{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Post-processing\n",
    "In this section, we will look at how to access and visualize data from saved files.\n",
    "We will also look at warm/hot starting of an optimization using\n",
    "previously saved data.\n",
    "\n",
    "## Saving optimization data to a file\n",
    "We saw earlier in [Basic User Guide](./basic.ipynb) how to save optimization data.\n",
    "Since our goal is to understand how we can use the saved files, we begin by\n",
    "running our standard problem and saving it to a file named `postprocessing.hdf5`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization terminated successfully    (Exit mode 0)\n",
      "            Final objective value                : 5.000000e-01\n",
      "            Final optimality                     : 1.232595e-31\n",
      "            Final feasibility                    : 0.000000e+00\n",
      "            Number of major iterations           : 4\n",
      "            Number of function evaluations       : 4\n",
      "            Number of derivative evaluations     : 4\n",
      "            Average Derivative evaluation time   : 0.000042 s per evaluation\n",
      "            Average Function evaluation time     : 0.000043 s per evaluation\n",
      "            Total Function evaluation time       : 0.000169 s [  0.68%]\n",
      "            Total Derivative evaluation time     : 0.000173 s [  0.70%]\n",
      "            Optimizer time                       : 0.000115 s [  0.46%]\n",
      "            Processing time                      : 0.024273 s [ 98.15%]\n",
      "            Visualization time                   : 0.000000 s [  0.00%]\n",
      "            Total optimization time              : 0.024730 s [100.00%]\n",
      "            Summary saved to                     : slsqp_summary.out\n",
      "            Iteration data saved to              : postprocessing.hdf5\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "def objective(x):\n",
    "    return x[0]**2 + x[1]**2\n",
    "\n",
    "def gradient(x):\n",
    "    return np.array([2*x[0], 2*x[1]])\n",
    "\n",
    "def constraints(x):\n",
    "    return  np.array([x[0] + x[1] - 1, 3*x[0] + 2*x[1] - 1])\n",
    "\n",
    "def jacobian(x):\n",
    "    return np.array([[1, 1], [3, 2]])\n",
    "\n",
    "x_lower = np.array([0.4, -np.inf])\n",
    "x_upper = np.array([np.inf, 0.6])\n",
    "num_eqcon = 1\n",
    "x0 = np.array([2,3])\n",
    "\n",
    "from pyslsqp import optimize\n",
    "results = optimize(x0, obj=objective, grad=gradient, con=constraints, jac=jacobian, meq=num_eqcon, xl=x_lower, xu=x_upper,\n",
    "                   save_itr='all', save_vars=['x', 'objective', 'optimality', 'feasibility', 'step', 'mode', 'iter', 'majiter', 'ismajor', 'constraints', 'gradient', 'multipliers', 'jacobian'],\n",
    "                   save_filename=\"postprocessing.hdf5\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Viewing saved file contents\n",
    "To view what data is available in a saved file, we call the `print_file_contents()` utility with the saved file name `postprocessing.hdf5`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Available data in the file:\n",
      "---------------------------\n",
      "     Attributes of optimization  : ['acc', 'con_scaler', 'finite_diff_abs_step', 'finite_diff_rel_step', 'hot_start', 'iprint', 'keep_plot_open', 'load_filename', 'm', 'maxiter', 'meq', 'n', 'obj_scaler', 'save_figname', 'save_filename', 'save_itr', 'save_vars', 'summary_filename', 'visualize', 'visualize_vars', 'warm_start', 'x0', 'x_scaler', 'xl', 'xu']\n",
      "     Saved variable iterates     : ['constraints', 'feasibility', 'gradient', 'ismajor', 'iter', 'jacobian', 'majiter', 'mode', 'multipliers', 'objective', 'optimality', 'step', 'x']\n",
      "     Results of Optimization     : ['constraints', 'feasibility', 'fev_time', 'gev_time', 'gradient', 'jacobian', 'message', 'multipliers', 'nfev', 'ngev', 'num_majiter', 'objective', 'optimality', 'optimizer_time', 'processing_time', 'save_filename', 'status', 'success', 'summary_filename', 'total_time', 'visualization_time', 'x']\n"
     ]
    }
   ],
   "source": [
    "from pyslsqp.postprocessing import print_file_contents\n",
    "print_file_contents('postprocessing.hdf5')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loading results and attributes\n",
