full-simulations-using-adam.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "719f172f",
   "metadata": {},
   "source": [
    "# SVGD for quantum tomography using PyMC\n",
    "\n",
    "## This notebook is a tutorial of the report document from the project Crowdfund:\n",
    "## [$Quantum](https://mirror.xyz/csferrie.eth/gUyt_o1KkZH-c3OHU85a8mfmLioNg4_uPHbtoAhcrR8)\n",
    "\n",
    "In this notebook we are going to use PyMC in order to simulate the estimationn of random quantum states.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "20c8863c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running on PyMC3 v3.11.2\n"
     ]
    }
   ],
   "source": [
    "## import the libraries\n",
    "import pymc3 as pm\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns ## we are going to use Kernel Density Estimation to plot the modified priors\n",
    "import arviz as az\n",
    "import time\n",
    "print(f\"Running on PyMC3 v{pm.__version__}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "2c0854e1",
   "metadata": {},
   "outputs": [],
   "source": [
    "## this cell is to have nice plots\n",
    "########## \n",
    "SMALL_SIZE = 8\n",
    "MEDIUM_SIZE = 10\n",
    "BIGGER_SIZE = 12\n",
    "\n",
    "plt.rc('font', size=SMALL_SIZE)          # controls default text sizes\n",
    "plt.rc('axes', titlesize=MEDIUM_SIZE)     # fontsize of the axes title\n",
    "plt.rc('axes', labelsize=MEDIUM_SIZE)    # fontsize of the x and y labels\n",
    "plt.rc('xtick', labelsize=MEDIUM_SIZE)    # fontsize of the tick labels\n",
    "plt.rc('ytick', labelsize=MEDIUM_SIZE)    # fontsize of the tick labels\n",
    "plt.rc('legend', fontsize=SMALL_SIZE)    # legend fontsize\n",
    "plt.rc('figure', titlesize=BIGGER_SIZE)  # fontsize of the figure title\n",
    "\n",
    "#############"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "00521ce3",
   "metadata": {},
   "source": [
    "In this notebook, a random quantum state is given by 4 random numbers coming from a normal distribution.  \n",
    "\n",
    "Let $\\vec{x}$ be a vector containing the 4 random numbers, \n",
    "\n",
    "$$\\vec{x}=(x_0,x_1,x_2,x_3),$$  \n",
    "\n",
    "and every $x_i$ is normal random variable. We want to get a random quantum state of the following way:\n",
    "\n",
    "$$\\tilde{\\rho}= \\frac{1}{||\\tilde{\\rho}||}\\begin{bmatrix} x_0^2 & x_0(x_1-ix_2) \\\\ \n",
    "x_0 (x_1+ix_2)& x_1^2+x_2^2+x_3^2 \\end{bmatrix}. $$\n",
    "\n",
    "$\\tilde{\\rho}$ comes from\n",
    "\n",
    "$$X= \\begin{bmatrix} x_0 & 0 \\\\\n",
    "x_1+ix_2 & x_3 \\end{bmatrix},$$ \n",
    "\n",
    "using the Cholesky decomposition. \n",
    "Thus,\n",
    "\n",
    "$$\\tilde{\\rho}=\\frac{1}{||\\tilde{\\rho}||}X X^\\dagger $$\n",
    "\n",
    "This is the parameterization that we are going to use in order to estimate the quantum state. In other words, we are going to estimate each $x_i$ using SVGD method build in PyMC3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "9cb06ee3",
   "metadata": {},
   "outputs": [],
   "source": [
    "## We define the fucntion that will convert the vector x into a density matrix\n",
    "## the input is the x0, x1,x2,x3 from the Cholesky decompostion\n",
    "def vec2rho(x): \n",
    "    '''Generate a matrix using the Cholesky decomposition.  '''\n",
    "    ##INPUT\n",
    "    #x(list): list of 4 random numbers coming from a normal distribution\n",
    "    #OUTPUT\n",
    "    # rho(numpy matrix): normalized quantum state \n",
    "    \n",
    "    d = int(np.sqrt(x.shape[0])) \n",
    "    X = 1j*np.zeros([d,d]) \n",
    "    # the first d-1 entries are the diagonal of X (the dth entry is fixed  by normalisation) \n",
    "    # the first d entries are the diagonal of X \n",
    "    np.fill_diagonal(X,x[:d]) \n",
    "    # indices of the upper triangle of d x d matrix \n",
    "    idu = np.triu_indices(d,1) \n",
    "    # take the next d (d-1)/2 entries and make them the real part of the upper triangle \n",
    "    X[idu] = x[d:d+int(d*(d-1)/2)] \n",
    "    # take the final d (d-1)/2 entrie\n",
    "    # take the final d (d-1)/2 entries and make them the imaginary part of the upper triangle \n",
    "    X[idu] += -1j * x[-int(d*(d-1)/2):]  ## modified in order to get our tilde rho\n",
    "    # now do the X^dag X thing \n",
    "    rho = X.T.conj() @ X \n",
    "    #rho = X@ X.T.conj() \n",
    "    return rho/np.trace(rho)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "fe4defe7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.03333333+0.j        , 0.1       -0.13333333j],\n",
       "       [0.1       +0.13333333j, 0.96666667+0.j        ]])"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "##testing function cell: this is cell test the function that we had created\n",
    "x = np.array([1,2,3,4])\n",
    "vec2rho(x) "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ec6f0719",
   "metadata": {},
   "source": [
    "## Fedelity is our metric\n",
    "\n",
    "We need to know if we are estimating well the random quantum state. Thus, we are using fidelity. We are going to measure the fidelity of the random quatum state and the state given by SVGD. For denisity matrices, fedelity is given by:\n",
    "\n",
    "$$ F(\\rho,\\sigma)=\\biggr(Tr\\sqrt{\\sqrt{\\rho} \\sigma \\sqrt{\\rho} }\\biggr)^2$$\n",
    "\n",
    "For $2\\times 2$ matrices, the square root of matrix is given by,\n",
    "\n",
    "$$\\sqrt\\rho=\\frac{1}{\\tau} (\\rho+det(\\rho)^{0.5}\\times I), $$\n",
    "where $I$ is the identity matrix and \n",
    "$$\\tau = (Tr(\\rho)+2*det(\\rho)^{0.5})^{0.5} $$ and $Tr(\\rho)$ is the trace of $\\rho$ and $det(\\rho)$ is the determinant of $\\rho$."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "07a0b043",
   "metadata": {},
   "outputs": [],
   "source": [
    "### fidelity of between mixed states \n",
    "# F(rho1, rho2) = (tr(sqrt( sqrt(rho1) rho2 sqrt(rho1))  )**2, \n",
    "#where\n",
    "# sqrt A = 1/tau (A +det (A)**(0.5) * Identity) \n",
    "#and\n",
    "# tau = (Tr A + 2 det(A)**(0.5) )**(0.5)\n",
    "\n",
    "def sqrtmatrix(rho1):\n",
    "    '''Generates the square root of a matrix 2x2'''\n",
    "    #INPUT\n",
    "    #rho1(numpy matrix 2x2 )\n",
    "    #OUTPUT\n",
    "    # (numpy matrix 2x2): the square root of rho1\n",
    "    tracerho1 = np.trace(rho1)\n",
    "    detrho1 = np.linalg.det(rho1)\n",
    "    tau = (tracerho1+2.0*detrho1**(0.5))**0.5\n",
    "    identity = np.array([[1.0,0.0],[0.0,1.0]])\n",
    "    return 1/tau*(rho1+detrho1**(0.5)*identity)\n",
    "\n",
    "def fidelity(rho1, rho2):\n",
    "    '''Gives the fidelity of two density matrices'''\n",
    "    #INPUT\n",
    "    #rho1(numpy matrix 2x2 ): density matrix \n",
    "    #rho2(numpy matrix 2x2 ): density matrix\n",
    "    #OUTPUT\n",
    "    # (float): fidelity between rho1 and rho2\n",
    "    sqrt1 = sqrtmatrix(rho1)\n",
    "    aux = np.matmul(sqrt1,rho2)\n",
    "    aux2 = np.matmul(aux,sqrt1)\n",
    "    bigsqrt=sqrtmatrix(aux2)\n",
    "\n",
    "    return ((np.trace(bigsqrt))**2).real"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ae113d94",
   "metadata": {},
   "source": [
    "## Using SVGD to estimate a random quatum state\n",
    "\n",
    "This is where PyMC will estimate $x_i$ with $i \\in \\lbrace0,1,2,3\\rbrace$ using SVGD. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "86ac9096",
   "metadata": {},
   "outputs": [],
   "source": [
    "## function that uses PyMC3  and estimates a random quantum state using SVGD and SMC \n",
    "def svgd_prestates(N=1000,z_observed=700,x_observed=200,y_observed=400,\n",
    "                   svgd_num_part = 1000,svgd_lear_rate = 0.01,svgd_iter = 200):\n",
    "  ''' Generate the posteior of the estimated xi with i\\in{0,1,2,3} with svgd approximation and SMC '''\n",
    "##INPUT\n",
    "# N(int): number of total coin tosses. This is for the 3 axis: x,y,z\n",
    "# z_observed(int): number of succes (heads) in N number of tosses of z coin\n",
    "# x_observed(int): number of succes (heads) in N number of tosses of x coin\n",
    "# y_observed(int): number of succes (heads) in N number of tosses of y coin\n",
    "#svgd_num_part(int): number of particles for SVGD \n",
    "#svgd_lear_rate(float): learning rate of we are going to update the particles\n",
    "#svgd_iter(int): total number of SVGD algorithm. This is max_iterations in the SVGD algorithm\n",
    "##OUTPUT\n",
    "#svgd_approx(pymc approximation): this is a pymc3 approximation result. This contains the modified priors\n",
    "#\n",
    "#sms(pymc approximation): this is a pymc3 sampling using SMC. This contains the modified priors \n",
    "\n",
    "  with pm.Model() as Binconstrain:\n",
    "    #start with xi normal random variables\n",
    "    x0 = pm.Normal(\"x0\",mu = 0.0, sigma = 1.0)\n",
    "    x1 = pm.Normal(\"x1\",mu = 0.0, sigma = 1.0)\n",
    "    x2 = pm.Normal(\"x2\",mu = 0.0, sigma = 1.0)\n",
    "    x3 = pm.Normal(\"x3\",mu = 0.0, sigma = 1.0)\n",
    "\n",
    "    #define the probabilities\n",
    "    pz0 = (1+1/(x0**2+x1**2+x2**2+x3**2)*(x0**2-(x1**2+x2**2+x3**2)))/2\n",
    "    px0 = (1+1/(x0**2+x1**2+x2**2+x3**2)*2*x0*x1)/2\n",
    "    py0 = (1+1/(x0**2+x1**2+x2**2+x3**2)*2*x0*x2)/2\n",
    "    \n",
    "    #\n",
    "    pz = pm.Deterministic('pz',pz0)\n",
    "    px = pm.Deterministic('px',px0)\n",
    "    py = pm.Deterministic('py',py0)\n",
    "\n",
    "    zxy = pm.Binomial(\"z\",n=N,p=pz,observed=z_observed)*pm.Binomial(\n",
    "        \"x\",n=N,p=px,observed=x_observed)*pm.Binomial(\"y\",n=N,p=py,observed=y_observed)\n",
    "\n",
    "    ##SVGD approx\n",
    "    svgd_approx = pm.fit(svgd_iter,method=\"svgd\",inf_kwargs=dict(\n",
    "        n_particles=svgd_num_part),obj_optimizer = pm.adam(learning_rate=svgd_lear_rate),progressbar=False)\n",
    "    \n",
    "    #comparison with SMS\n",
    "    sms = pm.sample_smc(2000,parallel=True,cores=4)\n",
    "    #az_trace = az.from_pymc3(sms)\n",
    "    \n",
    "    \n",
    "    return svgd_approx, sms\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d0e19bb6",
   "metadata": {},
   "source": [
    "## From SVGD or SMC to a quatum state \n",
    "\n",
    "The following fucntions will transform the output of SVGD and SMC into quantum density matrices."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "4ade2df2",
   "metadata": {},
   "outputs": [],
   "source": [
    "## from svgd_fit to a density matrix\n",
    "def svgd_rho_mean(svgd_state, n_samples = 2000):\n",
    "    '''This function will generate an estimated quantum state using SVGD'''\n",
    "    ### INPUT\n",
    "    #svgd_state(PyMC 3 approximation): appoximation of a variational posterior given by SVGD\n",
    "    #n_samples(int): Draw samples from variational posterior.\n",
    "    ##OUTPUT\n",
    "    # rho(numpy matrix): density matrix or quantum state using SVGD \n",
    "    #\n",
    "    fullsamples = svgd_state.sample(n_samples)\n",
    "    x0_matrix = fullsamples['x0'] # from the posterior we take the estimated prior ['x0']\n",
    "    x1_matrix = fullsamples['x1'] # from the posterior we take the estimated prior ['x1']\n",
    "    x2_matrix = fullsamples['x2'] # from the posterior we take the estimated prior ['x2']\n",
    "    x3_matrix = fullsamples['x3'] # from the posterior we take the estimated prior ['x3']\n",
    "    rho = 1j * np.zeros([2,2])\n",
    "    for i in range(n_samples):\n",
    "        vec = np.array([x0_matrix[i],x3_matrix[i],x1_matrix[i],x2_matrix[i]])\n",
    "        rho += vec2rho(vec)\n",
    "    rho = rho/n_samples\n",
    "    return rho \n",
    "\n",
    "### from sms to a density matrix\n",
    "\n",
    "def sms_rho(sms_sample, n_samples=2000):\n",
    "    '''This function will generate an estimated quantum state using SMC'''\n",
    "    ### INPUT\n",
    "    #sms_sample(PYMC approximation): appoximation of a variational posterior given by SVGD\n",
    "    #n_samples(int): Draw samples from variational posterior.\n",
    "    ##OUTPUT\n",
    "    # rho(numpy matrix): density matrix or quantum state using SMC \n",
    "    \n",
    "    x0_matrix = sms_sample['x0']\n",
    "    x1_matrix = sms_sample['x1']\n",
    "    x2_matrix = sms_sample['x2']\n",
    "    x3_matrix = sms_sample['x3']\n",
    "    rho = 1j * np.zeros([2,2])\n",
    "    for i in range(n_samples):\n",
    "        vec = np.array([x0_matrix[i],x3_matrix[i],x1_matrix[i],x2_matrix[i]])\n",
    "        rho += vec2rho(vec)\n",
    "    rho = rho/n_samples\n",
    "    return rho"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "751e2c8d",
   "metadata": {},
   "source": [
    "## Generate a random state\n",
    "\n",
    "In the following lines, we are going to estimate random quantum states and plot the results."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "581c0d7a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "the 4 random numbers are:\n",
      "[-0.93699007 -0.98076478 -1.77350299  0.89323058]\n",
      " \n",
      "The random density matrix is :\n",
      "[[0.15181511+0.j         0.15890768-0.28735049j]\n",
      " [0.15890768+0.28735049j 0.84818489+0.j        ]]\n"
     ]
    }
   ],
   "source": [
    "# generate the 4 random numbers\n",
    "x = np.random.randn(4)\n",
    "print(\"the 4 random numbers are:\")\n",
    "print(x)\n",
    "print(\" \")\n",
    "print(\"The random density matrix is :\")\n",
    "print(vec2rho(np.array([x[0],x[3],x[1],x[2]])))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "740e9e3f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Probabilities of measuring the random density matrix in the x, y and z axis:\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([0.65890768, 0.78735049, 0.15181511])"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## measure the density matrix in the 3 axis: x,y,z\n",
    "def get_probs(x):\n",
    "    '''Given the 4 random numbers, we get the probabilities of the density matrix '''\n",
    "    pz = (1+1/(x[0]**2+x[1]**2+x[2]**2+x[3]**2)*(x[0]**2-(x[1]**2+x[2]**2+x[3]**2)))/2\n",
    "    px = (1+1/(x[0]**2+x[1]**2+x[2]**2+x[3]**2)*2*x[0]*x[1])/2\n",
    "    py = (1+1/(x[0]**2+x[1]**2+x[2]**2+x[3]**2)*2*x[0]*x[2])/2\n",
    "    return np.array([px,py,pz])\n",
    "\n",
    "print(\"Probabilities of measuring the random density matrix in the x, y and z axis:\")\n",
    "get_probs(x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "e5ec1690",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of successes in 1000 toses is\n",
      "[637 806 149]\n"
     ]
    }
   ],
   "source": [
    "## Using the probabilities to get the number of successes \n",
    "def sim_experiment(x, N = 1000):\n",
    "    '''Get the number of successes of 3 binomial distributions using the 4 random numbers'''\n",
