{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# How to tune the parameters of an optimization\n",
    "**Defining parameters of the optimization using the cost history and early halt conditions**\n",
    "\n",
    "Boulder Opal provides a highly flexible [optimization engine](https://docs.q-ctrl.com/boulder-opal/toolkit/discover/adopt/compare-control-design-optimization-strategies-in-boulder-opal) for general-purpose optimization.\n",
    "It can be directly applied to model-based control optimization and for model-based system identification, where the model of the system is represented by an acyclic graph.\n",
    "You can adjust the parameters in the graph representation to customize many aspects of the optimization, such as the maximum value of the optimizable pulses, and the number of segments used in the optimization.\n",
    "\n",
    "When tuning the parameters of an optimization, you can use the additional information provided by the history of costs obtained during the course of the optimization to see which choices of these parameters are the best for your purposes.\n",
    "By halting the optimization before it runs for too long and extracting the histories, you can determine which choices of parameters have values of costs that are still improving in time, and which ones stagnate at a high value.\n",
    "\n",
    "This user guide will show how to use the cost histories returned by the optimization engine to select the number of segments for a non-stochastic optimization of a large system.\n",
    "For general instructions about how to perform optimization of a large system, see the user guide [How to optimize controls on large sparse Hamiltonians](https://docs.q-ctrl.com/boulder-opal/toolkit/design/design-model-based-controls/how-to-optimize-controls-on-large-sparse-hamiltonians).\n",
    "For an example of how to use the cost history to tune parameters of a stochastic optimization, see the user guide [How to tune the learning rate of a stochastic optimization](https://docs.q-ctrl.com/boulder-opal/toolkit/design/design-model-based-controls/tune-optimization-hyperparameters/how-to-tune-the-learning-rate-of-a-stochastic-optimization)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Summary workflow\n",
    "### 1. Define a function to create the computational graphs\n",
    "\n",
    "The Boulder Opal optimization engine expresses all optimization problems as [data flow graphs](https://docs.q-ctrl.com/boulder-opal/toolkit/design/calculate-with-graphs/get-an-introduction-to-graphs-in-boulder-opal), which you can create with `boulderopal.Graph`.\n",
    "The methods of the graph object allow you to represent the mathematical structure of the problem that you want to solve.\n",
    "You can define a function that creates a graph that represents your system, and which takes the parameter that you want to vary as an argument.\n",
    "\n",
    "### 2. Execute optimization and retrieve cost history\n",
    "\n",
    "You can calculate an optimization from an input graph using the functions `boulderopal.run_optimization`, `boulderopal.run_stochastic_optimization`, and `boulderopal.run_gradient_free_optimization`.\n",
    "Both functions allow you to retrieve the history of all the costs for each step of the optimization, or the history of the best value of the costs up to each iteration.\n",
    "You can request the history of costs with the `cost_history_scope` argument of these functions.\n",
    "\n",
    "If you are interested in just testing an optimization that might take a while to finish, you can limit the number of iterations with the parameter `max_iteration_count` (for `boulderopal.run_optimization`) and `iteration_count` (for `boulderopal.run_stochastic_optimization` and `boulderopal.run_gradient_free_optimization`).\n",
    "This will allow you to retrieve the beginning of the cost history, without spending too much time in the optimization.\n",