    "In the previous code block, we saw that there are mainly three types of information available in a saved file: *attributes*, *variable iterates*, and *results*.\n",
    "Attributes and results of optimization can be loaded as dictionaries by simply calling the `load_attributes()` and `load_results()` utility functions with the saved file name as shown below."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Attributes:\n",
      "--------------------------------------------------\n",
      "        acc                      : 1e-06\n",
      "        con_scaler               : 1.0\n",
      "        finite_diff_abs_step     : None (undefined)\n",
      "        finite_diff_rel_step     : 1.4901161193847656e-08\n",
      "        hot_start                : False\n",
      "        iprint                   : 1\n",
      "        keep_plot_open           : False\n",
      "        load_filename            : None (undefined)\n",
      "        m                        : 2\n",
      "        maxiter                  : 100\n",
      "        meq                      : 1\n",
      "        n                        : 2\n",
      "        obj_scaler               : 1.0\n",
      "        save_figname             : slsqp_plot.pdf\n",
      "        save_filename            : postprocessing.hdf5\n",
      "        save_itr                 : all\n",
      "        save_vars                : ['x', 'objective', 'optimality', 'feasibility', 'step', 'mode', 'iter', 'majiter', 'ismajor', 'constraints', 'gradient', 'multipliers', 'jacobian']\n",
      "        summary_filename         : slsqp_summary.out\n",
      "        visualize                : False\n",
      "        visualize_vars           : ['objective', 'optimality', 'feasibility']\n",
      "        warm_start               : False\n",
      "        x0                       : [2 3]\n",
      "        x_scaler                 : 1.0\n",
      "        xl                       : [ 0.4 -inf]\n",
      "        xu                       : [inf 0.6]\n",
      "--------------------------------------------------\n",
      "Results:\n",
      "--------------------------------------------------\n",
      "        constraints              : [0.  1.5]\n",
      "        feasibility              : 0.0\n",
      "        fev_time                 : 0.00016927719116210938\n",
      "        gev_time                 : 0.00017261505126953125\n",
      "        gradient                 : [1. 1.]\n",
      "        jacobian                 : [[1. 1.]\n",
      " [3. 2.]]\n",
      "        message                  : Optimization terminated successfully\n",
      "        multipliers              : [1. 0.]\n",
      "        nfev                     : 4\n",
      "        ngev                     : 4\n",
      "        num_majiter              : 4\n",
      "        objective                : 0.5\n",
      "        optimality               : 1.232595164407831e-31\n",
      "        optimizer_time           : 0.00011491775512695312\n",
      "        processing_time          : 0.024273395538330078\n",
      "        save_filename            : postprocessing.hdf5\n",
      "        status                   : 0\n",
      "        success                  : True\n",
      "        summary_filename         : slsqp_summary.out\n",
      "        total_time               : 0.024730205535888672\n",
      "        visualization_time       : 0.0\n",
      "        x                        : [0.5 0.5]\n",
      "--------------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "from pyslsqp.postprocessing import load_attributes, load_results, print_dict_as_table\n",
    "attributes = load_attributes('postprocessing.hdf5')\n",
    "results = load_results('postprocessing.hdf5')\n",
    "\n",
    "print(\"Attributes:\")\n",
    "print_dict_as_table(attributes)\n",
    "\n",
    "print(\"Results:\")\n",
    "print_dict_as_table(results)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loading variable iterates\n",
    "To load variable iterates from a saved file, we can use the `load_variable()` utility function from `pyslsqp.postprocessing`.\n",
    "This function needs two arguments. \n",
    "The first argument is the file name and the second argument is the list of variable names to load.\n",
