{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "2522791a",
   "metadata": {},
   "source": [
    "Non-Negative Basis Pursuit DeNoising (ADMM)\n",
    "===========================================\n",
    "\n",
    "This example demonstrates the solution of a non-negative sparse coding\n",
    "problem\n",
    "\n",
    "  $$\\mathrm{argmin}_{\\mathbf{x}} \\; (1/2) \\| \\mathbf{y} - D \\mathbf{x} \\|_2^2\n",
    "  + \\lambda \\| \\mathbf{x} \\|_1 + I(\\mathbf{x} \\geq 0) \\;,$$\n",
    "\n",
    "where $D$ the dictionary, $\\mathbf{y}$ the signal to be represented,\n",
    "$\\mathbf{x}$ is the sparse representation, and $I(\\mathbf{x} \\geq 0)$\n",
    "is the non-negative indicator."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "7b28a87a",
   "metadata": {
    "collapsed": false,
    "execution": {
     "iopub.execute_input": "2021-12-20T19:54:14.459537Z",
     "iopub.status.busy": "2021-12-20T19:54:14.458987Z",
     "iopub.status.idle": "2021-12-20T19:54:16.972003Z",
     "shell.execute_reply": "2021-12-20T19:54:16.972517Z"
    }
   },
   "outputs": [],
   "source": [
    "# This scico project Jupyter notebook has been automatically modified\n",
    "# to install the dependencies required for running it on Google Colab.\n",
    "# If you encounter any problems in running it, please open an issue at\n",
    "#   https://github.com/lanl/scico-data/issues\n",
    "\n",
    "!pip install 'scico[examples] @ git+https://github.com/lanl/scico'\n",
    "\n",
    "# This scico project Jupyter notebook has been automatically modified\n",
    "# to install the dependencies required for running it on Google Colab.\n",
    "# If you encounter any problems in running it, please open an issue at\n",
    "#   https://github.com/lanl/scico-data/issues\n",
    "\n",
    "!pip install 'scico[examples] @ git+https://github.com/lanl/scico'\n",
    "\n",
    "# This scico project Jupyter notebook has been automatically modified\n",
    "# to install the dependencies required for running it on Google Colab.\n",
    "# If you encounter any problems in running it, please open an issue at\n",
    "#   https://github.com/lanl/scico-data/issues\n",
    "\n",
    "!pip install 'scico[examples] @ git+https://github.com/lanl/scico\n'",
    "\n",
    "# This scico project Jupyter notebook has been automatically modified\n",
    "# to install the dependencies required for running it on Google Colab.\n",
    "# If you encounter any problems in running it, please open an issue at\n",
    "#   https://github.com/lanl/scico-data/issues\n",
    "\n",
    "!pip install git+https://github.com/lanl/scico\n",
    "\n",
    "# This scico project Jupyter notebook has been automatically modified\n",
    "# to install the dependencies required for running it on Google Colab.\n",
    "# If you encounter any problems in running it, please open an issue at\n",
    "#   https://github.com/lanl/scico-data/issues\n",
    "\n",
    "!pip install git+https://github.com/lanl/scico\n",
    "\n",
    "# This scico project Jupyter notebook has been automatically modified\n",
    "# to install the dependencies required for running it on Google Colab.\n",
    "# If you encounter any problems in running it, please open an issue at\n",
    "#   https://github.com/lanl/scico-data/issues\n",
    "\n",
    "!pip install git+https://github.com/lanl/scico\n",
    "\n",
