examples/nested/example.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Example notebook showing how to use the nested sampler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import sys\n",
    "import torch\n",
    "from getdist import plots, MCSamples\n",
    "import getdist\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "path = os.path.realpath(os.path.join(os.getcwd(), '../..'))\n",
    "sys.path.insert(0, path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from nnest import NestedSampler\n",
    "from nnest.likelihoods import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "%config InlineBackend.figure_format = 'retina'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Likelihood + prior\n",
    "#like = Himmelblau(2)\n",
    "#transform = lambda x: 5*x\n",
    "like = Rosenbrock(10)\n",
    "transform = lambda x: 5*x\n",
    "#like = Gaussian(2, 0.9)\n",
    "#transform = lambda x: 3*x\n",
    "#like = Eggbox(2)\n",
    "#transform = lambda x: 5*np.pi*x\n",
    "#like = GaussianShell(2)\n",
    "#transform = lambda x: 5*x\n",
    "#like = GaussianMix(2)\n",
    "#transform = lambda x: 5*x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creating directory for new run logs/test/run39\n",
      "[nnest.trainer] [INFO] SingleSpeedSpline(\n",
      "  (flow): NormalizingFlow(\n",
      "    (flows): ModuleList(\n",
      "      (0): ActNorm()\n",
      "      (1): Invertible1x1Conv()\n",
      "      (2): NSF_CL(\n",
      "        (f1): MLP(\n",
      "          (net): Sequential(\n",
      "            (0): Linear(in_features=5, out_features=16, bias=True)\n",
      "            (1): LeakyReLU(negative_slope=0.2)\n",
      "            (2): Linear(in_features=16, out_features=16, bias=True)\n",
      "            (3): LeakyReLU(negative_slope=0.2)\n",
      "            (4): Linear(in_features=16, out_features=16, bias=True)\n",
      "            (5): LeakyReLU(negative_slope=0.2)\n",
      "            (6): Linear(in_features=16, out_features=115, bias=True)\n",
      "          )\n",
      "        )\n",
      "        (f2): MLP(\n",
      "          (net): Sequential(\n",
      "            (0): Linear(in_features=5, out_features=16, bias=True)\n",
      "            (1): LeakyReLU(negative_slope=0.2)\n",
      "            (2): Linear(in_features=16, out_features=16, bias=True)\n",
      "            (3): LeakyReLU(negative_slope=0.2)\n",
      "            (4): Linear(in_features=16, out_features=16, bias=True)\n",
      "            (5): LeakyReLU(negative_slope=0.2)\n",
      "            (6): Linear(in_features=16, out_features=115, bias=True)\n",
      "          )\n",
      "        )\n",
      "      )\n",
      "      (3): ActNorm()\n",
      "      (4): Invertible1x1Conv()\n",
      "      (5): NSF_CL(\n",
      "        (f1): MLP(\n",
      "          (net): Sequential(\n",
      "            (0): Linear(in_features=5, out_features=16, bias=True)\n",
      "            (1): LeakyReLU(negative_slope=0.2)\n",
      "            (2): Linear(in_features=16, out_features=16, bias=True)\n",
      "            (3): LeakyReLU(negative_slope=0.2)\n",
      "            (4): Linear(in_features=16, out_features=16, bias=True)\n",
      "            (5): LeakyReLU(negative_slope=0.2)\n",
      "            (6): Linear(in_features=16, out_features=115, bias=True)\n",
      "          )\n",
      "        )\n",
      "        (f2): MLP(\n",
      "          (net): Sequential(\n",
      "            (0): Linear(in_features=5, out_features=16, bias=True)\n",
      "            (1): LeakyReLU(negative_slope=0.2)\n",
      "            (2): Linear(in_features=16, out_features=16, bias=True)\n",
      "            (3): LeakyReLU(negative_slope=0.2)\n",
      "            (4): Linear(in_features=16, out_features=16, bias=True)\n",
