errorscaling.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from datetime import date\n",
    "import time\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings(action='once')\n",
    "\n",
    "\n",
    "from fidelity import *\n",
    "from povm import *\n",
    "from mpl_toolkits.axes_grid1.inset_locator import inset_axes\n",
    "%matplotlib inline\n",
    "\n",
    "\n",
    "today = date.today()\n",
    "datestr = today.strftime(\"%y%m%d\")\n",
    "\n",
    "\n",
    "# Hyperparameters\n",
    "Nq = 6\n",
    "split = 0.8\n",
    "# Nf = int(sys.argv[3])\n",
    "\n",
    "povm = POVM('pauli6')\n",
    "Na = povm.Na"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# GHZ state density matrix\n",
    "Ne = 4\n",
    "rhoKB = np.zeros((Ne, 2*Nq), dtype=int)\n",
    "rhoAmp = np.zeros((Ne), dtype=complex)\n",
    "\n",
    "rhoKB[1, :] = 1\n",
    "rhoKB[2, :Nq] = 1\n",
    "rhoKB[3, Nq:] = 1\n",
    "\n",
    "rhoAmp[0] = 0.5\n",
    "rhoAmp[1] = 0.5\n",
    "rhoAmp[2] = 0.5\n",
    "rhoAmp[3] = np.conj(rhoAmp[2])\n",
    "\n",
    "\n",
    "density = {}\n",
    "density['KB'] = rhoKB\n",
    "density['Amp'] = rhoAmp\n",
    "\n",
    "\n",
    "ghz = SampleDM(Nq, povm, density)\n",
    "\n",
    "# GHZ state density matrix\n",
    "Ne = 4\n",
    "rhoKB = np.zeros((Ne, 2*Nq), dtype=int)\n",
    "rhoAmp = np.zeros((Ne), dtype=complex)\n",
    "\n",
    "rhoKB[1, :] = 1\n",
    "rhoKB[2, :Nq] = 1\n",
    "rhoKB[3, Nq:] = 1\n",
    "\n",
    "rhoAmp[0] = 0.5\n",
    "rhoAmp[1] = 0.5\n",
    "rhoAmp[2] = 0.\n",
    "rhoAmp[3] = np.conj(rhoAmp[2])\n",
    "\n",
    "\n",
    "density = {}\n",
    "density['KB'] = rhoKB\n",
    "density['Amp'] = rhoAmp\n",
    "\n",
    "\n",
    "ghz_cl = SampleDM(Nq, povm, density)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Generate Probability table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "111.020447 seconds\n",
      "113.158533 seconds\n",
      "112.497137 seconds\n",
      "114.401576 seconds\n",
      "115.499567 seconds\n"
     ]
    }
   ],
   "source": [
    "from ann import *\n",
    "\n",
    "Ns = 40000\n",
    "\n",
    "tag = 'errorscaling/ghz-errorscaling'\n",
    "\n",
    "Nep = 200\n",
    "for Ns in [40000, 80000, 130000, 200000, 400000]:\n",
    "    model = torch.load(f='%s-%i_2-64-4_6-%i.mod'%(tag, Nep, Ns))\n",
    "\n",
    "    t = time.time()\n",
    "    pt_model = POVMProbTable(model)\n",
    "\n",
    "    np.save('%s-%i_2-64-4_6-%i_pt.npy'%(tag, Nep, Ns), pt_model)\n",
    "    print('%f seconds'%(time.time()-t))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Error scaling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "pt_ghz = POVMProbTable(ghz)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "Ns_list = np.array([40000, 80000, 130000, 200000, 400000])\n",
    "Nep_list = np.array([200, 400, 600])\n",
    "\n",
    "\n",
    "pt_model = np.zeros((len(Ns_list), len(Nep_list), len(pt_ghz)))\n",
    "pt_data = np.zeros((len(Ns_list), len(pt_ghz)))\n",
    "\n",
    "tag = 'errorscaling/ghz-errorscaling'\n",
    "\n",
    "\n",
    "for i in range(len(Ns_list)):\n",
    "    for j in range(len(Nep_list)):\n",
    "        Ns = Ns_list[i]\n",
    "        Nep = Nep_list[j]\n",
    "\n",
    "        pt_model[i, j] = np.load('%s-%i_2-64-4_6-%i_pt.npy'%(tag, Nep, Ns))\n",
    "\n",
    "    data = np.load('data/ghz-sim-Pauli6-Nq6-Ns%i.npy'%Ns)\n",
    "\n",
    "    for d in data:\n",
