docs/tutorial.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "85d047e3-18d0-4931-904a-39ed9b8adfc2",
   "metadata": {},
   "source": [
    "# Tutorial\n",
    "This guide explains how to use **ionBench**. It gives practical advice on setting up and editing problems classes and running optimisers. If you want more theoretical information, such as what the performance metrics are, check out the **introduction**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "4e8ab68e-e916-42e3-a5a9-ba37386acc7a",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-11-08T11:38:07.491466Z",
     "start_time": "2024-11-08T11:37:44.277238Z"
    }
   },
   "outputs": [],
   "source": [
    "import ionbench\n",
    "import numpy as np\n",
    "import os"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b02b6c28-da5b-4336-af52-1b20a8f73502",
   "metadata": {},
   "source": [
    "## Benchmarkers and Test Problems\n",
    "Each of the benchmarker problem classes are loaded in a similar way. They may take a few seconds to initialise while they compile the Myokit simulation objects to run the models. Typically, you would only need one benchmarker at once, here we will use the staircase Hodgkin-Huxley benchmarker, but the functions to load the other benchmarkers are shown below, just commented out."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "4814dbd7-89b1-4425-9b4b-31d2da399faf",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-11-08T11:38:20.937898Z",
     "start_time": "2024-11-08T11:38:07.493618Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Initialising Hodgkin-Huxley IKr benchmark\n",
      "Benchmarker initialised\n"
     ]
    }
   ],
   "source": [
    "hh = ionbench.problems.staircase.HH()\n",
    "# mm = ionbench.problems.staircase.MM()\n",
    "# ikr = ionbench.problems.loewe2016.IKr()\n",
    "# ikur = ionbench.problems.loewe2016.IKur()\n",
    "# ina = ionbench.problems.moreno2016.INa()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4fa051e4-604b-42aa-b8d0-ae9518fcdab3",
   "metadata": {},
   "source": [
    "### Sampling Parameters\n",
    "Now that we have a problem initialised, we can take a look at some of the functions we can use with it.\n",
    "The `.sample()` method will randomly sample a set of parameters in a region defined for each problem. This is the main method for generating parameters. The sample method for each function also has an optional input `n`. This allows you to generate more than one parameter vector, returning a list of parameter vectors."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "633311c5-8e84-47ed-baab-b9a2dfeae0f2",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-11-08T11:38:21.287772Z",
     "start_time": "2024-11-08T11:38:20.944563Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sampled parameters: \n",
      "[8.13441954e-06 1.98113696e-01 1.33855751e-02 1.87667087e-02\n",
      " 1.37266965e-07 3.28452991e-01 4.82009847e-07 1.02899418e-01\n",
      " 4.62973771e-02]\n",
      "Sampling more than one parameter vector:\n",
      "[array([1.39426294e-02, 5.93973056e-02, 3.85369056e-03, 8.06044927e-02,\n",
      "       8.97993685e-04, 1.07478345e-01, 1.93367474e-05, 1.72241154e-02,\n",
      "       9.54871908e-02]), array([2.89720290e-03, 7.76566118e-02, 1.02928373e-05, 1.26903856e-01,\n",
      "       2.14436968e-07, 8.83204046e-02, 2.40140082e-05, 1.45116473e-01,\n",
      "       8.47472849e-02]), array([1.49068960e-07, 3.15766496e-01, 1.46423537e-05, 8.08163190e-02,\n",
      "       3.63301500e-01, 6.82866942e-02, 8.90911918e-04, 9.54385950e-02,\n",
      "       4.06730375e-02])]\n"
     ]
    }
   ],
   "source": [
    "print('Sampled parameters: ')\n",
    "print(hh.sample())\n",
    "print('Sampling more than one parameter vector:')\n",