    "# This scico project Jupyter notebook has been automatically modified\n",
    "# to install the dependencies required for running it on Google Colab.\n",
    "# If you encounter any problems in running it, please open an issue at\n",
    "#   https://github.com/lanl/scico-data/issues\n",
    "\n",
    "!pip install git+https://github.com/lanl/scico\n",
    "\n",
    "# This scico project Jupyter notebook has been automatically modified\n",
    "# to install the dependencies required for running it on Google Colab.\n",
    "# If you encounter any problems in running it, please open an issue at\n",
    "#   https://github.com/lanl/scico-data/issues\n",
    "\n",
    "!pip install git+https://github.com/lanl/scico\n",
    "\n",
    "import numpy as np\n",
    "\n",
    "import jax\n",
    "\n",
    "from scico import functional, linop, loss, plot\n",
    "from scico.optimize.admm import ADMM, MatrixSubproblemSolver\n",
    "from scico.util import device_info\n",
    "plot.config_notebook_plotting()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a5b824f6",
   "metadata": {},
   "source": [
    "Create random dictionary, reference random sparse representation, and\n",
    "test signal consisting of the synthesis of the reference sparse\n",
    "representation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "0b2a4327",
   "metadata": {
    "collapsed": false,
    "execution": {
     "iopub.execute_input": "2021-12-20T19:54:16.978559Z",
     "iopub.status.busy": "2021-12-20T19:54:16.978115Z",
     "iopub.status.idle": "2021-12-20T19:54:18.254884Z",
     "shell.execute_reply": "2021-12-20T19:54:18.253629Z"
    }
   },
   "outputs": [],
   "source": [
    "m = 32  # signal size\n",
    "n = 128  # dictionary size\n",
    "s = 10  # sparsity level\n",
    "\n",
    "np.random.seed(1)\n",
    "D = np.random.randn(m, n)\n",
    "D = D / np.linalg.norm(D, axis=0, keepdims=True)  # normalize dictionary\n",
    "\n",
    "xt = np.zeros(n)  # true signal\n",
    "idx = np.random.randint(low=0, high=n, size=s)  # support of xt\n",
    "xt[idx] = np.random.rand(s)\n",
    "y = D @ xt + 5e-2 * np.random.randn(m)  # synthetic signal\n",
    "\n",
    "xt = jax.device_put(xt)  # convert to jax array, push to GPU\n",
    "y = jax.device_put(y)  # convert to jax array, push to GPU"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6b93e839",
   "metadata": {},
   "source": [
    "Set up the forward operator and ADMM solver object."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "49d9f0b8",
   "metadata": {
    "collapsed": false,
    "execution": {
     "iopub.execute_input": "2021-12-20T19:54:18.258831Z",
     "iopub.status.busy": "2021-12-20T19:54:18.258087Z",
     "iopub.status.idle": "2021-12-20T19:54:18.826467Z",
     "shell.execute_reply": "2021-12-20T19:54:18.827280Z"
    }
   },
   "outputs": [],
   "source": [
    "lmbda = 1e-1\n",
    "A = linop.MatrixOperator(D)\n",
    "f = loss.SquaredL2Loss(y=y, A=A)\n",
    "g_list = [lmbda * functional.L1Norm(), functional.NonNegativeIndicator()]\n",
    "C_list = [linop.Identity((n)), linop.Identity((n))]\n",
    "rho_list = [1.0, 1.0]\n",
    "maxiter = 100  # number of ADMM iterations\n",
    "\n",
    "solver = ADMM(\n",
    "    f=f,\n",
    "    g_list=g_list,\n",
    "    C_list=C_list,\n",
    "    rho_list=rho_list,\n",
    "    x0=A.adj(y),\n",
    "    maxiter=maxiter,\n",
    "    subproblem_solver=MatrixSubproblemSolver(),\n",