      "            (5): LeakyReLU(negative_slope=0.2)\n",
      "            (6): Linear(in_features=16, out_features=115, bias=True)\n",
      "          )\n",
      "        )\n",
      "      )\n",
      "      (6): ActNorm()\n",
      "      (7): Invertible1x1Conv()\n",
      "      (8): NSF_CL(\n",
      "        (f1): MLP(\n",
      "          (net): Sequential(\n",
      "            (0): Linear(in_features=5, out_features=16, bias=True)\n",
      "            (1): LeakyReLU(negative_slope=0.2)\n",
      "            (2): Linear(in_features=16, out_features=16, bias=True)\n",
      "            (3): LeakyReLU(negative_slope=0.2)\n",
      "            (4): Linear(in_features=16, out_features=16, bias=True)\n",
      "            (5): LeakyReLU(negative_slope=0.2)\n",
      "            (6): Linear(in_features=16, out_features=115, bias=True)\n",
      "          )\n",
      "        )\n",
      "        (f2): MLP(\n",
      "          (net): Sequential(\n",
      "            (0): Linear(in_features=5, out_features=16, bias=True)\n",
      "            (1): LeakyReLU(negative_slope=0.2)\n",
      "            (2): Linear(in_features=16, out_features=16, bias=True)\n",
      "            (3): LeakyReLU(negative_slope=0.2)\n",
      "            (4): Linear(in_features=16, out_features=16, bias=True)\n",
      "            (5): LeakyReLU(negative_slope=0.2)\n",
      "            (6): Linear(in_features=16, out_features=115, bias=True)\n",
      "          )\n",
      "        )\n",
      "      )\n",
      "    )\n",
      "  )\n",
      ")\n",
      "[nnest.trainer] [INFO] Number of network params: [16260]\n",
      "[nnest.trainer] [INFO] Device [cpu]\n",
      "[nnest.sampler] [INFO] Num base params [10]\n",
      "[nnest.sampler] [INFO] Num derived params [0]\n",
      "[nnest.sampler] [INFO] Total params [10]\n",
      "[nnest.sampler] [INFO] Num live points [1000]\n"
     ]
    }
   ],
   "source": [
    "sampler = NestedSampler(like.x_dim, like, transform=transform, num_live_points=1000, hidden_dim=16, num_blocks=3, flow='spline')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[nnest.sampler] [INFO] MCMC steps [50]\n",
      "[nnest.sampler] [INFO] Initial scale [0.6325]\n",
      "[nnest.sampler] [INFO] Volume switch [-1.0000]\n",
      "[nnest.sampler] [INFO] Step [0] loglstar [-3.4611e+05] max logl [-5.9317e+03] logz [-3.4611e+05] vol [1.00000e+00] ncalls [1001] mean calls [0.0000]\n",
      "[nnest.sampler] [INFO] Step [200] loglstar [-1.7580e+05] max logl [-5.9317e+03] logz [-1.7580e+05] vol [8.18731e-01] ncalls [1215] mean calls [1.1000]\n",
      "[nnest.sampler] [INFO] Step [400] loglstar [-1.4460e+05] max logl [-5.9317e+03] logz [-1.4461e+05] vol [6.70320e-01] ncalls [1473] mean calls [1.4000]\n",
      "[nnest.sampler] [INFO] Step [600] loglstar [-1.2492e+05] max logl [-5.9317e+03] logz [-1.2493e+05] vol [5.48812e-01] ncalls [1777] mean calls [1.7000]\n",
      "[nnest.sampler] [INFO] Step [800] loglstar [-1.0860e+05] max logl [-5.9317e+03] logz [-1.0861e+05] vol [4.49329e-01] ncalls [2175] mean calls [2.8000]\n",
      "[nnest.sampler] [INFO] Step [1000] loglstar [-9.6207e+04] max logl [-5.1524e+03] logz [-9.6215e+04] vol [3.67879e-01] ncalls [2689] mean calls [2.5000]\n",
      "[nnest.sampler] [INFO] Step [1200] loglstar [-8.6079e+04] max logl [-5.1524e+03] logz [-8.6087e+04] vol [3.01194e-01] ncalls [3300] mean calls [4.0500]\n",
      "[nnest.sampler] [INFO] Step [1400] loglstar [-7.8701e+04] max logl [-5.1524e+03] logz [-7.8709e+04] vol [2.46597e-01] ncalls [3979] mean calls [3.3000]\n",
      "[nnest.sampler] [INFO] Step [1600] loglstar [-7.1208e+04] max logl [-5.1524e+03] logz [-7.1216e+04] vol [2.01897e-01] ncalls [4798] mean calls [4.3500]\n",
