{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Gate Sets Tutorial\n",
    "This tutorial will show you how to create and use `GateSet` objects.  `GateSet` objects are fundamental to pyGSTi, as each represents a set of quantum gates along with state preparation and measurement (i.e. POVM effect) operations.  In pyGSTi, a \"state space\" refers to a Hilbert space of *pure* quantum states (often thought of as length-$d$ vectors, where $d=2^N$ for $N$ qubits). A \"density matrix space\" refers to a Hilbert space of density matrices, which while often thought of as $d \\times d$ matrices can also be represented by length $d^2$ vectors.  Mathematically, these vectors live in Hilbert-Schmidt space, the space of linear operators on the original $d\\times d$ density matrix space.  pyGSTi uses this Hilbert-Schmidt vector-representation for density matrices and POVM effects, since this allows quantum gates to be represented by $d^2 \\times d^2$ matrices which act on Hilbert-Schmidt vectors.\n",
    "\n",
    "The basis used for Hilbert-Schmidt space can be any set of $d\\times d$ matrices which span the density matrix space.  pyGSTi constains support for three basis sets:\n",
    "- the matrix unit, or \"standard\" basis, consisting of the matrices with a single unit (1.0) element and otherwise zero.  This basis is selected by passing `\"std\"` to appropriate function arguments.\n",
    "- the Pauli-product basis, consisting of tensor products of the four Pauli matrices {I, X, Y, Z} normalized so that $Tr(B_i B_j) = \\delta_{ij}$.  All of these matrices are Hermitian, so that Hilbert-Schmidt vectors and matrices are real when this basis is used. This basis can only be used when the $d = 4^i$ for integer $i$, and is selected using the string `\"pp\"`.\n",
    "- the Gell-Mann basis, consisting of the normalized Gell-Mann matrices (see Wikipedia if you don't know what these are).  Similar to the Pauli-product case, these matrices are also Hermitian, so that Hilbert-Schmidt vectors and matrices are real when this basis is used.  Unlike the Pauli-product case, since Gell-Mann matrices are well defined in any dimension, the Gell-Mann basis is *not* restricted to cases when $d=4^i$.  This basis is selected using the string `\"gm\"`.\n",
    "\n",
    "`GateSet` objects have the look and feel of Python dictionaries which hold $d^2\\times d^2$ gate matrices and length-$d^2$ state preparation and POVM effect vectors (collectively referred to as \"SPAM\" vectors). "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from __future__ import print_function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "#Import the pyGSTi module -- you probably want this at the beginning of every notebook\n",
    "import pygsti\n",
    "import pygsti.construction as pc"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "## Creating gate sets\n",
    "There are more or less three ways to create `GateSet` objects in pyGSTi:\n",
    "\n",
    "*  By creating an empty `GateSet` and setting its elements directly, possibly with the help of `pygsti.construction`'s `build_gate` and `build_vector` functions.\n",
    "* By a single call to `build_gateset`, which automates the above approach.\n",
    "* By loading from a text-format gateset file using `pygsti.io.load_gateset`."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Creating a `GateSet` from scratch\n",
    "\n",
    "Gates and SPAM vectors can be assigned to a `GateSet` object as to an ordinary python dictionary.  Internally a `GateSet` holds these quantities as `Gate`- and `SPAMVec`-derived objects, but you may assign lists, Numpy arrays, or other types of Python iterables to a `GateSet` key and a conversion will be performed automatically.  To keep gates, state preparations, and POVM effects separate, the `GateSet` object looks at the beginning of the dictionary key being assigned: keys beginning with `rho`, `E`, and `G` are categorized as state preparations, POVM effects, and gates, respectively.  To avoid ambiguity, each key *must* begin with one of these three prefixes with the exception of `identity` which we will mention later.\n",
