Fall2022/seminars/seminar5.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import cvxpy as cp\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['CVXOPT',\n",
       " 'ECOS',\n",
       " 'ECOS_BB',\n",
       " 'GLPK',\n",
       " 'GLPK_MI',\n",
       " 'GUROBI',\n",
       " 'MOSEK',\n",
       " 'OSQP',\n",
       " 'SCIPY',\n",
       " 'SCS']"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cp.installed_solvers()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Sparse index fund"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "# number of stocks\n",
    "n = 500\n",
    "# number of factors (not described or needed in problem statement)\n",
    "k = 20\n",
    "\n",
    "# generate rbar, Sigma\n",
    "# factor matrix, entries uniform on [-0.5,1]\n",
    "F = 1.5 * np.random.rand(n, k) - 0.5\n",
    "Sigma = 0.5 * F @ np.diag(np.random.rand(k)) @ F.T + np.diag(0.1 * np.random.rand(n))\n",
    "# risk-free return (weekly, 5% annual)\n",
    "mu_rf = 0.2\n",
    "# Sharpe ratio\n",
    "SR = 0.4\n",
    "# expected return\n",
    "rbar = mu_rf + SR * np.sqrt(np.diag(Sigma))\n",
    "# market capitalization (index weights)\n",
    "c = 5 + np.exp(2 * np.random.randn(n))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(10000, 500)\n"
     ]
    }
   ],
   "source": [
    "num_samples = 10000\n",
    "r = np.random.multivariate_normal(rbar, Sigma, num_samples)\n",
    "print(r.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "rhs = r @ c\n",
    "threshold = 1e-1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "===============================================================================\n",
      "                                     CVXPY                                     \n",
      "                                     v1.2.2                                    \n",
      "===============================================================================\n",
      "(CVXPY) Nov 22 07:32:13 PM: Your problem has 500 variables, 1 constraints, and 0 parameters.\n",
      "(CVXPY) Nov 22 07:32:13 PM: It is compliant with the following grammars: DCP, DQCP\n",
      "(CVXPY) Nov 22 07:32:13 PM: (If you need to solve this problem multiple times, but with different data, consider using parameters.)\n",
      "(CVXPY) Nov 22 07:32:13 PM: CVXPY will first compile your problem; then, it will invoke a numerical solver to obtain a solution.\n",
      "-------------------------------------------------------------------------------\n",
      "                                  Compilation                                  \n",
      "-------------------------------------------------------------------------------\n",
      "(CVXPY) Nov 22 07:32:13 PM: Compiling problem (target solver=MOSEK).\n",
      "(CVXPY) Nov 22 07:32:13 PM: Reduction chain: Dcp2Cone -> CvxAttr2Constr -> ConeMatrixStuffing -> MOSEK\n",
      "(CVXPY) Nov 22 07:32:13 PM: Applying reduction Dcp2Cone\n",
      "(CVXPY) Nov 22 07:32:13 PM: Applying reduction CvxAttr2Constr\n",
      "(CVXPY) Nov 22 07:32:13 PM: Applying reduction ConeMatrixStuffing\n",
      "(CVXPY) Nov 22 07:32:13 PM: Applying reduction MOSEK\n",
      "(CVXPY) Nov 22 07:32:14 PM: Finished problem compilation (took 1.224e+00 seconds).\n",
      "-------------------------------------------------------------------------------\n",
      "                                Numerical solver                               \n",
      "-------------------------------------------------------------------------------\n",
      "(CVXPY) Nov 22 07:32:14 PM: Invoking solver MOSEK  to obtain a solution.\n",
      "\n",
      "\n",
      "(CVXPY) Nov 22 07:32:15 PM: Problem\n",
      "(CVXPY) Nov 22 07:32:15 PM:   Name                   :                 \n",
