simple_2-2-2_identity.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "Collapsed": "false"
   },
   "source": [
    "# 2-2-2 Identity PC Network\n",
    "Purpose: A baseline implementation to compare to."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "Collapsed": "false"
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from copy import deepcopy\n",
    "import torch\n",
    "from torch.utils.data import Dataset, DataLoader\n",
    "from tqdm import tqdm\n",
    "\n",
    "%load_ext autoreload\n",
    "%autoreload 1\n",
    "%aimport PCNetwork\n",
    "%aimport PCLayer\n",
    "%aimport PCConnection\n",
    "\n",
    "np.set_printoptions(precision=4)\n",
    "\n",
    "dtype = torch.float32\n",
    "if torch.cuda.is_available():\n",
    "    device = torch.device(\"cuda:5\") # Uncomment this to run on GPU\n",
    "else:\n",
    "    device = torch.device(\"cpu\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "Collapsed": "true"
   },
   "source": [
    "# Dataset class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "Collapsed": "false"
   },
   "outputs": [],
   "source": [
    "class SimpleDataset(Dataset):\n",
    "    def __init__(self, A, B, n=300, Anoise=0.1, Bnoise=0.):\n",
    "        self.samples = []\n",
    "        self.n_classes = len(A)\n",
    "        self.input_dim = len(A[0])\n",
    "        for i in range(n):\n",
    "            r = np.random.randint(self.n_classes)\n",
    "            sample = [A[r]+Anoise*torch.randn_like(A[r]), B[r]+Bnoise*torch.randn_like(B[r])]\n",
    "            #sample = [A[r]+noise*torch.randn_like(A[r]), B[r]+noise*torch.randn_like(B[r])]\n",
    "            self.samples.append(sample)\n",
    "        \n",
    "    def __getitem__(self, idx):\n",
    "        return self.samples[idx]\n",
    "        \n",
    "    def __len__(self):\n",
    "        return len(self.samples)\n",
    "    \n",
    "    def Inputs(self):\n",
    "        x = []\n",
    "        for s in self.samples:\n",
    "            x.append(s[0])\n",
    "        return torch.stack(x)\n",
    "    \n",
    "    def Targets(self):\n",
    "        t = []\n",
    "        for s in self.samples:\n",
    "            t.append(s[1])\n",
    "        return torch.stack(t)\n",
    "            \n",
    "    def InputsOfClass(self, c):\n",
    "        x = []\n",
    "        for s in self.samples:\n",
    "            if torch.argmax(s[1])==c:\n",
    "                x.append(s[0])\n",
    "        return torch.stack(x)\n",
    "    \n",
    "    def ClassMean(self):\n",