    "\n",
    "To separately access the state preparations, POVM effects, and gates contained in a `GateSet` use the `preps`, `effects`, and `gates` members respectively.  Each one provides dictionary-like access to the underlying objects.  For example, `myGateset.gates['Gx']` accesses the same underlying `Gate` object as `myGateset['Gx']`, and similarly for `myGateset.preps['rho0']` and `myGateset['rho0']`.  The values of gates and SPAM vectors can be read and written in this way. \n",
    "\n",
    "In addition to SPAM vectors and gate matrices, a `GateSet` holds a mapping between (state preparation, POVM effect) pairs and strings called \"SPAM labels\".  Each SPAM label identifies an experimental outcome, meaning \"I prepared state A and then measured outcome X\".  Experimental data is tabulated according to SPAM label, that is, each experimental count is assigned a particular SPAM label (this is explained further in the Dataset tutorial).  The map between (state preparation, POVM effect) pairs and \"SPAM labels\" is a dictionary called `spamdefs` whose keys are the SPAM labels and whose values are 2-tuples containing a state preparation label and POVM effect label.  The special POVM effect label \"remainder\" can be used to mean the identity minus all of the other POVM effects.  This \"remainder\" POVM effect is not properly contained within the `GateSet` (it is not parameterized during optimizations), but should rather be thought of as a quantity that can straightforwardly be *computed* from `GateSet` quantities.\n",
    "\n",
    "When the \"remainder\" label is used, the `GateSet` must then know what the identity vector is, and so one must set the special `'identity'` key of the `GateSet` to the identity vector in whatever basis is being used for the SPAM vectors and gate matrices.\n",
    "\n",
    "(Aside: Usually there is only a single state preparation, in which case the SPAM labels correspond directly with the POVM effects typically thought of as experimental outcomes.  However, if there *are* multiple state preparations, it is important that we treat the experiment counts for \"preparing state A and measuring outcome X\" and \"preparing state B and measuring outcome X\" differently.)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from math import sqrt\n",
    "import numpy as np\n",
    "\n",
    "#Initialize an empty GateSet object\n",
    "gateset1 = pygsti.objects.GateSet()\n",
    "\n",
    "#Populate the GateSet object with states, effects, gates,\n",
    "# all in the *normalized* Pauli basis: { I/sqrt(2), X/sqrt(2), Y/sqrt(2), Z/sqrt(2) }\n",
    "# where I, X, Y, and Z are the standard Pauli matrices.\n",
    "gateset1['rho0'] = [ 1/sqrt(2), 0, 0, 1/sqrt(2) ] # density matrix [[1, 0], [0, 0]] in Pauli basis\n",
    "gateset1['E0'] = [ 1/sqrt(2), 0, 0, -1/sqrt(2) ]  # projector onto [[0, 0], [0, 1]] in Pauli basis\n",
    "gateset1['Gi'] = np.identity(4,'d') # 4x4 identity matrix\n",
    "gateset1['Gx'] = [[1, 0, 0, 0],\n",
    "                  [0, 1, 0, 0],\n",
    "                  [0, 0, 0,-1],\n",
    "                  [0, 0, 1, 0]] # pi/2 X-rotation in Pauli basis\n",
    "\n",
    "gateset1['Gy'] = [[1, 0, 0, 0],\n",
    "                  [0, 0, 0, 1],\n",
    "                  [0, 0, 1, 0],\n",
    "                  [0,-1, 0, 0]] # pi/2 Y-rotation in Pauli basis\n",
    "\n",
    "#Create SPAM labels \"plus\" and \"minus\" using the special \"remainder\" label,\n",
    "# and set the then-needed identity vector.\n",
    "gateset1.spamdefs['plus'] = ('rho0','E0')\n",
    "gateset1.spamdefs['minus'] = ('rho0','remainder')\n",
    "gateset1['identity'] = [ sqrt(2), 0, 0, 0 ]  # [[1, 0], [0, 1]] in Pauli basis\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Creating a `GateSet` from scratch using `build_gate` and `build_vector`\n",