      "(CVXPY) Nov 22 07:32:15 PM:   Objective sense        : maximize        \n",
      "(CVXPY) Nov 22 07:32:15 PM:   Type                   : CONIC (conic optimization problem)\n",
      "(CVXPY) Nov 22 07:32:15 PM:   Constraints            : 1001            \n",
      "(CVXPY) Nov 22 07:32:15 PM:   Affine conic cons.     : 0               \n",
      "(CVXPY) Nov 22 07:32:15 PM:   Disjunctive cons.      : 0               \n",
      "(CVXPY) Nov 22 07:32:15 PM:   Cones                  : 1               \n",
      "(CVXPY) Nov 22 07:32:15 PM:   Scalar variables       : 11002           \n",
      "(CVXPY) Nov 22 07:32:15 PM:   Matrix variables       : 0               \n",
      "(CVXPY) Nov 22 07:32:15 PM:   Integer variables      : 0               \n",
      "(CVXPY) Nov 22 07:32:15 PM: \n",
      "(CVXPY) Nov 22 07:32:15 PM: Optimizer started.\n",
      "(CVXPY) Nov 22 07:32:15 PM: Presolve started.\n",
      "(CVXPY) Nov 22 07:32:15 PM: Linear dependency checker started.\n",
      "(CVXPY) Nov 22 07:32:15 PM: Linear dependency checker terminated.\n",
      "(CVXPY) Nov 22 07:32:15 PM: Eliminator started.\n",
      "(CVXPY) Nov 22 07:32:15 PM: Freed constraints in eliminator : 501\n",
      "(CVXPY) Nov 22 07:32:15 PM: Eliminator terminated.\n",
      "(CVXPY) Nov 22 07:32:15 PM: Eliminator started.\n",
      "(CVXPY) Nov 22 07:32:16 PM: Freed constraints in eliminator : 0\n",
      "(CVXPY) Nov 22 07:32:16 PM: Eliminator terminated.\n",
      "(CVXPY) Nov 22 07:32:16 PM: Eliminator - tries                  : 2                 time                   : 0.00            \n",
      "(CVXPY) Nov 22 07:32:16 PM: Lin. dep.  - tries                  : 1                 time                   : 0.02            \n",
      "(CVXPY) Nov 22 07:32:16 PM: Lin. dep.  - number                 : 0               \n",
      "(CVXPY) Nov 22 07:32:16 PM: Presolve terminated. Time: 0.44    \n",
      "(CVXPY) Nov 22 07:32:16 PM: GP based matrix reordering started.\n",
      "(CVXPY) Nov 22 07:32:16 PM: GP based matrix reordering terminated.\n",
      "(CVXPY) Nov 22 07:32:16 PM: Problem\n",
      "(CVXPY) Nov 22 07:32:16 PM:   Name                   :                 \n",
      "(CVXPY) Nov 22 07:32:16 PM:   Objective sense        : maximize        \n",
      "(CVXPY) Nov 22 07:32:16 PM:   Type                   : CONIC (conic optimization problem)\n",
      "(CVXPY) Nov 22 07:32:16 PM:   Constraints            : 1001            \n",
      "(CVXPY) Nov 22 07:32:16 PM:   Affine conic cons.     : 0               \n",
      "(CVXPY) Nov 22 07:32:16 PM:   Disjunctive cons.      : 0               \n",
      "(CVXPY) Nov 22 07:32:16 PM:   Cones                  : 1               \n",
      "(CVXPY) Nov 22 07:32:16 PM:   Scalar variables       : 11002           \n",
      "(CVXPY) Nov 22 07:32:16 PM:   Matrix variables       : 0               \n",
      "(CVXPY) Nov 22 07:32:16 PM:   Integer variables      : 0               \n",
      "(CVXPY) Nov 22 07:32:16 PM: \n",
      "(CVXPY) Nov 22 07:32:16 PM: Optimizer  - threads                : 8               \n",
      "(CVXPY) Nov 22 07:32:16 PM: Optimizer  - solved problem         : the primal      \n",
      "(CVXPY) Nov 22 07:32:16 PM: Optimizer  - Constraints            : 500\n",
      "(CVXPY) Nov 22 07:32:16 PM: Optimizer  - Cones                  : 1\n",
      "(CVXPY) Nov 22 07:32:16 PM: Optimizer  - Scalar variables       : 10501             conic                  : 10001           \n",
      "(CVXPY) Nov 22 07:32:16 PM: Optimizer  - Semi-definite variables: 0                 scalarized             : 0               \n",
      "(CVXPY) Nov 22 07:32:16 PM: Factor     - setup time             : 0.11              dense det. time        : 0.00            \n",