    "    itstat_options={\"display\": True, \"period\": 10},\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "27d797b8",
   "metadata": {},
   "source": [
    "Run the solver."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "213cbf3f",
   "metadata": {
    "collapsed": false,
    "execution": {
     "iopub.execute_input": "2021-12-20T19:54:18.831329Z",
     "iopub.status.busy": "2021-12-20T19:54:18.829918Z",
     "iopub.status.idle": "2021-12-20T19:54:23.475714Z",
     "shell.execute_reply": "2021-12-20T19:54:23.475162Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Solving on GPU (NVIDIA GeForce RTX 2080 Ti)\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Iter  Time      Objective  Prml Rsdl  Dual Rsdl  CG It  CG Res   \n",
      "-----------------------------------------------------------------\n",
      "   0  1.47e+00  2.810e+00  1.435e+00  4.750e+00      7  5.959e-05\n",
      "   1  2.21e+00  6.810e-01  4.430e-01  8.297e-01      9  8.225e-05\r",
      "   2  2.24e+00  6.931e-01  3.791e-01  2.579e-01      7  7.293e-05\r",
      "   3  2.27e+00  6.060e-01  2.025e-01  1.891e-01      7  7.780e-05\r",
      "   4  2.30e+00  5.527e-01  1.014e-01  1.657e-01      7  2.822e-05\r",
      "   5  2.33e+00  5.219e-01  7.434e-02  1.383e-01      6  7.671e-05\r",
      "   6  2.36e+00  5.079e-01  7.709e-02  1.088e-01      6  5.791e-05\r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   7  2.39e+00  4.989e-01  5.198e-02  9.590e-02      6  5.759e-05\r",
      "   8  2.42e+00  4.928e-01  4.370e-02  8.704e-02      6  4.320e-05\r",
      "   9  2.45e+00  4.904e-01  5.082e-02  7.129e-02      6  3.517e-05\r",
      "  10  2.48e+00  4.879e-01  3.590e-02  6.160e-02      6  3.352e-05\n",
      "  11  2.51e+00  4.858e-01  3.085e-02  5.368e-02      5  9.916e-05\r",
      "  12  2.53e+00  4.847e-01  3.253e-02  4.530e-02      5  7.902e-05\r",
      "  13  2.55e+00  4.829e-01  2.272e-02  3.870e-02      6  2.912e-05\r",
      "  14  2.58e+00  4.811e-01  1.595e-02  3.524e-02      5  7.934e-05\r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  15  2.61e+00  4.797e-01  1.319e-02  3.197e-02      5  6.346e-05\r",
      "  16  2.64e+00  4.785e-01  1.168e-02  2.921e-02      5  5.498e-05\r",
      "  17  2.67e+00  4.775e-01  1.092e-02  2.672e-02      5  5.073e-05\r",
      "  18  2.69e+00  4.765e-01  1.012e-02  2.467e-02      5  4.331e-05\r",
      "  19  2.72e+00  4.757e-01  9.883e-03  2.269e-02      5  4.049e-05\r",
      "  20  2.75e+00  4.753e-01  1.017e-02  2.036e-02      5  3.832e-05\n",
      "  21  2.78e+00  4.750e-01  9.117e-03  1.848e-02      5  3.448e-05\r",
      "  22  2.80e+00  4.749e-01  8.487e-03  1.640e-02      5  3.111e-05\r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  23  2.82e+00  4.750e-01  8.096e-03  1.461e-02      4  9.554e-05\r",
      "  24  2.85e+00  4.748e-01  6.643e-03  1.307e-02      4  9.588e-05\r",
      "  25  2.88e+00  4.746e-01  5.796e-03  1.176e-02      4  8.330e-05\r",
      "  26  2.90e+00  4.744e-01  5.057e-03  1.065e-02      4  7.405e-05\r",
      "  27  2.93e+00  4.742e-01  4.410e-03  9.722e-03      4  6.696e-05\r",
      "  28  2.96e+00  4.740e-01  3.857e-03  8.936e-03      4  5.982e-05\r",
      "  29  2.99e+00  4.738e-01  3.391e-03  8.266e-03      4  5.352e-05\r",
      "  30  3.02e+00  4.737e-01  3.005e-03  7.686e-03      4  4.805e-05\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  31  3.04e+00  4.736e-01  2.688e-03  7.175e-03      4  4.409e-05\r",