    "print(hh.sample(n=3))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f8456af1-834e-4ad6-a456-b9cc5ade437e",
   "metadata": {},
   "source": [
    "### Cost Functions\n",
    "Now that we have seen how to sample parameters, we can use the benchmarker to see how good sampled parameter are. \n",
    "Each benchmarker has a `.cost(param)` method which takes an input of a set of parameters and will simulate the model with those parameters, compare against the data stored in `hh.data` and calculate the RMSE."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "3c46a403-6a77-4059-aa27-e01877da6fc9",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-11-08T11:38:21.444056Z",
     "start_time": "2024-11-08T11:38:21.294582Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sampled parameters:\n",
      "[3.21500685e+00 3.78733201e-02 7.35437576e+01 1.05827992e-02\n",
      " 6.14812368e-06 2.94766813e-01 2.24701207e-01 4.84113872e-02\n",
      " 1.31454078e-01]\n",
      "Cost of sampled parameters:\n",
      "0.5522137111808383\n"
     ]
    }
   ],
   "source": [
    "param = hh.sample()\n",
    "print('Sampled parameters:')\n",
    "print(param)\n",
    "print('Cost of sampled parameters:')\n",
    "print(hh.cost(param))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fb700772-3d36-49bd-8d26-a4037912283b",
   "metadata": {},
   "source": [
    "Sometimes we don't want to know the RMSE cost and instead want a vector of residuals. There are methods in the benchmarker for exactly this, providing both residuals and squared residuals. These are sometimes required by certain optimisers such as scipy's least squares optimisers."
   ]
  },
  {
   "cell_type": "code",
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Residuals of sampled parameters:\n",
      "[-0.00267373 -0.01115909  0.01181726 ... -0.0303321   0.00667567\n",
      " -0.00461445]\n",
      "Squared residuals of sampled parameters:\n",
      "[7.14880838e-06 1.24525319e-04 1.39647747e-04 ... 9.20036052e-04\n",
      " 4.45646290e-05 2.12931168e-05]\n"
     ]
    }
   ],
   "source": [
    "print('Residuals of sampled parameters:')\n",
    "print(hh.signed_error(param))\n",
    "print('Squared residuals of sampled parameters:')\n",
    "print(hh.squared_error(param))"
   ],
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-11-08T11:38:21.549377Z",
     "start_time": "2024-11-08T11:38:21.448406Z"
    }
   },
   "id": "39c72c2e-fc91-47fd-904c-35f0075a64b3",
   "execution_count": 5
  },
  {
   "cell_type": "markdown",
   "source": [
    "We ran into a warning here. This is because we evaluated the cost of `param` in the previous cell and then evaluated the `signed_error(param)` in this cell. Since this required us to solve the model twice for the same parameters, we could have been more efficient by first finding `signed_error(param)` and then calculating the cost from its result. ***ionBench*** will track which parameters are used to solve the model and warn you if its thinks your optimisation is inefficient. Since we are just running a tutorial, we can ignore this warning, but it is something to be aware of when running optimisations."
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "a0a4bf37f260876"
  },
  {
   "cell_type": "markdown",
   "source": [
    "### Helper Functions\n",
    "There are also some other generally useful functions in the benchmarkers. Each benchmarker has a function which returns the number of parameter in its model, called `.n_parameters()`. Each benchmarker also stores a name attribute under `.NAME`. Try to keep the use of `.NAME` as limited as possible (as optimisers doing something problem-specific can lead to biases), but sometimes it may be necessary to include problem-specific features, for example, the modifications use `.NAME` to identify the bounds for the sampler."