    "        pt_data[i, basestr2int(d, Na)] += 1/len(data)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9999999999997934\n",
      "[array([0.99999092, 0.99999214, 0.99999228]), array([0.99999901, 0.9999987 , 0.99999558]), array([0.99999548, 0.99999652, 0.99999852]), array([0.99999917, 0.99999741, 0.99999993]), array([0.99999979, 0.99999992, 0.99999997])]\n",
      "[1.0000000000002296, 1.0000000000001676, 1.0000000000000282, 1.0000000000000286, 1.000000000000022]\n"
     ]
    }
   ],
   "source": [
    "print(sum(pt_ghz))\n",
    "print([np.sum(a, 1) for a in pt_model])\n",
    "print([sum(a) for a in pt_data])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "sd_model = np.zeros((len(Ns_list), len(Nep_list)))\n",
    "sd_data = np.zeros((Ns_list.shape))\n",
    "\n",
    "for i in range(len(Ns_list)):\n",
    "    for j in range(len(Nep_list)):\n",
    "        sd_model[i, j] = np.sqrt(np.mean((pt_ghz-pt_model[i, j])**2))\n",
    "        sd_data[i] = np.sqrt(np.mean((pt_ghz-pt_data[i])**2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[2.31271058e-05 1.62340819e-05 1.29398586e-05 1.03590042e-05\n",
      " 7.37763182e-06]\n",
      "[[6.49325631e-06 8.56895141e-06 1.10010821e-05]\n",
      " [5.50261509e-06 7.32833451e-06 8.12960595e-06]\n",
      " [5.29506069e-06 5.05774709e-06 6.54834306e-06]\n",
      " [3.44628533e-06 3.98212593e-06 5.50637080e-06]\n",
      " [2.01583940e-06 3.36154706e-06 4.05666493e-06]]\n"
     ]
    }
   ],
   "source": [
    "print(sd_data)\n",
    "print(sd_model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1.84222884e-06 1.09931131e-06 6.53954703e-07 8.72696948e-07\n",
      " 8.47157785e-07]\n"
     ]
    }
   ],
   "source": [
    "print(np.std(sd_model, 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[-0.49534174 -5.42861318]\n",
      "[-0.45256985 -6.80792894]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "\n",
    "p_data = np.polyfit(np.log(Ns_list), np.log(sd_data), deg=1)\n",
    "\n",
    "sd_model_0 = np.mean(sd_model, 1)\n",
    "sd_model_1 = np.std(sd_model, 1)\n",
    "p_model = np.polyfit(np.log(Ns_list), np.log(sd_model_0), deg=1)\n",
    "\n",
    "fx = np.array([10, 15])\n",
    "ax.plot(fx, np.polyval(p_data, fx), linestyle='--', color='tab:blue', linewidth=2)\n",
    "ax.plot(fx, np.polyval(p_model, fx), linestyle='--', color='tab:orange', linewidth=2)\n",
    "\n",
    "ax.plot(np.log(Ns_list), np.log(sd_data), marker='.', markersize=12, linestyle='', color='tab:blue', label='Data')\n",
    "# ax.plot(np.log(Ns_list), np.log(sd_model), marker='.', markersize=12, linestyle='', color='tab:orange', label='Transformer')\n",
    "ax.errorbar(np.log(Ns_list), np.log(sd_model_0), yerr=sd_model_1/sd_model_0, marker='.', markersize=12, linestyle='', color='tab:orange', label='Transformer')\n",
    "\n",
    "\n",
    "xticks_major = np.array([4, 8, 16, 32])*10**4\n",
    "xticklabels_major = np.array(['%i'%x for x in [4, 8, 16, 32]])\n",
    "\n",
    "xticks_minor = np.array([5, 6, 7, 10, 12, 14, 20, 24, 28, 40])*10**4\n",
    "ax.set_xticks(np.log(xticks_major))\n",
    "ax.set_xticklabels(xticklabels_major, size=14)\n",
    "ax.set_xticks(np.log(xticks_minor), minor=True)\n",
    "ax.xaxis.set_tick_params(which='minor', length=3)\n",
    "\n",
    "ax.set_xlabel('$N_s\\\\ (\\\\times10^4)$', size=14)\n",
    "\n",
    "yticks_major = np.array([2, 4, 8, 16])*10**(-6)\n",