    "    p = get_probs(x)\n",
    "    nx = np.random.binomial(N,p[0])\n",
    "    ny = np.random.binomial(N,p[1])\n",
    "    nz = np.random.binomial(N,p[2])\n",
    "    return np.array([nx,ny,nz])\n",
    "\n",
    "n = sim_experiment(x,N = 1000)\n",
    "print(\"Number of successes in \"+str(1000)+\" toses is\")\n",
    "print(n)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "026c4745",
   "metadata": {},
   "source": [
    "## The whole simulation process\n",
    "\n",
    "In the following line, we are simulating the estimation process. We are using only 2 number of simulations because if we use grater numbers like 50 or 100 it will take more time the simulation. The reader is free to change \n",
    "n_sim\n",
    "\n",
    "in order to generate more data and get better results. \n",
    "\n",
    "We want to explore:\n",
    "- how SVGD performs with 10, 100 and 1000 number of measurements $N$ (or total of tosses) and\n",
    "- how SVGD performs with 10, 100, 1000 number of particles\n",
    "- how SVGD performs againts SMC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "853876b3",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.347\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.477\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.013\n",
      "Stage:   1 Beta: 0.042\n",
      "Stage:   2 Beta: 0.116\n",
      "Stage:   3 Beta: 0.326\n",
      "Stage:   4 Beta: 0.902\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.015\n",
      "Stage:   1 Beta: 0.049\n",
      "Stage:   2 Beta: 0.140\n",
      "Stage:   3 Beta: 0.400\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.026\n",
      "Stage:   2 Beta: 0.070\n",
      "Stage:   3 Beta: 0.163\n",
      "Stage:   4 Beta: 0.413\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.003\n",
      "Stage:   2 Beta: 0.007\n",
      "Stage:   3 Beta: 0.019\n",
      "Stage:   4 Beta: 0.054\n",
      "Stage:   5 Beta: 0.155\n",
      "Stage:   6 Beta: 0.430\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.172\n",
      "Stage:   1 Beta: 0.473\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.386\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.038\n",
      "Stage:   1 Beta: 0.128\n",
      "Stage:   2 Beta: 0.385\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.050\n",
      "Stage:   1 Beta: 0.200\n",
      "Stage:   2 Beta: 0.606\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.021\n",
      "Stage:   2 Beta: 0.062\n",
      "Stage:   3 Beta: 0.185\n",
      "Stage:   4 Beta: 0.552\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.009\n",
      "Stage:   2 Beta: 0.027\n",
      "Stage:   3 Beta: 0.078\n",
      "Stage:   4 Beta: 0.236\n",
      "Stage:   5 Beta: 0.637\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.533\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.391\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.013\n",
      "Stage:   1 Beta: 0.043\n",
      "Stage:   2 Beta: 0.117\n",
      "Stage:   3 Beta: 0.316\n",
      "Stage:   4 Beta: 0.796\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.024\n",
      "Stage:   1 Beta: 0.073\n",
      "Stage:   2 Beta: 0.201\n",
      "Stage:   3 Beta: 0.549\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.015\n",
      "Stage:   2 Beta: 0.044\n",
      "Stage:   3 Beta: 0.123\n",
      "Stage:   4 Beta: 0.334\n",
      "Stage:   5 Beta: 0.894\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.011\n",
      "Stage:   2 Beta: 0.032\n",
      "Stage:   3 Beta: 0.093\n",
      "Stage:   4 Beta: 0.262\n",
      "Stage:   5 Beta: 0.784\n",
      "Stage:   6 Beta: 1.000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--- 892.1599922180176 seconds ---\n"
     ]
    }
   ],
   "source": [
    "## running the simulations\n",
    "start_time = time.time()\n",
    "\n",
    "n_sim = 2\n",
    "N = np.array([10,100,1000])  ## number of measurements \n",
    "#N = np.array([100,1000])  ## number of measurements \n",
    "\n",
    "particle_num = np.array([10,100,1000]) ## number of particles in SVGD\n",
    "\n",
    "statistics = np.zeros([n_sim,N.shape[0],particle_num.shape[0]])\n",
    "statistics_sms = np.zeros([n_sim,N.shape[0],particle_num.shape[0]])\n",
    "\n",
    "for idP in range(particle_num.shape[0]):\n",
    "    for idN in range(N.shape[0]):\n",
    "        for idx in range(n_sim):\n",
    "            true_state = np.random.randn(4)\n",
    "            data = sim_experiment(true_state, N = N[idN])\n",
    "            svgd, sms_sample = svgd_prestates(N[idN],z_observed=data[2],x_observed=data[0],y_observed=data[1], svgd_num_part = particle_num[idP], svgd_lear_rate = 0.01, svgd_iter = 400)\n",
    "            svgd_state = svgd_rho_mean(svgd)\n",
    "            sms_state = sms_rho(sms_sample)\n",
    "            \n",
    "            #print(\"svgd_state:\",svgd_state)\n",
    "            #print(\"sms_state\",sms_state)\n",
    "            #print(\"true_state:\",vec2rho(np.array([true_state[0],true_state[3],true_state[1],true_state[2]]) ) )\n",
    "            statistics[idx,idN,idP] = fidelity(vec2rho(np.array([true_state[0],true_state[3],true_state[1],true_state[2]])),svgd_state)\n",
    "            statistics_sms[idx,idN,idP] = fidelity(vec2rho(np.array([true_state[0],true_state[3],true_state[1],true_state[2]])),sms_state)\n",
    "\n",
    "print(\"--- %s seconds ---\" % (time.time() - start_time))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "471f6beb",
   "metadata": {},
   "source": [
    "## Results\n",
    "\n",
    "Print all the simulations:\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "d8baef1e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "svgd-fidelity: [[[0.93658131 0.92045192 0.97902348]\n",
      "  [0.98845669 0.99085365 0.99754292]\n",
      "  [0.9939418  0.99354045 0.97610322]]\n",
      "\n",
      " [[0.92631615 0.93792516 0.94577097]\n",
      "  [0.97720706 0.99587413 0.99337525]\n",
      "  [0.99931794 0.99735132 0.99782105]]]\n",
      "\n",
      "sms-fidelity: [[[0.91882275 0.91799373 0.97955582]\n",
      "  [0.98690599 0.99539279 0.99728804]\n",
      "  [0.99595753 0.99961864 0.99834183]]\n",
      "\n",
      " [[0.9306649  0.93576403 0.94675815]\n",
      "  [0.98193459 0.99930122 0.99607755]\n",
      "  [0.99992763 0.99949577 0.99888651]]]\n"
     ]
    }
   ],
   "source": [
    "print(\"svgd-fidelity:\",statistics)\n",
    "print(\"\")\n",
    "print(\"sms-fidelity:\",statistics_sms)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7cfb00b2",
   "metadata": {},
   "source": [
    "## Average of fidelilty\n",
    "\n",
    "In the case of SVGD, we get the mean with respect of the number of simulations. Deu to the fact that we are interested in the performance of SVGD with respect to particles and measurements. \n",
    "\n",
    "In the case of SMC, we get the mean with respect of the number of simulations and with respect of particles. SMC does not work with particles"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "id": "2fe60a44",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "svgd: [[0.93144873 0.92918854 0.96239722]\n",
      " [0.98283187 0.99336389 0.99545908]\n",
      " [0.99662987 0.99544588 0.98696214]]\n",
      "\n",
      "ave_sms [0.9382599  0.9928167  0.99870465]\n"
     ]
    }
   ],
   "source": [
    "ave_statistics = np.mean(statistics, axis = 0)\n",
    "ave_statistics_sms=np.mean(np.mean(statistics_sms, axis = 0),axis=1)\n",
    "\n",
    "print(\"svgd:\",ave_statistics)\n",
    "print(\"\")\n",
    "print(\"ave_sms\",ave_statistics_sms)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eaa74678",
   "metadata": {},
   "source": [
    "## Plots\n",
    "\n",
    "Plot the results:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "id": "74ae4dd3",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams[\"figure.figsize\"] = 8, 5\n",
    "plt.plot(N, ave_statistics[:,0], label = str(particle_num[0])+\" particles-SVGD\")\n",
    "plt.plot(N, ave_statistics[:,1], label = str(particle_num[1])+\" particles-SVGD\")\n",
    "plt.plot(N, ave_statistics[:,2], label = str(particle_num[2])+\" particles-SVGD\")\n",
    "\n",
    "plt.plot(N, ave_statistics_sms, label = \"SMC\")\n",
    "#plt.plot(N, ave_statistics_sms[:,1], label = \"SMC and SVGD uses \"+str(particle_num[1])+\" part.\")\n",
    "#plt.plot(N, ave_statistics_sms[:,2], label = \"SMC and SVGD uses \"+str(particle_num[2])+\" part.\")\n",
    "plt.legend(fontsize=10,loc=0)\n",
    "plt.ylabel(\"Average Fidelity\",fontsize=16)\n",
    "plt.xlabel(\"Number of measurements\",fontsize=16)\n",
    "plt.title(\"SVGD using adam vs SMC: Average Fidelity vs $N$\",fontsize=16)\n",
    "plt.legend(fontsize=13)\n",
    "plt.grid()\n",
    "plt.savefig('svgd-adam-measurements.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "id": "a06375a8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(particle_num, ave_statistics[0,:], label = str(N[0])+\" measurements-SVGD\")\n",
    "plt.plot(particle_num, ave_statistics[1,:], label = str(N[1])+\" measurements-SVGD\")\n",
    "plt.plot(particle_num, ave_statistics[2,:], label = str(N[2])+\" measurements-SVGD\")\n",
    "\n",
    "#plt.plot(particle_num, ave_statistics_sms[0,:], label = str(N[0])+\" measurements-SMC\")\n",
    "#plt.plot(particle_num, ave_statistics_sms[1,:], label = str(N[1])+\" measurements-SMC\")\n",
    "#plt.plot(particle_num, ave_statistics_sms[2,:], label = str(N[2])+\" measurements-SMC\")\n",
    "plt.legend(fontsize=10,loc=0)\n",
    "plt.ylabel(\"Average Fidelity\",fontsize=16)\n",
    "plt.xlabel(\"Number of particles\",fontsize=16)\n",
    "plt.title(\"Aver. fidelity vs num. particles :using adam \",fontsize=16)\n",
    "plt.legend(fontsize=11)\n",
    "plt.grid()\n",
    "plt.savefig('svgd-adam-particles.png')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f9f25551",
   "metadata": {},
   "source": [
    "## Extra plots\n",
    "\n",
    "The following plots are the modified priors of $x_0, x_1, x_2, x_3$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "id": "4a11f073",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "n_samples = 1000\n",
    "fullsamples = svgd.sample(n_samples)\n",
    "x0_matrix = fullsamples['x0']\n",
    "x1_matrix = fullsamples['x1']\n",
    "x2_matrix = fullsamples['x2']\n",
    "x3_matrix = fullsamples['x3']\n",
    "ax =sns.kdeplot(x0_matrix, label=\"SVGD\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "id": "0492e605",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "bx =sns.kdeplot(x1_matrix, label=\"SVGD\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "id": "65dbdad3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "cx =sns.kdeplot(x2_matrix, label=\"SVGD\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "id": "b0ec2264",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "dx =sns.kdeplot(x3_matrix, label=\"SVGD\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "bd058d2b",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "011b2646",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "8ca1168d",
   "metadata": {},
   "source": [
    "## More simulations\n",
    "\n",
    "### Be careful:\n",
    "The following simulation is with n_sim=70. This will take 30676.97034263611 seconds (approx 8 hours), be careful with your time."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "d3f33629",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.176\n",
      "Stage:   1 Beta: 0.517\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.213\n",
      "Stage:   1 Beta: 0.613\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.539\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.231\n",
      "Stage:   1 Beta: 0.740\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.458\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.345\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.167\n",
      "Stage:   1 Beta: 0.480\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.242\n",
      "Stage:   1 Beta: 0.731\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.343\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.329\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.395\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.393\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.230\n",
      "Stage:   1 Beta: 0.756\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.293\n",
      "Stage:   1 Beta: 0.953\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.245\n",
      "Stage:   1 Beta: 0.756\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.396\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.266\n",
      "Stage:   1 Beta: 0.802\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.502\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.253\n",
      "Stage:   1 Beta: 0.798\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.421\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.375\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.323\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.385\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.317\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.376\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.212\n",
      "Stage:   1 Beta: 0.681\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.335\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.245\n",
      "Stage:   1 Beta: 0.726\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.225\n",
      "Stage:   1 Beta: 0.680\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.109\n",
      "Stage:   1 Beta: 0.326\n",
      "Stage:   2 Beta: 0.956\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.388\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.424\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.127\n",
      "Stage:   1 Beta: 0.397\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.402\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.389\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.138\n",
      "Stage:   1 Beta: 0.429\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.209\n",
      "Stage:   1 Beta: 0.619\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.388\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.369\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.224\n",
      "Stage:   1 Beta: 0.685\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.084\n",
      "Stage:   1 Beta: 0.241\n",
      "Stage:   2 Beta: 0.620\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.357\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.089\n",
      "Stage:   1 Beta: 0.251\n",
      "Stage:   2 Beta: 0.646\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.433\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.396\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.372\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.098\n",