    "The `build_gate` and `build_vector` functions take a human-readable string representation of a gate or SPAM vector, and return a `Gate` or `SPAMVector` object that gets stored in the dictionary-like `GateSet` object.  To use these functions, you must specify what state space you're working with.  This is done via two quantities:\n",
    "1. **State space dimensions**: a list of integers specifying the dimension of each *block* in a direct-sum decomposition of the total state space.  For example, `[2]` means just a 2-dimensional Hilbert space, and `[2,2]` means the direct sum of two 2-dimensional Hilbert spaces.\n",
    "2. **State space labels**: a list of tuples of (string) labels.  Each tuple describes how to label the corresponding term of the direct-sum decomposition of the state space.  Thus, the length of the state-space-labels list must be equal to the length of the state-space-dimensions list.  The elements of a tuple must be strings that start with either \"Q\" or \"L\", and are followed by any letters or numbers of your choosing.  A label beginning with \"Q\" denotes a 2-dimensional space, whereas a label beginning with \"L\" denotes a 1-dimensional space.  The tuple itself represents a tensor product of the spaces denoted by it's elements, and so describes how to interpret a given dimension Hilbert space as the tensor product of 1- and 2-dimensional spaces.  For example, the tuple `('Q0',)` describes a 2-dimensional Hilbert space as that of a single qubit, and the tuple `('Q0','Q1')` describes a 4-dimensional Hilbert space as that of two qubit spaces tensored together.  Each tuple describes a single Hilbert-space *term* in the direct-sum decomposition of the entire Hilbert space, so the list `[('Q0','Q1'),('L0',)]` represents a Hilbert space that is the direct sum of a 4-dimensional and a 1-dimensional space; the 4-dimensional space is the a tensor product of two qubit spaces labelled 'Q0' and 'Q1' while the 1-dimensional space is labeled 'L0'.  (In this case, the corresponding state space dimensions list *must* be `[4,1]`, and is required as an argument to `build_vector` and `build_gate` just as a consistency check.)\n",
    "\n",
    "While specifying the state space in this way can seem overly cumbersome for small Hilbert spaces, it allows for great flexibility when moving to more complex spaces.  It is worthwhile to note that the state space labels described above are *only* used when interpreting the human-readable string used to specify gates and SPAM vectors in calls to `build_gate` and `build_vector`, respectively.  \n",
    "\n",
    "`build_vector` currently only understands strings which are integers (e.g. \"1\"), for which it creates a vector performing state preparation of (or, equivalently, a state projection onto) the $i^{th}$ state of the Hilbert space, that is, the state corresponding to the $i^{th}$ row and column of the $d\\times d$ density matrix.\n",
    "\n",
    "`build_gate` accepts a wider range of descriptor strings, which take the form of *functionName*(*args*) and include:\n",
    "- `I(label0, label1, ...)` : the identity on the spaces labeled by `label0`, `label1`, etc.\n",
    "- `X(theta,Qlabel)`, `Y(theta,Qlabel)`, `Z(theta,Qlabel)` : single qubit X-, Y-, and Z-axis rotations by angle `theta` (in radians) on the qubit labeled by `Qlabel`.  Note that `pi` can be used within an expression for `theta`, e.g. `X(pi/2,Q0)`.\n",
    "- `CX(theta, Qlabel1, Qlabel2)`, `CY(theta, Qlabel1, Qlabel2)`, `CZ(theta, Qlabel1, Qlabel2)` : two-qubit controlled rotations by angle `theta` (in radians) on qubits `Qlabel1` (the control) and `Qlabel2` (the target).\n",
    "\n",
    "When the special \"remainder\" label is used, the needed identity vector can be generated by a call to `build_identity_vec`. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "#Specify the state space\n",
    "stateSpace = [2] # Hilbert space has dimension 2; density matrix is a 2x2 matrix\n",
    "spaceLabels = [('Q0',)] #interpret the 2x2 density matrix as a single qubit named 'Q0'\n",
    "\n",
    "#Initialize an empty GateSet object\n",
    "gateset2 = pygsti.objects.GateSet()\n",