   ],
   "metadata": {},
   "id": "bb813a49-6002-4dba-b285-7d256963ca03"
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "95458d6b-41e2-4c2e-9e53-94621678db32",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-11-08T11:38:21.566232Z",
     "start_time": "2024-11-08T11:38:21.552720Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of parameter in the model:\n",
      "9\n",
      "Name of the benchmarker:\n",
      "staircase.hh\n"
     ]
    }
   ],
   "source": [
    "print('Number of parameter in the model:')\n",
    "print(hh.n_parameters())\n",
    "print('Name of the benchmarker:')\n",
    "print(hh.NAME)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "64fac929-53fc-4fdd-9ec1-d28d8676e6a7",
   "metadata": {},
   "source": [
    "The names for the benchmarker problems are:\n",
    "* staircase.hh\n",
    "* staircase.mm\n",
    "* loewe2016.ikr\n",
    "* loewe2016.ikur\n",
    "* moreno2016.ina"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d88ef52e-3afc-41d7-b276-8a291977f055",
   "metadata": {},
   "source": [
    "### Gradient\n",
    "We also provide a quick way to calculate the gradient using ***myokit*** sensitivities. The `benchmarker.grad(param)` method takes an input of a parameter vector and returns the gradient of the cost function with respect to the parameters. It also has options to return the evaluated cost at those parameters, or can work with residuals/signed error to find the jacobian of the residual vector instead of the gradient of the cost function. It can also handle different parameter spaces and transforms, described in the next section. \n",
    "\n",
    "In order to use the gradient calculator, you should turn on sensitivities, either by compiling a benchmarker with sensitivities turned on `hh = ionbench.problems.staircase.HH(sensitivities=True)`, or by turning it on for an already compiled benchmarker using `benchmarker.use_sensitivities()`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "5a463ded-f226-40e6-a1b9-53a64556936a",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-11-08T11:38:33.771695Z",
     "start_time": "2024-11-08T11:38:21.572912Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Gradient of the cost function:\n",
      "[ 6.84368233e-02  5.19968127e+00 -2.99174780e-03  5.19970317e+00\n",
      " -1.15782339e+03 -1.53356991e-01  3.04480878e-02 -1.19305753e-01\n",
      "  1.88075833e+00]\n"
     ]
    }
   ],
   "source": [
    "print('Gradient of the cost function:')\n",
    "hh.use_sensitivities()\n",
    "print(hh.grad(param))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e37bb8e0-945d-4cf5-89fc-c7fdc8d7fbaa",
   "metadata": {},
   "source": [
    "## Transforms and Bounds\n",
    "It is possible to apply transforms or bounds in the benchmarkers. These will automatically update the benchmarkers methods (`.sample()` and `.grad(param)`, for example) to account for these transforms. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2973a461-6891-454e-af4d-64f349dc670a",
   "metadata": {},
   "source": [
    "### Log Transforms\n",
    "You can apply log transforms for specific parameters on a benchmarker object. The parameters which are log transformed are stored in the `._logTransformParams` variable in the benchmarker as a list of bools. This can be changed, and therefore log transform some parameters, by calling the `.log_transform()` method. By default, this method will log transform all parameters unless the optional input whichParams is set as a list of bools, of length `n_parameters()`, where `True` is a parameter to log transform, and `False` is a parameter to not log transform. \n",
    "\n",
    "Beware using log transforms for the Loewe 2016 problems. Since the sampler can return negative parameters you need to be careful with which parameters to log transform. Each benchmarker has a `.STANDARD_LOG_TRANSFORM` variable which stores typical parameters that could be log transformed, especially useful for the Loewe problems but should generally be used for all problems. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "51e6db44-1557-45f7-8587-32b6f6e8aa2c",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-11-08T11:38:33.880584Z",
     "start_time": "2024-11-08T11:38:33.773893Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sampled parameters:\n",
      "[1.99471155e+01 4.48372185e-02 2.46488662e-06 1.42155239e-01\n",
