 125.9669,-220.5609 115.3891,-221.1622 118.0728,-227.6273\"/>\n",
       "</g>\n",
       "<!-- 140161764565680 -->\n",
       "<g id=\"node5\" class=\"node\">\n",
       "<title>140161764565680</title>\n",
       "<polygon fill=\"none\" stroke=\"#000000\" points=\"865.9942,-132.5 865.9942,-201.5 951.9942,-201.5 951.9942,-132.5 865.9942,-132.5\"/>\n",
       "<text text-anchor=\"middle\" x=\"908.9942\" y=\"-186.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">ReLU_res</text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"865.9942,-178.5 951.9942,-178.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"908.9942\" y=\"-163.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">data: 0.1122</text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"865.9942,-155.5 951.9942,-155.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"908.9942\" y=\"-140.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">grad: 1.0000</text>\n",
       "</g>\n",
       "<!-- 140161764565680ReLU&#45;&gt;140161764565680 -->\n",
       "<g id=\"edge2\" class=\"edge\">\n",
       "<title>140161764565680ReLU&#45;&gt;140161764565680</title>\n",
       "<path fill=\"none\" stroke=\"#000000\" d=\"M830.0464,-167C838.0851,-167 846.907,-167 855.65,-167\"/>\n",
       "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"855.8255,-170.5001 865.8255,-167 855.8254,-163.5001 855.8255,-170.5001\"/>\n",
       "</g>\n",
       "<!-- 140161764565176 -->\n",
       "<g id=\"node7\" class=\"node\">\n",
       "<title>140161764565176</title>\n",
       "<polygon fill=\"none\" stroke=\"#000000\" points=\"214,-264.5 214,-333.5 300,-333.5 300,-264.5 214,-264.5\"/>\n",
       "<text text-anchor=\"middle\" x=\"257\" y=\"-318.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\"> </text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"214,-310.5 300,-310.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"257\" y=\"-295.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">data: 0.1581</text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"214,-287.5 300,-287.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"257\" y=\"-272.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">grad: 1.0000</text>\n",
       "</g>\n",
       "<!-- 140161764565792+ -->\n",
       "<g id=\"node10\" class=\"node\">\n",
       "<title>140161764565792+</title>\n",
       "<ellipse fill=\"none\" stroke=\"#000000\" cx=\"365\" cy=\"-211\" rx=\"27\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"365\" y=\"-207.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">+</text>\n",
       "</g>\n",
       "<!-- 140161764565176&#45;&gt;140161764565792+ -->\n",
       "<g id=\"edge13\" class=\"edge\">\n",
       "<title>140161764565176&#45;&gt;140161764565792+</title>\n",
       "<path fill=\"none\" stroke=\"#000000\" d=\"M299.3745,-264.4726C313.1068,-253.2834 327.892,-241.2361 339.9458,-231.4145\"/>\n",
       "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"342.2663,-234.0385 347.8078,-225.0085 337.8446,-228.6119 342.2663,-234.0385\"/>\n",
       "</g>\n",
       "<!-- 140161764565736 -->\n",
       "<g id=\"node8\" class=\"node\">\n",
       "<title>140161764565736</title>\n",
       "<polygon fill=\"none\" stroke=\"#000000\" points=\"212.5,-.5 212.5,-69.5 301.5,-69.5 301.5,-.5 212.5,-.5\"/>\n",
       "<text text-anchor=\"middle\" x=\"257\" y=\"-54.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\"> </text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"212.5,-46.5 301.5,-46.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"257\" y=\"-31.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">data: &#45;2.0000</text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"212.5,-23.5 301.5,-23.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"257\" y=\"-8.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">grad: 0.0621</text>\n",
       "</g>\n",