      "  32  3.07e+00  4.735e-01  2.428e-03  6.718e-03      4  4.162e-05\r",
      "  33  3.10e+00  4.734e-01  2.215e-03  6.303e-03      4  3.903e-05\r",
      "  34  3.13e+00  4.733e-01  2.038e-03  5.922e-03      4  3.636e-05\r",
      "  35  3.15e+00  4.733e-01  1.865e-03  5.554e-03      3  9.948e-05\r",
      "  36  3.18e+00  4.732e-01  1.754e-03  5.243e-03      3  6.122e-05\r",
      "  37  3.20e+00  4.732e-01  1.630e-03  4.929e-03      3  8.706e-05\r",
      "  38  3.23e+00  4.731e-01  1.540e-03  4.651e-03      3  5.910e-05\r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  39  3.26e+00  4.731e-01  1.435e-03  4.374e-03      3  7.613e-05\r",
      "  40  3.28e+00  4.732e-01  1.731e-03  3.707e-03      3  5.594e-05\n",
      "  41  3.31e+00  4.732e-01  1.758e-03  3.223e-03      3  6.391e-05\r",
      "  42  3.33e+00  4.732e-01  1.495e-03  2.781e-03      3  7.975e-05\r",
      "  43  3.36e+00  4.732e-01  1.411e-03  2.401e-03      3  6.982e-05\r",
      "  44  3.38e+00  4.732e-01  1.296e-03  2.075e-03      3  6.177e-05\r",
      "  45  3.41e+00  4.732e-01  1.166e-03  1.800e-03      3  5.345e-05\r",
      "  46  3.43e+00  4.731e-01  1.036e-03  1.566e-03      3  4.550e-05\r",
      "  47  3.45e+00  4.731e-01  9.215e-04  1.343e-03      3  3.840e-05\r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  48  3.48e+00  4.731e-01  7.847e-04  1.084e-03      2  9.732e-05\r",
      "  49  3.50e+00  4.731e-01  6.643e-04  9.493e-04      2  9.293e-05\r",
      "  50  3.53e+00  4.731e-01  5.812e-04  8.044e-04      2  7.877e-05\n",
      "  51  3.55e+00  4.731e-01  5.317e-04  7.038e-04      2  6.443e-05\r",
      "  52  3.57e+00  4.730e-01  4.626e-04  5.887e-04      2  5.985e-05\r",
      "  53  3.60e+00  4.730e-01  4.065e-04  5.070e-04      2  5.060e-05\r",
      "  54  3.62e+00  4.730e-01  3.520e-04  4.284e-04      2  4.389e-05\r",
      "  55  3.64e+00  4.730e-01  2.558e-04  1.825e-04      1  8.865e-05\r",
      "  56  3.66e+00  4.730e-01  2.483e-04  1.764e-04      1  8.417e-05\r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  57  3.68e+00  4.730e-01  1.851e-04  1.027e-04      1  6.973e-05\r",
      "  58  3.70e+00  4.730e-01  1.851e-04  0.000e+00      0  9.308e-05\r",
      "  59  3.73e+00  4.730e-01  9.734e-05  1.078e-04      1  5.527e-05\r",
      "  60  3.75e+00  4.730e-01  9.734e-05  0.000e+00      0  7.698e-05\n",
      "  61  3.77e+00  4.730e-01  9.734e-05  0.000e+00      0  8.566e-05\r",
      "  62  3.79e+00  4.730e-01  9.734e-05  0.000e+00      0  9.952e-05\r",
      "  63  3.82e+00  4.730e-01  3.825e-05  6.954e-05      1  4.892e-05\r",
      "  64  3.84e+00  4.730e-01  3.825e-05  0.000e+00      0  6.494e-05\r",
      "  65  3.87e+00  4.730e-01  3.825e-05  0.000e+00      0  6.899e-05\r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  66  3.89e+00  4.730e-01  3.825e-05  0.000e+00      0  7.402e-05\r",
      "  67  3.92e+00  4.730e-01  3.825e-05  0.000e+00      0  7.983e-05\r",
      "  68  3.95e+00  4.730e-01  3.825e-05  0.000e+00      0  8.627e-05\r",
      "  69  3.97e+00  4.730e-01  3.825e-05  0.000e+00      0  9.321e-05\r",
      "  70  4.00e+00  4.730e-01  5.895e-05  6.213e-05      1  4.818e-05\n",
      "  71  4.02e+00  4.730e-01  5.895e-05  0.000e+00      0  6.711e-05\r",
      "  72  4.04e+00  4.730e-01  5.895e-05  0.000e+00      0  7.636e-05\r",