      " 2.39165097e+00 5.12524511e-02 3.92186280e-04 2.43352415e-02\n",
      " 1.65833442e-01]\n",
      "By default, no parameters are log transformed\n",
      "[False, False, False, False, False, False, False, False, False]\n",
      "Standard parameters to log transform for HH benchmarker:\n",
      "(True, False, True, False, True, False, True, False, False)\n",
      "The parameters now log transformed:\n",
      "(True, False, True, False, True, False, True, False, False)\n",
      "Sampled log transformed parameters:\n",
      "[ -8.90819971   0.23494448 -14.89825303   0.10011974  -4.65044416\n",
      "   0.18713122  -7.74939295   0.08859756   0.19729961]\n"
     ]
    }
   ],
   "source": [
    "print('Sampled parameters:')\n",
    "print(hh.sample())\n",
    "print('By default, no parameters are log transformed')\n",
    "print(hh.logTransformParams)\n",
    "print('Standard parameters to log transform for HH benchmarker:')\n",
    "print(hh.STANDARD_LOG_TRANSFORM)\n",
    "hh.log_transform(whichParams = hh.STANDARD_LOG_TRANSFORM)\n",
    "print('The parameters now log transformed:')\n",
    "print(hh.logTransformParams)\n",
    "print('Sampled log transformed parameters:')\n",
    "print(hh.sample())\n",
    "hh.log_transform(whichParams = [False]*hh.n_parameters()) #Revert log transforms"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7d966c6b-24cc-43f6-8f31-24f5618b81ee",
   "metadata": {},
   "source": [
    "### Scale Factors\n",
    "You can also apply scale factor transforms. This will set the default or true rates to be a vector of all ones and recognise inputted parameters as factors to these. \n",
    "\n",
    "Scale factor transformations can only be applied to all parameters at once."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "cf10c80b-0fd1-44d6-9271-ffc42558adff",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-11-08T11:38:33.976275Z",
     "start_time": "2024-11-08T11:38:33.882818Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sampled parameters:\n",
      "[6.60082684e-05 4.25691297e-02 2.01880579e-07 1.11247146e-01\n",
      " 1.36189282e-04 1.38184056e-01 2.43060349e-05 1.04049622e-01\n",
      " 1.00920248e-01]\n",
      "Sampled parameters when applying scale factors:\n",
      "[1.39880440e-05 2.10911106e-01 2.62998619e-06 3.91251013e-02\n",
      " 3.37918786e-01 9.83002688e-02 9.94660298e-02 7.66574577e-02\n",
      " 2.11971027e-02]\n"
     ]
    }
   ],
   "source": [
    "print('Sampled parameters:')\n",
    "print(hh.sample())\n",
    "print('Sampled parameters when applying scale factors:')\n",
    "hh._useScaleFactors = True\n",
    "print(hh.sample())\n",
    "hh._useScaleFactors = False #Revert scale factor transforms"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e27f27eb-3387-462c-bbb2-c9c3dc4b07fa",
   "metadata": {},
   "source": [
    "### Input and Original Parameter Spaces\n",
    "When using transforms, it may be useful to map between different parameter spaces. We define the input space as including any transformations applied to the benchmarker, while the original parameter space is the one used in model simulations. Each benchmarker has two helper functions which map between these parameter spaces. These are used frequently in the inner-workings of **ionBench**.\n",
    "\n",
    "Generally, optimisers will work exclusively in input parameter space. The cost functions should only be evaluated in input spaced parameters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "7cc2c21e-7a6e-4eed-8f97-ef4b872816a6",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-11-08T11:38:34.265569Z",
     "start_time": "2024-11-08T11:38:33.989980Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sampled parameters are given in input parameter space:\n",
      "[8.51462449e-03 1.33710475e-01 4.68870587e-07 6.14672569e-02\n",
      " 6.90207254e+00 3.99149207e-02 7.11672659e+00 3.37224126e-02\n",
      " 9.53630436e-02]\n",
      "When mapped to the original parameter space:\n",
      "[8.51462449e-03 1.33710475e-01 4.68870587e-07 6.14672569e-02\n",
      " 6.90207254e+00 3.99149207e-02 7.11672659e+00 3.37224126e-02\n",
      " 9.53630436e-02]\n",
      "Input space of these parameters will return the original sampled parameters:\n",
      "[8.51462449e-03 1.33710475e-01 4.68870587e-07 6.14672569e-02\n",
      " 6.90207254e+00 3.99149207e-02 7.11672659e+00 3.37224126e-02\n",