      "Stage:   1 Beta: 0.290\n",
      "Stage:   2 Beta: 0.761\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.383\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.391\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.190\n",
      "Stage:   1 Beta: 0.586\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.328\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.387\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.206\n",
      "Stage:   1 Beta: 0.583\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.190\n",
      "Stage:   1 Beta: 0.545\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.302\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.194\n",
      "Stage:   1 Beta: 0.588\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.167\n",
      "Stage:   1 Beta: 0.482\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.272\n",
      "Stage:   1 Beta: 0.893\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.278\n",
      "Stage:   1 Beta: 0.878\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.123\n",
      "Stage:   1 Beta: 0.391\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.185\n",
      "Stage:   1 Beta: 0.567\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.227\n",
      "Stage:   1 Beta: 0.686\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.555\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.376\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.645\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.502\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.385\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.424\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.235\n",
      "Stage:   1 Beta: 0.776\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.160\n",
      "Stage:   1 Beta: 0.454\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.039\n",
      "Stage:   1 Beta: 0.313\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.047\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Stage:   1 Beta: 0.170\n",
      "Stage:   2 Beta: 0.449\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.021\n",
      "Stage:   1 Beta: 0.068\n",
      "Stage:   2 Beta: 0.210\n",
      "Stage:   3 Beta: 0.565\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.034\n",
      "Stage:   1 Beta: 0.100\n",
      "Stage:   2 Beta: 0.235\n",
      "Stage:   3 Beta: 0.586\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.040\n",
      "Stage:   1 Beta: 0.139\n",
      "Stage:   2 Beta: 0.383\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.012\n",
      "Stage:   1 Beta: 0.036\n",
      "Stage:   2 Beta: 0.097\n",
      "Stage:   3 Beta: 0.253\n",
      "Stage:   4 Beta: 0.662\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.021\n",
      "Stage:   1 Beta: 0.065\n",
      "Stage:   2 Beta: 0.191\n",
      "Stage:   3 Beta: 0.589\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.024\n",
      "Stage:   1 Beta: 0.074\n",
      "Stage:   2 Beta: 0.206\n",
      "Stage:   3 Beta: 0.609\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.045\n",
      "Stage:   1 Beta: 0.168\n",
      "Stage:   2 Beta: 0.496\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.049\n",
      "Stage:   1 Beta: 0.168\n",
      "Stage:   2 Beta: 0.436\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.026\n",
      "Stage:   1 Beta: 0.085\n",
      "Stage:   2 Beta: 0.227\n",
      "Stage:   3 Beta: 0.562\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.029\n",
      "Stage:   1 Beta: 0.093\n",
      "Stage:   2 Beta: 0.273\n",
      "Stage:   3 Beta: 0.796\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.036\n",
      "Stage:   1 Beta: 0.117\n",
      "Stage:   2 Beta: 0.334\n",
      "Stage:   3 Beta: 0.999\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.036\n",
      "Stage:   1 Beta: 0.124\n",
      "Stage:   2 Beta: 0.386\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.023\n",
      "Stage:   1 Beta: 0.075\n",
      "Stage:   2 Beta: 0.190\n",
      "Stage:   3 Beta: 0.463\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.028\n",
      "Stage:   1 Beta: 0.085\n",
      "Stage:   2 Beta: 0.232\n",
      "Stage:   3 Beta: 0.645\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.039\n",
      "Stage:   1 Beta: 0.299\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.028\n",
      "Stage:   1 Beta: 0.087\n",
      "Stage:   2 Beta: 0.256\n",
      "Stage:   3 Beta: 0.759\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.039\n",
      "Stage:   1 Beta: 0.136\n",
      "Stage:   2 Beta: 0.372\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.014\n",
      "Stage:   1 Beta: 0.045\n",
      "Stage:   2 Beta: 0.126\n",
      "Stage:   3 Beta: 0.357\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.034\n",
      "Stage:   1 Beta: 0.110\n",
      "Stage:   2 Beta: 0.325\n",
      "Stage:   3 Beta: 0.947\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.042\n",
      "Stage:   1 Beta: 0.378\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.041\n",
      "Stage:   1 Beta: 0.189\n",
      "Stage:   2 Beta: 0.522\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.042\n",
      "Stage:   1 Beta: 0.142\n",
      "Stage:   2 Beta: 0.386\n",
      "Stage:   3 Beta: 0.913\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.042\n",
      "Stage:   1 Beta: 0.350\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.029\n",
      "Stage:   1 Beta: 0.094\n",
      "Stage:   2 Beta: 0.249\n",
      "Stage:   3 Beta: 0.595\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.014\n",
      "Stage:   1 Beta: 0.043\n",
      "Stage:   2 Beta: 0.121\n",
      "Stage:   3 Beta: 0.338\n",
      "Stage:   4 Beta: 0.969\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.015\n",
      "Stage:   1 Beta: 0.045\n",
      "Stage:   2 Beta: 0.128\n",
      "Stage:   3 Beta: 0.362\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.031\n",
      "Stage:   1 Beta: 0.100\n",
      "Stage:   2 Beta: 0.297\n",
      "Stage:   3 Beta: 0.892\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.028\n",
      "Stage:   1 Beta: 0.087\n",
      "Stage:   2 Beta: 0.235\n",
      "Stage:   3 Beta: 0.657\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.020\n",
      "Stage:   1 Beta: 0.064\n",
      "Stage:   2 Beta: 0.184\n",
      "Stage:   3 Beta: 0.528\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.016\n",
      "Stage:   1 Beta: 0.050\n",
      "Stage:   2 Beta: 0.138\n",
      "Stage:   3 Beta: 0.363\n",
      "Stage:   4 Beta: 0.905\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.044\n",
      "Stage:   1 Beta: 0.159\n",
      "Stage:   2 Beta: 0.481\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.029\n",
      "Stage:   1 Beta: 0.089\n",
      "Stage:   2 Beta: 0.248\n",
      "Stage:   3 Beta: 0.724\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.021\n",
      "Stage:   1 Beta: 0.065\n",
      "Stage:   2 Beta: 0.187\n",
      "Stage:   3 Beta: 0.554\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.014\n",
      "Stage:   1 Beta: 0.041\n",
      "Stage:   2 Beta: 0.112\n",
      "Stage:   3 Beta: 0.302\n",
      "Stage:   4 Beta: 0.818\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.040\n",
      "Stage:   1 Beta: 0.317\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.037\n",
      "Stage:   1 Beta: 0.126\n",
      "Stage:   2 Beta: 0.368\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.037\n",
      "Stage:   1 Beta: 0.122\n",
      "Stage:   2 Beta: 0.354\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.018\n",
      "Stage:   1 Beta: 0.050\n",
      "Stage:   2 Beta: 0.128\n",
      "Stage:   3 Beta: 0.339\n",
      "Stage:   4 Beta: 0.882\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.025\n",
      "Stage:   1 Beta: 0.080\n",
      "Stage:   2 Beta: 0.238\n",
      "Stage:   3 Beta: 0.692\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.039\n",
      "Stage:   1 Beta: 0.129\n",
      "Stage:   2 Beta: 0.379\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.040\n",
      "Stage:   1 Beta: 0.333\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.012\n",
      "Stage:   1 Beta: 0.039\n",
      "Stage:   2 Beta: 0.111\n",
      "Stage:   3 Beta: 0.308\n",
      "Stage:   4 Beta: 0.846\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.047\n",
      "Stage:   1 Beta: 0.210\n",
      "Stage:   2 Beta: 0.642\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.040\n",
      "Stage:   1 Beta: 0.141\n",
      "Stage:   2 Beta: 0.415\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.012\n",
      "Stage:   1 Beta: 0.040\n",
      "Stage:   2 Beta: 0.110\n",
      "Stage:   3 Beta: 0.315\n",
      "Stage:   4 Beta: 0.888\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.028\n",
      "Stage:   1 Beta: 0.090\n",
      "Stage:   2 Beta: 0.259\n",
      "Stage:   3 Beta: 0.780\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.048\n",
      "Stage:   1 Beta: 0.194\n",
      "Stage:   2 Beta: 0.589\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.025\n",
      "Stage:   1 Beta: 0.077\n",
      "Stage:   2 Beta: 0.229\n",
      "Stage:   3 Beta: 0.683\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.038\n",
      "Stage:   1 Beta: 0.295\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.036\n",
      "Stage:   1 Beta: 0.120\n",
      "Stage:   2 Beta: 0.346\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.024\n",
      "Stage:   1 Beta: 0.077\n",
      "Stage:   2 Beta: 0.217\n",
      "Stage:   3 Beta: 0.631\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.029\n",
      "Stage:   1 Beta: 0.095\n",
      "Stage:   2 Beta: 0.278\n",
      "Stage:   3 Beta: 0.723\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.039\n",
      "Stage:   1 Beta: 0.309\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.039\n",
      "Stage:   1 Beta: 0.133\n",
      "Stage:   2 Beta: 0.393\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.016\n",
      "Stage:   1 Beta: 0.053\n",
      "Stage:   2 Beta: 0.145\n",
      "Stage:   3 Beta: 0.368\n",
      "Stage:   4 Beta: 0.921\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.011\n",
      "Stage:   1 Beta: 0.035\n",
      "Stage:   2 Beta: 0.091\n",
      "Stage:   3 Beta: 0.251\n",
      "Stage:   4 Beta: 0.719\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.040\n",
      "Stage:   1 Beta: 0.341\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.024\n",
      "Stage:   1 Beta: 0.074\n",
      "Stage:   2 Beta: 0.202\n",
      "Stage:   3 Beta: 0.559\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.042\n",
      "Stage:   1 Beta: 0.352\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.025\n",
      "Stage:   1 Beta: 0.078\n",
      "Stage:   2 Beta: 0.221\n",
      "Stage:   3 Beta: 0.667\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.043\n",
      "Stage:   1 Beta: 0.423\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.040\n",
      "Stage:   1 Beta: 0.324\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.050\n",
      "Stage:   1 Beta: 0.217\n",
      "Stage:   2 Beta: 0.649\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.014\n",
      "Stage:   1 Beta: 0.043\n",
      "Stage:   2 Beta: 0.132\n",
      "Stage:   3 Beta: 0.367\n",
      "Stage:   4 Beta: 0.917\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.034\n",
      "Stage:   1 Beta: 0.116\n",
      "Stage:   2 Beta: 0.317\n",
      "Stage:   3 Beta: 0.864\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.012\n",
      "Stage:   1 Beta: 0.038\n",
      "Stage:   2 Beta: 0.105\n",
      "Stage:   3 Beta: 0.291\n",
      "Stage:   4 Beta: 0.824\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.039\n",
      "Stage:   1 Beta: 0.296\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.038\n",
      "Stage:   1 Beta: 0.127\n",
      "Stage:   2 Beta: 0.376\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.006\n",
      "Stage:   2 Beta: 0.019\n",
      "Stage:   3 Beta: 0.056\n",
      "Stage:   4 Beta: 0.166\n",
      "Stage:   5 Beta: 0.496\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.021\n",
      "Stage:   2 Beta: 0.066\n",
      "Stage:   3 Beta: 0.182\n",
      "Stage:   4 Beta: 0.436\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.005\n",
      "Stage:   2 Beta: 0.016\n",
      "Stage:   3 Beta: 0.048\n",
      "Stage:   4 Beta: 0.143\n",
      "Stage:   5 Beta: 0.362\n",
      "Stage:   6 Beta: 0.857\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.004\n",
      "Stage:   2 Beta: 0.014\n",
      "Stage:   3 Beta: 0.039\n",
      "Stage:   4 Beta: 0.096\n",
      "Stage:   5 Beta: 0.238\n",
      "Stage:   6 Beta: 0.602\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.017\n",
      "Stage:   2 Beta: 0.049\n",
      "Stage:   3 Beta: 0.140\n",
      "Stage:   4 Beta: 0.400\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.012\n",
      "Stage:   2 Beta: 0.030\n",
      "Stage:   3 Beta: 0.072\n",
      "Stage:   4 Beta: 0.175\n",
      "Stage:   5 Beta: 0.443\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.004\n",
      "Stage:   2 Beta: 0.010\n",
      "Stage:   3 Beta: 0.026\n",
      "Stage:   4 Beta: 0.076\n",
      "Stage:   5 Beta: 0.218\n",
      "Stage:   6 Beta: 0.618\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.008\n",
      "Stage:   2 Beta: 0.024\n",
      "Stage:   3 Beta: 0.069\n",
      "Stage:   4 Beta: 0.193\n",
      "Stage:   5 Beta: 0.559\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.035\n",
      "Stage:   2 Beta: 0.281\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.007\n",
      "Stage:   2 Beta: 0.019\n",
      "Stage:   3 Beta: 0.056\n",
      "Stage:   4 Beta: 0.167\n",
      "Stage:   5 Beta: 0.477\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.024\n",
      "Stage:   2 Beta: 0.073\n",
      "Stage:   3 Beta: 0.199\n",
      "Stage:   4 Beta: 0.477\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.007\n",
      "Stage:   2 Beta: 0.021\n",
      "Stage:   3 Beta: 0.060\n",
      "Stage:   4 Beta: 0.161\n",
      "Stage:   5 Beta: 0.389\n",
      "Stage:   6 Beta: 0.967\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.035\n",
      "Stage:   2 Beta: 0.112\n",
      "Stage:   3 Beta: 0.317\n",
      "Stage:   4 Beta: 0.873\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.004\n",
      "Stage:   2 Beta: 0.012\n",
      "Stage:   3 Beta: 0.036\n",
      "Stage:   4 Beta: 0.094\n",
      "Stage:   5 Beta: 0.230\n",
      "Stage:   6 Beta: 0.565\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.035\n",
      "Stage:   2 Beta: 0.123\n",
      "Stage:   3 Beta: 0.386\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.015\n",
      "Stage:   2 Beta: 0.044\n",
      "Stage:   3 Beta: 0.127\n",