    "\n",
    "### 3. Compare the cost histories for different choices of parameters\n",
    "\n",
    "You can retrieve the list of costs by extracting the values from the `\"cost_history\"` element in the dictionary returned by the optimization.\n",
    "By plotting these results, you can then verify which choices of parameters resulted in costs that stagnated, and which ones kept improving.\n",
    "You can then keep refining the choices, and requesting more iterations for the parameter values that did not stagnate."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Example: Tuning the number of segments in an optimization with six Rydberg atoms\n",
    "\n",
    "Consider a chain of 6 atoms, each one modeled as a qubit, with $|0\\rangle$ representing the ground state and $|1\\rangle$ representing a Rydberg state.\n",
    "The total Hamiltonian of the system, is given by\n",
    "\n",
    "$$ H = \\frac{\\Omega(t)}{2} \\sum_{i=1}^6 \\sigma_x^{(i)} - \\sum_{i=1}^6 (\\Delta(t) +\\delta_i) n_i + \\sum_{i<j}^6 \\frac{V}{\\vert i-j \\vert^6}n_i n_j , $$\n",
    "\n",
    "with $\\sigma_x^{(i)} = \\vert 0 \\rangle \\langle 1 \\vert_i + \\vert 1 \\rangle \\langle 0 \\vert_i$, and $n_i = \\vert 1\\rangle\\langle 1 \\vert_i$.\n",
    "The fixed system parameters are the interaction strength between excited atoms, $V$, and the local energy shifts, $\\{ \\delta_i \\}$.\n",
    "The control parameters are the effective coupling strength to the Rydberg state and the detuning, $\\Omega(t)$ and $\\Delta(t)$ respectively.\n",
    "\n",
    "Suppose that you want to use the optimizer to find the best pulses to obtain a GHZ state from an initial ground state, but you don't know which choice for the number of segments will yield the best result.\n",
    "In this case, you can run the optimization for different values of the number of segments, and retrieve the history of the best cost for each of them.\n",
    "By plotting these best costs as a function of the iteration, you can then see which choices of the parameters stagnated at a certain value of the cost, and which ones kept improving.\n",
    "For those that kept improving without stagnating, you can increase the iteration count to obtain better results."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from time import time\n",
    "import numpy as np\n",
    "import scipy\n",
    "import matplotlib.pyplot as plt\n",
    "import qctrlvisualizer as qv\n",
    "import boulderopal as bo\n",
    "\n",
    "plt.style.use(qv.get_qctrl_style())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define optimization parameters.\n",
    "duration = 1.1e-6  # s\n",
    "omega_max = 5.0e6 * (2 * np.pi)  # Hz\n",
    "delta_range = 15.0e6 * (2 * np.pi)  # Hz\n",
    "\n",
    "# Define physical parameters of the quantum system.\n",
    "interaction_strength = 24.0e6 * (2 * np.pi)  # Hz\n",
    "local_shifts = np.zeros((6,))\n",
    "local_shifts[[0, -1]] = -4.5e6 * (2 * np.pi)  # Hz"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Build the terms of the Hamiltonian.\n",
    "h_omega_values = [0.5] * 2**6 * 6\n",
    "h_omega_indices = []\n",
    "h_delta_values = []\n",
    "h_delta_indices = []\n",
    "h_fixed_values = []\n",
    "h_fixed_indices = []\n",
    "\n",
    "for j in range(2**6):\n",
    "    state = np.array([int(m) for m in np.binary_repr(j, width=6)])\n",
    "\n",
    "    # Global shift term (big delta).\n",
    "    n_exc = np.sum(state)  # Number of excited atoms.\n",
    "    h_delta_values.append(-n_exc)\n",
    "    h_delta_indices.append([j, j])\n",
    "\n",
    "    # Omega term.\n",
    "    for i in range(6):\n",
    "        # Couple \"state\" and \"state with qubit i flipped\".\n",
    "        h_omega_indices.append([j, j ^ 2 ** (5 - i)])\n",
    "\n",
    "    # Local shift term (small deltas).\n",
    "    h_fixed = -np.sum(local_shifts * state)\n",
    "\n",