      " 9.53630436e-02]\n"
     ]
    }
   ],
   "source": [
    "hh._useScaleFactors = True #Apply a transform\n",
    "print('Sampled parameters are given in input parameter space:')\n",
    "param = hh.sample()\n",
    "print(param)\n",
    "print('When mapped to the original parameter space:')\n",
    "output_space = hh.original_parameter_space(param)\n",
    "print(output_space)\n",
    "print('Input space of these parameters will return the original sampled parameters:')\n",
    "print(hh.input_parameter_space(output_space))\n",
    "hh._useScaleFactors = False # Revert scale factor transforms"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "10663027-0552-4880-ac85-51f9802eb73b",
   "metadata": {},
   "source": [
    "### Parameter Bounds\n",
    "Parameter bounds can be applied by using the `.add_parameter_bounds()` method. If the cost function (or its alternatives) are called with parameters that are outside of these bounds, the penalty function will be applied (guaranteeing a cost of over 100,000). \n",
    "\n",
    "To determine whether bounds are active (or to disable them), you can read (or set to `False`) the attribute `bm._parameters_bounded` in the benchmarker problem object."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "05e29294-6feb-4374-a11f-5600471e0deb",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-11-08T11:38:34.317847Z",
     "start_time": "2024-11-08T11:38:34.267784Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "By default the HH benchmarker is bounded:\n",
      "True\n",
      "When a parameter is out of bounds, the cost will use the penalty function\n",
      "170309.75114131137\n",
      "Bounds can be disabled by setting the parametersBounded attribute\n",
      "Now the cost will be calculated with the negative parameter\n",
      "1.5283532384561032\n"
     ]
    }
   ],
   "source": [
    "print('By default the HH benchmarker is bounded:')\n",
    "print(hh.parametersBounded)\n",
    "param = hh.sample()\n",
    "param[-1] = -1 #Parameter out of bounds \n",
    "print('When a parameter is out of bounds, the cost will use the penalty function')\n",
    "print(hh.cost(param))\n",
    "print('Bounds can be disabled by setting the parametersBounded attribute')\n",
    "hh.parametersBounded = False\n",
    "print('Now the cost will be calculated with the negative parameter')\n",
    "print(hh.cost(param))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fa39b43c-7362-4ccc-af67-50a77360f896",
   "metadata": {},
   "source": [
    "## Modifications\n",
    "Sometimes you might want to apply transforms to multiple different benchmarkers. In that case, it may be best to generate a modification. A modification is an object which stores settings for bounds, log transforms and scale factor transforms. \n",
    "\n",
    "Each modification stores three strings from when it is generated. These strings are the settings for log transforms, bounds and scale factors. \n",
    "\n",
    "The available settings for log transforms are `'off'`, `'on'`, or `'Custom'`.\n",
    "\n",
    "The available settings for parameter bounds are `'off'`, `'on'`, or `'Custom'`.\n",
    "\n",
    "The available settings for rate bounds are `'off'`, or `'on'`.\n",
    "\n",
    "The available settings for scale factors are `'off'` or `'on'`.\n",
    "\n",
    "Once the modification has been constructed and the settings have been specified, the `.apply()` method can be used to apply these settings to a benchmarker."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "323228bc-c4c5-466c-8eef-29f1cf0777af",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-11-08T11:38:34.331781Z",
     "start_time": "2024-11-08T11:38:34.320123Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The benchmarker is not bounded before\n",
      "False\n",
      "The benchmarker is bounded after applying the modification\n",
      "True\n",
      "The bounds are different for each problem\n",
      "Lower bounds:\n",
      "[1.e-07 1.e-07 1.e-07 1.e-07 1.e-07 1.e-07 1.e-07 1.e-07 2.e-02]\n",
      "Upper bounds:\n",
      "[1.e+03 4.e-01 1.e+03 4.e-01 1.e+03 4.e-01 1.e+03 4.e-01 2.e-01]\n"
     ]
    }
   ],
   "source": [
    "mod = ionbench.modification.Modification(parameterBounds = 'on') #Generate a modification with parameter bounds, no rate bounds and default transforms (no log and scale factor off)\n",