       "<!-- 140161764565736&#45;&gt;140161764565960* -->\n",
       "<g id=\"edge8\" class=\"edge\">\n",
       "<title>140161764565736&#45;&gt;140161764565960*</title>\n",
       "<path fill=\"none\" stroke=\"#000000\" d=\"M299.3745,-69.5274C313.1068,-80.7166 327.892,-92.7639 339.9458,-102.5855\"/>\n",
       "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"337.8446,-105.3881 347.8078,-108.9915 342.2663,-99.9615 337.8446,-105.3881\"/>\n",
       "</g>\n",
       "<!-- 140161764565792 -->\n",
       "<g id=\"node9\" class=\"node\">\n",
       "<title>140161764565792</title>\n",
       "<polygon fill=\"none\" stroke=\"#000000\" points=\"429.5,-176.5 429.5,-245.5 515.5,-245.5 515.5,-176.5 429.5,-176.5\"/>\n",
       "<text text-anchor=\"middle\" x=\"472.5\" y=\"-230.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">+_res</text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"429.5,-222.5 515.5,-222.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"472.5\" y=\"-207.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">data: 0.2364</text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"429.5,-199.5 515.5,-199.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"472.5\" y=\"-184.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">grad: 1.0000</text>\n",
       "</g>\n",
       "<!-- 140161764565792&#45;&gt;140161764566016+ -->\n",
       "<g id=\"edge6\" class=\"edge\">\n",
       "<title>140161764565792&#45;&gt;140161764566016+</title>\n",
       "<path fill=\"none\" stroke=\"#000000\" d=\"M515.605,-193.357C526.2069,-189.0177 537.3844,-184.4427 547.3971,-180.3444\"/>\n",
       "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"549.0281,-183.4587 556.9571,-176.4315 546.3765,-176.9804 549.0281,-183.4587\"/>\n",
       "</g>\n",
       "<!-- 140161764565792+&#45;&gt;140161764565792 -->\n",
       "<g id=\"edge3\" class=\"edge\">\n",
       "<title>140161764565792+&#45;&gt;140161764565792</title>\n",
       "<path fill=\"none\" stroke=\"#000000\" d=\"M392.1268,-211C400.4401,-211 409.8902,-211 419.317,-211\"/>\n",
       "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"419.4821,-214.5001 429.482,-211 419.482,-207.5001 419.4821,-214.5001\"/>\n",
       "</g>\n",
       "<!-- 140161764897120 -->\n",
       "<g id=\"node11\" class=\"node\">\n",
       "<title>140161764897120</title>\n",
       "<polygon fill=\"none\" stroke=\"#000000\" points=\"0,-132.5 0,-201.5 86,-201.5 86,-132.5 0,-132.5\"/>\n",
       "<text text-anchor=\"middle\" x=\"43\" y=\"-186.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\"> </text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"0,-178.5 86,-178.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"43\" y=\"-163.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">data: 0.0782</text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"0,-155.5 86,-155.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"43\" y=\"-140.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">grad: 1.0000</text>\n",
       "</g>\n",
       "<!-- 140161764897120&#45;&gt;140161764565904* -->\n",
       "<g id=\"edge7\" class=\"edge\">\n",
       "<title>140161764897120&#45;&gt;140161764565904*</title>\n",
       "<path fill=\"none\" stroke=\"#000000\" d=\"M86.1143,-184.8965C96.2389,-189.0992 106.8625,-193.509 116.4358,-197.4828\"/>\n",
       "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"115.3891,-200.8378 125.9669,-201.4391 118.0728,-194.3727 115.3891,-200.8378\"/>\n",
       "</g>\n",
       "<!-- 140161764565904 -->\n",
       "<g id=\"node12\" class=\"node\">\n",
       "<title>140161764565904</title>\n",
       "<polygon fill=\"none\" stroke=\"#000000\" points=\"214,-176.5 214,-245.5 300,-245.5 300,-176.5 214,-176.5\"/>\n",