    "    # Interaction term.\n",
    "    for d in range(1, 6):\n",
    "        # Number of excited atoms at distance d.\n",
    "        n_d = np.dot(((state[:-d] - state[d:]) == 0), state[:-d])\n",
    "        h_fixed += interaction_strength * n_d / d**6\n",
    "\n",
    "    if h_fixed != 0.0:\n",
    "        h_fixed_values.append(h_fixed)\n",
    "        h_fixed_indices.append([j, j])\n",
    "\n",
    "H_omega = scipy.sparse.coo_matrix(\n",
    "    (h_omega_values, np.array(h_omega_indices).T), shape=[2**6, 2**6]\n",
    ")\n",
    "H_delta = scipy.sparse.coo_matrix(\n",
    "    (h_delta_values, np.array(h_delta_indices).T), shape=[2**6, 2**6]\n",
    ")\n",
    "H_fixed = scipy.sparse.coo_matrix(\n",
    "    (h_fixed_values, np.array(h_fixed_indices).T), shape=[2**6, 2**6]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_recommended_k(segment_count, H_omega, H_delta, H_fixed):\n",
    "    \"\"\"\n",
    "    Calculate the appropriate Krylov subspace dimension (k)\n",
    "    for a given set of parameters, as required for the integration.\n",
    "    This is obtained for the Hamiltonian with the largest spectral range.\n",
    "    \"\"\"\n",
    "    graph = bo.Graph()\n",
    "\n",
    "    maximal_range_hamiltonian = (\n",
    "        omega_max * H_omega - delta_range * H_delta + H_fixed\n",
    "    ).tocoo()\n",
    "\n",
    "    # Calculate spectral range.\n",
    "    spectral_range = graph.spectral_range(operator=maximal_range_hamiltonian)\n",
    "\n",
    "    # Calculate the appropriate Krylov subspace dimension k.\n",
    "    graph.estimated_krylov_subspace_dimension_lanczos(\n",
    "        spectral_range=spectral_range,\n",
    "        error_tolerance=1e-6,\n",
    "        duration=duration,\n",
    "        maximum_segment_duration=duration / segment_count,\n",
    "        name=\"k\",\n",
    "    )\n",
    "\n",
    "    result = bo.execute_graph(graph=graph, output_node_names=\"k\")\n",
    "\n",
    "    return result[\"output\"][\"k\"][\"value\"]\n",
    "\n",
    "\n",
    "def get_optimization_graph(segment_count, krylov_subspace_dimension):\n",
    "    \"\"\"\n",
    "    Create an optimization graph for the requested number of segments.\n",
    "    \"\"\"\n",
    "    graph = bo.Graph()\n",
    "\n",
    "    # Define pulses.\n",
    "    omega_segments = graph.real_optimizable_pwc_signal(\n",
    "        segment_count=segment_count,\n",
    "        minimum=0.0,\n",
    "        maximum=omega_max,\n",
    "        duration=duration,\n",
    "        name=\"omega\",\n",
    "    )\n",
    "    delta_segments = graph.real_optimizable_pwc_signal(\n",
    "        segment_count=segment_count,\n",
    "        minimum=-delta_range,\n",
    "        maximum=delta_range,\n",
    "        duration=duration,\n",
    "        name=\"delta\",\n",
    "    )\n",
    "    fixed_segment = graph.pwc_signal(\n",
    "        values=np.array([1]), duration=duration, name=\"fixed\"\n",
    "    )\n",
    "\n",
    "    shift_delta = graph.sparse_pwc_operator(signal=delta_segments, operator=H_delta)\n",
    "    shift_omega = graph.sparse_pwc_operator(signal=omega_segments, operator=H_omega)\n",
    "    shift_fixed = graph.sparse_pwc_operator(signal=fixed_segment, operator=H_fixed)\n",
    "\n",
    "    # Define initial state\n",
    "    initial_state = graph.fock_state(2**6, 0)\n",
    "\n",
    "    # Create target state.\n",
    "    idx_even = sum(2**n for n in range(0, 6, 2))\n",
    "    idx_odd = sum(2**n for n in range(1, 6, 2))\n",
    "    target_state = np.zeros([2**6])\n",
    "    target_state[[idx_even, idx_odd]] = 1.0 / np.sqrt(2.0)\n",
    "\n",
    "    # Evolve the initial state.\n",
    "    evolved_state = graph.state_evolution_pwc(\n",
    "        initial_state=initial_state,\n",
    "        hamiltonian=graph.sparse_pwc_sum([shift_fixed, shift_omega, shift_delta]),\n",
    "        krylov_subspace_dimension=krylov_subspace_dimension,\n",