    "yticks_minor = np.array([2.5, 3, 3.5, 5, 6, 7, 10, 12, 14, 20, 24])*10**(-6)\n",
    "yticklabels_major = np.array(['%i'%y for y in [2, 4, 8, 16]])\n",
    "\n",
    "ax.set_yticks(np.log(yticks_major))\n",
    "ax.set_yticks(np.log(yticks_minor), minor=True)\n",
    "ax.set_yticklabels(yticklabels_major, size=14)\n",
    "ax.yaxis.set_tick_params(which='minor', length=3)\n",
    "ax.set_ylabel('MSE $(\\\\times10^{-6})$', size=14)\n",
    "\n",
    "ax.set_xbound(10.5, 13)\n",
    "ax.set_ybound(-13.2, -10.5)\n",
    "ax.legend(fontsize=14)\n",
    "\n",
    "\n",
    "print(p_data)\n",
    "print(p_model)\n",
    "\n",
    "# fig.savefig('figures/povm_error_scaling.pdf', dpi=300, frameon=True,bbox_inches=\"tight\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Histogram"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "pt_ghz = POVMProbTable(ghz)\n",
    "pt_ghzcl = POVMProbTable(ghz_cl)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3.982125934364559e-06\n",
      "1.0359004206496893e-05\n"
     ]
    }
   ],
   "source": [
    "Ns = 200000\n",
    "Nep = 400\n",
    "\n",
    "tag = 'errorscaling/ghz-errorscaling'\n",
    "# tag = 'data/aqt-ghz-large'\n",
    "\n",
    "pt_model = np.load('%s-%i_2-64-4_6-%i_pt.npy'%(tag, Nep, Ns))\n",
    "\n",
    "data = np.load('data/ghz-sim-Pauli6-Nq6-Ns%i.npy'%Ns)\n",
    "pt_data = np.zeros(pt_model.shape)\n",
    "\n",
    "for d in data:\n",
    "    pt_data[basestr2int(d, Na)] += 1./len(data)\n",
    "print(np.sqrt(np.mean((pt_ghz-pt_model)**2)))\n",
    "print(np.sqrt(np.mean((pt_ghz-pt_data)**2)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f0e3d4aa128>"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ghz_as = np.flip(np.argsort(pt_ghz))\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(12, 5))\n",
    "\n",
    "x = np.arange(start=1, stop=len(pt_ghz)+1)\n",
    "\n",
    "ax.plot(x, pt_data[ghz_as], label='Data')\n",
    "ax.plot(x, pt_model[ghz_as], label='Transformer')\n",
    "ax.plot(x, pt_ghz[ghz_as], label='GHZ', color='black')\n",
    "\n",
    "# axins = inset_axes(ax,6,2,bbox_to_anchor=(-0.06,-0.02, 1,1),bbox_transform=ax.transAxes)\n",
    "# ghz_as_pos = (pt_ghz[ghz_as] > 0)\n",
    "\n",
    "# axins.plot(pt_data[ghz_as][ghz_as_pos])\n",
    "# axins.plot(pt_model[ghz_as][ghz_as_pos])\n",
    "# axins.plot(pt_ghz[ghz_as][ghz_as_pos], color='black')\n",
    "# axins.set_xticks([])\n",
    "\n",
    "ax.set_xticks([])\n",
    "ax.set_ylabel('p', size=14)\n",
    "ax.tick_params(axis='y', labelsize=14)\n",
    "\n",
    "ax.legend(fontsize=14)\n",
    "# fig.savefig('figures/povm_sorted.pdf', dpi=300, frameon=True,bbox_inches=\"tight\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/peter/.local/lib/python3.6/site-packages/ipykernel_launcher.py:16: RuntimeWarning: divide by zero encountered in log\n",
      "  app.launch_new_instance()\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f0e3d489fd0>"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "Nbin = 100\n",
    "\n",
    "xbin, dx = np.linspace(start=min(pt_ghz), stop=max(pt_ghz), num=Nbin+1, retstep=True)\n",
    "\n",
    "x = (xbin[:-1]+xbin[1:])/2\n",
    "\n",
    "h_ghz, _ = np.histogram(pt_ghz, bins=xbin)\n",
    "h_data, _ = np.histogram(pt_data, bins=xbin)\n",
    "h_model, _ = np.histogram(pt_model, bins=xbin)\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(12,5))\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "ax.bar(x, np.log(h_data), width=dx, label='Data')\n",