    "        xmean = []\n",
    "        for c_idx in range(self.n_classes):\n",
    "            classmean = torch.mean(self.InputsOfClass(c_idx), axis=0)\n",
    "            xmean.append(classmean)\n",
    "        return torch.stack(xmean)\n",
    "            \n",
    "    def Plot(self, labels=[], idx=(1,2), equal=True):\n",
    "        X = self.Inputs()\n",
    "        if len(labels)==0:\n",
    "            labels = self.Targets()\n",
    "        colour_options = ['y', 'r', 'g', 'b', 'k']\n",
    "        cidx = torch.argmax(labels, axis=1)\n",
    "        colours = [colour_options[k] for k in cidx]\n",
    "        plt.scatter(X.cpu()[:,idx[0]], X.cpu()[:,idx[1]], color=colours, marker='.')\n",
    "\n",
    "        if equal:\n",
    "            plt.axis('equal');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "Collapsed": "false"
   },
   "outputs": [],
   "source": [
    "A = torch.tensor([[-1,-2],[3,4]], dtype=torch.float)\n",
    "blah = SimpleDataset(A, A, n=200, Anoise=0.2, Bnoise=0.0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "Collapsed": "false"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "blah.Plot(idx=(0,1));"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "Collapsed": "false"
   },
   "source": [
    "# Create Network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "Collapsed": "false"
   },
   "outputs": [],
   "source": [
    "# 2-2-2 network\n",
    "net = PCNetwork.PCNetwork(device=device)\n",
    "net.AddLayer(PCLayer.PCLayer(n=2, device=device))  # [0] v0 input\n",
    "\n",
    "net.AddLayer(PCLayer.PCLayer(n=2, device=device))  # [1] e1 error\n",
    "net.AddLayer(PCLayer.PCLayer(n=2, device=device))  # [2] v1 class\n",
    "\n",
    "net.Connect(v_idx=0, e_idx=1, type='general', sym=True, act_text='identity')\n",
    "net.Connect(e_idx=1, v_idx=2, type='1to1')\n",
    "\n",
    "net.SetTau(0.05)\n",
    "net.SetGamma(0.1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "Collapsed": "false"
   },
   "source": [
    "# Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "Collapsed": "false"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXIAAAD4CAYAAADxeG0DAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAXVklEQVR4nO3deZRcZZnH8e9TVb2kl3QItCShkzSRUbKwJLTAHEA8oA4ognCcUWfEnYgreAAROApuuKGjozKeiLhA1COCiAoqiqAcINBNICYTEzqhsxGSDiFLp9eqeuaP22vSIZ3UrS7eqt/nnD5U3Xvrvc976Pz6rbfee8vcHRERCVei0AWIiEhuFOQiIoFTkIuIBE5BLiISOAW5iEjgUoU46RFHHOGNjY2FOLWISLBaWlq2uXv93tsLEuSNjY00NzcX4tQiIsEys3WjbdfUiohI4BTkIiKBU5CLiAROQS4iEjgFuYhI4BTkIiKBU5CLlII//