    "\n",
    "#Populate the GateSet object with states, effects, and gates using \n",
    "# build_vector, build_gate, and build_identity_vec.   \n",
    "gateset2['rho0'] = pc.build_vector(stateSpace,spaceLabels,\"0\")\n",
    "gateset2['E0'] = pc.build_vector(stateSpace,spaceLabels,\"1\")\n",
    "gateset2['Gi'] = pc.build_gate(stateSpace,spaceLabels,\"I(Q0)\")\n",
    "gateset2['Gx'] = pc.build_gate(stateSpace,spaceLabels,\"X(pi/2,Q0)\")\n",
    "gateset2['Gy'] = pc.build_gate(stateSpace,spaceLabels,\"Y(pi/2,Q0)\")\n",
    "gateset2['identity'] = pc.build_identity_vec(stateSpace)\n",
    "\n",
    "#Create SPAM labels \"plus\" and \"minus\" using the special \"remainder\" label.\n",
    "gateset2.spamdefs['plus'] = ('rho0','E0')\n",
    "gateset2.spamdefs['minus'] = ('rho0','remainder')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Create a `GateSet` in a single call to build_gateset\n",
    "The approach illustrated above using calls to `build_vector`, `build_gate`, and `build_identity_vec` can be performed in a single call to `build_gateset`.  You will notice that all of the arguments to `build_gateset` corrspond to those used to construct a gate set using `build_vector` and `build_gate`; the `build_gateset` function is merely a convenience function which allows you to specify everything at once.  These arguments are:\n",
    "- Args 1 & 2 : the state-space-dimensions and state-space-labels, familiar from before.\n",
    "- Args 3 & 4 : list-of-gate-labels, list-of-gate-expressions (labels *must* begin with 'G'; \"expressions\" being the descriptor strings passed to `build_gate`)\n",
    "- Args 5 & 6 : list-of-prep-labels, list-of-prep-expressions (labels *must* begin with 'rho'; \"expressions\" being the descriptor strings passed to `build_vector`)\n",
    "- Args 7 & 8 : list-of-effect-labels, list-of-effect-expressions (labels *must* begin with 'E'; \"expressions\" being the descriptor strings passed to `build_vector`)\n",
    "- Arg 9 : the dictionary of SPAM label definitions.\n",
    "\n",
    "Note that the optional parameter `basis` can be set to `\"gm\"` (the default), `\"pp\"`, or `\"std\"` to select the basis for the gate matrices and SPAM vectors."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "gateset3 = pc.build_gateset( [2], [('Q0',)],\n",
    "                             ['Gi','Gx','Gy'], [ \"I(Q0)\",\"X(pi/2,Q0)\", \"Y(pi/2,Q0)\"],\n",
    "                             prepLabels = ['rho0'], prepExpressions=[\"0\"], \n",
    "                             effectLabels = ['E0'], effectExpressions=[\"1\"], \n",
    "                             spamdefs={'plus': ('rho0','E0'), 'minus': ('rho0','remainder') }) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load a `GateSet` from a file\n",
    "You can also construct a `GateSet` object from a file using `pygsti.io.load_gateset`.  The format of the text file should be fairly self-evident given the above discussion.  Note that vector and matrix elements need not be simple numbers, but can be any mathematical expression parseable by the Python interpreter, and in addition to numbers can include \"sqrt\" and \"pi\". "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "#3) Write a text-format gateset file and read it in.\n",
    "gateset4_txt = \\\n",
    "\"\"\"\n",
    "# Example text file describing a gateset\n",
    "\n",
    "# State prepared, specified as a state in the Pauli basis (I,X,Y,Z)\n",
    "rho0\n",
    "PauliVec\n",
    "1/sqrt(2) 0 0 1/sqrt(2)\n",
    "\n",
    "# State measured as yes outcome, also specified as a state in the Pauli basis\n",
    "E0\n",
    "PauliVec\n",
    "1/sqrt(2) 0 0 -1/sqrt(2)\n",
    "\n",
    "Gi\n",
    "PauliMx\n",
    "1 0 0 0\n",
    "0 1 0 0\n",
    "0 0 1 0\n",
    "0 0 0 1\n",
    "\n",
    "Gx\n",
    "PauliMx\n",
    "1 0 0 0\n",
    "0 1 0 0\n",
    "0 0 0 -1\n",
    "0 0 1 0\n",
    "\n",
    "Gy\n",
    "PauliMx\n",
    "1 0 0 0\n",
    "0 0 0 1\n",
    "0 0 1 0\n",
    "0 -1 0 0\n",
    "\n",
    "IDENTITYVEC sqrt(2) 0 0 0\n",