      "Stage:   4 Beta: 0.364\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.020\n",
      "Stage:   2 Beta: 0.059\n",
      "Stage:   3 Beta: 0.174\n",
      "Stage:   4 Beta: 0.497\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.005\n",
      "Stage:   2 Beta: 0.013\n",
      "Stage:   3 Beta: 0.035\n",
      "Stage:   4 Beta: 0.098\n",
      "Stage:   5 Beta: 0.262\n",
      "Stage:   6 Beta: 0.724\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.004\n",
      "Stage:   2 Beta: 0.012\n",
      "Stage:   3 Beta: 0.034\n",
      "Stage:   4 Beta: 0.096\n",
      "Stage:   5 Beta: 0.276\n",
      "Stage:   6 Beta: 0.797\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.007\n",
      "Stage:   2 Beta: 0.019\n",
      "Stage:   3 Beta: 0.054\n",
      "Stage:   4 Beta: 0.156\n",
      "Stage:   5 Beta: 0.438\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.008\n",
      "Stage:   2 Beta: 0.025\n",
      "Stage:   3 Beta: 0.067\n",
      "Stage:   4 Beta: 0.158\n",
      "Stage:   5 Beta: 0.385\n",
      "Stage:   6 Beta: 0.939\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.012\n",
      "Stage:   2 Beta: 0.034\n",
      "Stage:   3 Beta: 0.097\n",
      "Stage:   4 Beta: 0.255\n",
      "Stage:   5 Beta: 0.658\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.031\n",
      "Stage:   2 Beta: 0.227\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.005\n",
      "Stage:   2 Beta: 0.014\n",
      "Stage:   3 Beta: 0.040\n",
      "Stage:   4 Beta: 0.115\n",
      "Stage:   5 Beta: 0.303\n",
      "Stage:   6 Beta: 0.755\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.039\n",
      "Stage:   2 Beta: 0.162\n",
      "Stage:   3 Beta: 0.465\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.012\n",
      "Stage:   2 Beta: 0.034\n",
      "Stage:   3 Beta: 0.104\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Stage:   4 Beta: 0.312\n",
      "Stage:   5 Beta: 0.910\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.004\n",
      "Stage:   2 Beta: 0.010\n",
      "Stage:   3 Beta: 0.028\n",
      "Stage:   4 Beta: 0.079\n",
      "Stage:   5 Beta: 0.207\n",
      "Stage:   6 Beta: 0.502\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.004\n",
      "Stage:   2 Beta: 0.011\n",
      "Stage:   3 Beta: 0.031\n",
      "Stage:   4 Beta: 0.085\n",
      "Stage:   5 Beta: 0.236\n",
      "Stage:   6 Beta: 0.655\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.013\n",
      "Stage:   2 Beta: 0.037\n",
      "Stage:   3 Beta: 0.095\n",
      "Stage:   4 Beta: 0.229\n",
      "Stage:   5 Beta: 0.579\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.023\n",
      "Stage:   2 Beta: 0.070\n",
      "Stage:   3 Beta: 0.208\n",
      "Stage:   4 Beta: 0.624\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.012\n",
      "Stage:   2 Beta: 0.033\n",
      "Stage:   3 Beta: 0.095\n",
      "Stage:   4 Beta: 0.261\n",
      "Stage:   5 Beta: 0.726\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.034\n",
      "Stage:   2 Beta: 0.233\n",
      "Stage:   3 Beta: 0.945\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.006\n",
      "Stage:   1 Beta: 0.028\n",
      "Stage:   2 Beta: 0.079\n",
      "Stage:   3 Beta: 0.182\n",
      "Stage:   4 Beta: 0.444\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.005\n",
      "Stage:   2 Beta: 0.015\n",
      "Stage:   3 Beta: 0.043\n",
      "Stage:   4 Beta: 0.109\n",
      "Stage:   5 Beta: 0.273\n",
      "Stage:   6 Beta: 0.678\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.005\n",
      "Stage:   2 Beta: 0.015\n",
      "Stage:   3 Beta: 0.037\n",
      "Stage:   4 Beta: 0.095\n",
      "Stage:   5 Beta: 0.249\n",
      "Stage:   6 Beta: 0.639\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.003\n",
      "Stage:   2 Beta: 0.008\n",
      "Stage:   3 Beta: 0.022\n",
      "Stage:   4 Beta: 0.060\n",
      "Stage:   5 Beta: 0.165\n",
      "Stage:   6 Beta: 0.448\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.004\n",
      "Stage:   2 Beta: 0.011\n",
      "Stage:   3 Beta: 0.031\n",
      "Stage:   4 Beta: 0.089\n",
      "Stage:   5 Beta: 0.239\n",
      "Stage:   6 Beta: 0.592\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.036\n",
      "Stage:   2 Beta: 0.114\n",
      "Stage:   3 Beta: 0.328\n",
      "Stage:   4 Beta: 0.838\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.029\n",
      "Stage:   2 Beta: 0.090\n",
      "Stage:   3 Beta: 0.270\n",
      "Stage:   4 Beta: 0.823\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.030\n",
      "Stage:   2 Beta: 0.100\n",
      "Stage:   3 Beta: 0.302\n",
      "Stage:   4 Beta: 0.861\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.018\n",
      "Stage:   2 Beta: 0.053\n",
      "Stage:   3 Beta: 0.154\n",
      "Stage:   4 Beta: 0.461\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.041\n",
      "Stage:   2 Beta: 0.148\n",
      "Stage:   3 Beta: 0.445\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.029\n",
      "Stage:   2 Beta: 0.089\n",
      "Stage:   3 Beta: 0.216\n",
      "Stage:   4 Beta: 0.486\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.006\n",
      "Stage:   2 Beta: 0.018\n",
      "Stage:   3 Beta: 0.044\n",
      "Stage:   4 Beta: 0.110\n",
      "Stage:   5 Beta: 0.272\n",
      "Stage:   6 Beta: 0.701\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.013\n",
      "Stage:   2 Beta: 0.037\n",
      "Stage:   3 Beta: 0.106\n",
      "Stage:   4 Beta: 0.301\n",
      "Stage:   5 Beta: 0.866\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.006\n",
      "Stage:   1 Beta: 0.027\n",
      "Stage:   2 Beta: 0.082\n",
      "Stage:   3 Beta: 0.230\n",
      "Stage:   4 Beta: 0.562\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.003\n",
      "Stage:   2 Beta: 0.007\n",
      "Stage:   3 Beta: 0.020\n",
      "Stage:   4 Beta: 0.056\n",
      "Stage:   5 Beta: 0.155\n",
      "Stage:   6 Beta: 0.451\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.010\n",
      "Stage:   2 Beta: 0.027\n",
      "Stage:   3 Beta: 0.074\n",
      "Stage:   4 Beta: 0.212\n",
      "Stage:   5 Beta: 0.606\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.019\n",
      "Stage:   2 Beta: 0.056\n",
      "Stage:   3 Beta: 0.161\n",
      "Stage:   4 Beta: 0.448\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.004\n",
      "Stage:   2 Beta: 0.011\n",
      "Stage:   3 Beta: 0.029\n",
      "Stage:   4 Beta: 0.083\n",
      "Stage:   5 Beta: 0.245\n",
      "Stage:   6 Beta: 0.686\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.007\n",
      "Stage:   2 Beta: 0.018\n",
      "Stage:   3 Beta: 0.045\n",
      "Stage:   4 Beta: 0.112\n",
      "Stage:   5 Beta: 0.284\n",
      "Stage:   6 Beta: 0.740\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.043\n",
      "Stage:   2 Beta: 0.259\n",
      "Stage:   3 Beta: 0.729\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.014\n",
      "Stage:   2 Beta: 0.040\n",
      "Stage:   3 Beta: 0.097\n",
      "Stage:   4 Beta: 0.236\n",
      "Stage:   5 Beta: 0.578\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.008\n",
      "Stage:   2 Beta: 0.023\n",
      "Stage:   3 Beta: 0.057\n",
      "Stage:   4 Beta: 0.139\n",
      "Stage:   5 Beta: 0.347\n",
      "Stage:   6 Beta: 0.879\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.016\n",
      "Stage:   2 Beta: 0.047\n",
      "Stage:   3 Beta: 0.123\n",
      "Stage:   4 Beta: 0.299\n",
      "Stage:   5 Beta: 0.761\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.024\n",
      "Stage:   2 Beta: 0.072\n",
      "Stage:   3 Beta: 0.211\n",
      "Stage:   4 Beta: 0.625\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.009\n",
      "Stage:   2 Beta: 0.026\n",
      "Stage:   3 Beta: 0.067\n",
      "Stage:   4 Beta: 0.168\n",
      "Stage:   5 Beta: 0.422\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.005\n",
      "Stage:   2 Beta: 0.013\n",
      "Stage:   3 Beta: 0.038\n",
      "Stage:   4 Beta: 0.105\n",
      "Stage:   5 Beta: 0.262\n",
      "Stage:   6 Beta: 0.644\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.019\n",
      "Stage:   2 Beta: 0.053\n",
      "Stage:   3 Beta: 0.127\n",
      "Stage:   4 Beta: 0.314\n",
      "Stage:   5 Beta: 0.811\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.004\n",
      "Stage:   2 Beta: 0.012\n",
      "Stage:   3 Beta: 0.032\n",
      "Stage:   4 Beta: 0.079\n",
      "Stage:   5 Beta: 0.195\n",
      "Stage:   6 Beta: 0.486\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.013\n",
      "Stage:   2 Beta: 0.036\n",
      "Stage:   3 Beta: 0.087\n",
      "Stage:   4 Beta: 0.212\n",
      "Stage:   5 Beta: 0.529\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.014\n",
      "Stage:   2 Beta: 0.042\n",
      "Stage:   3 Beta: 0.122\n",
      "Stage:   4 Beta: 0.364\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.005\n",
      "Stage:   2 Beta: 0.014\n",
      "Stage:   3 Beta: 0.042\n",
      "Stage:   4 Beta: 0.113\n",
      "Stage:   5 Beta: 0.277\n",
      "Stage:   6 Beta: 0.708\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.009\n",
      "Stage:   2 Beta: 0.025\n",
      "Stage:   3 Beta: 0.074\n",
      "Stage:   4 Beta: 0.221\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Stage:   5 Beta: 0.639\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.005\n",
      "Stage:   2 Beta: 0.015\n",
      "Stage:   3 Beta: 0.038\n",
      "Stage:   4 Beta: 0.095\n",
      "Stage:   5 Beta: 0.249\n",
      "Stage:   6 Beta: 0.623\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.009\n",
      "Stage:   2 Beta: 0.026\n",
      "Stage:   3 Beta: 0.076\n",
      "Stage:   4 Beta: 0.226\n",
      "Stage:   5 Beta: 0.708\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.004\n",
      "Stage:   2 Beta: 0.012\n",
      "Stage:   3 Beta: 0.033\n",
      "Stage:   4 Beta: 0.099\n",
      "Stage:   5 Beta: 0.298\n",
      "Stage:   6 Beta: 0.875\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.009\n",
      "Stage:   2 Beta: 0.025\n",
      "Stage:   3 Beta: 0.074\n",
      "Stage:   4 Beta: 0.214\n",
      "Stage:   5 Beta: 0.626\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.014\n",
      "Stage:   2 Beta: 0.042\n",
      "Stage:   3 Beta: 0.120\n",
      "Stage:   4 Beta: 0.313\n",
      "Stage:   5 Beta: 0.748\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.008\n",
      "Stage:   2 Beta: 0.023\n",
      "Stage:   3 Beta: 0.060\n",
      "Stage:   4 Beta: 0.144\n",
      "Stage:   5 Beta: 0.350\n",
      "Stage:   6 Beta: 0.879\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.552\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.346\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.104\n",
      "Stage:   1 Beta: 0.306\n",
      "Stage:   2 Beta: 0.806\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.389\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.265\n",
      "Stage:   1 Beta: 0.798\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.127\n",
      "Stage:   1 Beta: 0.364\n",
      "Stage:   2 Beta: 0.955\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.407\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.257\n",
      "Stage:   1 Beta: 0.813\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.279\n",
      "Stage:   1 Beta: 0.890\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.363\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.369\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.283\n",
      "Stage:   1 Beta: 0.859\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.392\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.321\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.413\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.418\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.204\n",
      "Stage:   1 Beta: 0.660\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.119\n",
      "Stage:   1 Beta: 0.360\n",
      "Stage:   2 Beta: 0.986\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.201\n",
      "Stage:   1 Beta: 0.648\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.376\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.513\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.187\n",
      "Stage:   1 Beta: 0.523\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.422\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.384\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.181\n",
      "Stage:   1 Beta: 0.589\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.177\n",
      "Stage:   1 Beta: 0.573\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.089\n",
      "Stage:   1 Beta: 0.260\n",
      "Stage:   2 Beta: 0.658\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.323\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.117\n",
      "Stage:   1 Beta: 0.358\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.394\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.185\n",
      "Stage:   1 Beta: 0.548\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.103\n",
      "Stage:   1 Beta: 0.295\n",
      "Stage:   2 Beta: 0.771\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.350\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.576\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.438\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.385\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.157\n",
      "Stage:   1 Beta: 0.496\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.342\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.322\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.504\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.377\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.177\n",
      "Stage:   1 Beta: 0.553\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.295\n",
      "Stage:   1 Beta: 0.958\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.386\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.420\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.373\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.085\n",
      "Stage:   1 Beta: 0.242\n",
      "Stage:   2 Beta: 0.604\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.406\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.086\n",
      "Stage:   1 Beta: 0.249\n",
      "Stage:   2 Beta: 0.654\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.109\n",
      "Stage:   1 Beta: 0.323\n",
      "Stage:   2 Beta: 0.853\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.321\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.485\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.106\n",
      "Stage:   1 Beta: 0.313\n",
      "Stage:   2 Beta: 0.810\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.376\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.400\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.459\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.319\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.374\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.215\n",