AFOOw3e/W7YubPQ1UjMCrKOXETGSV8frFoFF1wAvb3w+OOQycDixYWuTGKkEblIsfrnP2HqVDjxxCjEAdJpeOwx6O4ubG0SKwW5SDHJZuGqq2DOHLjoInjhhWgEPlxbGxx5JLS0FKREiZ+CXKSY3HYb3HwzrFwJzzwDyeTQvoHH2Szs2gVXX12YGiV2CnKRYrJ589A0SiIBs2bBlCnwoQ/BlVdCefnQsQ8+CHffvW8bN90Es2fDBz4APT3jUrbkxgrxnZ1NTU2um2aJ5MFzz8GCBbBnD9TWRtMn5eVw2GFRsH/+83D99UPHz5oFa9YMPX/wQTj33KE59KYmeOKJce2C7J+Ztbh7097bNSIXKSYdHXD77fD3v0crVF772ugDz+OOi5YdfvzjI4/fti367wMPwNlnwxe+MPKD0OZmuPba8atfDomCXKRY3H57tELlwgvhox+Nnq9bFy1BbG2Fn/wEKiqikfkAd9iyBd7ylijMH35433a1VPFlT+vIRYrFV78KXV3R45YWWLEiCnGI5s1/+lPo7ISTToInn4w+9DzmGNi0CcyGjtvbqaeOT/1yyBTkIsVi9mxYvToK456efT+obGmBp56KHptFo/FVq6IR+Zw50UqXTCb6cPTZZ6Pj3vpWuPXW8e2HHLTYgtzMkkAzsMndz4urXREZox/8AOrqYO3a6EPLbHZo30BwZzLRMsR0OtruDrt3R1Mqjz8OGzfCJZcMva63F6qrx7UbcvDinCO/DFgZY3sicjDq6qIw/+1vYcKEoe1mMG0aVFZGIT4wjQLRB5udnVBWBqefDvPmDf0BKCuL9l11FSxZMr59kYMSS5CbWQPwZuCWONoTkRysXTsyrN2jeXCAT35y5Jy3O3z4w3DFFdHzefPgy1+GhgaYOTO6nP+mm+Css6KpF3lZimtq5VvAp4Da/R1gZguBhQAzZsyI6bQiso9XvhJqaqKliMN1d8N3vhON1pPJoUv3u7vhRz+KVrxUVEQrXrZtgxtvHDomkYiWIs6ePb59kTHJOcjN7Dxgq7u3mNnr9necuy8CFkF0QVCu5xWR/Vi5MgrxRGLkPDmM/BA0kYhG5GVl0XEf+Ui0/Z574M47R96jpacnWpMuL0txTK2cBpxvZm3AL4CzzOz2GNoVkUNxww1RkO8d4nvLZqMgP+GE6APPPXuin3vuiS4iGi6RGPqAVF52cg5yd7/G3RvcvRF4B/CAu78r58pE5NBMmTLynioQjbqHz5sPt/cl+LW1UZgfdtjQNrORH6DKy4qu7BQpNl/7GrzxjXD00dF89w03RMsLL7po/68Zfs+lF1+Mli9eeWUU5pMmwbe+Fa18kZelWC8IcvcHgQfjbFNEDlJdXXTjrA0boiD+7GejEfUXvgD33jt09ef+pNNw333w0EPR8sOqKjjzzPGpXQ6JruwUKTa//CV84xvRfHdra7TSpKEBFi2K1oQ//3z0eEB1dTQiT6ejwJ88OZqa6eyM9ieT8I9/wLHHFqY/ckAKcpFis2HD0D1WenrgrruiH/coqM89N5pmuftumDgRDj88urkWREsXly2LbqB1//3R8akUnHFG4fojB6T7kYsUm02boqmVrq7oas6amqF7p0AU3jt3Dj1PpYaWGqZSQ38Eli6F5cuj29tqfvxlQfcjFykVRx0VfVnEQw9FV3nOnz9yxcoFF4w8/u1vj+bBKytH7ps/Hy6+WCEeAAW5SDGqqYmCuKYmunvhe98bfbnEl74EP/7xyGMXLoxG5GbRlMrmzYWoWHKgOXKRYldX99K3or3ttqGrPZNJ+Mtf4F26FCQkGpGLlKrbb4d3vCN6XFUV/TebhblzC1eTHBKNyEVK0e9/D5deGi1RrKqC97wHtm+P5sTnzy90dXKQFOQipeipp4YuDOrri64CvfnmwtYkh0xTKyKl6K1vje6dUlsb3br2/PMLXZHkQCNykVI0d260RvyJJ+CUU0DfERA0BblIqWpsjH4keJpaEREJnIJcRCRwCnIRkcApyEVEAqcgFxEJnIJcRCRwCnIRkcApyEVEApdzkJtZpZk9bmZPm9kKM/tcHIWJiMjYxHFlZw9wlrt3mFkZ8LCZ3efuj8XQtoiIHEDOQe7Rl3529D8t6/8Z/y8CFREpUbHMkZtZ0syeArYC97v7klGOWWhmzWbW3N7eHsdpRUSEmILc3TPufiLQAJxsZvNGOWaRuze5e1N9fX0cpxUREWJeteLuO4AHgXPibFdERPYvjlUr9WY2qf/xBOD1wD9zbVdERMYmjlUrU4GfmFmS6A/DL939dzG0KyIiYxDHqpVlgL6tVUSkQHRlp4hI4BTkIiKBU5CLiAROQS4iEjgFuYhI4BTkIiKBU5CLiAROQS4iEjgFuYhI4BTkIiKBU5CLiAROQS4iEjgFuYhI4BTkIiKBU5CLiAROQS4iEjgFuYhI4BTkIiKBU5CLiAROQS4iEricg9zMppvZX81spZmtMLPL4ihMRETGJhVDG2ngCnd/0sxqgRYzu9/d/y+GtkVE5AByHpG7+2Z3f7L/8W5gJXBUru2KiMjYxDpHbmaNwHxgySj7FppZs5k1t7e3x3laEZGSFluQm1kNcCdwubvv2nu/uy9y9yZ3b6qvr4/rtCIiJS+WIDezMqIQX+zud8XRpoiIjE0cq1YM+CGw0t2/mXtJIiJyMOIYkZ8GXAycZWZP9f+8KYZ2RURkDHJefujuDwMWQy0iInIIdGWniEjgFOQiIoFTkIuIBE5BLiISOAW5iEjgFOQiIoFTkIuIBE5BLiISOAW5iEjgFOQiIoFTkIuIBE5BLiISOAW5iEjgFOQiIoFTkIuIBE5BLiISOAW5iEjgFOQiIoFTkIuIBE5BLiISuFiC3MxuNbOtZrY8jvZERGTs4hqR/xg4J6a2RETkIMQS5O7+N2B7HG2JiMjBGbc5cjNbaGbNZtbc3t4+XqcVESl64xbk7r7I3Zvcvam+vn68TisiUvS0akVEJHAKchGRwMW1/PDnwKPAq81so5l9II52RUTkwFJxNOLu74yjHREROXiaWhERCZyCXEQkcApyEZHAKchFRAKnIBcRCZyCXEQkcApyEZHAKchFRAKnIBcRCZyCXEQkcApyEZHAKchFRAKnIBcRCZyCXEQkcApyEZHAKchFRAKnIBcRCZyCXEQkcApyEZHAKchFRAIXS5Cb2TlmtsrMWs3s03G0KSIiY5NzkJtZEvgecC4wB3inmc3JtV0RERmbOEbkJwOt7r7W3XuBXwAXxNCuiIiMQRxBfhSwYdjzjf3bRjCzhWbWbGbN7e3tMZxWREQgniC3Ubb5PhvcF7l7k7s31dfXx3BaERGBeIJ8IzB92PMG4LkY2hURkTGII8ifAP7FzI42s3LgHcA9MbQrIiJjkMq1AXdPm9nHgD8CSeBWd1+Rc2UiIjImOQc5gLvfC9wbR1siInJwdGWniEjgFOQiIoFTkIuIBE5BLiISOAW5iEjgFOQiIoFTkIuIBE5BLiISOAW5iEjgFOQiIoFTkIuIBE5BLiISOAW5iEjgFOQiIoFTkIuIBE5BLiISOAW5iEjgFOQiIoFTkIuIBE5BLiISuJyC3Mz+3cxWmFnWzJriKkpERMYu1xH5cuAi4G8x1CIiIocglcuL3X0lgJnFU42IiBw0zZGLiATugCNyM/szMGWUXde5+2/GeiIzWwgsBJgxY8aYCxQRkZd2wCB399fHcSJ3XwQsAmhqavI42hQREU2tiIgEL9flhxea2UbgX4Hfm9kf4ylLRETGKtdVK78Gfh1TLSIicgg0tSIiEjgFuYhI4BTkIiKBU5CLiAROQS4iEjgFuYhI4BTkIiKBU5DLIdm48Ts88sg0li49g56e5wtdjkhJU5DLQevsXMXatVfT27uZnTsfpbX1E4UuSaSkKcjloKXTuxn61cnQ17e9kOWIlDwFuezXzp2P0N5+N5lM14jttbULmDz53zArJ5mcyKxZXylQhSICOd5rRYrXhg3f4Nlnr8fMqKw8hqamZsySAJglmDfvTnp7t5FK1ZJIVNDZuZrt2/9Ibe1rqKs7tcDVi5QWBbmM6rnnvk82uweArq7VdHU9S1XVMSOOKS8/on//GlpaTiKbTfeH/K+ZPPmN416zSKnS1IqMqrb2NZhVAIZZORUVU/d77I4dD+HuuHeTzXaybduYvzhKRGKgEbmM6tWvvoXKykZ6ejZRX/821qy5ioqKBqZPv4JEooJ0ehcdHcuorp5Nbe3JuGcBMJvAYYedVeDqRUqLglz20dHxD5YtexPp9HYgwZYtPwXArJxt237LjBlXs3r1h8hmuwFYsOAxamvns2vXkv459cbCFS9SgjS1IvtYuvRMens3ks12ks12DG5372X37sdYseI/SKd3kMnsIpPZxbp1N7JnzzIgQzbbydq11wHQ1dVGS8spPProTLZs+VmBeiNS/DQilxH6+raTybx4oKPwYV+fnU7vxD0DgFkKszKWLn0tHR1Pk8nsBpxVqz7AYYe9gfLy+rzVLlKqNCIvQb297TQ3z+ehhypZufJ9g/PbAMlkDYnEhGFHW/968cP3296OHQ8wa9ZXqKhooK7uDPbsWcbOnQ+TyewChhI/k+nYbxsicug0Ii9BbW2fZ8+e5bin2bbtV2zf/nZ6e7eyYcPXMCtn1qyvs3nzD0mlJjJnzmLKy6exZMmxZDIvDGslAUR/AFKpOhoaPkFDQ3Sp/t/+Vs3wAIcyXvGK/9LcuUieKMhLkHvf4FRINpvhxRcfYOPGbwLRttbWT5BIVALOM898kkmTTqeycibd3auHtZLFrILKypnMnr14RPszZ36Gdes+B8C0aR9hxoxPa0pFJI9yCnIz+zrwFqAXWAO8z913xFGY5E9Z2eTBx+699PSsYyDEI1my2U4Atm27g+3bf0tZ2b5B7N5DV9dq2to+x3HH3cPmzbfQ1nY95eXTmDr1UsrKJjF9+qdIJifs81oRiU+uc+T3A/Pc/XhgNXBN7iVJvu3evZSBqY9ksprq6uMwKx92RHLE8dlsNz09W9nfr8v27b/j+edvo7X1E/T2bqajo4VNm77N+vVfZfnyC/LTCREZlFOQu/uf3D3d//QxoCH3kiTfpkx5L4lEFWZVZLPdtLVdT2VlI9OnX8OcOXewYMHj1NTMB2zwNWZJZs78DFDBaL82PT1tI44HJ5vtYseOh/LbGRGJddXK+4H79rfTzBaaWbOZNbe3t8d4WjlYRx75dk444S9MnfpuzFJAlu7uZ0kkUrziFW9j4sQFNDbeQCJRNfga917AmDixiWSyhlTqiMF9ZlWUlzdQV3c6ZmVAArNyEokJTJp05rj3T6TUHHCO3Mz+DEwZZdd17v6b/mOuA9LA4lGOA8DdFwGLAJqamnx/x8n4qKs7le7uNp5//rb+Lcbwv+uTJ7+JsrLJ9PTs6d+SZt26LxL9bx4uRSo1kTVrLsPdmTr1gxx99BfZsuVnmCWZMuV9+e+MSIk7YJC7++tfar+ZvQc4Dzjb3RXQgejre5GtW3/ev4Y8QXX18TQ0XA5AOt3BM898jJ6eTXu9au8Qj7b19W1hYM5969Zf8qpX3UxDw8fyWb6IDJPrqpVzgKuBM929M56SZDy0tl7OCy/8joG14NnsHpLJGgBWrbqE9va7Bve9tOHz6OXU1p4Uf7Ei8pJynSP/LlAL3G9mT5nZ92OoScZBT896hgd1d/ezdHW1ArBnz9NEK0oHGPsyEokJ/be6dcCorp7L3Ll3DB6RyXSyfPlFPPJIA2vWXIvesInkR66rVo5x9+nufmL/z6VxFSb51dj4eYYvMzQro6JiGgANDZ/EbAIDb9hGXrI/cHyS6uoTMRv6Faqqmk0qNXHw+fr1X+WFF+6lt3cTmzb9Dy+++Of8dEakxOnKzhI1adIZnHrqetrarieT2c3MmdcMhvC0aZeQSExg9epLyGbTZLNdRKPyoRG1e5rdux/tX/WSoqxsEo2Nnx1xjr6+F3Dv639m/bfFFZG4KchLWGXlNI499geDz7u62li9eiHp9C6mTbuUoTdsif5L9AeuAB0KdbMURx55MdOnX0FV1atHtD99+hW0t99BJrOLqqpXcfjh549Ht0RKjhVi3rKpqcmbm5vH/bzy0pqbF9DR8TSQJZmsZerUhTz33P9SWXk0xx9/H8lkLUuXnk5n5yqiFSwDc+dJkslqTjnlmX3uqZLN9tLXt43y8ikjpmFE5OCZWYu7N+29XSNyGdTTs5GhVSxdNDZezzHH3DS4v7PzGbq7n2VoGeLAICANOB0dTzN58sjVqolE+eDcu4jkh4ZIMqix8XMkEpUkElUceeR7SaVqR+wvL5/af+XmaIyamhPyX6SI7EMjchl01FEf5vDD30wm07nPfDdAKlXD/Pl/Z/36G9m69Q4G7piYSh3OSSc9rlvVihSIRuQyQmXlDKqrj8VstLXjUFNzHHPm/Jy5c++gvHwKlZXHcMIJf2LChFnjXKmIDNCIXA5Jff2F1NdfWOgyRASNyEVEgqcgFxEJnIJcRCRwCnIRkcApyEVEAqcgFxEJnIJcRCRwBblplpm1A+vy1PwRwLY8tf1yUir9hNLpq/pZfOLu60x33+cS6oIEeT6ZWfNodwcrNqXSTyidvqqfxWe8+qqpFRGRwCnIRUQCV4xBvqjQBYyTUuknlE5f1c/iMy59Lbo5chGRUlOMI3IRkZKiIBcRCVxRB7mZXWlmbmZHFLqWfDCzr5vZP81smZn92swmFbqmOJnZOWa2ysxazezTha4nH8xsupn91cxWmtkKM7us0DXlk5klzWypmf2u0LXkk5lNMrNf9f/7XGlm/5rP8xVtkJvZdOANwPpC15JH9wPz3P14YDVwTYHriY2ZJYHvAecCc4B3mtmcwlaVF2ngCnefDZwKfLRI+zngMmBloYsYB98G/uDuxwInkOc+F22QA/8NfIqhr3ovOu7+J3cf+Er7x4CGQtYTs5OBVndf6+69wC+ACwpcU+zcfbO7P9n/eDfRP/ijCltVfphZA/Bm4JZC15JPZjYReC3wQwB373X3Hfk8Z1EGuZmdD2xy96cLXcs4ej9wX6GLiNFRwIZhzzdSpAE3wMwagfnAksJWkjffIhpcZQtdSJ7NAtqBH/VPI91iZtX5PGGw39lpZn8Gpoyy6zrgWuCN41tRfrxUP939N/3HXEf0Fn3xeNaWZ6N9+3PRvrsysxrgTuByd99V6HriZmbnAVvdvcXMXlfoevIsBSwAPu7uS8zs28Cngc/k84RBcvfXj7bdzI4Djgae7v8m+AbgSTM72d2fH8cSY7G/fg4ws/cA5wFne3FdFLARmD7seQPwXIFqySszKyMK8cXufleh68mT04DzzexNQCUw0cxud/d3FbiufNgIbHT3gXdWvyIK8rwp+guCzKwNaHL3orvbmpmdA3wTONPd2wtdT5zMLEX0Ae7ZwCbgCeA/3X1FQQuLmUWjjZ8A29398kLXMx76R+RXuvt5ha4lX8zs78AH3X2Vmd0AVLv7Vfk6X7AjcgHgu0AFcH//u4/H3P3SwpYUD3dPm9nHgD8CSeDWYgvxfqcBFwP/MLOn+rdd6+73FrAmyd3HgcVmVg6sBd6Xz5MV/YhcRKTYFeWqFRGRUqIgFxEJnIJcRCRwCnIRkcApyEVEAqcgFxEJnIJcRCRw/w/JmKCG4r1+EQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Dataset\n",