       "<text text-anchor=\"middle\" x=\"257\" y=\"-230.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">*_res</text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"214,-222.5 300,-222.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"257\" y=\"-207.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">data: 0.0782</text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"214,-199.5 300,-199.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"257\" y=\"-184.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">grad: 1.0000</text>\n",
       "</g>\n",
       "<!-- 140161764565904&#45;&gt;140161764565792+ -->\n",
       "<g id=\"edge9\" class=\"edge\">\n",
       "<title>140161764565904&#45;&gt;140161764565792+</title>\n",
       "<path fill=\"none\" stroke=\"#000000\" d=\"M300.3054,-211C309.4162,-211 318.9501,-211 327.8127,-211\"/>\n",
       "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"327.8429,-214.5001 337.8429,-211 327.8429,-207.5001 327.8429,-214.5001\"/>\n",
       "</g>\n",
       "<!-- 140161764565904*&#45;&gt;140161764565904 -->\n",
       "<g id=\"edge4\" class=\"edge\">\n",
       "<title>140161764565904*&#45;&gt;140161764565904</title>\n",
       "<path fill=\"none\" stroke=\"#000000\" d=\"M176.253,-211C184.6049,-211 194.099,-211 203.5696,-211\"/>\n",
       "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"203.782,-214.5001 213.782,-211 203.7819,-207.5001 203.782,-214.5001\"/>\n",
       "</g>\n",
       "<!-- 140161764565960 -->\n",
       "<g id=\"node14\" class=\"node\">\n",
       "<title>140161764565960</title>\n",
       "<polygon fill=\"none\" stroke=\"#000000\" points=\"428,-88.5 428,-157.5 517,-157.5 517,-88.5 428,-88.5\"/>\n",
       "<text text-anchor=\"middle\" x=\"472.5\" y=\"-142.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">*_res</text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"428,-134.5 517,-134.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"472.5\" y=\"-119.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">data: &#45;0.1242</text>\n",
       "<polyline fill=\"none\" stroke=\"#000000\" points=\"428,-111.5 517,-111.5 \"/>\n",
       "<text text-anchor=\"middle\" x=\"472.5\" y=\"-96.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">grad: 1.0000</text>\n",
       "</g>\n",
       "<!-- 140161764565960&#45;&gt;140161764566016+ -->\n",
       "<g id=\"edge12\" class=\"edge\">\n",
       "<title>140161764565960&#45;&gt;140161764566016+</title>\n",
       "<path fill=\"none\" stroke=\"#000000\" d=\"M517.1558,-141.2777C527.2599,-145.4134 537.8057,-149.7298 547.3083,-153.6192\"/>\n",
       "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"546.1907,-156.9436 556.7713,-157.4924 548.8424,-150.4652 546.1907,-156.9436\"/>\n",
       "</g>\n",
       "<!-- 140161764565960*&#45;&gt;140161764565960 -->\n",
       "<g id=\"edge5\" class=\"edge\">\n",
       "<title>140161764565960*&#45;&gt;140161764565960</title>\n",
       "<path fill=\"none\" stroke=\"#000000\" d=\"M392.1268,-123C399.9901,-123 408.8705,-123 417.7862,-123\"/>\n",
       "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"417.8151,-126.5001 427.8151,-123 417.815,-119.5001 417.8151,-126.5001\"/>\n",
       "</g>\n",
       "</g>\n",
       "</svg>\n"
      ],
      "text/plain": [
       "<graphviz.graphs.Digraph at 0x7f79f4302470>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "n.zero_grad()\n",
    "x = [Var(1.0), Var(-2.0)]\n",
    "y = n(x)\n",
    "y.backward()\n",
    "plot_computational_graph(y)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "violent-louisville",
   "metadata": {},
   "source": [
    "## Multi-Layer Perceptron\n",
    "\n",
    "### Learning XOR, using an explicit training loop"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "subject-irish",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[Var(data=0), Var(data=0)], [Var(data=0), Var(data=1)], [Var(data=1), Var(data=0)], [Var(data=1), Var(data=1)]]\n"
     ]
    }
   ],
   "source": [
    "from picograd.engine import Var\n",