    "SPAMLABEL plus = rho0 E0\n",
    "SPAMLABEL minus = rho0 remainder\n",
    "\"\"\"\n",
    "with open(\"tutorial_files/Example_Gateset.txt\",\"w\") as gsetfile:\n",
    "    gsetfile.write(gateset4_txt)\n",
    "\n",
    "gateset4 = pygsti.io.load_gateset(\"tutorial_files/Example_Gateset.txt\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "#All four of the above gatesets are identical.  See this by taking the frobenius differences between them:\n",
    "assert(gateset1.frobeniusdist(gateset2) < 1e-8)\n",
    "assert(gateset1.frobeniusdist(gateset3) < 1e-8)\n",
    "assert(gateset1.frobeniusdist(gateset4) < 1e-8)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Viewing gate sets\n",
    "In the cells below, we demonstrate how to print and access information within a `GateSet`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Gateset 1:\n",
      " rho0 =    0.7071        0        0   0.7071\n",
      "\n",
      "\n",
      "E0 =    0.7071        0        0  -0.7071\n",
      "\n",
      "\n",
      "Gi = \n",
      "   1.0000        0        0        0\n",
      "        0   1.0000        0        0\n",
      "        0        0   1.0000        0\n",
      "        0        0        0   1.0000\n",
      "\n",
      "\n",
      "Gx = \n",
      "   1.0000        0        0        0\n",
      "        0   1.0000        0        0\n",
      "        0        0        0  -1.0000\n",
      "        0        0   1.0000        0\n",
      "\n",
      "\n",
      "Gy = \n",
      "   1.0000        0        0        0\n",
      "        0        0        0   1.0000\n",
      "        0        0   1.0000        0\n",
      "        0  -1.0000        0        0\n",
      "\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#Printing the contents of a GateSet is easy\n",
    "print(\"Gateset 1:\\n\", gateset1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Gx =  Fully Parameterized gate with shape (4, 4)\n",
      " 1.00   0   0   0\n",
      "   0 1.00   0   0\n",
      "   0   0   0-1.00\n",
      "   0   0 1.00   0\n",
      "\n",
      "Array-like printout\n",
      " [[ 1.  0.  0.  0.]\n",
      " [ 0.  1.  0.  0.]\n",
      " [ 0.  0.  0. -1.]\n",
      " [ 0.  0.  1.  0.]] \n",
      "\n",
      "First row\n",
      " [ 1.  0.  0.  0.] \n",
      "\n",
      "Element [2,3] =  -1.0 \n",
      "\n",
      "Id_dot_Gx\n",
      " [[ 1.  0.  0.  0.]\n",
      " [ 0.  1.  0.  0.]\n",
      " [ 0.  0.  0. -1.]\n",
      " [ 0.  0.  1.  0.]] \n",
      "\n"
     ]
    }
   ],
   "source": [
    "#You can also access individual gates like they're numpy arrays:\n",
    "Gx = gateset1['Gx'] # a Gate object, but behaves like a numpy array\n",
    "\n",
    "#By printing a gate, you can see that it's not just a numpy array\n",
    "print(\"Gx = \", Gx)\n",
    "\n",
    "#But can be accessed as one:\n",
    "print(\"Array-like printout\\n\", Gx[:,:],\"\\n\")\n",
    "print(\"First row\\n\", Gx[0,:],\"\\n\")\n",
    "print(\"Element [2,3] = \",Gx[2,3], \"\\n\")\n",
    "\n",
    "Id = np.identity(4,'d')\n",
    "Id_dot_Gx = np.dot(Id,Gx)\n",
    "print(\"Id_dot_Gx\\n\", Id_dot_Gx, \"\\n\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Basic Operations with Gatesets\n",
    "\n",
    "`GateSet` objects have a number of methods that support a variety of operations, including:\n",
    "\n",
    "* Depolarizing or rotating every gate\n",
    "* Writing the gate set to a file\n",
    "* Computing products of gate matrices\n",
    "* Printing more information about the gate set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "#Add 10% depolarization noise to the gates\n",
    "depol_gateset3 = gateset3.depolarize(gate_noise=0.1)\n",
    "\n",
    "#Add 10% depolarization noise to the gates\n",
    "rot_gateset3 = gateset3.rotate(rotate=0.1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "#Writing a gateset as a text file\n",
    "pygsti.io.write_gateset(depol_gateset3, \"tutorial_files/Example_depolarizedGateset.txt\", title=\"My Gateset\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Product of Gx * Gx = \n",
