jupyter_notebooks/Examples/ModelMemberGraph.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# ModelMemberGraph and Serialization\n",
    "\n",
    "Example notebook of ModelMemberGraph functionality"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "\n",
    "import pygsti"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pygsti.modelpacks import smq2Q_XYICNOT"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Similar/Equivalent"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ex_mdl1 = smq2Q_XYICNOT.target_model()\n",
    "ex_mdl2 = ex_mdl1.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ex_mmg1 = ex_mdl1.create_modelmember_graph()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ex_mmg1.print_graph()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ex_mmg1.mm_nodes['operations']['Gxpi2', 0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ex_mmg2 = ex_mdl2.create_modelmember_graph()\n",
    "print(ex_mmg1.is_similar(ex_mmg2))\n",
    "print(ex_mmg1.is_equivalent(ex_mmg2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ex_mdl2.operations['Gxpi2', 0][0, 0] = 0.0\n",
    "ex_mmg2 = ex_mdl2.create_modelmember_graph()\n",
    "print(ex_mmg1.is_similar(ex_mmg2))\n",
    "print(ex_mmg1.is_equivalent(ex_mmg2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ex_mdl2.operations['Gxpi2', 0] = pygsti.modelmembers.operations.StaticArbitraryOp(ex_mdl2.operations['Gxpi2', 0])\n",
    "ex_mmg2 = ex_mdl2.create_modelmember_graph()\n",
    "print(ex_mmg1.is_similar(ex_mmg2))\n",
    "print(ex_mmg1.is_equivalent(ex_mmg2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pspec = pygsti.processors.QubitProcessorSpec(2, ['Gi', 'Gxpi2', 'Gypi2', 'mygate'], geometry='line', nonstd_gate_unitaries={'mygate': np.eye(2, dtype='complex')})\n",
    "\n",
    "ln_mdl1 = pygsti.models.create_crosstalk_free_model(pspec,\n",
    "                                                    depolarization_strengths={('Gxpi2', 0): 0.1, ('mygate', 0): 0.2},\n",
    "                                                    lindblad_error_coeffs={('Gypi2', 1): {('H', 'X'): 0.2, ('S', 'Y'): 0.3}})\n",
    "print(ln_mdl1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ln_mmg1 = ln_mdl1.create_modelmember_graph()\n",
    "ln_mmg1.print_graph()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Should be exactly the same\n",
    "ln_mdl2 = pygsti.models.create_crosstalk_free_model(pspec,\n",
    "                                                    depolarization_strengths={('Gxpi2', 0): 0.1, ('mygate', 0): 0.2},\n",
    "                                                    lindblad_error_coeffs={('Gypi2', 1): {('H', 'X'): 0.2, ('S', 'Y'): 0.3}})\n",
    "ln_mmg2 = ln_mdl2.create_modelmember_graph()\n",
    "print(ln_mmg1.is_similar(ln_mmg2))\n",
    "print(ln_mmg1.is_equivalent(ln_mmg2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Should be similar if we change params\n",
    "ln_mdl3 = pygsti.models.create_crosstalk_free_model(pspec,\n",
    "                                                    depolarization_strengths={('Gxpi2', 0): 0.01, ('mygate', 0): 0.02},\n",
    "                                                    lindblad_error_coeffs={('Gypi2', 1): {('H', 'X'): 0.5, ('S', 'Y'): 0.1}})\n",
    "ln_mmg3 = ln_mdl3.create_modelmember_graph()\n",
    "print(ln_mmg1.is_similar(ln_mmg3))\n",
    "print(ln_mmg1.is_equivalent(ln_mmg3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Should fail both, depolarize is on different gate\n",
    "ln_mdl4 = pygsti.models.create_crosstalk_free_model(pspec,\n",
    "                                                    depolarization_strengths={('Gypi2', 0): 0.1},\n",
    "                                                    lindblad_error_coeffs={('Gypi2', 1): {('H', \"X\"): 0.2, ('S', 'Y'): 0.3}})\n",
    "ln_mmg4 = ln_mdl4.create_modelmember_graph()\n",
    "print(ln_mmg1.is_similar(ln_mmg4))\n",
    "print(ln_mmg1.is_equivalent(ln_mmg4))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Serialization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ex_mdl1.write('example_files/ex_mdl1.json')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ln_mdl1.write('example_files/ln_mdl1.json')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "38bc3ecdefa6d4ad98c3e3909d821405273dd289f85d5c490bf102ec408e5058"
  },
  "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
}