    "\n",
    "x_train = [list(map(Var, x)) for x in [[0, 0], [0, 1], [1, 0], [1, 1]]]\n",
    "print(x_train)\n",
    "y_train = [Var(0), Var(1), Var(1), Var(0)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "frozen-reunion",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MLP of [Layer of [Neuron(2, relu), Neuron(2, relu), Neuron(2, relu), Neuron(2, relu)], Layer of [Neuron(4, linear)]]\n",
      "Number of parameters: 17\n",
      "Epoch [0/10000], loss: 1.4299425249394688\n",
      "Epoch [100/10000], loss: 0.2792828142137011\n",
      "Epoch [200/10000], loss: 0.2235485288422007\n",
      "Epoch [300/10000], loss: 0.17934672002547392\n",
      "Epoch [400/10000], loss: 0.15320426626822\n",
      "Epoch [500/10000], loss: 0.1341904930563754\n",
      "Epoch [600/10000], loss: 0.11853619146152647\n",
      "Epoch [700/10000], loss: 0.10486788583038205\n",
      "Epoch [800/10000], loss: 0.09311760445132325\n",
      "Epoch [900/10000], loss: 0.08213031180430044\n",
      "Epoch [1000/10000], loss: 0.07161372369356092\n",
      "Epoch [1100/10000], loss: 0.06212125098508617\n",
      "Epoch [1200/10000], loss: 0.053104030724664304\n",
      "Epoch [1300/10000], loss: 0.04482746863669852\n",
      "Epoch [1400/10000], loss: 0.03720344979371739\n",
      "Epoch [1500/10000], loss: 0.03059736744604425\n",
      "Epoch [1600/10000], loss: 0.024790459047215798\n",
      "Epoch [1700/10000], loss: 0.019816903158239706\n",
      "Epoch [1800/10000], loss: 0.015666073679820564\n",
      "Epoch [1900/10000], loss: 0.012197822202859516\n",
      "Epoch [2000/10000], loss: 0.009456363165101973\n",
      "Epoch [2100/10000], loss: 0.007224839559199983\n",
      "Epoch [2200/10000], loss: 0.005493088148877028\n",
      "Epoch [2300/10000], loss: 0.004128668318731865\n",
      "Epoch [2400/10000], loss: 0.0030668110748517112\n",
      "Epoch [2500/10000], loss: 0.0022897591223689597\n",
      "Epoch [2600/10000], loss: 0.001688710563764231\n",
      "Epoch [2700/10000], loss: 0.0012334547040952623\n",
      "Epoch [2800/10000], loss: 0.0009061558347720507\n",
      "Epoch [2900/10000], loss: 0.0006588033550179328\n",
      "Epoch [3000/10000], loss: 0.00048096426979822837\n",
      "Epoch [3100/10000], loss: 0.0003484466597970812\n",
      "Epoch [3200/10000], loss: 0.00025092896423255294\n",
      "Epoch [3300/10000], loss: 0.00018123922191197337\n",
      "Epoch [3400/10000], loss: 0.00013008618731874958\n",
      "Epoch [3500/10000], loss: 9.408220882989879e-05\n",
      "Epoch [3600/10000], loss: 6.765548827215806e-05\n",
      "Epoch [3700/10000], loss: 4.827518205504623e-05\n",
      "Epoch [3800/10000], loss: 3.483263237112168e-05\n",
      "Epoch [3900/10000], loss: 2.4814216091818728e-05\n",
      "Epoch [4000/10000], loss: 1.7730909854878462e-05\n",
      "Epoch [4100/10000], loss: 1.2759650792841357e-05\n",
      "Epoch [4200/10000], loss: 9.08135261548331e-06\n",
      "Epoch [4300/10000], loss: 6.486719639260885e-06\n",
      "Epoch [4400/10000], loss: 4.650439105225085e-06\n",
      "Epoch [4500/10000], loss: 3.3184461852686543e-06\n",
      "Epoch [4600/10000], loss: 2.365051828597933e-06\n",
      "Epoch [4700/10000], loss: 1.6922664544811265e-06\n",
      "Epoch [4800/10000], loss: 1.2114680826182485e-06\n",
      "Epoch [4900/10000], loss: 8.667155711140078e-07\n",
      "Epoch [5000/10000], loss: 6.149143381893174e-07\n",
      "Epoch [5100/10000], loss: 4.408746951480862e-07\n",
      "Epoch [5200/10000], loss: 3.1461555995051716e-07\n",
      "Epoch [5300/10000], loss: 2.250417998190062e-07\n",
      "Epoch [5400/10000], loss: 1.596119181602337e-07\n",
      "Epoch [5500/10000], loss: 1.1433118807078408e-07\n",
      "Epoch [5600/10000], loss: 8.133040889676093e-08\n",
      "Epoch [5700/10000], loss: 5.8324991374249075e-08\n",
      "Epoch [5800/10000], loss: 4.140751900501509e-08\n",
