final_decoderv2.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "b31eda61",
   "metadata": {},
   "outputs": [],
   "source": [
    "# REF: https://arxiv.org/pdf/0704.1317.pdf\n",
    "\n",
    "# import libraries\n",
    "import numpy as np\n",
    "import scipy\n",
    "from scipy import interpolate, sparse\n",
    "import utilsv2 as utils\n",
    "\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "2df19741",
   "metadata": {},
   "outputs": [],
   "source": [
    "matrix = np.loadtxt('./generator_matrix_d3.txt')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "554f8196",
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_sparse_matrix(mat,data):\n",
    "    \"\"\"\n",
    "    Creates the generator matrix\n",
    "    \"\"\"\n",
    "    rows = mat.shape[0]\n",
    "    cols = mat.shape[1]\n",
    "    row_id = mat.ravel()\n",
    "    col_id = np.tile(np.arange(cols),rows)\n",
    "    mat_data = np.repeat(data,cols)\n",
    "    mat_data = [each*np.random.choice([-1,1]) for each in mat_data]\n",
    "    \n",
    "    matrix = sparse.csr_matrix((mat_data, (row_id, col_id)), shape=(cols, cols)).toarray()\n",
    "    \n",
    "    return matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "dec8daa8",
   "metadata": {},
   "outputs": [],
   "source": [
    "vals = [1/2.31 , 1/3.17, 1/5.11 , 1/7.33 , 1/11.71]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "79aa2fd0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(3, 128)"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "matrix.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "315949e7",
   "metadata": {},
   "outputs": [],
   "source": [
    "H = create_sparse_matrix(matrix,vals[:3])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "a396ba88",
   "metadata": {},
   "outputs": [],
   "source": [
    "H = H/np.abs(np.linalg.det(H))**(1/128.) # normalize H\n",
    "G = np.linalg.inv(H)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "808c0590",
   "metadata": {},
   "outputs": [],
   "source": [
    "np.random.seed(1234)\n",
    "b = np.random.randint(10,size = (128,1))\n",
    "x = np.dot(G,b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "c90190fa",
   "metadata": {},
   "outputs": [],
   "source": [
    "mu, sigma = 0, 0.05 # mean and standard deviation\n",
    "w = np.random.normal(mu, sigma, x.shape)\n",
    "y = x + w"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "55ead470",
   "metadata": {},
   "outputs": [],
   "source": [
    "x_input = np.linspace(-40.0,40,201)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "fad22d89",
   "metadata": {},
   "outputs": [],
   "source": [
    "node = utils.CheckNode(H,x_input,x,sigma)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "56537c19",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(201,)"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_input.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "96f9414f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(128, 128, 401)"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "node.cnode_message.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "dc60b93d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 96.2 ms, sys: 58.8 ms, total: 155 ms\n",
      "Wall time: 154 ms\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "result = utils.init_message(x_input, H,y,sigma)\n",
    "convolution = node.pj(0,28,vnode_message = result,x_in = x_input)\n",
    "stretch = node.stretch_pj(0,28,vnode_message = result)\n",
    "# extension = node.periodic_extension(0,28,vnode_message = result)\n",
    "# q = node.Q(vnode_message=result)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "9b2d66ef",
   "metadata": {},
   "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": [
    "plt.plot(range(len(stretch)),stretch,'r');\n",
    "plt.plot(range(len(stretch)),convolution,'b');\n",
    "# plt.plot(range(len(stretch)),extension,'g');\n",
    "# plt.plot(range(len(stretch)),q[0,28],'c');\n",
    "# plt.plot(range(len(stretch)),q[0,33],'r');\n",
    "# plt.plot(range(len(stretch)),q[0,112],'g');"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a3c6c72a",
   "metadata": {},
   "source": [
    "## Testting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "28623933",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[[28, 33, 112],\n",
       " [6, 40, 125],\n",
       " [19, 109, 120],\n",
       " [31, 76, 121],\n",
       " [98, 109, 114],\n",
       " [7, 12, 56],\n",
       " [42, 52, 69],\n",
       " [88, 99, 104],\n",
       " [9, 62, 81],\n",
       " [25, 47, 74],\n",
       " [94, 115, 120],\n",
       " [14, 21, 27],\n",
       " [77, 117, 122],\n",
       " [4, 49, 63],\n",
       " [43, 96, 106],\n",
       " [39, 84, 113],\n",
       " [17, 36, 86],\n",
       " [15, 51, 120],\n",
       " [57, 83, 125],\n",
       " [9, 24, 84],\n",
       " [18, 82, 119],\n",
       " [10, 19, 101],\n",
       " [48, 70, 111],\n",
       " [11, 67, 115],\n",
       " [26, 37, 100],\n",
       " [45, 69, 81],\n",