      "  73  4.06e+00  4.730e-01  5.895e-05  0.000e+00      0  8.711e-05\r",
      "  74  4.08e+00  4.730e-01  5.895e-05  0.000e+00      0  9.886e-05\r",
      "  75  4.10e+00  4.730e-01  3.793e-05  5.370e-05      1  4.677e-05\r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  76  4.12e+00  4.730e-01  3.793e-05  0.000e+00      0  5.983e-05\r",
      "  77  4.14e+00  4.730e-01  3.793e-05  0.000e+00      0  6.447e-05\r",
      "  78  4.15e+00  4.730e-01  3.793e-05  0.000e+00      0  7.008e-05\r",
      "  79  4.17e+00  4.730e-01  3.793e-05  0.000e+00      0  7.644e-05\r",
      "  80  4.19e+00  4.730e-01  3.793e-05  0.000e+00      0  8.337e-05\n",
      "  81  4.21e+00  4.730e-01  3.793e-05  0.000e+00      0  9.076e-05\r",
      "  82  4.24e+00  4.730e-01  3.793e-05  0.000e+00      0  9.849e-05\r",
      "  83  4.27e+00  4.730e-01  5.727e-05  5.079e-05      1  4.646e-05\r",
      "  84  4.29e+00  4.730e-01  5.727e-05  0.000e+00      0  6.218e-05\r",
      "  85  4.31e+00  4.730e-01  5.727e-05  0.000e+00      0  7.058e-05\r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  86  4.33e+00  4.730e-01  5.727e-05  0.000e+00      0  8.064e-05\r",
      "  87  4.35e+00  4.730e-01  5.727e-05  0.000e+00      0  9.179e-05\r",
      "  88  4.38e+00  4.730e-01  2.943e-05  5.005e-05      1  4.083e-05\r",
      "  89  4.40e+00  4.730e-01  2.943e-05  0.000e+00      0  5.311e-05\r",
      "  90  4.42e+00  4.730e-01  2.943e-05  0.000e+00      0  5.663e-05\n",
      "  91  4.44e+00  4.730e-01  2.943e-05  0.000e+00      0  6.083e-05\r",
      "  92  4.47e+00  4.730e-01  2.943e-05  0.000e+00      0  6.558e-05\r",
      "  93  4.49e+00  4.730e-01  2.943e-05  0.000e+00      0  7.077e-05\r",
      "  94  4.51e+00  4.730e-01  2.943e-05  0.000e+00      0  7.630e-05\r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  95  4.53e+00  4.730e-01  2.943e-05  0.000e+00      0  8.211e-05\r",
      "  96  4.56e+00  4.730e-01  2.943e-05  0.000e+00      0  8.815e-05\r",
      "  97  4.58e+00  4.730e-01  2.928e-05  1.322e-06      0  9.436e-05\r",
      "  98  4.60e+00  4.730e-01  6.007e-05  4.695e-05      1  4.417e-05\r",
      "  99  4.62e+00  4.730e-01  6.007e-05  0.000e+00      0  5.957e-05\r\n"
     ]
    }
   ],
   "source": [
    "print(f\"Solving on {device_info()}\\n\")\n",
    "x = solver.solve()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e3f03fd6",
   "metadata": {},
   "source": [
    "Plot the recovered coefficients and signal."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "c2e8b2c4",
   "metadata": {
    "collapsed": false,
    "execution": {
     "iopub.execute_input": "2021-12-20T19:54:23.492471Z",
     "iopub.status.busy": "2021-12-20T19:54:23.482292Z",
     "iopub.status.idle": "2021-12-20T19:54:23.952319Z",
     "shell.execute_reply": "2021-12-20T19:54:23.952764Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plot.subplots(nrows=1, ncols=2, figsize=(12, 5))\n",
    "plot.plot(\n",
    "    np.vstack((xt, solver.x)).T,\n",
    "    title=\"Coefficients\",\n",
    "    lgnd=(\"Ground Truth\", \"Recovered\"),\n",
    "    fig=fig,\n",
    "    ax=ax[0],\n",
    ")\n",
    "plot.plot(\n",
    "    np.vstack((D @ xt, y, D @ solver.x)).T,\n",
    "    title=\"Signal\",\n",
    "    lgnd=(\"Ground Truth\", \"Noisy\", \"Recovered\"),\n",
    "    fig=fig,\n",
    "    ax=ax[1],\n",
    ")\n",
    "fig.show()"
   ]
  }
 ],
 "metadata": {
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}