      "Epoch [5900/10000], loss: 2.964190389520097e-08\n",
      "Epoch [6000/10000], loss: 2.108522409569712e-08\n",
      "Epoch [6100/10000], loss: 1.5105631778611913e-08\n",
      "Epoch [6200/10000], loss: 1.0720953271124732e-08\n",
      "Epoch [6300/10000], loss: 7.688421682400142e-09\n",
      "Epoch [6400/10000], loss: 5.465826452419515e-09\n",
      "Epoch [6500/10000], loss: 3.8903064119308715e-09\n",
      "Epoch [6600/10000], loss: 2.7756012766948205e-09\n",
      "Epoch [6700/10000], loss: 1.9938634199187977e-09\n",
      "Epoch [6800/10000], loss: 1.4179779178003249e-09\n",
      "Epoch [6900/10000], loss: 1.0087427724484165e-09\n",
      "Epoch [7000/10000], loss: 7.184643937205386e-10\n",
      "Epoch [7100/10000], loss: 5.128769114299914e-10\n",
      "Epoch [7200/10000], loss: 3.675216602802475e-10\n",
      "Epoch [7300/10000], loss: 2.6141264244068636e-10\n",
      "Epoch [7400/10000], loss: 1.8594620961309786e-10\n",
      "Epoch [7500/10000], loss: 1.3276893340896104e-10\n",
      "Epoch [7600/10000], loss: 9.528171966807836e-11\n",
      "Epoch [7700/10000], loss: 6.785402834875762e-11\n",
      "Epoch [7800/10000], loss: 4.8189111032269654e-11\n",
      "Epoch [7900/10000], loss: 3.4360954283277636e-11\n",
      "Epoch [8000/10000], loss: 2.456171341132925e-11\n",
      "Epoch [8100/10000], loss: 1.7564367273343313e-11\n",
      "Epoch [8200/10000], loss: 1.2498591562125707e-11\n",
      "Epoch [8300/10000], loss: 8.897000659252059e-12\n",
      "Epoch [8400/10000], loss: 6.3541744339634595e-12\n",
      "Epoch [8500/10000], loss: 4.544528973770064e-12\n",
      "Epoch [8600/10000], loss: 3.247216772454529e-12\n",
      "Epoch [8700/10000], loss: 2.3061238623718614e-12\n",
      "Epoch [8800/10000], loss: 1.6447948951803252e-12\n",
      "Epoch [8900/10000], loss: 1.1763888085029327e-12\n",
      "Epoch [9000/10000], loss: 8.405940314242829e-13\n",
      "Epoch [9100/10000], loss: 6.008330449261915e-13\n",
      "Epoch [9200/10000], loss: 4.259170307374398e-13\n",
      "Epoch [9300/10000], loss: 3.042786875748597e-13\n",
      "Epoch [9400/10000], loss: 2.17592961803541e-13\n",
      "Epoch [9500/10000], loss: 1.5551538613309814e-13\n",
      "Epoch [9600/10000], loss: 1.1040035913116744e-13\n",
      "Epoch [9700/10000], loss: 7.873490859974654e-14\n",
      "Epoch [9800/10000], loss: 5.6318663324280575e-14\n",
      "Epoch [9900/10000], loss: 4.011143798064417e-14\n"
     ]
    }
   ],
   "source": [
    "from picograd.nn import MLP\n",
    "from picograd.optim import SGD, Adam\n",
    "from picograd.metrics import mean_squared_error\n",
    "\n",
    "model = MLP(in_features=2, layers=[4, 1], activations=['relu', 'linear']) # 1 hidden layer\n",
    "print(model)\n",
    "print(f\"Number of parameters: {len(model.parameters())}\")\n",
    "\n",
    "optimizer = SGD(model.parameters(), lr=0.01)\n",
    "num_epochs = 10000\n",
    "\n",
    "for epoch in range(num_epochs):\n",
    "    # Reset the gradients of model parameters\n",
    "    optimizer.zero_grad()\n",
    "    \n",
    "    # Forward pass\n",
    "    outputs = list(map(model, x_train))\n",
    "    \n",
    "    # Loss computation\n",
    "    y_pred = [item for sublist in outputs for item in sublist]\n",
    "    loss = mean_squared_error(y_train, y_pred)\n",
    "    \n",
    "    # Backprop and gradient descent\n",
    "    loss.backward()\n",
    "    optimizer.step()\n",
    "    \n",
    "    if epoch % 100 == 0:\n",
    "        print(f\"Epoch [{epoch}/{num_epochs}], loss: {loss.data}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "deluxe-cabinet",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predicted: 0.00000, actual: 0\n",
      "Predicted: 1.00000, actual: 1\n",
      "Predicted: 1.00000, actual: 1\n",
      "Predicted: 0.00000, actual: 0\n"
     ]
    }
   ],
   "source": [
    "# Compare predictions and ground truth\n",