      " [[  1.00000000e+00   0.00000000e+00   1.51390444e-16  -2.78344767e-16]\n",
      " [  0.00000000e+00   8.10000000e-01   0.00000000e+00   0.00000000e+00]\n",
      " [ -1.09201969e-16   0.00000000e+00  -8.10000000e-01  -3.17943723e-16]\n",
      " [  2.63428889e-16   0.00000000e+00   3.17943723e-16  -8.10000000e-01]]\n",
      "\n",
      "Probability of 'plus' spam label of gate string GxGx =  0.9049999999999999\n",
      "Probability of 'minus' spam label of gate string GxGx =  0.09499999999999997\n",
      "Probabilities as a dict =  {'minus': 0.09499999999999997, 'plus': 0.9049999999999999}\n"
     ]
    }
   ],
   "source": [
    "#Computing the product of gate matrices (more on this in the next tutorial on gate strings)\n",
    "print(\"Product of Gx * Gx = \\n\",depol_gateset3.product((\"Gx\", \"Gx\")), end='\\n\\n')\n",
    "print(\"Probability of 'plus' spam label of gate string GxGx = \",depol_gateset3.pr('plus', (\"Gx\", \"Gx\")))\n",
    "print(\"Probability of 'minus' spam label of gate string GxGx = \",depol_gateset3.pr('minus', (\"Gx\", \"Gx\")))\n",
    "print(\"Probabilities as a dict = \",depol_gateset3.probs((\"Gx\", \"Gx\")))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "rho0 =    0.7071        0        0   0.7071\n",
      "\n",
      "\n",
      "E0 =    0.7071        0        0  -0.7071\n",
      "\n",
      "\n",
      "Gi = \n",
      "   1.0000        0        0        0\n",
      "        0   0.9000        0        0\n",
      "        0        0   0.9000        0\n",
      "        0        0        0   0.9000\n",
      "\n",
      "\n",
      "Gx = \n",
      "   1.0000        0        0        0\n",
      "        0   0.9000        0        0\n",
      "        0        0        0  -0.9000\n",
      "        0        0   0.9000        0\n",
      "\n",
      "\n",
      "Gy = \n",
      "   1.0000        0        0        0\n",
      "        0        0        0   0.9000\n",
      "        0        0   0.9000        0\n",
      "        0  -0.9000        0        0\n",
      "\n",
      "\n",
      "\n",
      "\n",
      "\n",
      "Choi Matrices:\n",
      "('Choi(Gi) in pauli basis = \\n', '   0.9250       +0j        0       +0j        0       +0j        0       +0j\\n        0       +0j   0.0250       +0j        0       +0j        0       +0j\\n        0       +0j        0       +0j   0.0250       +0j        0       +0j\\n        0       +0j        0       +0j        0       +0j   0.0250       +0j\\n')\n",
      "('  --eigenvals = ', [0.024999999999999977, 0.024999999999999998, 0.024999999999999998, 0.92500000000000027], '\\n')\n",
      "('Choi(Gx) in pauli basis = \\n', '   0.4750       +0j        0  +0.4500j        0       +0j        0       +0j\\n        0  -0.4500j   0.4750       +0j        0       +0j        0       +0j\\n        0       +0j        0       +0j   0.0250       +0j        0       +0j\\n        0       +0j        0       +0j        0       +0j   0.0250       +0j\\n')\n",
      "('  --eigenvals = ', [0.024999999999999974, 0.024999999999999991, 0.025000000000000026, 0.92500000000000016], '\\n')\n",
      "('Choi(Gy) in pauli basis = \\n', '   0.4750       +0j        0       +0j        0  +0.4500j        0       +0j\\n        0       +0j   0.0250       +0j        0       +0j        0       +0j\\n        0  -0.4500j        0       +0j   0.4750       +0j        0       +0j\\n        0       +0j        0       +0j        0       +0j   0.0250       +0j\\n')\n",
      "('  --eigenvals = ', [0.024999999999999932, 0.025000000000000008, 0.025000000000000099, 0.92500000000000082], '\\n')\n",
      "('Sum of negative Choi eigenvalues = ', 0.0)\n",
      "('rhoVec Penalty (>0 if invalid rhoVecs) = ', 1.1102230246251565e-16)\n",
      "('EVec Penalty (>0 if invalid EVecs) = ', 0)\n"
     ]
    }
   ],
   "source": [
    "#Printing more detailed information about a gateset\n",
    "depol_gateset3.print_info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}