      "[nnest.sampler] [INFO] Step [1800] loglstar [-6.5142e+04] max logl [-5.1524e+03] logz [-6.5151e+04] vol [1.65299e-01] ncalls [5808] mean calls [4.2500]\n",
      "[nnest.sampler] [INFO] Step [2000] loglstar [-5.9881e+04] max logl [-2.7636e+03] logz [-5.9890e+04] vol [1.35335e-01] ncalls [7230] mean calls [8.2500]\n",
      "[nnest.sampler] [INFO] Step [2200] loglstar [-5.5125e+04] max logl [-2.7636e+03] logz [-5.5134e+04] vol [1.10803e-01] ncalls [8780] mean calls [6.9500]\n",
      "[nnest.sampler] [INFO] Step [2400] loglstar [-5.0426e+04] max logl [-2.7636e+03] logz [-5.0435e+04] vol [9.07180e-02] ncalls [10642] mean calls [10.2000]\n",
      "[nnest.sampler] [INFO] Step [2600] loglstar [-4.6522e+04] max logl [-2.7636e+03] logz [-4.6531e+04] vol [7.42736e-02] ncalls [13020] mean calls [14.2500]\n",
      "[nnest.sampler] [INFO] Step [2800] loglstar [-4.2585e+04] max logl [-2.7636e+03] logz [-4.2594e+04] vol [6.08101e-02] ncalls [15591] mean calls [16.4500]\n",
      "[nnest.sampler] [INFO] Step [3000] loglstar [-3.8554e+04] max logl [-2.7636e+03] logz [-3.8564e+04] vol [4.97871e-02] ncalls [19577] mean calls [22.7500]\n",
      "[nnest.sampler] [INFO] Step [3200] loglstar [-3.5490e+04] max logl [-2.7636e+03] logz [-3.5500e+04] vol [4.07622e-02] ncalls [24891] mean calls [22.7000]\n",
      "[nnest.sampler] [INFO] Step [3400] loglstar [-3.2824e+04] max logl [-2.7636e+03] logz [-3.2834e+04] vol [3.33733e-02] ncalls [31012] mean calls [35.1500]\n",
      "[nnest.sampler] [INFO] Step [3600] loglstar [-3.0110e+04] max logl [-2.1637e+03] logz [-3.0120e+04] vol [2.73237e-02] ncalls [37121] mean calls [45.8500]\n",
      "[nnest.sampler] [INFO] Rejection prior no longer efficient, switching sampling method\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0675]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [0.0526] validation loss [0.0502]\n",
      "[nnest.trainer] [INFO] Epoch [76] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [26] validation loss [0.0467] train time (s) [46.2190]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5360] min ESS [2.7570] max ESS [5.2856] average jump [0.1742]\n",
      "[nnest.sampler] [INFO] Step [3800] loglstar [-2.7859e+04] maxlogl [-2.1637e+03] logz [-2.7870e+04] vol [2.23708e-02] ncalls [44991] scale [0.4606]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0646]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [0.0412] validation loss [0.0379]\n",
      "[nnest.trainer] [INFO] Epoch [52] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [2] validation loss [0.0379] train time (s) [21.6525]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4580] min ESS [2.3328] max ESS [4.6992] average jump [0.1371]\n",
      "[nnest.sampler] [INFO] Step [4000] loglstar [-2.5500e+04] maxlogl [-2.1637e+03] logz [-2.5511e+04] vol [1.83156e-02] ncalls [52639] scale [0.2129]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5140] min ESS [2.0878] max ESS [7.5164] average jump [0.1510]\n",
      "[nnest.sampler] [INFO] Step [4200] loglstar [-2.3371e+04] maxlogl [-2.1637e+03] logz [-2.3382e+04] vol [1.49956e-02] ncalls [60228] scale [0.3172]\n",
      "[nnest.sampler] [INFO] Acceptance [0.6840] min ESS [1.6548] max ESS [4.4941] average jump [0.0974]\n",
      "[nnest.sampler] [INFO] Step [4400] loglstar [-2.1491e+04] maxlogl [-2.1637e+03] logz [-2.1503e+04] vol [1.22773e-02] ncalls [67667] scale [0.2246]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5080] min ESS [2.0072] max ESS [7.3968] average jump [0.1271]\n",
      "[nnest.sampler] [INFO] Step [4600] loglstar [-1.9819e+04] maxlogl [-2.1637e+03] logz [-1.9831e+04] vol [1.00518e-02] ncalls [75252] scale [0.1764]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5580] min ESS [1.4733] max ESS [5.2034] average jump [0.1408]\n",