    "for x, y in zip(x_train, y_train):\n",
    "    y_pred = model(x)\n",
    "    print(f\"Predicted: {y_pred[0].data:.5f}, actual: {y.data}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "pediatric-lightweight",
   "metadata": {},
   "source": [
    "## Classifying 2D data, using the training API\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "alleged-bones",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from sklearn.datasets import make_moons"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "arranged-sympathy",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Setup plots\n",
    "%matplotlib inline\n",
    "plt.rcParams['figure.figsize'] = 10, 8\n",
    "%config InlineBackend.figure_format = 'retina'\n",
    "sns.set()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "consecutive-monte",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "x_train: (200, 2). y_train: (200,)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 472,
       "width": 615
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_planar_data(X, y):\n",
    "    \"\"\"Plot some planar data\"\"\"\n",
    "\n",
    "    plt.figure()\n",
    "    plt.plot(X[y == 0, 0], X[y == 0, 1], 'or', alpha=0.5, label=0)\n",
    "    plt.plot(X[y == 1, 0], X[y == 1, 1], 'ob', alpha=0.5, label=1)\n",
    "    plt.legend()\n",
    "\n",
    "# Generate moon-shaped, non-linearly separable data\n",
    "x_train, y_train = make_moons(n_samples=200, noise=0.10, random_state=0)\n",
    "\n",
    "print(f'x_train: {x_train.shape}. y_train: {y_train.shape}')\n",
    "plot_planar_data(x_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "specific-graduation",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MLP of [Layer of [Neuron(2, relu), Neuron(2, relu), Neuron(2, relu), Neuron(2, relu), Neuron(2, relu), Neuron(2, relu), Neuron(2, relu), Neuron(2, relu), Neuron(2, relu), Neuron(2, relu), Neuron(2, relu), Neuron(2, relu), Neuron(2, relu), Neuron(2, relu), Neuron(2, relu), Neuron(2, relu)], Layer of [Neuron(16, relu), Neuron(16, relu), Neuron(16, relu), Neuron(16, relu), Neuron(16, relu), Neuron(16, relu), Neuron(16, relu), Neuron(16, relu), Neuron(16, relu), Neuron(16, relu), Neuron(16, relu), Neuron(16, relu), Neuron(16, relu), Neuron(16, relu), Neuron(16, relu), Neuron(16, relu)], Layer of [Neuron(16, linear)]]\n",
      "Number of parameters: 337\n",
      "Epoch [1/70], loss: 11.169831, accuracy: 36.50%\n",
      "Epoch [2/70], loss: 2.362962, accuracy: 49.50%\n",
      "Epoch [3/70], loss: 1.046105, accuracy: 80.00%\n",
      "Epoch [4/70], loss: 0.622207, accuracy: 91.50%\n",
      "Epoch [5/70], loss: 0.423670, accuracy: 94.50%\n",
      "Epoch [6/70], loss: 0.368302, accuracy: 96.50%\n",
      "Epoch [7/70], loss: 0.271892, accuracy: 96.00%\n",
      "Epoch [8/70], loss: 0.241696, accuracy: 97.00%\n",
      "Epoch [9/70], loss: 0.203948, accuracy: 97.50%\n",
      "Epoch [10/70], loss: 0.188012, accuracy: 97.50%\n",
      "Epoch [11/70], loss: 0.156890, accuracy: 98.00%\n",
      "Epoch [12/70], loss: 0.133703, accuracy: 98.50%\n",
      "Epoch [13/70], loss: 0.112515, accuracy: 99.00%\n",
      "Epoch [14/70], loss: 0.101089, accuracy: 99.50%\n",
      "Epoch [15/70], loss: 0.095818, accuracy: 99.50%\n",
      "Epoch [16/70], loss: 0.097507, accuracy: 99.50%\n",
      "Epoch [17/70], loss: 0.104294, accuracy: 98.50%\n",
      "Epoch [18/70], loss: 0.086418, accuracy: 99.50%\n",
      "Epoch [19/70], loss: 0.095117, accuracy: 99.50%\n",
      "Epoch [20/70], loss: 0.081014, accuracy: 99.50%\n",
      "Epoch [21/70], loss: 0.075559, accuracy: 100.00%\n",
      "Epoch [22/70], loss: 0.074256, accuracy: 99.50%\n",
      "Epoch [23/70], loss: 0.069721, accuracy: 99.50%\n",
      "Epoch [24/70], loss: 0.078404, accuracy: 100.00%\n",
      "Epoch [25/70], loss: 0.070333, accuracy: 100.00%\n",
      "Epoch [26/70], loss: 0.079556, accuracy: 99.50%\n",