    "print('The benchmarker is not bounded before')\n",
    "print(hh.parametersBounded)\n",
    "mod.apply(hh)\n",
    "print('The benchmarker is bounded after applying the modification')\n",
    "print(hh.parametersBounded)\n",
    "print('The bounds are different for each problem')\n",
    "print('Lower bounds:')\n",
    "print(hh.lb)\n",
    "print('Upper bounds:')\n",
    "print(hh.ub)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e9f21ac1-9778-420d-94eb-946b107d6369",
   "metadata": {},
   "source": [
    "There are also a range of predefined modifications from different papers."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "f5555646-6c90-4ffc-9d8c-622863a3253d",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-11-08T11:38:34.347582Z",
     "start_time": "2024-11-08T11:38:34.336234Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The settings for the Clerx 2019 modification are:\n",
      "{'log transform': 'on', 'parameterBounds': 'on', 'rateBounds': 'on', 'scale factors': 'off'}\n"
     ]
    }
   ],
   "source": [
    "mod = ionbench.modification.Clerx2019()\n",
    "print('The settings for the Clerx 2019 modification are:')\n",
    "print(mod.dict)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "69caf9ad-bf6f-4ff9-ae3e-12f9dd8fb429",
   "metadata": {},
   "source": [
    "## Optimisers\n",
    "There are a variety of optimisers included in **ionBench** and if you are using it, it's likely that you would like to test your own optimiser. If you do wish to test your own optimiser, it is important to understand how the optimisers are formatted in **ionBench**. \n",
    "\n",
    "The main thing to note is that each optimiser has a run function which has only one required input, a benchmarker object. The optimiser will then use this benchmarker object to find its bounds (if your optimiser has addition features for bounds such as ensuring gradient calculations are inside parameter bounds), use its `.sample()` method to get parameter values and use its `.cost()` method to check the performance of those parameters. Other inputs can be included, it's recommended to include an optional initial point argument named `x0`, but all non-benchmarker inputs should be optional (If no initial point is specified, the optimiser should use a sampled point).\n",
    "\n",
    "Another detail, although not one you need to account for to measure the performance of your own optimisers, is that each optimiser has an associated modification class that can be retrieved using the `get_modification()` function.\n",
    "\n",
    "Below we retrieve the default modification for the Trust Region Reflective scipy optimiser, apply it to a benchmarker, and run the Trust Region Reflective scipy optimiser on the benchmarker.\n",
    "\n",
    "We specify the optional parameter `maxIter` here to ensure the optimisation doesn't run for too long."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "abaa26e5-2723-4a70-9cfd-8538b5616344",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-11-08T11:38:38.809306Z",
     "start_time": "2024-11-08T11:38:34.351776Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "`xtol` termination condition is satisfied.\n",
      "Function evaluations 24, initial cost 2.8727e+05, final cost 4.9544e+04, first-order optimality 1.34e+12.\n",
      "\n",
      "========================================\n",
      "===      Evaluating Performance      ===\n",
      "========================================\n",
      "\n",
      "Number of cost evaluations:      8\n",
      "Number of grad evaluations:      8\n",
      "Best cost:                       1.793635\n",
      "Convergence reason:              Unknown optimiser specific termination.\n",
      "Cost evaluations at convergence: 8\n",
      "Grad evaluations at convergence: 8\n",
      "Best cost at convergence:        1.793635\n",
      "Model solve time at convergence: 0.173000\n",
      "Grad solve time at convergence:  1.562000\n"
     ]
    },
    {
     "data": {
      "text/plain": "<Figure size 640x480 with 1 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "<Figure size 640x480 with 1 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "<Figure size 640x480 with 1 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "<Figure size 640x480 with 1 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "<Figure size 640x480 with 1 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitted parameters are:\n",