      "[nnest.sampler] [INFO] Step [4800] loglstar [-1.8116e+04] maxlogl [-2.1637e+03] logz [-1.8127e+04] vol [8.22975e-03] ncalls [82901] scale [0.2159]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0589]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [0.0341] validation loss [0.0318]\n",
      "[nnest.trainer] [INFO] Epoch [56] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [6] validation loss [0.0311] train time (s) [29.5174]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5200] min ESS [2.0554] max ESS [5.9744] average jump [0.1351]\n",
      "[nnest.sampler] [INFO] Step [5000] loglstar [-1.6814e+04] maxlogl [-2.1637e+03] logz [-1.6826e+04] vol [6.73795e-03] ncalls [90538] scale [0.1300]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5380] min ESS [3.3664] max ESS [9.6332] average jump [0.1706]\n",
      "[nnest.sampler] [INFO] Step [5200] loglstar [-1.5526e+04] maxlogl [-1.3128e+03] logz [-1.5538e+04] vol [5.51656e-03] ncalls [98200] scale [0.4784]\n",
      "[nnest.sampler] [INFO] Acceptance [0.6140] min ESS [2.3689] max ESS [6.2089] average jump [0.1635]\n",
      "[nnest.sampler] [INFO] Step [5400] loglstar [-1.4327e+04] maxlogl [-1.3128e+03] logz [-1.4339e+04] vol [4.51658e-03] ncalls [105881] scale [0.3172]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5560] min ESS [1.9869] max ESS [11.8416] average jump [0.1262]\n",
      "[nnest.sampler] [INFO] Step [5600] loglstar [-1.3137e+04] maxlogl [-1.2002e+03] logz [-1.3150e+04] vol [3.69786e-03] ncalls [113516] scale [0.3028]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5620] min ESS [2.4363] max ESS [13.2369] average jump [0.1529]\n",
      "[nnest.sampler] [INFO] Step [5800] loglstar [-1.2101e+04] maxlogl [-9.9343e+02] logz [-1.2114e+04] vol [3.02755e-03] ncalls [121308] scale [0.3383]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0534]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [0.0242] validation loss [0.0237]\n",
      "[nnest.trainer] [INFO] Epoch [69] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [19] validation loss [0.0222] train time (s) [31.6180]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4740] min ESS [2.8276] max ESS [6.0755] average jump [0.1184]\n",
      "[nnest.sampler] [INFO] Step [6000] loglstar [-1.1179e+04] maxlogl [-8.0213e+02] logz [-1.1192e+04] vol [2.47875e-03] ncalls [129617] scale [0.4040]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5600] min ESS [2.7267] max ESS [6.2980] average jump [0.1284]\n",
      "[nnest.sampler] [INFO] Step [6200] loglstar [-1.0346e+04] maxlogl [-5.4193e+02] logz [-1.0358e+04] vol [2.02943e-03] ncalls [137249] scale [0.2842]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5420] min ESS [2.6744] max ESS [10.0685] average jump [0.1307]\n",
      "[nnest.sampler] [INFO] Step [6400] loglstar [-9.4711e+03] maxlogl [-5.4193e+02] logz [-9.4844e+03] vol [1.66156e-03] ncalls [144944] scale [0.2319]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5000] min ESS [2.1230] max ESS [8.3358] average jump [0.1359]\n",
      "[nnest.sampler] [INFO] Step [6600] loglstar [-8.8278e+03] maxlogl [-5.4193e+02] logz [-8.8411e+03] vol [1.36037e-03] ncalls [152625] scale [0.2721]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5060] min ESS [3.0141] max ESS [14.9117] average jump [0.1431]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[nnest.sampler] [INFO] Step [6800] loglstar [-8.1809e+03] maxlogl [-5.4193e+02] logz [-8.1942e+03] vol [1.11378e-03] ncalls [160229] scale [0.3008]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0485]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [0.0134] validation loss [0.0103]\n",
      "[nnest.trainer] [INFO] Epoch [53] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [3] validation loss [0.0097] train time (s) [19.9540]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5640] min ESS [2.9880] max ESS [9.6553] average jump [0.1244]\n",