    "The function returns a dictionary with *keys* as variable names and *values* as the list of variable iterates corresponding to a variable name."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--------------------------------------------------\n",
      "        x                        : [array([2. , 0.6]), array([0.4, 0.6]), array([0.4, 0.6]), array([0.53333333, 0.46666667]), array([0.53333333, 0.46666667]), array([0.5, 0.5]), array([0.5, 0.5]), array([0.5, 0.5])]\n",
      "        objective                : [4.36, 0.5200000000000002, 0.5200000000000002, 0.5022222222222221, 0.5022222222222221, 0.5, 0.5, 0.5]\n",
      "        optimality               : [99.0, 3.8399999999999985, 0.0, 7.697546304067748e-16, 0.0, 1.5173048003210468e-16, 0.0, 1.232595164407831e-31]\n",
      "        feasibility              : [99.0, 4.440892098500626e-16, 4.440892098500626e-16, 1.1102230246251565e-16, 1.1102230246251565e-16, 0.0, 0.0, 0.0]\n",
      "        step                     : [99.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0]\n",
      "        mode                     : [0, 1, -1, 1, -1, 1, -1, 0]\n",
      "        iter                     : [0, 1, 2, 3, 4, 5, 6, 7]\n",
      "        majiter                  : [0, 1, 1, 2, 2, 3, 3, 4]\n",
      "        ismajor                  : [True, True, False, True, False, True, False, True]\n",
      "        constraints              : [array([1.6, 6.2]), array([4.4408921e-16, 1.4000000e+00]), array([4.4408921e-16, 1.4000000e+00]), array([-1.11022302e-16,  1.53333333e+00]), array([-1.11022302e-16,  1.53333333e+00]), array([0. , 1.5]), array([0. , 1.5]), array([0. , 1.5])]\n",
      "        gradient                 : [array([4. , 1.2]), array([4. , 1.2]), array([0.8, 1.2]), array([0.8, 1.2]), array([1.06666667, 0.93333333]), array([1.06666667, 0.93333333]), array([1., 1.]), array([1., 1.])]\n",
      "        multipliers              : [array([0., 0.]), array([2.4, 0. ]), array([2.4, 0. ]), array([1.06666667, 0.        ]), array([1.06666667, 0.        ]), array([1., 0.]), array([1., 0.]), array([1., 0.])]\n",
      "        jacobian                 : [array([[1., 1.],\n",
      "       [3., 2.]]), array([[1., 1.],\n",
      "       [3., 2.]]), array([[1., 1.],\n",
      "       [3., 2.]]), array([[1., 1.],\n",
      "       [3., 2.]]), array([[1., 1.],\n",
      "       [3., 2.]]), array([[1., 1.],\n",
      "       [3., 2.]]), array([[1., 1.],\n",
      "       [3., 2.]]), array([[1., 1.],\n",
      "       [3., 2.]])]\n",
      "--------------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "from pyslsqp.postprocessing import load_variables\n",
    "vars = load_variables('postprocessing.hdf5', ['x', 'objective', 'optimality', 'feasibility', 'step', 'mode', 'iter', 'majiter', 'ismajor', 'constraints', 'gradient', 'multipliers', 'jacobian'])\n",
    "print_dict_as_table(vars)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Users have the option to specify `itr_start` and `itr_end`, which will load all variable iterates between and including these two points.\n",
    "By default, `itr_start` is set to *0* and `itr_end` to *-1*.\n",
    "If the saved file contains all iterations of the optimization algorithm, but the user only needs the variables from major iterations between `itr_start` and `itr_end`,\n",
    "they can set `major_only=True` in the function call.\n",
    "In the following code, we only load the major iterations 1 ,2, and 3. \n",
    "The variables we load are the major iteration and optimization iteration number, \n",
    "the *ismajor* indicator, $x_1$, the objective, optimality, feasibililty, and the constraint derivative $\\frac{\\partial c_1}{\\partial x_1}(x)$."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--------------------------------------------------\n",
      "        majiter                  : [1, 2, 3]\n",
      "        iter                     : [1, 3, 5]\n",
      "        ismajor                  : [True, True, True]\n",
      "        x[0]                     : [0.40000000000000036, 0.5333333333333333, 0.5000000000000002]\n",
      "        objective                : [0.5200000000000002, 0.5022222222222221, 0.5]\n",
      "        optimality               : [3.8399999999999985, 7.697546304067748e-16, 1.5173048003210468e-16]\n",
      "        feasibility              : [4.440892098500626e-16, 1.1102230246251565e-16, 0.0]\n",
      "        jacobian[0,0]            : [1.0, 1.0, 1.0]\n",