      "Epoch [27/70], loss: 0.101585, accuracy: 99.50%\n",
      "Epoch [28/70], loss: 0.103068, accuracy: 99.50%\n",
      "Epoch [29/70], loss: 0.095077, accuracy: 99.50%\n",
      "Epoch [30/70], loss: 0.077156, accuracy: 99.50%\n",
      "Epoch [31/70], loss: 0.085611, accuracy: 100.00%\n",
      "Epoch [32/70], loss: 0.068080, accuracy: 99.50%\n",
      "Epoch [33/70], loss: 0.068150, accuracy: 100.00%\n",
      "Epoch [34/70], loss: 0.066271, accuracy: 100.00%\n",
      "Epoch [35/70], loss: 0.064284, accuracy: 100.00%\n",
      "Epoch [36/70], loss: 0.074462, accuracy: 100.00%\n",
      "Epoch [37/70], loss: 0.064987, accuracy: 100.00%\n",
      "Epoch [38/70], loss: 0.072650, accuracy: 100.00%\n",
      "Epoch [39/70], loss: 0.064528, accuracy: 100.00%\n",
      "Epoch [40/70], loss: 0.058341, accuracy: 100.00%\n",
      "Epoch [41/70], loss: 0.061261, accuracy: 99.50%\n",
      "Epoch [42/70], loss: 0.057830, accuracy: 100.00%\n",
      "Epoch [43/70], loss: 0.056442, accuracy: 100.00%\n",
      "Epoch [44/70], loss: 0.058671, accuracy: 100.00%\n",
      "Epoch [45/70], loss: 0.055383, accuracy: 100.00%\n",
      "Epoch [46/70], loss: 0.055396, accuracy: 100.00%\n",
      "Epoch [47/70], loss: 0.058156, accuracy: 100.00%\n",
      "Epoch [48/70], loss: 0.059454, accuracy: 100.00%\n",
      "Epoch [49/70], loss: 0.057368, accuracy: 100.00%\n",
      "Epoch [50/70], loss: 0.063649, accuracy: 100.00%\n",
      "Epoch [51/70], loss: 0.054654, accuracy: 100.00%\n",
      "Epoch [52/70], loss: 0.056615, accuracy: 100.00%\n",
      "Epoch [53/70], loss: 0.060014, accuracy: 100.00%\n",
      "Epoch [54/70], loss: 0.056679, accuracy: 99.50%\n",
      "Epoch [55/70], loss: 0.052797, accuracy: 100.00%\n",
      "Epoch [56/70], loss: 0.056187, accuracy: 100.00%\n",
      "Epoch [57/70], loss: 0.052618, accuracy: 100.00%\n",
      "Epoch [58/70], loss: 0.055562, accuracy: 100.00%\n",
      "Epoch [59/70], loss: 0.067872, accuracy: 100.00%\n",
      "Epoch [60/70], loss: 0.062357, accuracy: 100.00%\n",
      "Epoch [61/70], loss: 0.058031, accuracy: 100.00%\n",
      "Epoch [62/70], loss: 0.057308, accuracy: 100.00%\n",
      "Epoch [63/70], loss: 0.056521, accuracy: 99.50%\n",
      "Epoch [64/70], loss: 0.053146, accuracy: 100.00%\n",
      "Epoch [65/70], loss: 0.052805, accuracy: 100.00%\n",
      "Epoch [66/70], loss: 0.058053, accuracy: 100.00%\n",
      "Epoch [67/70], loss: 0.056890, accuracy: 100.00%\n",
      "Epoch [68/70], loss: 0.061410, accuracy: 100.00%\n",
      "Epoch [69/70], loss: 0.064117, accuracy: 100.00%\n",
      "Epoch [70/70], loss: 0.059491, accuracy: 100.00%\n"
     ]
    }
   ],
   "source": [
    "from picograd.nn import MLP\n",
    "from picograd.engine import Var\n",
    "from picograd.data import BatchIterator\n",
    "from picograd.trainer import Trainer\n",
    "from picograd.optim import SGD, Adam\n",
    "from picograd.metrics import binary_accuracy, mean_squared_error\n",
    "\n",
    "\n",
    "model = MLP(in_features=2, layers=[16, 16, 1], activations=['relu', 'relu', 'linear']) # 2 hidden layers\n",
    "print(model)\n",
    "print(f\"Number of parameters: {len(model.parameters())}\")\n",
    "\n",
    "optimizer = SGD(model.parameters(), lr=0.05)\n",
    "data_iterator = BatchIterator(x_train, list(map(Var, y_train)))\n",
    "trainer = Trainer(model, optimizer, loss=mean_squared_error, acc_metric=binary_accuracy)\n",
    "\n",
    "history = trainer.fit(data_iterator, num_epochs=70, verbose=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "suspended-great",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 500,
       "width": 615
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_loss_acc(history):\n",
    "    \"\"\"Plot training and (optionally) validation loss and accuracy\"\"\"\n",
    "\n",
    "    loss = history['loss']\n",
    "    epochs = range(1, len(loss) + 1)\n",