    "    )\n",
    "\n",
    "    # Calculate state infidelity.\n",
    "    infidelity = graph.state_infidelity(target_state, evolved_state, name=\"infidelity\")\n",
    "\n",
    "    return graph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Your task (action_id=\"1829167\") is queued.\n",
      "Your task (action_id=\"1829167\") has completed.\n",
      "Your task (action_id=\"1829168\") has started.\n",
      "Your task (action_id=\"1829168\") has completed.\n",
      "Your task (action_id=\"1829169\") has started.\n",
      "Your task (action_id=\"1829169\") has completed.\n",
      "Your task (action_id=\"1829170\") has started.\n",
      "Your task (action_id=\"1829170\") has completed.\n",
      "Your task (action_id=\"1829171\") has started.\n",
      "Your task (action_id=\"1829171\") has completed.\n",
      "Your task (action_id=\"1829172\") has started.\n",
      "Your task (action_id=\"1829172\") has completed.\n",
      "Your task (action_id=\"1829173\") has started.\n",
      "Your task (action_id=\"1829173\") has completed.\n",
      "Your task (action_id=\"1829174\") has started.\n",
      "Your task (action_id=\"1829174\") has completed.\n",
      "Your task (action_id=\"1829177\") has started.\n",
      "Your task (action_id=\"1829177\") has completed.\n",
      "Your task (action_id=\"1829178\") has started.\n",
      "Your task (action_id=\"1829178\") has completed.\n",
      "Your task (action_id=\"1829179\") has started.\n",
      "Your task (action_id=\"1829179\") has completed.\n",
      "Your task (action_id=\"1829180\") has started.\n",
      "Your task (action_id=\"1829180\") has completed.\n"
     ]
    }
   ],
   "source": [
    "results = {}\n",
    "segment_counts = [2**n for n in range(4, 10)]\n",
    "calculation_times = []\n",
    "\n",
    "# Run optimizations.\n",
    "for segment_count in segment_counts:\n",
    "    krylov_subspace_dimension = get_recommended_k(\n",
    "        segment_count, H_omega, H_delta, H_fixed\n",
    "    )\n",
    "    graph = get_optimization_graph(segment_count, krylov_subspace_dimension)\n",
    "    time_reference = time()\n",
    "    results[segment_count] = bo.run_optimization(\n",
    "        graph=graph,\n",
    "        cost_node_name=\"infidelity\",\n",
    "        output_node_names=[\"infidelity\"],\n",
    "        optimization_count=1,\n",
    "        max_iteration_count=300,\n",
    "        cost_history_scope=\"HISTORICAL_BEST\",\n",
    "    )\n",
    "    calculation_times.append(time() - time_reference)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot history of best costs.\n",
    "qv.plot_cost_histories(\n",
    "    [\n",
    "        result[\"cost_history\"][\"historical_best\"][0]\n",
    "        for segment_count, result in results.items()\n",
    "    ],\n",
    "    labels=[str(n) for n in segment_counts],\n",
    "    y_axis_log=True,\n",
    ")\n",
    "plt.suptitle(\"Best cost values histories for different numbers of segments\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1080x360 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot final costs and calculation times.\n",
    "\n",
    "costs = [result[\"cost\"] for result in results.values()]\n",
    "\n",
    "fig, axes = plt.subplots(1, 2, figsize=(15, 5))\n",
    "labels = [str(s) for s in segment_counts]\n",
    "axes[0].bar(labels, costs, log=True)\n",
    "axes[1].bar(labels, calculation_times)\n",
    "\n",
    "axes[0].set_title(\"Best cost achieved\")\n",
    "axes[0].set_xlabel(\"Segment count\")\n",
    "axes[0].set_ylabel(\"Final cost\")\n",
    "\n",
    "axes[1].set_title(\"Calculation time\")\n",
    "axes[1].set_xlabel(\"Segment count\")\n",
    "axes[1].set_ylabel(\"Runtime (s)\")\n",
    "\n",
    "plt.show()"
   ]
  }
 ],
 "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": "3.11.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}