      "--------------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "vars = load_variables('postprocessing.hdf5', ['majiter', 'iter', 'ismajor', 'x[0]', 'objective', 'optimality', 'feasibility', 'jacobian[0,0]'], itr_start=1, itr_end=3, major_only=True)\n",
    "print_dict_as_table(vars)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visualizing saved optimization\n",
    "To visualize a saved optimization, we use the `visualize()` utility from `pyslsqp.postprocessing`.\n",
    "The two required arguments are the saved file name and the list of variable names to visualize.\n",
    "This function also optionally takes in keyword arguments `itr_start`, `itr_end`, and `major_only`,\n",
    "with their meanings exactly the same as discussed above for the `load_variables` function.\n",
    "The final optimization plot can be saved by setting the keyword argument\n",
    "`save_figname` which is `None` by default.\n",
    "\n",
    "In the following code, we plot **all** the **major** iterations for $x_1$, objective, optimality, feasibility, and $\\frac{\\partial c_1}{\\partial x_1}(x)$."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x1500 with 5 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "\n",
    "from pyslsqp.postprocessing import visualize\n",
    "visualize('postprocessing.hdf5', ['x[0]', 'objective', 'optimality', 'feasibility', 'jacobian[0,0]'], itr_start=0, itr_end=-1, major_only=True, save_figname='postprocessing_plot.pdf')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Restarting optimization - Warm start and Hot start\n",
    "In many situations, users may need to restart a previously run optimization.\n",
    "For instance, if the user wants SLSQP to converge to a solution with higher accuracy than before,\n",
    "they would need to run `optimize()` again, this time with a smaller value for the accuracy parameter `acc`.\n",
    "Additionally, they might need to increase the value of `maxiter` to ensure that \n",
    "the optimization doesn't stop prematurely due to reaching the iteration limit.\n",
    "\n",
    "In PySLSQP, ***warm starting*** refers to the process of restarting a previously run optimization \n",
    "using the most recent value of `x` from a saved file. \n",
    "If `warm_start=True` is set, the initial guess `x0` provided by the user is \n",
    "replaced with the value of `x` from the *results* in the file specified by `load_filename`. \n",
    "If for some reason the *results* data is not available in `load_filename`, \n",
    "PySLSQP will use the value of `x` from the last available saved iteration as `x0`.\n",
    "<!-- Setting `warm_start=True` overwrites the user input `x0` with the value of `x` from the *results*\n",
    "available in the user specified `load_filename`.\n",
    "For some reason, if no data on *results* are available in `load_filename`, PySLSQP will\n",
    "set `x0` as the value of `x` from the last saved iteration available. -->\n",
    "\n",
    "In the following example, we warm start the optimization from the `postprocessing.hdf5` file saved before."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Warm starting from previous optimization solution x from postprocessing.hdf5...\n",
      "Optimization terminated successfully    (Exit mode 0)\n",
      "            Final objective value                : 5.000000e-01\n",
      "            Final optimality                     : 2.465190e-31\n",
      "            Final feasibility                    : 0.000000e+00\n",
      "            Number of major iterations           : 1\n",
      "            Number of function evaluations       : 1\n",
      "            Number of derivative evaluations     : 1\n",
      "            Average Derivative evaluation time   : 0.000050 s per evaluation\n",
      "            Average Function evaluation time     : 0.000031 s per evaluation\n",
      "            Total Function evaluation time       : 0.000050 s [  1.73%]\n",
      "            Total Derivative evaluation time     : 0.000031 s [  1.08%]\n",
      "            Optimizer time                       : 0.000019 s [  0.67%]\n",
      "            Processing time                      : 0.002776 s [ 96.52%]\n",
      "            Visualization time                   : 0.000000 s [  0.00%]\n",
      "            Total optimization time              : 0.002876 s [100.00%]\n",