      "[nnest.sampler] [INFO] Step [7000] loglstar [-7.5988e+03] maxlogl [-5.4193e+02] logz [-7.6121e+03] vol [9.11882e-04] ncalls [167864] scale [0.2145]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5300] min ESS [2.6770] max ESS [7.9142] average jump [0.1241]\n",
      "[nnest.sampler] [INFO] Step [7200] loglstar [-7.0754e+03] maxlogl [-5.4193e+02] logz [-7.0892e+03] vol [7.46586e-04] ncalls [176117] scale [0.2287]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5440] min ESS [3.3144] max ESS [8.2445] average jump [0.1186]\n",
      "[nnest.sampler] [INFO] Step [7400] loglstar [-6.5646e+03] maxlogl [-5.4193e+02] logz [-6.5786e+03] vol [6.11253e-04] ncalls [184129] scale [0.1823]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5840] min ESS [2.6490] max ESS [11.3603] average jump [0.1332]\n",
      "[nnest.sampler] [INFO] Step [7600] loglstar [-6.0949e+03] maxlogl [-5.4193e+02] logz [-6.1089e+03] vol [5.00451e-04] ncalls [192063] scale [0.3299]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4700] min ESS [2.5175] max ESS [11.0411] average jump [0.1160]\n",
      "[nnest.sampler] [INFO] Step [7800] loglstar [-5.6352e+03] maxlogl [-5.4193e+02] logz [-5.6496e+03] vol [4.09735e-04] ncalls [200074] scale [0.2732]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0447]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [0.0049] validation loss [0.0029]\n",
      "[nnest.trainer] [INFO] Epoch [74] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [24] validation loss [0.0018] train time (s) [30.5739]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5760] min ESS [1.9122] max ESS [10.3533] average jump [0.1197]\n",
      "[nnest.sampler] [INFO] Step [8000] loglstar [-5.1859e+03] maxlogl [-5.3264e+02] logz [-5.2008e+03] vol [3.35463e-04] ncalls [207990] scale [0.3958]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5540] min ESS [1.8324] max ESS [9.4225] average jump [0.0931]\n",
      "[nnest.sampler] [INFO] Step [8200] loglstar [-4.8260e+03] maxlogl [-5.3264e+02] logz [-4.8407e+03] vol [2.74654e-04] ncalls [215480] scale [0.1640]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5340] min ESS [2.7156] max ESS [12.6801] average jump [0.1435]\n",
      "[nnest.sampler] [INFO] Step [8400] loglstar [-4.4946e+03] maxlogl [-5.3264e+02] logz [-4.5096e+03] vol [2.24867e-04] ncalls [223117] scale [0.2841]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5460] min ESS [3.8117] max ESS [18.3195] average jump [0.1106]\n",
      "[nnest.sampler] [INFO] Step [8600] loglstar [-4.1220e+03] maxlogl [-5.3264e+02] logz [-4.1372e+03] vol [1.84106e-04] ncalls [230663] scale [0.3526]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5300] min ESS [1.9132] max ESS [12.5913] average jump [0.1149]\n",
      "[nnest.sampler] [INFO] Step [8800] loglstar [-3.8609e+03] maxlogl [-5.3264e+02] logz [-3.8760e+03] vol [1.50733e-04] ncalls [238589] scale [0.3385]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0404]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.0067] validation loss [-0.0086]\n",
      "[nnest.trainer] [INFO] Epoch [62] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [12] validation loss [-0.0098] train time (s) [24.2311]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5380] min ESS [2.2873] max ESS [8.1687] average jump [0.1042]\n",
      "[nnest.sampler] [INFO] Step [9000] loglstar [-3.5570e+03] maxlogl [-5.3264e+02] logz [-3.5729e+03] vol [1.23410e-04] ncalls [246089] scale [0.2434]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5500] min ESS [3.8070] max ESS [11.6405] average jump [0.1049]\n",
      "[nnest.sampler] [INFO] Step [9200] loglstar [-3.2964e+03] maxlogl [-5.3264e+02] logz [-3.3119e+03] vol [1.01039e-04] ncalls [253598] scale [0.1915]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4380] min ESS [3.5885] max ESS [14.6704] average jump [0.1169]\n",