    "ax.bar(x, np.log(h_model), width=dx, label='Transformer')\n",
    "ax.bar(x, np.log(h_ghz), width=0.3*dx, label='GHZ', color='black')\n",
    "\n",
    "yticks = [1, 10, 100, 1000, 10000]\n",
    "ax.set_yticks(np.log(yticks))\n",
    "ax.set_yticklabels(['%i'%i for i in yticks])\n",
    "ax.set_ylabel('Frequency', size=14)\n",
    "ax.set_xlabel('p', size=14)\n",
    "ax.tick_params(axis='both', labelsize=14)\n",
    "ax.legend(fontsize=14)\n",
    "\n",
    "# fig.savefig('figures/povm_histogram.pdf', dpi=300, frameon=True,bbox_inches=\"tight\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/peter/.local/lib/python3.6/site-packages/ipykernel_launcher.py:23: RuntimeWarning: divide by zero encountered in log\n",
      "/home/peter/.local/lib/python3.6/site-packages/ipykernel_launcher.py:50: MatplotlibDeprecationWarning: \n",
      "The frameon kwarg was deprecated in Matplotlib 3.1 and will be removed in 3.3. Use facecolor instead.\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ghz_as = np.flip(np.argsort(pt_ghz))\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(12, 5))\n",
    "\n",
    "x = np.arange(start=1, stop=len(pt_ghz)+1)\n",
    "\n",
    "ax.plot(x, pt_data[ghz_as], label='Data')\n",
    "ax.plot(x, pt_model[ghz_as], label='Transformer')\n",
    "ax.plot(x, pt_ghz[ghz_as], label='GHZ', color='black')\n",
    "\n",
    "axins = inset_axes(ax,7.5,2.2,bbox_to_anchor=(-0.02,-0.02, 1,1),bbox_transform=ax.transAxes)\n",
    "\n",
    "Nbin = 100\n",
    "\n",
    "xbin, dx = np.linspace(start=min(pt_ghz), stop=max(pt_ghz), num=Nbin+1, retstep=True)\n",
    "\n",
    "x = (xbin[:-1]+xbin[1:])/2\n",
    "\n",
    "h_ghz, _ = np.histogram(pt_ghz, bins=xbin)\n",
    "h_data, _ = np.histogram(pt_data, bins=xbin)\n",
    "h_model, _ = np.histogram(pt_model, bins=xbin)\n",
    "\n",
    "axins.bar(x, np.log(h_data), width=dx, label='Data ($p_D$)')\n",
    "axins.bar(x, np.log(h_model), width=dx, label='Transformer ($p_T$)')\n",
    "axins.bar(x, np.log(h_ghz), width=0.3*dx, label='Exact GHZ ($p_Q$)', color='black')\n",
    "\n",
    "yticks = [1, 10, 100, 1000, 10000]\n",
    "axins.set_yticks(np.log(yticks))\n",
    "axins.set_yticklabels(['$10^%i$'%i for i in range(len(yticks))])\n",
    "axins.set_xticks(np.array([0, 1, 2, 3, 4, 5, 6, 7])*10**(-4))\n",
    "axins.set_xticklabels([0, 1, 2, 3, 4, 5, 6, 7])\n",
    "axins.set_ylabel('Frequency', size=14)\n",
    "axins.set_xlabel('$p$ $(\\\\times 10^{-4})$', size=14)\n",
    "axins.tick_params(axis='both', labelsize=14)\n",
    "axins.legend(fontsize=14)\n",
    "# axins.plot(pt_data[ghz_as][ghz_as_pos])\n",
    "# axins.plot(pt_model[ghz_as][ghz_as_pos])\n",
    "# axins.plot(pt_ghz[ghz_as][ghz_as_pos], color='black')\n",
    "# axins.set_xticks([])\n",
    "\n",
    "ax.set_ylim(bottom=0)\n",
    "ax.set_xticks([])\n",
    "ax.set_yticks(np.array([0, 1, 2, 3, 4, 5, 6, 7])*10**(-4))\n",
    "ax.set_yticklabels([0, 1, 2, 3, 4, 5, 6, 7])\n",
    "ax.set_ylabel('$p$ $(\\\\times 10^{-4})$', size=14)\n",
    "ax.set_xlabel('POVM outcomes (sorted)', size=14)\n",
    "ax.tick_params(axis='y', labelsize=14)\n",
    "\n",
    "# ax.legend(fontsize=14)\n",
    "fig.savefig('figures/povm_probabilities.pdf', dpi=300, frameon=True,bbox_inches=\"tight\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "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": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}