      "Stage:   1 Beta: 0.655\n",
      "Stage:   2 Beta: 1.000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.405\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.405\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.224\n",
      "Stage:   1 Beta: 0.640\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.405\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.266\n",
      "Stage:   1 Beta: 0.855\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.128\n",
      "Stage:   1 Beta: 0.382\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.390\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.107\n",
      "Stage:   1 Beta: 0.325\n",
      "Stage:   2 Beta: 0.901\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.156\n",
      "Stage:   1 Beta: 0.506\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.177\n",
      "Stage:   1 Beta: 0.554\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.308\n",
      "Stage:   1 Beta: 0.999\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.031\n",
      "Stage:   1 Beta: 0.100\n",
      "Stage:   2 Beta: 0.271\n",
      "Stage:   3 Beta: 0.711\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.039\n",
      "Stage:   1 Beta: 0.331\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.017\n",
      "Stage:   1 Beta: 0.053\n",
      "Stage:   2 Beta: 0.152\n",
      "Stage:   3 Beta: 0.426\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.016\n",
      "Stage:   1 Beta: 0.048\n",
      "Stage:   2 Beta: 0.134\n",
      "Stage:   3 Beta: 0.374\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.018\n",
      "Stage:   1 Beta: 0.059\n",
      "Stage:   2 Beta: 0.175\n",
      "Stage:   3 Beta: 0.495\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.036\n",
      "Stage:   1 Beta: 0.118\n",
      "Stage:   2 Beta: 0.318\n",
      "Stage:   3 Beta: 0.846\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.019\n",
      "Stage:   1 Beta: 0.061\n",
      "Stage:   2 Beta: 0.175\n",
      "Stage:   3 Beta: 0.505\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.014\n",
      "Stage:   1 Beta: 0.044\n",
      "Stage:   2 Beta: 0.121\n",
      "Stage:   3 Beta: 0.311\n",
      "Stage:   4 Beta: 0.782\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.014\n",
      "Stage:   1 Beta: 0.044\n",
      "Stage:   2 Beta: 0.122\n",
      "Stage:   3 Beta: 0.336\n",
      "Stage:   4 Beta: 0.941\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.042\n",
      "Stage:   1 Beta: 0.338\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.017\n",
      "Stage:   1 Beta: 0.051\n",
      "Stage:   2 Beta: 0.145\n",
      "Stage:   3 Beta: 0.406\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.045\n",
      "Stage:   1 Beta: 0.160\n",
      "Stage:   2 Beta: 0.466\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.042\n",
      "Stage:   1 Beta: 0.143\n",
      "Stage:   2 Beta: 0.392\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.044\n",
      "Stage:   1 Beta: 0.353\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.043\n",
      "Stage:   1 Beta: 0.144\n",
      "Stage:   2 Beta: 0.428\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.043\n",
      "Stage:   1 Beta: 0.409\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.022\n",
      "Stage:   1 Beta: 0.070\n",
      "Stage:   2 Beta: 0.204\n",
      "Stage:   3 Beta: 0.604\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.026\n",
      "Stage:   1 Beta: 0.080\n",
      "Stage:   2 Beta: 0.224\n",
      "Stage:   3 Beta: 0.616\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.038\n",
      "Stage:   1 Beta: 0.240\n",
      "Stage:   2 Beta: 0.738\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.037\n",
      "Stage:   1 Beta: 0.124\n",
      "Stage:   2 Beta: 0.356\n",
      "Stage:   3 Beta: 0.969\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.053\n",
      "Stage:   1 Beta: 0.233\n",
      "Stage:   2 Beta: 0.646\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.033\n",
      "Stage:   1 Beta: 0.105\n",
      "Stage:   2 Beta: 0.274\n",
      "Stage:   3 Beta: 0.656\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.038\n",
      "Stage:   1 Beta: 0.109\n",
      "Stage:   2 Beta: 0.257\n",
      "Stage:   3 Beta: 0.631\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.026\n",
      "Stage:   1 Beta: 0.081\n",
      "Stage:   2 Beta: 0.236\n",
      "Stage:   3 Beta: 0.680\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.042\n",
      "Stage:   1 Beta: 0.262\n",
      "Stage:   2 Beta: 0.755\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.011\n",
      "Stage:   1 Beta: 0.033\n",
      "Stage:   2 Beta: 0.091\n",
      "Stage:   3 Beta: 0.260\n",
      "Stage:   4 Beta: 0.746\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.042\n",
      "Stage:   1 Beta: 0.334\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.014\n",
      "Stage:   1 Beta: 0.043\n",
      "Stage:   2 Beta: 0.117\n",
      "Stage:   3 Beta: 0.303\n",
      "Stage:   4 Beta: 0.760\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.029\n",
      "Stage:   1 Beta: 0.093\n",
      "Stage:   2 Beta: 0.271\n",
      "Stage:   3 Beta: 0.780\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.023\n",
      "Stage:   1 Beta: 0.067\n",
      "Stage:   2 Beta: 0.184\n",
      "Stage:   3 Beta: 0.493\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.037\n",
      "Stage:   1 Beta: 0.124\n",
      "Stage:   2 Beta: 0.376\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.016\n",
      "Stage:   1 Beta: 0.048\n",
      "Stage:   2 Beta: 0.131\n",
      "Stage:   3 Beta: 0.360\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.057\n",
      "Stage:   1 Beta: 0.224\n",
      "Stage:   2 Beta: 0.556\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.038\n",
      "Stage:   1 Beta: 0.301\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.047\n",
      "Stage:   1 Beta: 0.229\n",
      "Stage:   2 Beta: 0.646\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.022\n",
      "Stage:   1 Beta: 0.068\n",
      "Stage:   2 Beta: 0.174\n",
      "Stage:   3 Beta: 0.434\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.014\n",
      "Stage:   1 Beta: 0.044\n",
      "Stage:   2 Beta: 0.123\n",
      "Stage:   3 Beta: 0.352\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.009\n",
      "Stage:   1 Beta: 0.026\n",
      "Stage:   2 Beta: 0.063\n",
      "Stage:   3 Beta: 0.166\n",
      "Stage:   4 Beta: 0.414\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.042\n",
      "Stage:   1 Beta: 0.341\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.017\n",
      "Stage:   1 Beta: 0.050\n",
      "Stage:   2 Beta: 0.139\n",
      "Stage:   3 Beta: 0.391\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.020\n",
      "Stage:   1 Beta: 0.062\n",
      "Stage:   2 Beta: 0.151\n",
      "Stage:   3 Beta: 0.388\n",
      "Stage:   4 Beta: 0.989\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.012\n",
      "Stage:   1 Beta: 0.038\n",
      "Stage:   2 Beta: 0.106\n",
      "Stage:   3 Beta: 0.297\n",
      "Stage:   4 Beta: 0.818\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.061\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Stage:   1 Beta: 0.275\n",
      "Stage:   2 Beta: 0.737\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.026\n",
      "Stage:   1 Beta: 0.081\n",
      "Stage:   2 Beta: 0.231\n",
      "Stage:   3 Beta: 0.674\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.043\n",
      "Stage:   1 Beta: 0.230\n",
      "Stage:   2 Beta: 0.665\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.041\n",
      "Stage:   1 Beta: 0.347\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.049\n",
      "Stage:   1 Beta: 0.189\n",
      "Stage:   2 Beta: 0.570\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.020\n",
      "Stage:   1 Beta: 0.064\n",
      "Stage:   2 Beta: 0.172\n",
      "Stage:   3 Beta: 0.415\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.024\n",
      "Stage:   1 Beta: 0.067\n",
      "Stage:   2 Beta: 0.159\n",
      "Stage:   3 Beta: 0.410\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.010\n",
      "Stage:   1 Beta: 0.028\n",
      "Stage:   2 Beta: 0.075\n",
      "Stage:   3 Beta: 0.223\n",
      "Stage:   4 Beta: 0.670\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.013\n",
      "Stage:   1 Beta: 0.042\n",
      "Stage:   2 Beta: 0.110\n",
      "Stage:   3 Beta: 0.324\n",
      "Stage:   4 Beta: 0.942\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.011\n",
      "Stage:   1 Beta: 0.034\n",
      "Stage:   2 Beta: 0.091\n",
      "Stage:   3 Beta: 0.251\n",
      "Stage:   4 Beta: 0.694\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.040\n",
      "Stage:   1 Beta: 0.139\n",
      "Stage:   2 Beta: 0.403\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.028\n",
      "Stage:   1 Beta: 0.088\n",
      "Stage:   2 Beta: 0.242\n",
      "Stage:   3 Beta: 0.680\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.019\n",
      "Stage:   1 Beta: 0.059\n",
      "Stage:   2 Beta: 0.166\n",
      "Stage:   3 Beta: 0.481\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.059\n",
      "Stage:   1 Beta: 0.307\n",
      "Stage:   2 Beta: 0.833\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.041\n",
      "Stage:   1 Beta: 0.320\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.045\n",
      "Stage:   1 Beta: 0.164\n",
      "Stage:   2 Beta: 0.481\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.039\n",
      "Stage:   1 Beta: 0.307\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.015\n",
      "Stage:   1 Beta: 0.048\n",
      "Stage:   2 Beta: 0.148\n",
      "Stage:   3 Beta: 0.411\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.028\n",
      "Stage:   1 Beta: 0.087\n",
      "Stage:   2 Beta: 0.244\n",
      "Stage:   3 Beta: 0.683\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.045\n",
      "Stage:   1 Beta: 0.153\n",
      "Stage:   2 Beta: 0.394\n",
      "Stage:   3 Beta: 0.970\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.027\n",
      "Stage:   1 Beta: 0.087\n",
      "Stage:   2 Beta: 0.267\n",
      "Stage:   3 Beta: 0.806\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.016\n",
      "Stage:   1 Beta: 0.052\n",
      "Stage:   2 Beta: 0.144\n",
      "Stage:   3 Beta: 0.365\n",
      "Stage:   4 Beta: 0.912\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.039\n",
      "Stage:   1 Beta: 0.332\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.047\n",
      "Stage:   1 Beta: 0.370\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.039\n",
      "Stage:   1 Beta: 0.303\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.021\n",
      "Stage:   1 Beta: 0.067\n",
      "Stage:   2 Beta: 0.197\n",
      "Stage:   3 Beta: 0.578\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.042\n",
      "Stage:   1 Beta: 0.320\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.011\n",
      "Stage:   1 Beta: 0.032\n",
      "Stage:   2 Beta: 0.086\n",
      "Stage:   3 Beta: 0.246\n",
      "Stage:   4 Beta: 0.729\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.006\n",
      "Stage:   1 Beta: 0.028\n",
      "Stage:   2 Beta: 0.077\n",
      "Stage:   3 Beta: 0.180\n",
      "Stage:   4 Beta: 0.445\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.038\n",
      "Stage:   2 Beta: 0.152\n",
      "Stage:   3 Beta: 0.434\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.009\n",
      "Stage:   2 Beta: 0.024\n",
      "Stage:   3 Beta: 0.062\n",
      "Stage:   4 Beta: 0.154\n",
      "Stage:   5 Beta: 0.379\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.021\n",
      "Stage:   2 Beta: 0.059\n",
      "Stage:   3 Beta: 0.175\n",
      "Stage:   4 Beta: 0.510\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.019\n",
      "Stage:   2 Beta: 0.048\n",
      "Stage:   3 Beta: 0.111\n",
      "Stage:   4 Beta: 0.282\n",
      "Stage:   5 Beta: 0.731\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.007\n",
      "Stage:   2 Beta: 0.018\n",
      "Stage:   3 Beta: 0.044\n",
      "Stage:   4 Beta: 0.111\n",
      "Stage:   5 Beta: 0.285\n",
      "Stage:   6 Beta: 0.726\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.007\n",
      "Stage:   2 Beta: 0.021\n",
      "Stage:   3 Beta: 0.063\n",
      "Stage:   4 Beta: 0.181\n",
      "Stage:   5 Beta: 0.534\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.020\n",
      "Stage:   2 Beta: 0.060\n",
      "Stage:   3 Beta: 0.179\n",
      "Stage:   4 Beta: 0.514\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.007\n",
      "Stage:   2 Beta: 0.019\n",
      "Stage:   3 Beta: 0.046\n",
      "Stage:   4 Beta: 0.119\n",
      "Stage:   5 Beta: 0.304\n",
      "Stage:   6 Beta: 0.798\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.004\n",
      "Stage:   2 Beta: 0.011\n",
      "Stage:   3 Beta: 0.031\n",
      "Stage:   4 Beta: 0.085\n",
      "Stage:   5 Beta: 0.233\n",
      "Stage:   6 Beta: 0.592\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.033\n",
      "Stage:   2 Beta: 0.246\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.021\n",
      "Stage:   2 Beta: 0.065\n",
      "Stage:   3 Beta: 0.201\n",
      "Stage:   4 Beta: 0.578\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.014\n",
      "Stage:   2 Beta: 0.041\n",
      "Stage:   3 Beta: 0.120\n",
      "Stage:   4 Beta: 0.328\n",
      "Stage:   5 Beta: 0.783\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.033\n",
      "Stage:   2 Beta: 0.248\n",
      "Stage:   3 Beta: 0.954\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.035\n",
      "Stage:   2 Beta: 0.238\n",
      "Stage:   3 Beta: 0.743\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.008\n",
      "Stage:   2 Beta: 0.024\n",
      "Stage:   3 Beta: 0.069\n",
      "Stage:   4 Beta: 0.179\n",
      "Stage:   5 Beta: 0.449\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.013\n",
      "Stage:   2 Beta: 0.031\n",
      "Stage:   3 Beta: 0.076\n",
      "Stage:   4 Beta: 0.183\n",
      "Stage:   5 Beta: 0.470\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.036\n",
      "Stage:   2 Beta: 0.225\n",
      "Stage:   3 Beta: 0.720\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.033\n",
      "Stage:   2 Beta: 0.109\n",
      "Stage:   3 Beta: 0.334\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.009\n",
      "Stage:   2 Beta: 0.028\n",
      "Stage:   3 Beta: 0.082\n",
      "Stage:   4 Beta: 0.199\n",
      "Stage:   5 Beta: 0.481\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.012\n",
      "Stage:   2 Beta: 0.033\n",