      "[nnest.sampler] [INFO] Step [9400] loglstar [-3.0466e+03] maxlogl [-2.3106e+02] logz [-3.0617e+03] vol [8.27241e-05] ncalls [261270] scale [0.2974]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4700] min ESS [3.5670] max ESS [13.5232] average jump [0.1146]\n",
      "[nnest.sampler] [INFO] Step [9600] loglstar [-2.8587e+03] maxlogl [-2.3106e+02] logz [-2.8751e+03] vol [6.77287e-05] ncalls [268826] scale [0.3955]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5020] min ESS [2.8110] max ESS [22.0734] average jump [0.1036]\n",
      "[nnest.sampler] [INFO] Step [9800] loglstar [-2.6550e+03] maxlogl [-2.3106e+02] logz [-2.6712e+03] vol [5.54516e-05] ncalls [276200] scale [0.1761]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0365]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.0162] validation loss [-0.0197]\n",
      "[nnest.trainer] [INFO] Epoch [69] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [19] validation loss [-0.0212] train time (s) [28.3242]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4860] min ESS [1.8156] max ESS [16.9800] average jump [0.0874]\n",
      "[nnest.sampler] [INFO] Step [10000] loglstar [-2.4685e+03] maxlogl [-2.3106e+02] logz [-2.4849e+03] vol [4.53999e-05] ncalls [283852] scale [0.4827]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4680] min ESS [4.1499] max ESS [16.4302] average jump [0.1046]\n",
      "[nnest.sampler] [INFO] Step [10200] loglstar [-2.2961e+03] maxlogl [-2.3106e+02] logz [-2.3118e+03] vol [3.71703e-05] ncalls [291478] scale [0.4800]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5500] min ESS [2.9203] max ESS [7.8321] average jump [0.1047]\n",
      "[nnest.sampler] [INFO] Step [10400] loglstar [-2.1167e+03] maxlogl [-2.3106e+02] logz [-2.1332e+03] vol [3.04325e-05] ncalls [298998] scale [0.3568]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5540] min ESS [2.0273] max ESS [13.8440] average jump [0.1037]\n",
      "[nnest.sampler] [INFO] Step [10600] loglstar [-1.9706e+03] maxlogl [-2.3106e+02] logz [-1.9874e+03] vol [2.49160e-05] ncalls [306565] scale [0.2584]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5500] min ESS [3.4260] max ESS [10.2555] average jump [0.0905]\n",
      "[nnest.sampler] [INFO] Step [10800] loglstar [-1.8446e+03] maxlogl [-2.3106e+02] logz [-1.8611e+03] vol [2.03995e-05] ncalls [314169] scale [0.2184]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0330]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.0265] validation loss [-0.0281]\n",
      "[nnest.trainer] [INFO] Epoch [69] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [19] validation loss [-0.0288] train time (s) [28.2071]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5460] min ESS [3.0398] max ESS [9.3517] average jump [0.0887]\n",
      "[nnest.sampler] [INFO] Step [11000] loglstar [-1.7305e+03] maxlogl [-2.3106e+02] logz [-1.7480e+03] vol [1.67017e-05] ncalls [321822] scale [0.1506]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5380] min ESS [3.9193] max ESS [12.0152] average jump [0.0966]\n",
      "[nnest.sampler] [INFO] Step [11200] loglstar [-1.6144e+03] maxlogl [-2.3106e+02] logz [-1.6322e+03] vol [1.36742e-05] ncalls [329279] scale [0.2346]\n",
      "[nnest.sampler] [INFO] Acceptance [0.6380] min ESS [3.4756] max ESS [8.7069] average jump [0.0707]\n",
      "[nnest.sampler] [INFO] Step [11400] loglstar [-1.5197e+03] maxlogl [-1.9393e+02] logz [-1.5368e+03] vol [1.11955e-05] ncalls [336918] scale [0.3410]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5700] min ESS [2.5781] max ESS [17.5737] average jump [0.0830]\n",
      "[nnest.sampler] [INFO] Step [11600] loglstar [-1.4147e+03] maxlogl [-1.9393e+02] logz [-1.4318e+03] vol [9.16609e-06] ncalls [344457] scale [0.2021]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5720] min ESS [4.2201] max ESS [10.9822] average jump [0.0801]\n",