    "ds = SimpleDataset(A, A, n=100, Anoise=0.2, Bnoise=0.0)\n",
    "dl = DataLoader(ds, batch_size=100, shuffle=True)\n",
    "ds.Plot(idx=(0,1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "Collapsed": "false",
    "jupyter": {
     "source_hidden": true
    }
   },
   "outputs": [],
   "source": [
    "net.ClearHistory()\n",
    "net.Probe(True)\n",
    "net.Learning(True)\n",
    "\n",
    "# Connection (v0)--(e1)\n",
    "#net.con[0].SetIdentity() # v0-e1\n",
    "net.con[0].SetRandom(random=0.1) # v0-e1\n",
    "\n",
    "# Turn off bells and whistles\n",
    "net.SetDynamicWeightDecay(0.0, 0.0)\n",
    "net.SetActivityDecay(0.0)\n",
    "net.SetBlackout(0.)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[ 0.0231, -0.1668],\n",
       "        [ 0.0608, -0.0627]])"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "net.con[0].W"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[1., 0.],\n",
       "        [0., 1.]])"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "net.con[1].M"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "Collapsed": "true"
   },
   "source": [
    "## Train it"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "Collapsed": "false"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 10/10 [00:03<00:00,  3.19it/s]\n"
     ]
    }
   ],
   "source": [
    "net.Reset()\n",
    "net.Learn(dl, T=1., epochs=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "Collapsed": "false"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tensor([[0.7338, 0.1712],\n",
      "        [0.1921, 0.8815]])\n",
      "tensor([[0.7338, 0.1921],\n",
      "        [0.1712, 0.8815]])\n"
     ]
    }
   ],
   "source": [
    "print(net.con[0].M)\n",
    "print(net.con[0].W)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "Collapsed": "false"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tensor([[1., 0.],\n",
      "        [0., 1.]])\n",
      "tensor([[1., 0.],\n",
      "        [0., 1.]])\n"
     ]
    }
   ],
   "source": [
    "print(net.con[1].M)\n",
    "print(net.con[1].W)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "Collapsed": "false"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x288 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "net.Plot();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "Collapsed": "false"
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "pc",
   "language": "python",
   "name": "pc"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}