      "[6.58141203e-01 5.89344202e-03 4.81952338e-01 3.02110791e-05\n",
      " 5.60329461e-06 3.16663413e-01 2.94534351e-06 1.55420266e-01\n",
      " 1.09886338e-01]\n"
     ]
    }
   ],
   "source": [
    "mod = ionbench.optimisers.scipy_optimisers.trustRegionReflective_scipy.get_modification()\n",
    "mod.apply(hh)\n",
    "hh.reset() #Don't worry about this yet\n",
    "out = ionbench.optimisers.scipy_optimisers.trustRegionReflective_scipy.run(hh, maxIter=100)\n",
    "print('Fitted parameters are:')\n",
    "print(out)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "460dc5bf-421d-4262-8db4-b2eb8440bd26",
   "metadata": {},
   "source": [
    "Once the optimisation is complete, the optimiser calls `hh.evaluate()`. This function prints the performance metrics for the best parameters as well as the performance metrics when the optimiser converged. It then plots the performance metrics over time and the model output for the first parameters, the best parameters, and the data.\n",
    "\n",
    "These plots can be disabled by setting `hh.plotter = False`."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "76f862f8-feab-4834-9086-5031176c040a",
   "metadata": {},
   "source": [
    "## Tracker\n",
    "Each benchmarker stores a __Tracker__ object which calculates and stores the performance metrics over time as well as figuring out when the optimiser converged. You shouldn't normally need to interact with this __Tracker__ object, but you may wish to save the information from a __Tracker__ and store it to analyse later, for example if you are running an optimiser on an HPC and can't view the figures. You can use the __Tracker__'s `.save()` and `.load()` methods to do exactly that. \n",
    "\n",
    "You may have noticed the line `hh.reset()` in the previous code cell. This resets the __Tracker__ object, since otherwise all the times we called the cost function would have been included in the plots."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "48b12b47-0d38-4c72-95b1-9d592ac57a5d",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-11-08T11:38:38.907080Z",
     "start_time": "2024-11-08T11:38:38.818906Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of times the model has been solved: \n",
      "[1, 1, 2, 2, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 4, 4, 4, 5, 5, 5, 6, 6, 6, 7, 7, 7, 7, 7, 8, 8]\n",
      "Number of model solves after reset:\n",
      "[]\n",
      "Number of model solves after load:\n",
      "[1, 1, 2, 2, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 4, 4, 4, 5, 5, 5, 6, 6, 6, 7, 7, 7, 7, 7, 8, 8]\n"
     ]
    }
   ],
   "source": [
    "filename = 'temporary_tracker_file.pickle'\n",
    "print('Number of times the model has been solved: ')\n",
    "print(hh.tracker.costSolves)\n",
    "hh.tracker.save(filename)\n",
    "print('Number of model solves after reset:')\n",
    "hh.reset()\n",
    "print(hh.tracker.costSolves)\n",
    "print('Number of model solves after load:')\n",
    "hh.tracker.load(filename)\n",
    "print(hh.tracker.costSolves)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "b727bded-7afc-407e-b7e1-e8bf6aa3af47",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-11-08T11:38:38.921050Z",
     "start_time": "2024-11-08T11:38:38.912181Z"
    }
   },
   "outputs": [],
   "source": [
    "#Remove temporary file\n",
    "os.remove(filename)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "77f1f210-dd90-4f75-b71b-52c3736dd8b4",
   "metadata": {},
   "source": [
    "## Multistart\n",
    "It might be useful to start an optimisation multiple times. This can be done with the multistart function. It takes inputs of an optimisers `.run` function, a benchmarker, a list of initial parameters, and a filename to use for saving the tracker information.\n",
    "\n",
    "Additional keyword arguments, such as hyperparameters to pass into the optimiser can also be passed into the multistart function. \n",
    "\n",
    "The filename will be automatically appended with `'_run{i}.pickle'` where `{i}` is the index for which of the initial parameters was used for the optimisation.\n",
    "\n",
    "Multistart will save the tracker data and printing the evaluate output (which is done whenever an optimiser is called).\n",
    "\n",