      "[nnest.sampler] [INFO] Step [11800] loglstar [-1.3336e+03] maxlogl [-1.9393e+02] logz [-1.3505e+03] vol [7.50456e-06] ncalls [352164] scale [0.1846]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0298]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.0361] validation loss [-0.0385]\n",
      "[nnest.trainer] [INFO] Epoch [57] ran out of patience\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[nnest.trainer] [INFO] Best epoch [7] validation loss [-0.0403] train time (s) [21.7799]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5360] min ESS [4.5170] max ESS [17.8169] average jump [0.0931]\n",
      "[nnest.sampler] [INFO] Step [12000] loglstar [-1.2507e+03] maxlogl [-1.9393e+02] logz [-1.2683e+03] vol [6.14421e-06] ncalls [359689] scale [0.2569]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5400] min ESS [3.7656] max ESS [14.9908] average jump [0.0875]\n",
      "[nnest.sampler] [INFO] Step [12200] loglstar [-1.1715e+03] maxlogl [-1.9393e+02] logz [-1.1890e+03] vol [5.03046e-06] ncalls [367238] scale [0.1394]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5320] min ESS [2.8719] max ESS [13.4080] average jump [0.0845]\n",
      "[nnest.sampler] [INFO] Step [12400] loglstar [-1.0954e+03] maxlogl [-1.8747e+02] logz [-1.1140e+03] vol [4.11859e-06] ncalls [374983] scale [0.2498]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5880] min ESS [2.4388] max ESS [9.5460] average jump [0.0778]\n",
      "[nnest.sampler] [INFO] Step [12600] loglstar [-1.0245e+03] maxlogl [-1.8747e+02] logz [-1.0435e+03] vol [3.37202e-06] ncalls [382511] scale [0.2611]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4540] min ESS [3.8799] max ESS [15.3366] average jump [0.0745]\n",
      "[nnest.sampler] [INFO] Step [12800] loglstar [-9.6152e+02] maxlogl [-1.3454e+02] logz [-9.7966e+02] vol [2.76077e-06] ncalls [390198] scale [0.2542]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0273]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.0455] validation loss [-0.0480]\n",
      "[nnest.trainer] [INFO] Epoch [100] train loss [-0.0550] validation loss [-0.0494]\n",
      "[nnest.trainer] [INFO] Epoch [147] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [97] validation loss [-0.0500] train time (s) [72.4307]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.6100] min ESS [4.3775] max ESS [14.9111] average jump [0.0749]\n",
      "[nnest.sampler] [INFO] Step [13000] loglstar [-9.0443e+02] maxlogl [-1.3454e+02] logz [-9.2272e+02] vol [2.26033e-06] ncalls [398216] scale [0.1264]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5720] min ESS [4.5341] max ESS [11.9379] average jump [0.0644]\n",
      "[nnest.sampler] [INFO] Step [13200] loglstar [-8.5036e+02] maxlogl [-1.3454e+02] logz [-8.6921e+02] vol [1.85060e-06] ncalls [405317] scale [0.1022]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5140] min ESS [5.0059] max ESS [12.5340] average jump [0.0861]\n",
      "[nnest.sampler] [INFO] Step [13400] loglstar [-8.0332e+02] maxlogl [-1.3454e+02] logz [-8.2225e+02] vol [1.51514e-06] ncalls [412300] scale [0.1554]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4900] min ESS [6.7564] max ESS [29.3734] average jump [0.0928]\n",
      "[nnest.sampler] [INFO] Step [13600] loglstar [-7.5996e+02] maxlogl [-1.3241e+02] logz [-7.7853e+02] vol [1.24050e-06] ncalls [419364] scale [0.1943]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4680] min ESS [6.2853] max ESS [38.7445] average jump [0.0708]\n",
      "[nnest.sampler] [INFO] Step [13800] loglstar [-7.0842e+02] maxlogl [-1.3241e+02] logz [-7.2732e+02] vol [1.01563e-06] ncalls [426580] scale [0.2996]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0246]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.0559] validation loss [-0.0619]\n",