      "Stage:   3 Beta: 0.095\n",
      "Stage:   4 Beta: 0.282\n",
      "Stage:   5 Beta: 0.851\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.035\n",
      "Stage:   2 Beta: 0.141\n",
      "Stage:   3 Beta: 0.408\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.004\n",
      "Stage:   2 Beta: 0.012\n",
      "Stage:   3 Beta: 0.033\n",
      "Stage:   4 Beta: 0.094\n",
      "Stage:   5 Beta: 0.259\n",
      "Stage:   6 Beta: 0.698\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.003\n",
      "Stage:   2 Beta: 0.009\n",
      "Stage:   3 Beta: 0.027\n",
      "Stage:   4 Beta: 0.076\n",
      "Stage:   5 Beta: 0.230\n",
      "Stage:   6 Beta: 0.626\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.010\n",
      "Stage:   2 Beta: 0.025\n",
      "Stage:   3 Beta: 0.062\n",
      "Stage:   4 Beta: 0.157\n",
      "Stage:   5 Beta: 0.393\n",
      "Stage:   6 Beta: 0.965\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.003\n",
      "Stage:   2 Beta: 0.010\n",
      "Stage:   3 Beta: 0.026\n",
      "Stage:   4 Beta: 0.073\n",
      "Stage:   5 Beta: 0.202\n",
      "Stage:   6 Beta: 0.539\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.005\n",
      "Stage:   2 Beta: 0.015\n",
      "Stage:   3 Beta: 0.043\n",
      "Stage:   4 Beta: 0.126\n",
      "Stage:   5 Beta: 0.378\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.006\n",
      "Stage:   2 Beta: 0.015\n",
      "Stage:   3 Beta: 0.038\n",
      "Stage:   4 Beta: 0.092\n",
      "Stage:   5 Beta: 0.241\n",
      "Stage:   6 Beta: 0.630\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.008\n",
      "Stage:   2 Beta: 0.023\n",
      "Stage:   3 Beta: 0.069\n",
      "Stage:   4 Beta: 0.208\n",
      "Stage:   5 Beta: 0.603\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.008\n",
      "Stage:   2 Beta: 0.020\n",
      "Stage:   3 Beta: 0.052\n",
      "Stage:   4 Beta: 0.133\n",
      "Stage:   5 Beta: 0.349\n",
      "Stage:   6 Beta: 0.849\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.023\n",
      "Stage:   2 Beta: 0.063\n",
      "Stage:   3 Beta: 0.150\n",
      "Stage:   4 Beta: 0.374\n",
      "Stage:   5 Beta: 0.925\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.012\n",
      "Stage:   2 Beta: 0.033\n",
      "Stage:   3 Beta: 0.084\n",
      "Stage:   4 Beta: 0.204\n",
      "Stage:   5 Beta: 0.480\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.010\n",
      "Stage:   2 Beta: 0.029\n",
      "Stage:   3 Beta: 0.085\n",
      "Stage:   4 Beta: 0.249\n",
      "Stage:   5 Beta: 0.748\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.033\n",
      "Stage:   2 Beta: 0.107\n",
      "Stage:   3 Beta: 0.329\n",
      "Stage:   4 Beta: 0.983\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.004\n",
      "Stage:   2 Beta: 0.010\n",
      "Stage:   3 Beta: 0.027\n",
      "Stage:   4 Beta: 0.076\n",
      "Stage:   5 Beta: 0.220\n",
      "Stage:   6 Beta: 0.650\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.030\n",
      "Stage:   2 Beta: 0.099\n",
      "Stage:   3 Beta: 0.297\n",
      "Stage:   4 Beta: 0.907\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.006\n",
      "Stage:   2 Beta: 0.015\n",
      "Stage:   3 Beta: 0.038\n",
      "Stage:   4 Beta: 0.098\n",
      "Stage:   5 Beta: 0.244\n",
      "Stage:   6 Beta: 0.595\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.012\n",
      "Stage:   2 Beta: 0.035\n",
      "Stage:   3 Beta: 0.099\n",
      "Stage:   4 Beta: 0.273\n",
      "Stage:   5 Beta: 0.748\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.007\n",
      "Stage:   2 Beta: 0.020\n",
      "Stage:   3 Beta: 0.047\n",
      "Stage:   4 Beta: 0.118\n",
      "Stage:   5 Beta: 0.315\n",
      "Stage:   6 Beta: 0.802\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.038\n",
      "Stage:   2 Beta: 0.153\n",
      "Stage:   3 Beta: 0.461\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.042\n",
      "Stage:   2 Beta: 0.147\n",
      "Stage:   3 Beta: 0.398\n",
      "Stage:   4 Beta: 0.968\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.021\n",
      "Stage:   2 Beta: 0.059\n",
      "Stage:   3 Beta: 0.169\n",
      "Stage:   4 Beta: 0.495\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.019\n",
      "Stage:   2 Beta: 0.057\n",
      "Stage:   3 Beta: 0.172\n",
      "Stage:   4 Beta: 0.520\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.010\n",
      "Stage:   2 Beta: 0.028\n",
      "Stage:   3 Beta: 0.080\n",
      "Stage:   4 Beta: 0.243\n",
      "Stage:   5 Beta: 0.738\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.004\n",
      "Stage:   2 Beta: 0.011\n",
      "Stage:   3 Beta: 0.032\n",
      "Stage:   4 Beta: 0.083\n",
      "Stage:   5 Beta: 0.204\n",
      "Stage:   6 Beta: 0.507\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.005\n",
      "Stage:   2 Beta: 0.015\n",
      "Stage:   3 Beta: 0.043\n",
      "Stage:   4 Beta: 0.123\n",
      "Stage:   5 Beta: 0.367\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.009\n",
      "Stage:   2 Beta: 0.021\n",
      "Stage:   3 Beta: 0.051\n",
      "Stage:   4 Beta: 0.130\n",
      "Stage:   5 Beta: 0.334\n",
      "Stage:   6 Beta: 0.857\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.031\n",
      "Stage:   2 Beta: 0.099\n",
      "Stage:   3 Beta: 0.287\n",
      "Stage:   4 Beta: 0.805\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.032\n",
      "Stage:   2 Beta: 0.101\n",
      "Stage:   3 Beta: 0.285\n",
      "Stage:   4 Beta: 0.806\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.017\n",
      "Stage:   2 Beta: 0.051\n",
      "Stage:   3 Beta: 0.147\n",
      "Stage:   4 Beta: 0.435\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.010\n",
      "Stage:   2 Beta: 0.028\n",
      "Stage:   3 Beta: 0.079\n",
      "Stage:   4 Beta: 0.223\n",
      "Stage:   5 Beta: 0.617\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.037\n",
      "Stage:   2 Beta: 0.241\n",
      "Stage:   3 Beta: 0.780\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.007\n",
      "Stage:   2 Beta: 0.020\n",
      "Stage:   3 Beta: 0.058\n",
      "Stage:   4 Beta: 0.172\n",
      "Stage:   5 Beta: 0.478\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.007\n",
      "Stage:   2 Beta: 0.020\n",
      "Stage:   3 Beta: 0.059\n",
      "Stage:   4 Beta: 0.159\n",
      "Stage:   5 Beta: 0.393\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.004\n",
      "Stage:   2 Beta: 0.012\n",
      "Stage:   3 Beta: 0.035\n",
      "Stage:   4 Beta: 0.104\n",
      "Stage:   5 Beta: 0.296\n",
      "Stage:   6 Beta: 0.757\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.006\n",
      "Stage:   2 Beta: 0.016\n",
      "Stage:   3 Beta: 0.045\n",
      "Stage:   4 Beta: 0.128\n",
      "Stage:   5 Beta: 0.360\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.016\n",
      "Stage:   2 Beta: 0.046\n",
      "Stage:   3 Beta: 0.136\n",
      "Stage:   4 Beta: 0.391\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.006\n",
      "Stage:   2 Beta: 0.017\n",
      "Stage:   3 Beta: 0.042\n",
      "Stage:   4 Beta: 0.108\n",
      "Stage:   5 Beta: 0.274\n",
      "Stage:   6 Beta: 0.679\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.025\n",
      "Stage:   2 Beta: 0.076\n",
      "Stage:   3 Beta: 0.213\n",
      "Stage:   4 Beta: 0.609\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.009\n",
      "Stage:   2 Beta: 0.027\n",
      "Stage:   3 Beta: 0.081\n",
      "Stage:   4 Beta: 0.249\n",
      "Stage:   5 Beta: 0.714\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.004\n",
      "Stage:   2 Beta: 0.010\n",
      "Stage:   3 Beta: 0.028\n",
      "Stage:   4 Beta: 0.080\n",
      "Stage:   5 Beta: 0.217\n",
      "Stage:   6 Beta: 0.613\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.030\n",
      "Stage:   2 Beta: 0.087\n",
      "Stage:   3 Beta: 0.209\n",
      "Stage:   4 Beta: 0.512\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.015\n",
      "Stage:   2 Beta: 0.034\n",
      "Stage:   3 Beta: 0.084\n",
      "Stage:   4 Beta: 0.214\n",
      "Stage:   5 Beta: 0.535\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.009\n",
      "Stage:   2 Beta: 0.027\n",
      "Stage:   3 Beta: 0.085\n",
      "Stage:   4 Beta: 0.254\n",
      "Stage:   5 Beta: 0.711\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.014\n",
      "Stage:   2 Beta: 0.036\n",
      "Stage:   3 Beta: 0.086\n",
      "Stage:   4 Beta: 0.208\n",
      "Stage:   5 Beta: 0.508\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.006\n",
      "Stage:   2 Beta: 0.020\n",
      "Stage:   3 Beta: 0.054\n",
      "Stage:   4 Beta: 0.134\n",
      "Stage:   5 Beta: 0.337\n",
      "Stage:   6 Beta: 0.912\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.004\n",
      "Stage:   2 Beta: 0.010\n",
      "Stage:   3 Beta: 0.028\n",
      "Stage:   4 Beta: 0.078\n",
      "Stage:   5 Beta: 0.211\n",
      "Stage:   6 Beta: 0.587\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.012\n",
      "Stage:   2 Beta: 0.034\n",
      "Stage:   3 Beta: 0.106\n",
      "Stage:   4 Beta: 0.272\n",
      "Stage:   5 Beta: 0.632\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.042\n",
      "Stage:   2 Beta: 0.254\n",
      "Stage:   3 Beta: 0.736\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.010\n",
      "Stage:   2 Beta: 0.029\n",
      "Stage:   3 Beta: 0.081\n",
      "Stage:   4 Beta: 0.230\n",
      "Stage:   5 Beta: 0.663\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.427\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.088\n",
      "Stage:   1 Beta: 0.252\n",
      "Stage:   2 Beta: 0.631\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.444\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.163\n",
      "Stage:   1 Beta: 0.508\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.386\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.365\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.127\n",
      "Stage:   1 Beta: 0.384\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.242\n",
      "Stage:   1 Beta: 0.727\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.406\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.408\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.136\n",
      "Stage:   1 Beta: 0.421\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.293\n",
      "Stage:   1 Beta: 0.869\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.365\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.263\n",
      "Stage:   1 Beta: 0.859\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.177\n",
      "Stage:   1 Beta: 0.544\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.121\n",
      "Stage:   1 Beta: 0.359\n",
      "Stage:   2 Beta: 0.952\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.400\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.128\n",
      "Stage:   1 Beta: 0.380\n",
      "Stage:   2 Beta: 0.987\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.278\n",
      "Stage:   1 Beta: 0.861\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.415\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.327\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.164\n",
      "Stage:   1 Beta: 0.519\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.281\n",
      "Stage:   1 Beta: 0.878\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.231\n",
      "Stage:   1 Beta: 0.685\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.238\n",
      "Stage:   1 Beta: 0.726\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.089\n",
      "Stage:   1 Beta: 0.253\n",
      "Stage:   2 Beta: 0.636\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.249\n",
      "Stage:   1 Beta: 0.761\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.180\n",
      "Stage:   1 Beta: 0.580\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.380\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.406\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.308\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.415\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.379\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.252\n",
      "Stage:   1 Beta: 0.771\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.416\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.438\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.394\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.160\n",
      "Stage:   1 Beta: 0.450\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.211\n",
      "Stage:   1 Beta: 0.587\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.214\n",
      "Stage:   1 Beta: 0.695\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.175\n",
      "Stage:   1 Beta: 0.536\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.372\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.120\n",
      "Stage:   1 Beta: 0.378\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.386\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.107\n",
      "Stage:   1 Beta: 0.328\n",
      "Stage:   2 Beta: 0.905\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.437\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Stage:   0 Beta: 0.381\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.372\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.304\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.239\n",
      "Stage:   1 Beta: 0.721\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.326\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.377\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.213\n",
      "Stage:   1 Beta: 0.618\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.410\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.183\n",
      "Stage:   1 Beta: 0.559\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.398\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.357\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.378\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.212\n",
      "Stage:   1 Beta: 0.678\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.284\n",
      "Stage:   1 Beta: 0.858\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.141\n",
      "Stage:   1 Beta: 0.433\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.270\n",
      "Stage:   1 Beta: 0.855\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.195\n",
      "Stage:   1 Beta: 0.556\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.377\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.360\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.362\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.336\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.210\n",
      "Stage:   1 Beta: 0.693\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.332\n",
      "Stage:   1 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.162\n",