    "We also include the `maxIters` keyword argument to ensure the optimisation doesn't run for too long. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "ae3f5b72-e65e-4583-82d7-a345e339dd6d",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-11-08T11:39:00.965998Z",
     "start_time": "2024-11-08T11:38:38.926664Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "`ftol` termination condition is satisfied.\n",
      "Function evaluations 49, initial cost 1.6263e+05, final cost 2.6809e+03, first-order optimality 2.62e+05.\n",
      "\n",
      "========================================\n",
      "===      Evaluating Performance      ===\n",
      "========================================\n",
      "\n",
      "Number of cost evaluations:      25\n",
      "Number of grad evaluations:      25\n",
      "Best cost:                       0.417237\n",
      "Convergence reason:              Unknown optimiser specific termination.\n",
      "Cost evaluations at convergence: 25\n",
      "Grad evaluations at convergence: 25\n",
      "Best cost at convergence:        0.417237\n",
      "Model solve time at convergence: 0.875000\n",
      "Grad solve time at convergence:  6.861000\n"
     ]
    },
    {
     "data": {
      "text/plain": "<Figure size 640x480 with 1 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "<Figure size 640x480 with 1 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "<Figure size 640x480 with 1 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "<Figure size 640x480 with 1 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "<Figure size 640x480 with 1 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[2.58465989 0.09927133 0.01575325 0.04228588 0.43244112 0.12910107\n",
      " 0.01236179 0.00603362 0.03684349]\n",
      "`xtol` termination condition is satisfied.\n",
      "Function evaluations 33, initial cost 4.5843e+03, final cost 3.3961e+03, first-order optimality 1.30e+07.\n",
      "\n",
      "========================================\n",
      "===      Evaluating Performance      ===\n",
      "========================================\n",
      "\n",
      "Number of cost evaluations:      29\n",
      "Number of grad evaluations:      15\n",
      "Best cost:                       0.469603\n",
      "Convergence reason:              Unknown optimiser specific termination.\n",
      "Cost evaluations at convergence: 29\n",
      "Grad evaluations at convergence: 15\n",
      "Best cost at convergence:        0.469603\n",
      "Model solve time at convergence: 0.891000\n",
      "Grad solve time at convergence:  3.672000\n"
     ]
    },
    {
     "data": {
      "text/plain": "<Figure size 640x480 with 1 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "<Figure size 640x480 with 1 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "<Figure size 640x480 with 1 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "<Figure size 640x480 with 1 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "<Figure size 640x480 with 1 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[2.38431688e-01 1.35998296e-01 1.77806970e-06 6.46100843e-02\n",
      " 1.39425262e-03 1.72403017e-01 1.01495317e-04 8.02443211e-05\n",
      " 3.22083230e-02]\n"
     ]
    },
    {
     "data": {
      "text/plain": "[array([2.58465989, 0.09927133, 0.01575325, 0.04228588, 0.43244112,\n        0.12910107, 0.01236179, 0.00603362, 0.03684349]),\n array([2.38431688e-01, 1.35998296e-01, 1.77806970e-06, 6.46100843e-02,\n        1.39425262e-03, 1.72403017e-01, 1.01495317e-04, 8.02443211e-05,\n        3.22083230e-02])]"
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "opt = ionbench.optimisers.scipy_optimisers.trustRegionReflective_scipy.run\n",
    "hh.reset()\n",
    "initParams = hh.sample(2)\n",
    "filename = 'temporary_tracker_files'\n",
    "ionbench.utils.multistart(opt, hh, initParams, filename, maxIter=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "1dab5209-e02d-4905-9925-9246f858baf7",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-11-08T11:39:00.990534Z",
     "start_time": "2024-11-08T11:39:00.971034Z"
    }
   },
   "outputs": [],
   "source": [
    "os.remove(filename+'_run0.pickle')\n",
    "os.remove(filename+'_run1.pickle')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (Spyder)",
   "language": "python3",
   "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": "3.11.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}