      "(CVXPY) Nov 22 07:32:16 PM: Factor     - ML order time          : 0.00              GP order time          : 0.00            \n",
      "(CVXPY) Nov 22 07:32:16 PM: Factor     - nonzeros before factor : 1.25e+05          after factor           : 1.25e+05        \n",
      "(CVXPY) Nov 22 07:32:16 PM: Factor     - dense dim.             : 0                 flops                  : 2.55e+09        \n",
      "(CVXPY) Nov 22 07:32:16 PM: Factor     - GP saved nzs           : 0                 GP saved flops         : 2.50e+09        \n",
      "(CVXPY) Nov 22 07:32:16 PM: ITE PFEAS    DFEAS    GFEAS    PRSTATUS   POBJ              DOBJ              MU       TIME  \n",
      "(CVXPY) Nov 22 07:32:16 PM: 0   8.8e-01  5.3e+04  5.3e+04  0.00e+00   -5.260004533e+04  0.000000000e+00   1.0e+00  0.76  \n",
      "(CVXPY) Nov 22 07:32:16 PM: 1   4.4e-03  2.7e+02  3.7e+03  -1.00e+00  -2.642702952e+04  2.577989540e+04   5.0e-03  0.83  \n",
      "(CVXPY) Nov 22 07:32:16 PM: 2   2.3e-03  1.4e+02  2.7e+03  -9.85e-01  -3.267399224e+03  4.802856838e+04   2.7e-03  0.89  \n",
      "(CVXPY) Nov 22 07:32:16 PM: 3   2.3e-03  1.4e+02  2.3e+03  -4.70e-01  -2.621454877e+03  4.282958336e+04   2.6e-03  0.94  \n",
      "(CVXPY) Nov 22 07:32:16 PM: 4   2.1e-03  1.2e+02  3.4e+02  1.58e+00   -2.902997663e+03  1.882559088e+04   2.4e-03  0.99  \n",
      "(CVXPY) Nov 22 07:32:16 PM: 5   7.7e-04  4.6e+01  9.4e+01  1.65e+00   2.292366679e+03   7.058082953e+03   8.8e-04  1.05  \n",
      "(CVXPY) Nov 22 07:32:16 PM: 6   1.4e-04  8.4e+00  5.6e+00  1.16e+00   4.249664477e+03   5.019145762e+03   1.6e-04  1.12  \n",
      "(CVXPY) Nov 22 07:32:16 PM: 7   2.2e-05  1.3e+00  4.2e-01  1.04e+00   4.422906705e+03   4.537811840e+03   2.5e-05  1.19  \n",
      "(CVXPY) Nov 22 07:32:16 PM: 8   3.9e-06  2.3e-01  4.1e-02  1.02e+00   4.393791471e+03   4.414324875e+03   4.5e-06  1.27  \n",
      "(CVXPY) Nov 22 07:32:16 PM: 9   5.8e-07  3.5e-02  2.5e-03  1.00e+00   4.387287223e+03   4.390329394e+03   6.6e-07  1.32  \n",
      "(CVXPY) Nov 22 07:32:16 PM: 10  1.4e-07  8.7e-03  3.1e-04  1.00e+00   4.386200353e+03   4.386961847e+03   1.7e-07  1.39  \n",
      "(CVXPY) Nov 22 07:32:16 PM: 11  1.5e-08  8.9e-04  1.0e-05  1.00e+00   4.385864355e+03   4.385942432e+03   1.7e-08  1.44  \n",
      "(CVXPY) Nov 22 07:32:16 PM: 12  1.3e-11  7.6e-07  2.5e-10  1.00e+00   4.385823977e+03   4.385824043e+03   1.4e-11  1.51  \n",
      "(CVXPY) Nov 22 07:32:17 PM: 13  1.7e-11  2.9e-06  2.0e-13  1.00e+00   4.385823972e+03   4.385823943e+03   4.9e-17  1.58  \n",
      "(CVXPY) Nov 22 07:32:17 PM: Optimizer terminated. Time: 1.60    \n",
      "(CVXPY) Nov 22 07:32:17 PM: \n",
      "(CVXPY) Nov 22 07:32:17 PM: \n",
      "(CVXPY) Nov 22 07:32:17 PM: Interior-point solution summary\n",
      "(CVXPY) Nov 22 07:32:17 PM:   Problem status  : PRIMAL_AND_DUAL_FEASIBLE\n",
      "(CVXPY) Nov 22 07:32:17 PM:   Solution status : OPTIMAL\n",
      "(CVXPY) Nov 22 07:32:17 PM:   Primal.  obj: 4.3858239723e+03    nrm: 1e+00    Viol.  con: 7e-09    var: 1e-14    cones: 1e-17  \n",
      "(CVXPY) Nov 22 07:32:17 PM:   Dual.    obj: 4.3858239433e+03    nrm: 5e+04    Viol.  con: 0e+00    var: 3e-04    cones: 1e-10  \n",
      "-------------------------------------------------------------------------------\n",
      "                                    Summary                                    \n",
      "-------------------------------------------------------------------------------\n",
      "(CVXPY) Nov 22 07:32:17 PM: Problem status: optimal\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(CVXPY) Nov 22 07:32:17 PM: Optimal value: 4.386e+03\n",
      "(CVXPY) Nov 22 07:32:17 PM: Compilation took 1.224e+00 seconds\n",