      "Stage:   1 Beta: 0.510\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.037\n",
      "Stage:   1 Beta: 0.256\n",
      "Stage:   2 Beta: 0.834\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.032\n",
      "Stage:   1 Beta: 0.104\n",
      "Stage:   2 Beta: 0.305\n",
      "Stage:   3 Beta: 0.906\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.010\n",
      "Stage:   1 Beta: 0.028\n",
      "Stage:   2 Beta: 0.075\n",
      "Stage:   3 Beta: 0.215\n",
      "Stage:   4 Beta: 0.625\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.052\n",
      "Stage:   1 Beta: 0.244\n",
      "Stage:   2 Beta: 0.723\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.026\n",
      "Stage:   1 Beta: 0.079\n",
      "Stage:   2 Beta: 0.230\n",
      "Stage:   3 Beta: 0.704\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.036\n",
      "Stage:   1 Beta: 0.123\n",
      "Stage:   2 Beta: 0.372\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.010\n",
      "Stage:   1 Beta: 0.028\n",
      "Stage:   2 Beta: 0.075\n",
      "Stage:   3 Beta: 0.199\n",
      "Stage:   4 Beta: 0.526\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.011\n",
      "Stage:   1 Beta: 0.035\n",
      "Stage:   2 Beta: 0.097\n",
      "Stage:   3 Beta: 0.270\n",
      "Stage:   4 Beta: 0.769\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.053\n",
      "Stage:   1 Beta: 0.213\n",
      "Stage:   2 Beta: 0.643\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.015\n",
      "Stage:   1 Beta: 0.046\n",
      "Stage:   2 Beta: 0.134\n",
      "Stage:   3 Beta: 0.414\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.045\n",
      "Stage:   1 Beta: 0.166\n",
      "Stage:   2 Beta: 0.490\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.043\n",
      "Stage:   1 Beta: 0.161\n",
      "Stage:   2 Beta: 0.491\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.037\n",
      "Stage:   1 Beta: 0.119\n",
      "Stage:   2 Beta: 0.345\n",
      "Stage:   3 Beta: 0.970\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.043\n",
      "Stage:   1 Beta: 0.239\n",
      "Stage:   2 Beta: 0.674\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.025\n",
      "Stage:   1 Beta: 0.077\n",
      "Stage:   2 Beta: 0.222\n",
      "Stage:   3 Beta: 0.630\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.023\n",
      "Stage:   1 Beta: 0.074\n",
      "Stage:   2 Beta: 0.217\n",
      "Stage:   3 Beta: 0.631\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.024\n",
      "Stage:   1 Beta: 0.076\n",
      "Stage:   2 Beta: 0.221\n",
      "Stage:   3 Beta: 0.630\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.014\n",
      "Stage:   1 Beta: 0.043\n",
      "Stage:   2 Beta: 0.124\n",
      "Stage:   3 Beta: 0.353\n",
      "Stage:   4 Beta: 0.947\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.018\n",
      "Stage:   1 Beta: 0.055\n",
      "Stage:   2 Beta: 0.138\n",
      "Stage:   3 Beta: 0.348\n",
      "Stage:   4 Beta: 0.891\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.023\n",
      "Stage:   1 Beta: 0.071\n",
      "Stage:   2 Beta: 0.203\n",
      "Stage:   3 Beta: 0.590\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.038\n",
      "Stage:   1 Beta: 0.318\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.026\n",
      "Stage:   1 Beta: 0.074\n",
      "Stage:   2 Beta: 0.181\n",
      "Stage:   3 Beta: 0.458\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.044\n",
      "Stage:   1 Beta: 0.360\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.040\n",
      "Stage:   1 Beta: 0.335\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.037\n",
      "Stage:   1 Beta: 0.297\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.042\n",
      "Stage:   1 Beta: 0.156\n",
      "Stage:   2 Beta: 0.465\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.049\n",
      "Stage:   1 Beta: 0.346\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.011\n",
      "Stage:   1 Beta: 0.034\n",
      "Stage:   2 Beta: 0.092\n",
      "Stage:   3 Beta: 0.260\n",
      "Stage:   4 Beta: 0.757\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.048\n",
      "Stage:   1 Beta: 0.207\n",
      "Stage:   2 Beta: 0.621\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.043\n",
      "Stage:   1 Beta: 0.179\n",
      "Stage:   2 Beta: 0.510\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.047\n",
      "Stage:   1 Beta: 0.141\n",
      "Stage:   2 Beta: 0.331\n",
      "Stage:   3 Beta: 0.804\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.043\n",
      "Stage:   1 Beta: 0.246\n",
      "Stage:   2 Beta: 0.722\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.033\n",
      "Stage:   1 Beta: 0.109\n",
      "Stage:   2 Beta: 0.313\n",
      "Stage:   3 Beta: 0.958\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.011\n",
      "Stage:   1 Beta: 0.032\n",
      "Stage:   2 Beta: 0.090\n",
      "Stage:   3 Beta: 0.264\n",
      "Stage:   4 Beta: 0.700\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.015\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Stage:   1 Beta: 0.047\n",
      "Stage:   2 Beta: 0.130\n",
      "Stage:   3 Beta: 0.346\n",
      "Stage:   4 Beta: 0.889\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.041\n",
      "Stage:   1 Beta: 0.352\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.047\n",
      "Stage:   1 Beta: 0.208\n",
      "Stage:   2 Beta: 0.632\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.036\n",
      "Stage:   1 Beta: 0.123\n",
      "Stage:   2 Beta: 0.360\n",
      "Stage:   3 Beta: 0.914\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.016\n",
      "Stage:   1 Beta: 0.052\n",
      "Stage:   2 Beta: 0.142\n",
      "Stage:   3 Beta: 0.398\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.046\n",
      "Stage:   1 Beta: 0.196\n",
      "Stage:   2 Beta: 0.558\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.040\n",
      "Stage:   1 Beta: 0.144\n",
      "Stage:   2 Beta: 0.429\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.010\n",
      "Stage:   1 Beta: 0.031\n",
      "Stage:   2 Beta: 0.081\n",
      "Stage:   3 Beta: 0.218\n",
      "Stage:   4 Beta: 0.573\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.043\n",
      "Stage:   1 Beta: 0.191\n",
      "Stage:   2 Beta: 0.554\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.019\n",
      "Stage:   1 Beta: 0.057\n",
      "Stage:   2 Beta: 0.160\n",
      "Stage:   3 Beta: 0.438\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.018\n",
      "Stage:   1 Beta: 0.053\n",
      "Stage:   2 Beta: 0.154\n",
      "Stage:   3 Beta: 0.460\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.039\n",
      "Stage:   1 Beta: 0.133\n",
      "Stage:   2 Beta: 0.392\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.017\n",
      "Stage:   1 Beta: 0.053\n",
      "Stage:   2 Beta: 0.159\n",
      "Stage:   3 Beta: 0.463\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.037\n",
      "Stage:   1 Beta: 0.122\n",
      "Stage:   2 Beta: 0.352\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.036\n",
      "Stage:   1 Beta: 0.120\n",
      "Stage:   2 Beta: 0.344\n",
      "Stage:   3 Beta: 0.995\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.013\n",
      "Stage:   1 Beta: 0.040\n",
      "Stage:   2 Beta: 0.111\n",
      "Stage:   3 Beta: 0.313\n",
      "Stage:   4 Beta: 0.798\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.038\n",
      "Stage:   1 Beta: 0.114\n",
      "Stage:   2 Beta: 0.274\n",
      "Stage:   3 Beta: 0.654\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.014\n",
      "Stage:   1 Beta: 0.047\n",
      "Stage:   2 Beta: 0.133\n",
      "Stage:   3 Beta: 0.363\n",
      "Stage:   4 Beta: 0.903\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.022\n",
      "Stage:   1 Beta: 0.064\n",
      "Stage:   2 Beta: 0.173\n",
      "Stage:   3 Beta: 0.469\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.055\n",
      "Stage:   1 Beta: 0.232\n",
      "Stage:   2 Beta: 0.702\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.014\n",
      "Stage:   1 Beta: 0.041\n",
      "Stage:   2 Beta: 0.112\n",
      "Stage:   3 Beta: 0.317\n",
      "Stage:   4 Beta: 0.909\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.013\n",
      "Stage:   1 Beta: 0.039\n",
      "Stage:   2 Beta: 0.114\n",
      "Stage:   3 Beta: 0.300\n",
      "Stage:   4 Beta: 0.715\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.018\n",
      "Stage:   1 Beta: 0.053\n",
      "Stage:   2 Beta: 0.143\n",
      "Stage:   3 Beta: 0.391\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.024\n",
      "Stage:   1 Beta: 0.077\n",
      "Stage:   2 Beta: 0.206\n",
      "Stage:   3 Beta: 0.483\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.018\n",
      "Stage:   1 Beta: 0.053\n",
      "Stage:   2 Beta: 0.143\n",
      "Stage:   3 Beta: 0.380\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.041\n",
      "Stage:   1 Beta: 0.334\n",
      "Stage:   2 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.022\n",
      "Stage:   1 Beta: 0.068\n",
      "Stage:   2 Beta: 0.179\n",
      "Stage:   3 Beta: 0.422\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.032\n",
      "Stage:   1 Beta: 0.101\n",
      "Stage:   2 Beta: 0.292\n",
      "Stage:   3 Beta: 0.761\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.042\n",
      "Stage:   1 Beta: 0.136\n",
      "Stage:   2 Beta: 0.347\n",
      "Stage:   3 Beta: 0.827\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.025\n",
      "Stage:   1 Beta: 0.079\n",
      "Stage:   2 Beta: 0.237\n",
      "Stage:   3 Beta: 0.678\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.014\n",
      "Stage:   1 Beta: 0.041\n",
      "Stage:   2 Beta: 0.118\n",
      "Stage:   3 Beta: 0.332\n",
      "Stage:   4 Beta: 0.914\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.026\n",
      "Stage:   1 Beta: 0.083\n",
      "Stage:   2 Beta: 0.233\n",
      "Stage:   3 Beta: 0.607\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.012\n",
      "Stage:   1 Beta: 0.038\n",
      "Stage:   2 Beta: 0.100\n",
      "Stage:   3 Beta: 0.278\n",
      "Stage:   4 Beta: 0.798\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.049\n",
      "Stage:   1 Beta: 0.259\n",
      "Stage:   2 Beta: 0.805\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.020\n",
      "Stage:   1 Beta: 0.061\n",
      "Stage:   2 Beta: 0.190\n",
      "Stage:   3 Beta: 0.547\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.017\n",
      "Stage:   1 Beta: 0.050\n",
      "Stage:   2 Beta: 0.146\n",
      "Stage:   3 Beta: 0.440\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.004\n",
      "Stage:   2 Beta: 0.011\n",
      "Stage:   3 Beta: 0.032\n",
      "Stage:   4 Beta: 0.089\n",
      "Stage:   5 Beta: 0.247\n",
      "Stage:   6 Beta: 0.617\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.011\n",
      "Stage:   2 Beta: 0.033\n",
      "Stage:   3 Beta: 0.087\n",
      "Stage:   4 Beta: 0.207\n",
      "Stage:   5 Beta: 0.521\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.006\n",
      "Stage:   1 Beta: 0.025\n",
      "Stage:   2 Beta: 0.071\n",
      "Stage:   3 Beta: 0.173\n",
      "Stage:   4 Beta: 0.420\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.007\n",
      "Stage:   2 Beta: 0.017\n",
      "Stage:   3 Beta: 0.043\n",
      "Stage:   4 Beta: 0.110\n",
      "Stage:   5 Beta: 0.279\n",
      "Stage:   6 Beta: 0.735\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.006\n",
      "Stage:   2 Beta: 0.017\n",
      "Stage:   3 Beta: 0.048\n",
      "Stage:   4 Beta: 0.138\n",
      "Stage:   5 Beta: 0.410\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.004\n",
      "Stage:   2 Beta: 0.011\n",
      "Stage:   3 Beta: 0.031\n",
      "Stage:   4 Beta: 0.090\n",
      "Stage:   5 Beta: 0.255\n",
      "Stage:   6 Beta: 0.634\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.003\n",
      "Stage:   2 Beta: 0.009\n",
      "Stage:   3 Beta: 0.025\n",
      "Stage:   4 Beta: 0.071\n",
      "Stage:   5 Beta: 0.207\n",
      "Stage:   6 Beta: 0.581\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.019\n",
      "Stage:   2 Beta: 0.053\n",
      "Stage:   3 Beta: 0.162\n",
      "Stage:   4 Beta: 0.502\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.006\n",
      "Stage:   2 Beta: 0.016\n",
      "Stage:   3 Beta: 0.039\n",
      "Stage:   4 Beta: 0.098\n",
      "Stage:   5 Beta: 0.245\n",
      "Stage:   6 Beta: 0.634\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.031\n",
      "Stage:   2 Beta: 0.225\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.007\n",
      "Stage:   2 Beta: 0.021\n",
      "Stage:   3 Beta: 0.059\n",
      "Stage:   4 Beta: 0.171\n",
      "Stage:   5 Beta: 0.498\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.018\n",
      "Stage:   2 Beta: 0.054\n",
      "Stage:   3 Beta: 0.153\n",
      "Stage:   4 Beta: 0.372\n",
      "Stage:   5 Beta: 0.917\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.034\n",
      "Stage:   2 Beta: 0.119\n",
      "Stage:   3 Beta: 0.345\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.004\n",
      "Stage:   2 Beta: 0.010\n",
      "Stage:   3 Beta: 0.028\n",
      "Stage:   4 Beta: 0.080\n",
      "Stage:   5 Beta: 0.219\n",
      "Stage:   6 Beta: 0.631\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.011\n",
      "Stage:   2 Beta: 0.028\n",
      "Stage:   3 Beta: 0.070\n",
      "Stage:   4 Beta: 0.173\n",
      "Stage:   5 Beta: 0.427\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.021\n",
      "Stage:   2 Beta: 0.064\n",
      "Stage:   3 Beta: 0.196\n",
      "Stage:   4 Beta: 0.563\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.020\n",
      "Stage:   2 Beta: 0.060\n",
      "Stage:   3 Beta: 0.181\n",
      "Stage:   4 Beta: 0.524\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.032\n",
      "Stage:   2 Beta: 0.234\n",
      "Stage:   3 Beta: 0.949\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.006\n",
      "Stage:   2 Beta: 0.019\n",
      "Stage:   3 Beta: 0.051\n",
      "Stage:   4 Beta: 0.130\n",
      "Stage:   5 Beta: 0.325\n",
      "Stage:   6 Beta: 0.822\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.019\n",
      "Stage:   2 Beta: 0.051\n",
      "Stage:   3 Beta: 0.121\n",
      "Stage:   4 Beta: 0.293\n",
      "Stage:   5 Beta: 0.747\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.009\n",
      "Stage:   2 Beta: 0.025\n",
      "Stage:   3 Beta: 0.072\n",
      "Stage:   4 Beta: 0.216\n",
      "Stage:   5 Beta: 0.653\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.011\n",
      "Stage:   2 Beta: 0.032\n",
      "Stage:   3 Beta: 0.094\n",
      "Stage:   4 Beta: 0.245\n",