       " [54, 105, 118],\n",
       " [10, 78, 94],\n",
       " [30, 46, 105],\n",
       " [0, 72, 96],\n",
       " [12, 57, 95],\n",
       " [42, 105, 124],\n",
       " [99, 108, 126],\n",
       " [5, 24, 25],\n",
       " [28, 67, 91],\n",
       " [106, 111, 118],\n",
       " [43, 71, 89],\n",
       " [15, 63, 80],\n",
       " [33, 55, 64],\n",
       " [3, 22, 45],\n",
       " [16, 94, 97],\n",
       " [65, 101, 122],\n",
       " [52, 60, 95],\n",
       " [27, 30, 47],\n",
       " [16, 39, 64],\n",
       " [55, 118, 121],\n",
       " [9, 68, 87],\n",
       " [1, 33, 78],\n",
       " [38, 66, 116],\n",
       " [11, 25, 26],\n",
       " [20, 102, 115],\n",
       " [48, 55, 77],\n",
       " [22, 44, 97],\n",
       " [40, 61, 91],\n",
       " [53, 59, 85],\n",
       " [5, 29, 79],\n",
       " [45, 68, 93],\n",
       " [53, 73, 107],\n",
       " [54, 62, 92],\n",
       " [21, 23, 49],\n",
       " [35, 50, 75],\n",
       " [44, 58, 82],\n",
       " [4, 116, 122],\n",
       " [32, 34, 70],\n",
       " [73, 85, 125],\n",
       " [7, 16, 69],\n",
       " [18, 43, 66],\n",
       " [76, 86, 110],\n",
       " [7, 112, 113],\n",
       " [72, 108, 113],\n",
       " [8, 96, 107],\n",
       " [13, 20, 112],\n",
       " [13, 73, 93],\n",
       " [3, 75, 80],\n",
       " [17, 29, 67],\n",
       " [38, 90, 123],\n",
       " [40, 72, 114],\n",
       " [60, 84, 100],\n",
       " [5, 6, 34],\n",
       " [56, 83, 119],\n",
       " [65, 110, 127],\n",
       " [44, 119, 123],\n",
       " [17, 54, 108],\n",
       " [24, 50, 65],\n",
       " [27, 83, 117],\n",
       " [37, 46, 58],\n",
       " [35, 70, 123],\n",
       " [31, 61, 71],\n",
       " [12, 74, 124],\n",
       " [35, 79, 92],\n",
       " [28, 101, 126],\n",
       " [41, 42, 53],\n",
       " [23, 58, 104],\n",
       " [19, 90, 97],\n",
       " [2, 48, 126],\n",
       " [3, 79, 88],\n",
       " [109, 117, 127],\n",
       " [18, 51, 57],\n",
       " [75, 95, 111],\n",
       " [8, 32, 62],\n",
       " [31, 50, 80],\n",
       " [36, 59, 76],\n",
       " [41, 63, 68],\n",
       " [11, 107, 110],\n",
       " [14, 37, 47],\n",
       " [29, 78, 121],\n",
       " [1, 8, 114],\n",
       " [1, 26, 59],\n",
       " [52, 89, 102],\n",
       " [13, 15, 21],\n",
       " [2, 6, 89],\n",
       " [4, 23, 98],\n",
       " [0, 41, 103],\n",
       " [2, 93, 99],\n",
       " [30, 77, 87],\n",
       " [22, 46, 116],\n",
       " [49, 74, 100],\n",
       " [86, 88, 124],\n",
       " [64, 82, 103],\n",
       " [34, 91, 127],\n",
       " [10, 20, 90],\n",
       " [0, 60, 85],\n",
       " [56, 71, 103],\n",
       " [66, 81, 106],\n",
       " [14, 38, 98],\n",
       " [36, 39, 92],\n",
       " [32, 51, 87],\n",
       " [61, 102, 104]]"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "node.vnode_id# for each chk node"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "be517bde",
   "metadata": {},
   "outputs": [],
   "source": [
    "from IPython import display\n",
    "\n",
    "def visualize(x,y,z):\n",
    "    \n",
    "    plt.plot(x,y,'g',label='before iteration')\n",
    "    plt.plot(x,z,'r',label = 'after iteration')    \n",
    "    plt.legend()\n",
    "    display.clear_output(wait=True)\n",
    "    #display.display(plt.gcf())\n",
    "    plt.show();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "921afe17",
   "metadata": {},
   "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"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1082.086540504832 384.0\n",
      "iter0: 6.21936310223195\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "cnode = utils.CheckNode(H,x_input,x,sigma)\n",
    "vnode = utils.VariableNode(H,x_input,x,sigma)\n",
    "res = utils.init_message(x_input, H,x,sigma)\n",
    "for i in range(2):\n",
    "    q = cnode.Q(res)\n",
    "    f = vnode.f(q)\n",
    "    visualize(x_input,res[0][28],f[0][28])\n",
    "    print (res.sum() , f.sum())\n",
    "    print (f'iter{i}:',res[0][28].max()-f[0][28].max())\n",
    "    res = f\n",
    "plt.clf(); "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b3ada340",
   "metadata": {},
   "outputs": [],
   "source": [
    "f_final = vnode.final(q)\n",
    "x_b = np.argmax(f_final,axis = 1)\n",
    "b_cal = np.rint(np.dot(H,x_input[x_b]))\n",
    "        \n",
    "print (b_cal.astype(int),b.ravel())\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2a2ec9ec",
   "metadata": {},
   "outputs": [],
   "source": [
    "np.array_equal(b_cal.astype(int), b.ravel()) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "01bcc577",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[3],\n",
       "       [6],\n",
       "       [5],\n",
       "       [4],\n",
       "       [8],\n",
       "       [9],\n",
       "       [1],\n",
       "       [7],\n",
       "       [9],\n",
       "       [6]])"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6142dcf3",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}