      "(CVXPY) Nov 22 07:32:17 PM: Solver (including time spent in interface) took 2.735e+00 seconds\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "4385.823943324029"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "w = cp.Variable(n)\n",
    "\n",
    "obj = cp.Minimize(cp.norm1(w))\n",
    "constr = [cp.norm2(r @ w - rhs) <= threshold * np.linalg.norm(rhs)]\n",
    "\n",
    "problem = cp.Problem(obj, constr)\n",
    "problem.solve(solver=cp.MOSEK, verbose=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.10000000000000178"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.linalg.norm(r @ w.value - rhs) / np.linalg.norm(rhs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(39, 500)"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.sum(np.abs(w.value) > 0.05), np.sum(np.abs(c) > 0.05)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x14f45a490>"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x700 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,7))\n",
    "plt.plot(w.value, label=\"Sparse\")\n",
    "plt.plot(c, label=\"Initial\")\n",
    "plt.yscale(\"log\")\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Sparse covariance estimation for Gaussian variables\n",
    "\n",
    "## Introduction\n",
    "\n",
    "Suppose $y \\in \\mathbf{\\mbox{R}}^n$ is a Gaussian random variable with zero mean and\n",
    "covariance matrix $R = \\mathbf{\\mbox{E}}[yy^T]$, with sparse inverse $S = R^{-1}$\n",
    "($S_{ij} = 0$ means that $y_i$ and $y_j$ are conditionally independent).\n",
    "We want to estimate the covariance matrix $R$ based on $N$ independent\n",
    "samples $y_1,\\dots,y_N$ drawn from the distribution, and using prior knowledge\n",
    "that $S$ is sparse\n",
    "\n",
    "A good heuristic for estimating $R$ is to solve the problem\n",
    "  $$\\begin{array}{ll}\n",
    "    \\mbox{maximize}   & \\log \\det(S) - \\mbox{tr}(SY) \\\\\n",
    "    \\mbox{subject to} & \\sum_{i=1}^n \\sum_{j=1}^n |S_{ij}| \\le \\alpha \\\\\n",
    "                      & S \\succeq 0,\n",
    "    \\end{array}$$\n",
    "where $Y$ is the sample covariance of $y_1,\\dots,y_N$, and $\\alpha$ is a sparsity\n",
    "parameter to be chosen or tuned.\n",
    "\n",
    "**Remark**: more detailed description of the objective function is given [here](http://people.eecs.berkeley.edu/~jordan/courses/260-spring10/other-readings/chapter13.pdf), in particular eq. (13.50)\n",
    "\n",
    "## Generate problem data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "import cvxpy as cp\n",
    "import numpy as np\n",
    "import scipy as scipy\n",
    "\n",
    "# Dimension of matrix.\n",
    "n = 10\n",
    "\n",
    "# Number of samples, y_i\n",
    "N = 1000\n",
    "\n",
    "# Create sparse, symmetric PSD matrix S\n",
    "A = np.random.randn(n, n)  # Unit normal gaussian distribution.\n",
    "A[scipy.sparse.rand(n, n, 0.85).todense().nonzero()] = 0  # Sparsen the matrix.\n",
    "Strue = A.dot(A.T) + 0.05 * np.eye(n)  # Force strict pos. def.\n",
    "\n",
    "# Create the covariance matrix associated with S.\n",
    "R = np.linalg.inv(Strue)\n",
    "\n",
    "# Create samples y_i from the distribution with covariance R. \n",
    "y_sample = scipy.linalg.sqrtm(R).dot(np.random.randn(n, N))\n",
    "\n",
    "# Calculate the sample covariance matrix.\n",
    "Y = np.cov(y_sample)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "===============================================================================\n",
      "                                     CVXPY                                     \n",
      "                                     v1.2.2                                    \n",
      "===============================================================================\n",
      "(CVXPY) Nov 22 08:00:44 PM: Your problem has 100 variables, 1 constraints, and 0 parameters.\n",