      "Stage:   5 Beta: 0.592\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.009\n",
      "Stage:   2 Beta: 0.024\n",
      "Stage:   3 Beta: 0.068\n",
      "Stage:   4 Beta: 0.198\n",
      "Stage:   5 Beta: 0.549\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.010\n",
      "Stage:   2 Beta: 0.026\n",
      "Stage:   3 Beta: 0.063\n",
      "Stage:   4 Beta: 0.158\n",
      "Stage:   5 Beta: 0.408\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.013\n",
      "Stage:   2 Beta: 0.037\n",
      "Stage:   3 Beta: 0.109\n",
      "Stage:   4 Beta: 0.319\n",
      "Stage:   5 Beta: 0.914\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.007\n",
      "Stage:   2 Beta: 0.021\n",
      "Stage:   3 Beta: 0.056\n",
      "Stage:   4 Beta: 0.157\n",
      "Stage:   5 Beta: 0.425\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.029\n",
      "Stage:   2 Beta: 0.089\n",
      "Stage:   3 Beta: 0.245\n",
      "Stage:   4 Beta: 0.673\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.008\n",
      "Stage:   2 Beta: 0.022\n",
      "Stage:   3 Beta: 0.067\n",
      "Stage:   4 Beta: 0.193\n",
      "Stage:   5 Beta: 0.483\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.006\n",
      "Stage:   2 Beta: 0.015\n",
      "Stage:   3 Beta: 0.042\n",
      "Stage:   4 Beta: 0.113\n",
      "Stage:   5 Beta: 0.311\n",
      "Stage:   6 Beta: 0.844\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.007\n",
      "Stage:   2 Beta: 0.019\n",
      "Stage:   3 Beta: 0.045\n",
      "Stage:   4 Beta: 0.112\n",
      "Stage:   5 Beta: 0.281\n",
      "Stage:   6 Beta: 0.721\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.035\n",
      "Stage:   2 Beta: 0.262\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.028\n",
      "Stage:   2 Beta: 0.087\n",
      "Stage:   3 Beta: 0.252\n",
      "Stage:   4 Beta: 0.681\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.014\n",
      "Stage:   2 Beta: 0.038\n",
      "Stage:   3 Beta: 0.092\n",
      "Stage:   4 Beta: 0.218\n",
      "Stage:   5 Beta: 0.579\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.007\n",
      "Stage:   2 Beta: 0.019\n",
      "Stage:   3 Beta: 0.047\n",
      "Stage:   4 Beta: 0.119\n",
      "Stage:   5 Beta: 0.298\n",
      "Stage:   6 Beta: 0.763\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.003\n",
      "Stage:   2 Beta: 0.008\n",
      "Stage:   3 Beta: 0.021\n",
      "Stage:   4 Beta: 0.060\n",
      "Stage:   5 Beta: 0.178\n",
      "Stage:   6 Beta: 0.492\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.009\n",
      "Stage:   2 Beta: 0.022\n",
      "Stage:   3 Beta: 0.053\n",
      "Stage:   4 Beta: 0.135\n",
      "Stage:   5 Beta: 0.349\n",
      "Stage:   6 Beta: 0.878\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.040\n",
      "Stage:   2 Beta: 0.219\n",
      "Stage:   3 Beta: 0.668\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.007\n",
      "Stage:   2 Beta: 0.021\n",
      "Stage:   3 Beta: 0.061\n",
      "Stage:   4 Beta: 0.162\n",
      "Stage:   5 Beta: 0.396\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.005\n",
      "Stage:   2 Beta: 0.014\n",
      "Stage:   3 Beta: 0.038\n",
      "Stage:   4 Beta: 0.104\n",
      "Stage:   5 Beta: 0.281\n",
      "Stage:   6 Beta: 0.740\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.010\n",
      "Stage:   2 Beta: 0.030\n",
      "Stage:   3 Beta: 0.073\n",
      "Stage:   4 Beta: 0.177\n",
      "Stage:   5 Beta: 0.449\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.032\n",
      "Stage:   2 Beta: 0.104\n",
      "Stage:   3 Beta: 0.321\n",
      "Stage:   4 Beta: 0.973\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.017\n",
      "Stage:   2 Beta: 0.048\n",
      "Stage:   3 Beta: 0.137\n",
      "Stage:   4 Beta: 0.404\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.015\n",
      "Stage:   2 Beta: 0.044\n",
      "Stage:   3 Beta: 0.130\n",
      "Stage:   4 Beta: 0.376\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.025\n",
      "Stage:   2 Beta: 0.075\n",
      "Stage:   3 Beta: 0.226\n",
      "Stage:   4 Beta: 0.577\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.005\n",
      "Stage:   2 Beta: 0.014\n",
      "Stage:   3 Beta: 0.035\n",
      "Stage:   4 Beta: 0.090\n",
      "Stage:   5 Beta: 0.244\n",
      "Stage:   6 Beta: 0.615\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.010\n",
      "Stage:   2 Beta: 0.028\n",
      "Stage:   3 Beta: 0.081\n",
      "Stage:   4 Beta: 0.243\n",
      "Stage:   5 Beta: 0.716\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.006\n",
      "Stage:   1 Beta: 0.025\n",
      "Stage:   2 Beta: 0.077\n",
      "Stage:   3 Beta: 0.206\n",
      "Stage:   4 Beta: 0.508\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.022\n",
      "Stage:   2 Beta: 0.067\n",
      "Stage:   3 Beta: 0.197\n",
      "Stage:   4 Beta: 0.576\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.006\n",
      "Stage:   2 Beta: 0.017\n",
      "Stage:   3 Beta: 0.047\n",
      "Stage:   4 Beta: 0.136\n",
      "Stage:   5 Beta: 0.385\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.001\n",
      "Stage:   1 Beta: 0.005\n",
      "Stage:   2 Beta: 0.012\n",
      "Stage:   3 Beta: 0.029\n",
      "Stage:   4 Beta: 0.072\n",
      "Stage:   5 Beta: 0.189\n",
      "Stage:   6 Beta: 0.512\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.033\n",
      "Stage:   2 Beta: 0.265\n",
      "Stage:   3 Beta: 1.000\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.013\n",
      "Stage:   2 Beta: 0.037\n",
      "Stage:   3 Beta: 0.110\n",
      "Stage:   4 Beta: 0.323\n",
      "Stage:   5 Beta: 0.860\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.022\n",
      "Stage:   2 Beta: 0.065\n",
      "Stage:   3 Beta: 0.179\n",
      "Stage:   4 Beta: 0.492\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.005\n",
      "Stage:   2 Beta: 0.014\n",
      "Stage:   3 Beta: 0.042\n",
      "Stage:   4 Beta: 0.126\n",
      "Stage:   5 Beta: 0.343\n",
      "Stage:   6 Beta: 0.844\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.009\n",
      "Stage:   2 Beta: 0.028\n",
      "Stage:   3 Beta: 0.081\n",
      "Stage:   4 Beta: 0.226\n",
      "Stage:   5 Beta: 0.578\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.017\n",
      "Stage:   2 Beta: 0.048\n",
      "Stage:   3 Beta: 0.132\n",
      "Stage:   4 Beta: 0.364\n",
      "Stage:   5 Beta: 0.998\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.025\n",
      "Stage:   2 Beta: 0.075\n",
      "Stage:   3 Beta: 0.216\n",
      "Stage:   4 Beta: 0.601\n",
      "Stage:   5 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.041\n",
      "Stage:   2 Beta: 0.242\n",
      "Stage:   3 Beta: 0.759\n",
      "Stage:   4 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.020\n",
      "Stage:   2 Beta: 0.058\n",
      "Stage:   3 Beta: 0.138\n",
      "Stage:   4 Beta: 0.318\n",
      "Stage:   5 Beta: 0.768\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.006\n",
      "Stage:   2 Beta: 0.017\n",
      "Stage:   3 Beta: 0.046\n",
      "Stage:   4 Beta: 0.111\n",
      "Stage:   5 Beta: 0.270\n",
      "Stage:   6 Beta: 0.681\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.005\n",
      "Stage:   2 Beta: 0.013\n",
      "Stage:   3 Beta: 0.039\n",
      "Stage:   4 Beta: 0.114\n",
      "Stage:   5 Beta: 0.310\n",
      "Stage:   6 Beta: 0.768\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.006\n",
      "Stage:   1 Beta: 0.023\n",
      "Stage:   2 Beta: 0.061\n",
      "Stage:   3 Beta: 0.146\n",
      "Stage:   4 Beta: 0.363\n",
      "Stage:   5 Beta: 0.906\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.032\n",
      "Stage:   2 Beta: 0.235\n",
      "Stage:   3 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.011\n",
      "Stage:   2 Beta: 0.032\n",
      "Stage:   3 Beta: 0.094\n",
      "Stage:   4 Beta: 0.278\n",
      "Stage:   5 Beta: 0.856\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.006\n",
      "Stage:   2 Beta: 0.018\n",
      "Stage:   3 Beta: 0.045\n",
      "Stage:   4 Beta: 0.112\n",
      "Stage:   5 Beta: 0.295\n",
      "Stage:   6 Beta: 0.745\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.002\n",
      "Stage:   1 Beta: 0.007\n",
      "Stage:   2 Beta: 0.022\n",
      "Stage:   3 Beta: 0.065\n",
      "Stage:   4 Beta: 0.158\n",
      "Stage:   5 Beta: 0.388\n",
      "Stage:   6 Beta: 0.978\n",
      "Stage:   7 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.003\n",
      "Stage:   1 Beta: 0.010\n",
      "Stage:   2 Beta: 0.028\n",
      "Stage:   3 Beta: 0.067\n",
      "Stage:   4 Beta: 0.163\n",
      "Stage:   5 Beta: 0.414\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.013\n",
      "Stage:   2 Beta: 0.037\n",
      "Stage:   3 Beta: 0.111\n",
      "Stage:   4 Beta: 0.337\n",
      "Stage:   5 Beta: 0.868\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.004\n",
      "Stage:   1 Beta: 0.015\n",
      "Stage:   2 Beta: 0.040\n",
      "Stage:   3 Beta: 0.098\n",
      "Stage:   4 Beta: 0.243\n",
      "Stage:   5 Beta: 0.613\n",
      "Stage:   6 Beta: 1.000\n",
      "Initializing SMC sampler...\n",
      "Sampling 4 chains in 4 jobs\n",
      "Stage:   0 Beta: 0.005\n",
      "Stage:   1 Beta: 0.045\n",
      "Stage:   2 Beta: 0.186\n",
      "Stage:   3 Beta: 0.547\n",
      "Stage:   4 Beta: 1.000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--- 30676.97034263611 seconds ---\n"
     ]
    }
   ],
   "source": [
    "## running the simulations\n",
    "start_time = time.time()\n",
    "\n",
    "n_sim = 70\n",
    "N = np.array([10,100,1000])  ## number of measurements \n",
    "\n",
    "particle_num = np.array([10,100,1000]) ## number of particles in SVGD\n",
    "\n",
    "statistics = np.zeros([n_sim,N.shape[0],particle_num.shape[0]])\n",
    "statistics_sms = np.zeros([n_sim,N.shape[0],particle_num.shape[0]])\n",
    "\n",
    "for idP in range(particle_num.shape[0]):\n",
    "\n",
    "    for idN in range(N.shape[0]):\n",
    "\n",
    "        for idx in range(n_sim):\n",
    "            true_state = np.random.randn(4)\n",
    "            data = sim_experiment(true_state, N = N[idN])\n",
    "            svgd, sms_sample = svgd_prestates(N[idN],z_observed=data[2],x_observed=data[0],y_observed=data[1], svgd_num_part = particle_num[idP], svgd_lear_rate = 0.01, svgd_iter = 400)\n",
    "            svgd_state = svgd_rho_mean(svgd)\n",
    "            sms_state = sms_rho(sms_sample)\n",
    "            #print(\"svgd_state:\",svgd_state)\n",
    "            #print(\"sms_state\",sms_state)\n",
    "            #print(\"true_state:\",vec2rho(np.array([true_state[0],true_state[3],true_state[1],true_state[2]]) ) )\n",
    "            statistics[idx,idN,idP] = fidelity(vec2rho(np.array([true_state[0],true_state[3],true_state[1],true_state[2]])),svgd_state)\n",
    "            statistics_sms[idx,idN,idP] = fidelity(vec2rho(np.array([true_state[0],true_state[3],true_state[1],true_state[2]])),sms_state)\n",
    "\n",
    "            \n",
    "print(\"--- %s seconds ---\" % (time.time() - start_time))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c27be647",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "ea825f4e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "svgd-fidliety: [[[0.91335548 0.94771287 0.90030946]\n",
      "  [0.99048945 0.99545983 0.98978528]\n",
      "  [0.96866035 0.9879611  0.98840236]]\n",
      "\n",
      " [[0.87190897 0.91917359 0.9278272 ]\n",
      "  [0.9942537  0.99024017 0.99416084]\n",
      "  [0.98447624 0.99967302 0.99351064]]]\n",
      "\n",
      "sms-fidelity: [[[0.9456665  0.95037928 0.89806389]\n",
      "  [0.99686345 0.99655254 0.99462895]\n",
      "  [0.99900668 0.99946196 0.9991141 ]]\n",
      "\n",
      " [[0.86711117 0.92296207 0.92520817]\n",
      "  [0.99427463 0.97595666 0.99731269]\n",
      "  [0.99579379 0.99962931 0.99811669]]]\n",
      "svgd: [[0.89263222 0.93344323 0.91406833]\n",
      " [0.99237157 0.99285    0.99197306]\n",
      " [0.9765683  0.99381706 0.9909565 ]]\n",
      "\n",
      "ave_sms [0.91823185 0.99259815 0.99852042]\n"
     ]
    }
   ],
   "source": [
    "#print(\"svgd-fidliety:\",statistics)\n",
    "#print(\"\")\n",
    "#print(\"sms-fidelity:\",statistics_sms)\n",
    "\n",
    "ave_statistics = np.mean(statistics, axis = 0)\n",
    "ave_statistics_sms=np.mean(np.mean(statistics_sms, axis = 0),axis=1)\n",
    "print(\"svgd:\",ave_statistics)\n",
    "print(\"\")\n",
    "print(\"ave_sms\",ave_statistics_sms)\n",
    "#ave_statistics[0,:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "13462560",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams[\"figure.figsize\"] = 10, 7\n",
    "plt.plot(N, ave_statistics[:,0], label = str(particle_num[0])+\" particles-SVGD\")\n",
    "plt.plot(N, ave_statistics[:,1], label = str(particle_num[1])+\" particles-SVGD\")\n",
    "plt.plot(N, ave_statistics[:,2], label = str(particle_num[2])+\" particles-SVGD\")\n",
    "\n",
    "plt.plot(N, ave_statistics_sms, label = \"SMC\")\n",
    "\n",
    "plt.legend(fontsize=10,loc=0)\n",
    "plt.ylabel(\"Average Fidelity\",fontsize=16)\n",
    "plt.xlabel(\"Number of measurements\",fontsize=16)\n",
    "plt.title(\"SVGD using adam vs SMC: Average Fidelity vs $N$\",fontsize=16)\n",
    "plt.legend(fontsize=13)\n",
    "plt.grid()\n",
    "plt.savefig('svgd-adam-measurements.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "23cf15d4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(particle_num, ave_statistics[0,:], label = str(N[0])+\" measurements-SVGD\")\n",
    "plt.plot(particle_num, ave_statistics[1,:], label = str(N[1])+\" measurements-SVGD\")\n",
    "plt.plot(particle_num, ave_statistics[2,:], label = str(N[2])+\" measurements-SVGD\")\n",
    "\n",
    "plt.legend(fontsize=10,loc=0)\n",
    "plt.ylabel(\"Average Fidelity\",fontsize=16)\n",
    "plt.xlabel(\"Number of particles\",fontsize=16)\n",
    "plt.title(\"Aver. fidelity vs num. particles :using adam \",fontsize=16)\n",
    "plt.legend(fontsize=11)\n",
    "plt.grid()\n",
    "plt.savefig('svgd-adam-particles.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e226b1a4",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "92da2e51",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1e1a79fb",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "83a2878e",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "4a0b5b94",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running on PyMC3 v3.11.2\n"
     ]
    }
   ],
   "source": [
    "\n",
    "print(f\"Running on PyMC3 v{pm.__version__}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2cf01652",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "aa38a52a",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.10.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}