      "            Summary saved to                     : slsqp_summary.out\n"
     ]
    }
   ],
   "source": [
    "results = optimize(x0, obj=objective, grad=gradient, con=constraints, jac=jacobian, meq=num_eqcon, xl=x_lower, xu=x_upper,\n",
    "                   warm_start=True, load_filename=\"postprocessing.hdf5\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We see above that the optimization converged in a single iteration since we started from an already converged solution."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In PySLSQP, ***hot starting*** refers to the process of restarting a previously run optimization \n",
    "by reusing the function (objective and constraints) and derivative values available from a previously saved file.\n",
    "This approach is particularly beneficial when the functions and/or their derivatives are costly to evaluate.\n",
    "One advantage of hot starting over warm starting is that during a hot start,\n",
    "the BFGS Hessians approximated by the SLSQP algorithm will follow the same path as in\n",
    "the previous optimization while also saving the cost of function and derivative evaluations.\n",
    "In contrast, during a warm start, although the algorithm starts from the previous solution `x`, the Hessian\n",
    "is initialized as the identity matrix, which might result in more iterations before the\n",
    "algorithm can converge.\n",
    "\n",
    "It is very important to note that hot starting in PySLSQP will only work if `save_itr` was set to `\"all\"`\n",
    "and `save_vars` included all of `\"objective\", \"constraints\", \"gradient\", and \"jacobian\"` in the previous optimization run.\n",
    "This is because a hot start requires function and derivative data from *all* iterations and not just major iterations.\n",
    "To hot start a problem, set `hot_start=True` when calling `optimize()` and specify `load_filename`\n",
    "to indicate where the saved data will be reused from.\n",
    "The following code hot-starts our optimization from the previously saved file `postprocessing.hdf5`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Hot starting using saved x, objective, constraints, gradient, and jacobian from postprocessing.hdf5...\n",
      "Hot start is complete at iteration 7. Starting normal function evaluations...\n",
      "Optimization terminated successfully    (Exit mode 0)\n",
      "            Final objective value                : 5.000000e-01\n",
      "            Final optimality                     : 1.232595e-31\n",
      "            Final feasibility                    : 0.000000e+00\n",
      "            Number of major iterations           : 4\n",
      "            Num fun evals (reused in hotstart)   : 4 (4)\n",
      "            Num deriv evals (reused in hotstart) : 4 (4)\n",
      "            Average Derivative evaluation time   : 0.000015 s per evaluation\n",
      "            Average Function evaluation time     : 0.000008 s per evaluation\n",
      "            Total Function evaluation time       : 0.000060 s [  0.69%]\n",
      "            Total Derivative evaluation time     : 0.000031 s [  0.35%]\n",
      "            Optimizer time                       : 0.000096 s [  1.09%]\n",
      "            Processing time                      : 0.008577 s [ 97.87%]\n",
      "            Visualization time                   : 0.000000 s [  0.00%]\n",
      "            Total optimization time              : 0.008764 s [100.00%]\n",
      "            Summary saved to                     : slsqp_summary.out\n"
     ]
    }
   ],
   "source": [
    "results = optimize(x0, obj=objective, grad=gradient, con=constraints, jac=jacobian, meq=num_eqcon, xl=x_lower, xu=x_upper,\n",
    "                   hot_start=True, load_filename=\"postprocessing.hdf5\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see from the console output above that all 4 of the function and derivative evaluations required by SLSQP\n",
    "were reused from the saved file, and were never really computed using the user-provided functions and derivatives.\n",
    "\n",
    "For more details on any of the post-processing utilities, visit the [API Reference](./api.md) page."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "venv",
   "language": "python",
   "name": "venv"
  },
  "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": 2
}