      "(CVXPY) Nov 22 08:00:44 PM: It is compliant with the following grammars: DCP, DQCP\n",
      "(CVXPY) Nov 22 08:00:44 PM: (If you need to solve this problem multiple times, but with different data, consider using parameters.)\n",
      "(CVXPY) Nov 22 08:00:44 PM: CVXPY will first compile your problem; then, it will invoke a numerical solver to obtain a solution.\n",
      "-------------------------------------------------------------------------------\n",
      "                                  Compilation                                  \n",
      "-------------------------------------------------------------------------------\n",
      "(CVXPY) Nov 22 08:00:44 PM: Compiling problem (target solver=SCS).\n",
      "(CVXPY) Nov 22 08:00:44 PM: Reduction chain: FlipObjective -> Dcp2Cone -> CvxAttr2Constr -> ConeMatrixStuffing -> SCS\n",
      "(CVXPY) Nov 22 08:00:44 PM: Applying reduction FlipObjective\n",
      "(CVXPY) Nov 22 08:00:44 PM: Applying reduction Dcp2Cone\n",
      "(CVXPY) Nov 22 08:00:44 PM: Applying reduction CvxAttr2Constr\n",
      "(CVXPY) Nov 22 08:00:44 PM: Applying reduction ConeMatrixStuffing\n",
      "(CVXPY) Nov 22 08:00:44 PM: Applying reduction SCS\n",
      "(CVXPY) Nov 22 08:00:44 PM: Finished problem compilation (took 1.866e-02 seconds).\n",
      "-------------------------------------------------------------------------------\n",
      "                                Numerical solver                               \n",
      "-------------------------------------------------------------------------------\n",
      "(CVXPY) Nov 22 08:00:44 PM: Invoking solver SCS  to obtain a solution.\n",
      "------------------------------------------------------------------\n",
      "\t       SCS v3.2.2 - Splitting Conic Solver\n",
      "\t(c) Brendan O'Donoghue, Stanford University, 2012\n",
      "------------------------------------------------------------------\n",
      "problem:  variables n: 220, constraints m: 496\n",
      "cones: \t  l: linear vars: 201\n",
      "\t  s: psd vars: 265, ssize: 2\n",
      "\t  e: exp vars: 30, dual exp vars: 0\n",
      "settings: eps_abs: 1.0e-05, eps_rel: 1.0e-05, eps_infeas: 1.0e-07\n",
      "\t  alpha: 1.50, scale: 1.00e-01, adaptive_scale: 1\n",
      "\t  max_iters: 100000, normalize: 1, rho_x: 1.00e-06\n",
      "\t  acceleration_lookback: 10, acceleration_interval: 10\n",
      "lin-sys:  sparse-direct-amd-qdldl\n",
      "\t  nnz(A): 695, nnz(P): 0\n",
      "------------------------------------------------------------------\n",
      " iter | pri res | dua res |   gap   |   obj   |  scale  | time (s)\n",
      "------------------------------------------------------------------\n",
      "     0| 1.18e+02  1.19e+01  1.05e+04 -5.25e+03  1.00e-01  1.41e-03 \n",
      "   250| 3.28e-02  6.67e-03  1.75e+00  2.21e+01  8.87e-01  4.17e-02 \n",
      "   500| 2.88e-03  3.32e-03  1.92e-01  2.43e+01  2.82e+00  1.18e-01 \n",
      "   750| 1.52e-05  6.09e-06  3.77e-04  2.44e+01  2.82e+00  1.81e-01 \n",
      "   825| 6.06e-07  8.71e-06  7.01e-05  2.44e+01  2.82e+00  2.35e-01 \n",
      "------------------------------------------------------------------\n",
      "status:  solved\n",
      "timings: total: 2.35e-01s = setup: 6.91e-04s + solve: 2.34e-01s\n",
      "\t lin-sys: 9.63e-03s, cones: 2.03e-01s, accel: 1.57e-02s\n",
      "------------------------------------------------------------------\n",
      "objective = 24.403352\n",
      "------------------------------------------------------------------\n",
      "-------------------------------------------------------------------------------\n",
      "                                    Summary                                    \n",
      "-------------------------------------------------------------------------------\n",