    "\n",
    "    plt.figure()\n",
    "\n",
    "    plt.subplot(2, 1, 1)\n",
    "    plt.plot(epochs, loss, '.--', label='Training loss')\n",
    "    final_loss = loss[-1]\n",
    "    title = 'Training loss: {:.4f}'.format(final_loss)\n",
    "    plt.ylabel('Loss')\n",
    "    plt.title(title)\n",
    "    plt.legend()\n",
    "\n",
    "    acc = history['acc']\n",
    "\n",
    "    plt.subplot(2, 1, 2)\n",
    "    plt.plot(epochs, acc, '.--', label='Training acc')\n",
    "    final_acc = acc[-1]\n",
    "    title = 'Training accuracy: {:.2f}%'.format(final_acc * 100)\n",
    "    plt.xlabel('Epochs')\n",
    "    plt.ylabel('Accuracy')\n",
    "    plt.title(title)\n",
    "    plt.legend()\n",
    "\n",
    "plot_loss_acc(history)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "national-triumph",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 468,
       "width": 612
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_decision_boundary(model, X, y):\n",
    "    \"\"\"Plot a decision boundary\"\"\"\n",
    "\n",
    "    # Set min and max values and give it some padding\n",
    "    x_min, x_max = X[:, 0].min() - .5, X[:, 0].max() + .5\n",
    "    y_min, y_max = X[:, 1].min() - .5, X[:, 1].max() + .5\n",
    "    \n",
    "    h = 0.1\n",
    "    # Generate a grid of points with distance h between them\n",
    "    xx, yy = np.meshgrid(np.arange(x_min, x_max, h),\n",
    "                         np.arange(y_min, y_max, h))\n",
    "    \n",
    "    # Predict the function value for the whole grid\n",
    "    Xmesh = np.c_[xx.ravel(), yy.ravel()]\n",
    "    inputs = [list(map(Var, xrow)) for xrow in Xmesh]\n",
    "    outputs = list(map(model, inputs))\n",
    "    y_pred = [item for sublist in outputs for item in sublist]\n",
    "    Z = np.array([s.data > 0.5 for s in y_pred])\n",
    "    Z = Z.reshape(xx.shape)\n",
    "    \n",
    "    # Plot the contour and training examples \n",
    "    plt.contourf(xx, yy, Z, cmap=plt.cm.Spectral, alpha=0.8)\n",
    "    plt.scatter(X[:, 0], X[:, 1], c=y, s=40, cmap=plt.cm.Spectral)\n",
    "    plt.xlim(xx.min(), xx.max())\n",
    "    plt.ylim(yy.min(), yy.max())\n",
    "\n",
    "plot_decision_boundary(model, x_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "caroline-interview",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predicted: -0.03407, actual: 0\n",
      "Predicted: 1.00691, actual: 1\n",
      "Predicted: 0.93207, actual: 1\n",
      "Predicted: -0.24457, actual: 0\n",
      "Predicted: 0.95621, actual: 1\n",
      "Predicted: 0.95621, actual: 1\n",
      "Predicted: -0.02645, actual: 0\n",
      "Predicted: 0.99226, actual: 1\n",
      "Predicted: -0.03246, actual: 0\n",
      "Predicted: 1.06592, actual: 1\n",
      "Predicted: 0.19223, actual: 0\n",
      "Predicted: 0.93757, actual: 1\n",
      "Predicted: 0.95621, actual: 1\n",
      "Predicted: 0.94970, actual: 1\n",
      "Predicted: -0.18090, actual: 0\n",
      "Predicted: -0.01868, actual: 0\n",
      "Predicted: -0.04211, actual: 0\n",
      "Predicted: 0.91280, actual: 1\n",
      "Predicted: 0.01215, actual: 0\n",
      "Predicted: -0.24846, actual: 0\n",
      "Predicted: 0.95621, actual: 1\n",
      "Predicted: 0.97209, actual: 1\n",
      "Predicted: -0.10186, actual: 0\n",
      "Predicted: 0.91185, actual: 1\n",
      "Predicted: -0.12059, actual: 0\n",
      "Predicted: 0.96167, actual: 1\n",
      "Predicted: 0.95621, actual: 1\n",
      "Predicted: 0.95621, actual: 1\n",
      "Predicted: 1.03282, actual: 1\n",
      "Predicted: 0.12694, actual: 0\n"
     ]
    }
   ],
   "source": [
    "# Compare predictions and ground truth\n",
    "for x, y in zip(x_train[:30], y_train[:30]):\n",
    "    y_pred = model(x)\n",
    "    print(f\"Predicted: {y_pred[0].data:.5f}, actual: {y}\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}