      "(CVXPY) Nov 22 08:00:45 PM: Problem status: optimal\n",
      "(CVXPY) Nov 22 08:00:45 PM: Optimal value: -2.440e+01\n",
      "(CVXPY) Nov 22 08:00:45 PM: Compilation took 1.866e-02 seconds\n",
      "(CVXPY) Nov 22 08:00:45 PM: Solver (including time spent in interface) took 2.356e-01 seconds\n"
     ]
    }
   ],
   "source": [
    "alpha = 7\n",
    "S = cp.Variable(shape=(n,n), PSD=True)\n",
    "obj = cp.Maximize(cp.log_det(S) - cp.trace(S @ Y))\n",
    "\n",
    "# Set constraint.\n",
    "constraints = [cp.sum(cp.abs(S)) <= alpha]\n",
    "\n",
    "# Form and solve optimization problem\n",
    "prob = cp.Problem(obj, constraints)\n",
    "prob.solve(solver=cp.SCS, verbose=True)\n",
    "\n",
    "# If the covariance matrix R is desired, here is how it to create it.\n",
    "R_hat = np.linalg.inv(S.value)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Threshold S element values to enforce exact zeros:\n",
    "S = S.value\n",
    "S[abs(S) <= 1e-4] = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1200x1200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Estimated inv. cov matrix, $\\\\alpha$=7')"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZYAAAG4CAYAAABmXsHQAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAaHklEQVR4nO3dT3Kq6BrH8cdTXeXoRJKedXUGF3dAclZwyQ7IyQqO7kDL0b09ssgONCtIdAdyV5AjO5BJbvUsCXpGTpo7yIX2DybgeVRIvp8qq4+G1scX5McLL1CJoigSAACUfDp0AQCA94VgAQCoIlgAAKoIFgCAKoIFAKCKYAEAqCJYAACqCBYAgCqCBQCg6kMFSxiGcnl5KWdnZ1KpVKRSqcjl5eXS4+LiQur1ulQqFTk+Pt5JDfV6Xa6vr9Xfu0g8z5OzszM5Pj6Wy8vLzP/fR2kfpPuo89/3fbm8vJQgCA5dio7oA3p+fo5EJDJNc+M0vV4vEpHo+fl5q89otVqpr08mk0hEIsdxtnrfXdhUqwbLsnJ91yK2D/Lbdpn6qPN/MBhEIvLqo9frHbrMzH45SJodmGEYS/9N02g0ZDweSxAEYllW7s/YtOVhmqZEBbs82y63kk5OTnJNX8T2QX7bLlMfdf4HQSC2bYtpmmt/e3p6kiAIpNFoHKCy7XzIYMnq8vJSnp6ecv9/w+FQwjDUL2gHylQryoFlKr/Hx0cZjUapf2s2m3Jzc7Pnin7OhzrGkoXnecm/z8/Pc295BUEg37590y5rJ8pUK8qBZWo7FxcXqa8Ph0O5uLh4de9KEdFjWTEYDMS2bRF52VW22P0MgkB6vZ7U6/Vki2wymUiz2RTLsmQ4HMrt7a2IiHz//j05aG2apriuK0EQSLPZlO/fv4tpmjIej0Xk5cDdt2/fku7wzc2N9Pt9MQxDRqORnJycSK/XkzAMpd/vi4jI/f198r6r4jp//fVXeXx8FBFZm+6tWmNhGEq73ZZ6vS6Pj48SBIF0Op3U3YPX19cymUykXq+LYRip3frXbNs+izX3+33p9Xri+34y/+K/eZ4nl5eXEoah2La9cQtxG/1+X8bjcbICuLi4SJajRYsHpR8fH6VeryfLmGbtRVymPM+Tdrud7Na5uroSz/NkNBolv6G0+X99fS23t7fi+76IiLRareSz6/W6BEEgpmmKbdvS6/VebZes4vlZr9dlMpnI2dnZTndFpS0rQRDI/f196vwovAMf4zkYEYksy0qePz8/R67rvnrQcHH6mG3b0Xg8XnvNtu2N72Pb9sb3sm07cl136XXDMKJWq5X6+uprURRFjUZj6Xmr1Ur9vLdqnUwmkWEYS98v7bUoemmb1QO2o9EoMgwj94HYbdpn9TXTNFM/t9VqqR8EdRwntc1Ho9HSa5ZlRYPBYG261To1ay/aMhVFL9+v0Wgkn7P6fTfN/0ajsTagxnXdtdp+luM4a21gWdZS2+/jQHqZBzB86GCJV3qO40S2bb86GmU8HqeOIhsMBrmDpdFopP5wHMeJRCSaTCZr77fp9dX3GY1GkYgsrdTiUXCrK7W3arVtO7U94vaKtVqtyDCM1PfIOyosirZrn9XpXdeN0rabtEfAbRo9uLqybDQaG9vZMIyleaNZe9GWqfjvhmEkbbbadpvmfxS9LE/xez8/P6vPz3h9sKrX6y0t44uf22q1ItM0cz/S2i42GAxSA74sPvQxFtM0ZTAYyGAwkNFoJJPJ5NVpgyCQi4uLpeMwjuNsNWoszcnJSeouJNM0N76+epA03iWwOG28eybP8aIwDMXzPPny5cva3y4uLuT79+/J836/L1+/ft34nbS81j6rFncvxcIwlF9//VWtHhGRdrstjuOs7QO3LEuurq6S5/1+f+P5PLZtS7vdTp5r1l6kZSrt8xffK4vBYCCe50m/35dut6u6m8jzPBkOh9LpdNb+FrdL/LtYnLeu68pkMsn9cBxnYy3xclVWHGNZYJrm2g/N8zyxbVsMw5DBYCDfvn1LDrRZliWu66buH93WphVx1hW0aZrJ/vcwDCUIguTHH+8bzyIOjslksrSCi8U/6CAIkpPa9iFrOxiGkexzX1xRa+4nj1c0acE2GAySf8fHBl6bt4sraO3ai7JMbfPZabX0ej1pNptLbazBdV0xTTN1QzGuNwgCGY1GOz3u4ft+ctyorD50jyXN6tbK4kFSx3Hk+flZRqORtFotCcNwrQeTJs/Qy01bb3m26uKRJN1uV0QkdYt6k7jW+Id0cXEhjUYj9XEIedqh2WwmP1KRl5Wg5uiaeCj6Wz2JbYasa9ZelGVq289edX5+LqZpLvXyNMQbka+JQ22Xer1eqUNFhB7LmtUFPv5RxOFh23bycF1XLi8vl0aSpYlHuexjYen3+9Jut2U8Hm/1eXGt8VbbW7s64s94bTfiocQrv16vJ1dXV6m79X5G1u9+fn4uIpvb8unpaW1e7br2PLSWKa3l//b2VsbjsfzjH/+QZrOpMhIs/p2fnZ29Ol29Xl/7Hu12+82NyzSdTid1d5fneaUbXryKHssrVheWtK53s9lc2yJNCyfNYw2vabfb8vXr17WFf3GrcXHI62u1tlqtjT/axa22Vqu18Yf19PR00JPlGo2G9Pt9ub293ck+a8dxNn73eBeiYRjiOE4yFHfVcDhM3frede1ZaS5TGrV0Oh0xDEP+85//SL/f32qlvsowDDEMY+OyGv/G0zYgXdeV8Xic+7FpngZBsLf1xa4QLBv4vi/NZnNpC6bf769tdY5Go6UDeSLrB7eDINjYE1q0aSX82uurVvfXi7ysuGzbTn2P12p1XVcMw1i7IOBwOFw6EB1PNxwOl6aL22ubXUEa7SPyEoBvBVu9Xt/6GFF8RvRqMPi+v7RyuLm5WTpnZLE+27ZTdy1mqf0tRVumYm99r7S/x20cv5dlWdJoNFIv3nh8fJx7nnY6ndTzg+Ll2DCMpH00wizNu7liwaGHpe3T8/Nz5DhOZFlWcmG3+ByERqORDDVc/Hs8xHI0GkWu60a9Xi9yXTd5bBoy2Gq1kvMH4veYTCaR4ziRYRjJZ4/H49TXR6NR5tdt207qiL9j/Nmu60aTySSaTCaRbdtRq9VaG96ZVuvq3+NzHlzXXRtevThdPE2r1YrG43EytNSyrI3/X+xn22exHRbZtr02rHbR4hDWbbVareTcjF6vl9qO8XRxOzUajTfPh3ir9k2KuEyNx+Nk+HP82YtDajfN/3g4u/z/FIHF6RdfdxwnqcOyrMgwjNxtFy+7vV4vGfIbv+doNIoajUbq99UUn2NUZpUo+oBXfAPw7sWj8bROB0B27AoD8C7FgwawfwQLgHcpDMPSj64qK4IFwLvzbg6ClxTBAuDd6ff70mq1Dl3Gh8XBewCAKnosAABVBAsAQFWprhUWBIEMh8PkEvaNRoNRHyl830/ODL6/v5ebmxva6Q2LlwrBMs/zlq62q3k17/ciCALxPC+5SoHjOB97qPNhz8/MZ/HmP/FZuli3eDaz67obb5qEF+PxOPVmXfj7bPMoevnNpd3sDtHaTbm072pZNqXZFbZ6LSDTNHd2vZ4y830/ubS5yMtFEhcvv451Zb/3xS41m83k3iOL92XBsk0XGP2oShMscTdz0cnJSXLZBrywLCu5MKLI+v1VsGw4HJb6Tn27FF9A1DAM8X1/403N8PL7Ojs7S3aJxTcD/KhKEyxvXc4af1tcUd7e3iZ3wMQyzsx+XXyF5vi4Zr/fX7uCNV7Et9So1+syGAw+/MZKqQ7ep+EM283CMJThcCjj8fjQpRTS3d3dwe6EWQZPT08SBEGyYdJoNOT4+FgiTn1b43meuK4rQRAk9yrSuAFZWZWmx7J4L4RY3E1Huna7LaPRiDZK4XmefP369dBlFJppmskNsET+vg8Ku5+XBUEg9/f3yX11JpOJ3N3dfejjmqXpsdi2nboFEN/2Fcuur6+l3W6LaZpJr46AWXZ3d5f8OwgC6Xa7cnV1xWXW/4/jKdn4vr9062jTNKXT6XzovSml6bGsLuRBEMj5+TkryxTD4VAsy0pC5e7ujnZaEW9dxg+RlxFQhMrfTNOU8/PzZAUZj56jjZZZliX39/dLrz0+Pn7odirVtcKCIJBerydfvnyR+/t7TmhLEQTB2i1ZDcOQ5+fnA1VUbPHtgtvttjQaDcJlRRiG0m635ezsTMbjcdILxjLP88T3/WR9ZNv2h26nUgULAKD4SrMrDABQDgQLAEAVwQIAUEWwAABUESwAAFUECwBAFcECAFBVumCZz+fy73//W+bz+aFLKTTaKRvaKRvaKRva6UXpTpCczWZSq9VkOp3K0dHRocspLNopG9opG9opG9rpRel6LACAYiNYAACq9nbZ/L/++kv+/PNP+fz5s1Qqla3fZzabLf0X6WinbGinbGinbN5zO0VRJD9+/JDffvtNPn16vU+yt2Ms//3vf+X09HQfHwUA2JGHhwf5/fffX51mbz2Wz58/7+ujsAPT6fTQJeAdqdVqhy6hFIr0u5vNZnJ6epppXb63YPmZ3V84vI88wgU4lCL+7rKsyzl4DwBQRbAAAFQRLAAAVQQLAEAVwQIAUEWwAABUESwAAFUECwBAFcECAFBFsAAAVOW6pEsQBDIcDsU0TQmCQBqNhhiGsaPSAABllCtYLi8vZTwei8hLyHz79k0Gg8FOCgMAlFPmXWFBECw9N01TPM9TLwgAUG6Zg8XzPDk5OVl67eTkRHzfVy8KAFBemXeFhWGY+vrT01Pq6/P5XObzefL8Pd5RDQCw7qdHhW0KnG63K7VaLXlw90gA+BgyB4thGGu9k6enp42jwjqdjkyn0+Tx8PDwU4UCAMohc7DYtp36+vn5eerr1WpVjo6Olh4AgPcvc7CYprn0PAgCOT8/5zwWAMCSXOexDAYDabfb8uXLF7m/v+ccFgDAmkoURdE+Pmg2m0mtVtvHR2EH9rSY4IOoVCqHLqEUivS7i9fh0+n0zUMbXCsMAKCKYAEAqCJYAACqCBYAgCqCBQCgimABAKgiWAAAqggWAIAqggUAoIpgAQCoIlgAAKoIFgCAqlxXN8Z+FOnCc0XGhQzLi2X8faPHAgBQRbAAAFQRLAAAVQQLAEAVwQIAUEWwAABUESwAAFUECwBAFcECAFBFsAAAVBEsAABVBAsAQBXBAgBQRbAAAFQRLAAAVQQLAEAVwQIAUEWwAABUESwAAFUECwBAFcECAFBFsAAAVBEsAABVBAsAQBXBAgBQRbAAAFQRLAAAVQQLAEAVwQIAUEWwAABUESwAAFUECwBAFcECAFBFsAAAVBEsAABVBAsAQNUvhy7g0KIoOnQJayqVyqFLKIUizjtkwzKeTVmXcXosAABVBAsAQBXBAgBQRbAAAFQRLAAAVQQLAEAVwQIAUEWwAABUESwAAFUECwBAFcECAFBFsAAAVBEsAABVua5u7Pu+eJ4nIiL39/dyc3MjhmHsoi4AQEnlChbP86TVaomIyPX1tfzzn/+U8Xi8k8IAAOWUeVeY7/vS7XaT547jiO/7EgTBTgoDAJRT5mCxLEtubm6S52EYiojIycmJelEAgPLKtSvMcZzk37e3t2Lb9sZjLPP5XObzefJ8NpttVyEAoFS2GhUWhqEMh0MZDAYbp+l2u1Kr1ZLH6enp1kUCAMqjEm1xU+VmsyntdltM09w4TVqPpYjhUsR7SnM/8GyKOO+QDct4NkVaxmezmdRqNZlOp3J0dPTqtLl2hYm8jAaLQyU+zpK2O6xarUq1Ws379gCAksu1K2w4HIplWUmo3N3dcR4LAGBJ5l1hQRBIvV5fes0wDHl+fs70QXE3qmiK1NWMsZsgmyLOO2TDMp5NkZbxnewKM02zUF8SAFBMXCsMAKCKYAEAqCJYAACqCBYAgCqCBQCgimABAKgiWAAAqggWAIAqggUAoIpgAQCoIlgAAKoIFgCAqtz3Y/lZWa6MCWRRxCvkFvFCrbRTNkVsp7KixwIAUEWwAABUESwAAFUECwBAFcECAFBFsAAAVBEsAABVBAsAQBXBAgBQRbAAAFQRLAAAVQQLAEAVwQIAUEWwAABUESwAAFUECwBAFcECAFBFsAAAVBEsAABVBAsAQBXBAgBQRbAAAFQRLAAAVQQLAEAVwQIAUEWwAABUESwAAFUECwBAFcECAFBFsAAAVBEsAABVBAsAQBXBAgBQRbAAAFQRLAAAVQQLAEDVL4cu4NAqlcqhS1gTRdGhS1hTxHYqItoJmsq6PNFjAQCoIlgAAKoIFgCAKoIFAKCKYAEAqCJYAACqCBYAgCqCBQCgimABAKgiWAAAqggWAIAqggUAoIpgAQCo2jpY2u22hGGoWAoA4D3YKlh835fr62vtWgAA78BWwRIEgZimqV0LAOAdyB0sw+FQHMfZRS0AgHcg1x0kwzAUwzAyTTufz2U+nyfPZ7NZrsIAAOWUq8dyd3cntm1nmrbb7UqtVksep6enWxUIACiXzMHieZ58/fo18xt3Oh2ZTqfJ4+HhYasCAQDlkmtX2N3dXfLvIAik2+3K1dWVWJa1Nm21WpVqtfrzFQIASqUSRVG01f9YqchkMsk8Omw2m0mtVpPpdCpHR0fbfOROVCqVQ5ewZstZslNFbCeUF8t4eWVZh+ceFRaGYXIOi+u64vv+dtUBAN6lrXssedFjyY6tObx3LOPltZMeCwAAryFYAACqCBYAgCqCBQCgimABAKgiWAAAqggWAIAqggUAoIpgAQCoIlgAAKoIFgCAKoIFAKAq1/1Y8HFx0UBg/4r0u4svJJwFPRYAgCqCBQCgimABAKgiWAAAqggWAIAqggUAoIpgAQCoIlgAAKoIFgCAKoIFAKCKYAEAqCJYAACqCBYAgCqCBQCgimABAKgiWAAAqggWAIAqggUAoIpgAQCoIlgAAKoIFgCAKoIFAKCKYAEAqCJYAACqCBYAgCqCBQCgimABAKgiWAAAqggWAIAqggUAoIpgAQCoIlgAAKoIFgCAKoIFAKCKYAEAqCJYAACqfjl0ASiHSqVy6BLwjrA8ZVPWdqLHAgBQRbAAAFQRLAAAVQQLAEAVwQIAUEWwAABUESwAAFUECwBAFcECAFBFsAAAVBEsAABVBAsAQBXBAgBQlfvqxp7nSRAEYpqmiIjYtq1eFACgvHL1WDzPk8FgII1GQ0zTlGazuau6AAAllavH0mw2ZTwei4iIaZoyGo12UhQAoLwy91iCIJCnpycxDEN835cwDJPdYQAAxDIHi+/7cnJyIsPhUEzTlH6/L8PhcOP08/lcZrPZ0gMA8P5l3hX29PQkQRCIbdtiGIY0Gg05Pj6WKIpSp+92u/LHH3+oFQoAKIfMPRbTNMUwDDEMQ0Qk+a/v+6nTdzodmU6nyePh4eGniwUAFF/mHkve4ynValWq1WruggAA5Zarx3J+fi5hGIqIJOeyWJa1q9oAACWUa7jxYDCQdrstZ2dnMh6PGW4MAFhTiTYdfVc2m82kVqvJdDqVo6OjfXxkJpVK5dAlrNnTLMmliO0EYP+yrMO5VhgAQBXBAgBQRbAAAFQRLAAAVQQLAEAVwQIAUEWwAABUESwAAFUECwBAFcECAFBFsAAAVBEsAABVua5urKFWq+37I1/FBR+zKWI7IZsiLk/Ipki/u/hCwlnQYwEAqCJYAACqCBYAgCqCBQCgimABAKgiWAAAqggWAIAqggUAoIpgAQCoIlgAAKoIFgCAKoIFAKCKYAEAqCJYAACqCBYAgCqCBQCgimABAKgiWAAAqggWAIAqggUAoIpgAQCoIlgAAKoIFgCAKoIFAKCKYAEAqCJYAACqCBYAgCqCBQCgimABAKgiWAAAqggWAIAqggUAoIpgAQCoIlgAAKoIFgCAKoIFAKDql0MXgHKoVCqHLgFASdBjAQCoIlgAAKoIFgCAKoIFAKCKYAEAqCJYAACqCBYAgCqCBQCgimABAKgiWAAAqggWAIAqggUAoIpgAQCoynV14yAIxPM8OTk5kSAIxHEcMU1zV7UBAEooV7AMh0NptVrJ82azKb1eT70oAEB55doVdnt7u6s6AADvRK5gOTk5kbOzs2SX2MXFxa7qAgCUVK5gGQwGIiJSr9dlMBiI4zgbp53P5zKbzZYeAID3L9cxFs/zxHVdCYJAms2miMjGYyzdblf++OOPn68QAFAqlSiKoiwTBkEgvV5PXNdNnp+dncl4PE4dGTafz2U+nyfPZ7OZnJ6eKpWtJ+PX3yvuLw9ApFjrp9lsJrVaTabTqRwdHb06beYei+/78uXLl+S5aZrS6XQkDMPU6avVqlSr1axvDwB4JzIfY7EsS+7v75dee3x8FMuy1IsCAJRX5l1hIi/HWHzfF8MwRETEtu3MJ0jG3aiiKVJXM8auMAAixVo/5dkVlitYNIoqmiLNuBjBAkCkWOunPMHCtcIAAKoIFgCAKoIFAKCKYAEAqCJYAACqCBYAgCqCBQCgimABAKgiWAAAqggWAIAqggUAoIpgAQCoynUHSQ1ZLmD20RXpwnNFVsSLdRZx3tFO5VXEeZcFPRYAgCqCBQCgimABAKgiWAAAqggWAIAqggUAoIpgAQCoIlgAAKoIFgCAKoIFAKCKYAEAqCJYAACqCBYAgCqCBQCgimABAKgiWAAAqggWAIAqggUAoIpgAQCoIlgAAKoIFgCAKoIFAKCKYAEAqCJYAACqCBYAgCqCBQCgimABAKgiWAAAqggWAIAqggUAoIpgAQCoIlgAAKoIFgCAKoIFAKCKYAEAqPplXx8URZGIiMxms319JLB3LN/Z0E7lFa/LX7O3YPnx44eIiJyenu7rI4G9q9Vqhy6hFGin8vrx48eb868SZYkfBX/99Zf8+eef8vnzZ6lUKlu/z2w2k9PTU3l4eJCjoyPFCt8X2ikb2ikb2imb99xOURTJjx8/5LfffpNPn14/irK3HsunT5/k999/V3u/o6OjdzfjdoF2yoZ2yoZ2yua9tlPWniYH7wEAqggWAICq0gVLtVqVf/3rX1KtVg9dSqHRTtnQTtnQTtnQTi/2dvAeAPAxlK7HAgAoNoIFAKCKYAEAqCJYAACqCBYAgCqCBQCgimABAKgiWAAAqv4HwBVmmKdJ/wgAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Show plot inline in ipython.\n",
    "%matplotlib inline\n",
    "\n",
    "# Plot properties.\n",
    "plt.rc('text', usetex=True)\n",
    "plt.rc('font', family='serif')\n",
    "\n",
    "# Create figure.\n",
    "plt.figure()\n",
    "plt.figure(figsize=(12, 12))\n",
    "\n",
    "# Plot sparsity pattern for the true covariance matrix.\n",
    "plt.subplot(2, 2, 1)\n",
    "plt.spy(Strue)\n",
    "plt.title('Inverse of true covariance matrix', fontsize=16)\n",
    "plt.show()\n",
    "\n",
    "# Plot sparsity pattern for each result, corresponding to a specific alpha.\n",
    "plt.spy(S)\n",
    "plt.title('Estimated inv. cov matrix, $\\\\alpha$={}'.format(alpha), fontsize=16)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x = cp.Variable(2)\n",
    "# problem = cp.Problem(cp.Minimize(cp.norm(x)), [x[0] >= 0, x[1] * x[0] >= 0])\n",
    "problem = cp.Problem(cp.Minimize(cp.norm(x)), [cp.bmat([[x[0], 0], [0, x[1]]]) >> 0])\n",
    "problem.solve(solver=cp.SCS)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "x.value"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}