      "[nnest.trainer] [INFO] Epoch [65] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [15] validation loss [-0.0632] train time (s) [25.7361]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5040] min ESS [6.3800] max ESS [16.2485] average jump [0.0682]\n",
      "[nnest.sampler] [INFO] Step [14000] loglstar [-6.6590e+02] maxlogl [-8.5547e+01] logz [-6.8503e+02] vol [8.31529e-07] ncalls [433838] scale [0.1968]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5760] min ESS [4.3719] max ESS [13.2863] average jump [0.0723]\n",
      "[nnest.sampler] [INFO] Step [14200] loglstar [-6.2600e+02] maxlogl [-8.5547e+01] logz [-6.4561e+02] vol [6.80798e-07] ncalls [441294] scale [0.1446]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5420] min ESS [6.6257] max ESS [15.6188] average jump [0.0800]\n",
      "[nnest.sampler] [INFO] Step [14400] loglstar [-5.8778e+02] maxlogl [-8.5547e+01] logz [-6.0725e+02] vol [5.57390e-07] ncalls [448666] scale [0.3988]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5280] min ESS [3.9070] max ESS [33.3785] average jump [0.0711]\n",
      "[nnest.sampler] [INFO] Step [14600] loglstar [-5.5447e+02] maxlogl [-8.5547e+01] logz [-5.7408e+02] vol [4.56353e-07] ncalls [456641] scale [0.2121]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5420] min ESS [5.5862] max ESS [12.7806] average jump [0.0623]\n",
      "[nnest.sampler] [INFO] Step [14800] loglstar [-5.2029e+02] maxlogl [-8.5547e+01] logz [-5.4041e+02] vol [3.73630e-07] ncalls [464019] scale [0.1659]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0223]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.0647] validation loss [-0.0683]\n",
      "[nnest.trainer] [INFO] Epoch [100] train loss [-0.0762] validation loss [-0.0686]\n",
      "[nnest.trainer] [INFO] Epoch [110] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [60] validation loss [-0.0705] train time (s) [50.9532]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5280] min ESS [2.9585] max ESS [27.2945] average jump [0.0702]\n",
      "[nnest.sampler] [INFO] Step [15000] loglstar [-4.9370e+02] maxlogl [-8.5547e+01] logz [-5.1324e+02] vol [3.05902e-07] ncalls [471483] scale [0.2736]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5700] min ESS [8.9937] max ESS [51.0000] average jump [0.0710]\n",
      "[nnest.sampler] [INFO] Step [15200] loglstar [-4.6906e+02] maxlogl [-8.5547e+01] logz [-4.8883e+02] vol [2.50452e-07] ncalls [478652] scale [0.2280]\n",
      "[nnest.sampler] [INFO] Acceptance [0.6140] min ESS [4.5999] max ESS [35.7242] average jump [0.0664]\n",
      "[nnest.sampler] [INFO] Step [15400] loglstar [-4.4112e+02] maxlogl [-8.1445e+01] logz [-4.6133e+02] vol [2.05052e-07] ncalls [485967] scale [0.2130]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5860] min ESS [3.1209] max ESS [12.5451] average jump [0.0600]\n",
      "[nnest.sampler] [INFO] Step [15600] loglstar [-4.1598e+02] maxlogl [-8.1445e+01] logz [-4.3615e+02] vol [1.67883e-07] ncalls [493186] scale [0.2274]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5800] min ESS [3.9670] max ESS [26.8358] average jump [0.0592]\n",
      "[nnest.sampler] [INFO] Step [15800] loglstar [-3.9426e+02] maxlogl [-6.3384e+01] logz [-4.1478e+02] vol [1.37451e-07] ncalls [500505] scale [0.2969]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0203]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.0763] validation loss [-0.0790]\n",
      "[nnest.trainer] [INFO] Epoch [56] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [6] validation loss [-0.0795] train time (s) [21.6743]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5100] min ESS [6.1400] max ESS [19.7314] average jump [0.0590]\n",
      "[nnest.sampler] [INFO] Step [16000] loglstar [-3.7444e+02] maxlogl [-6.3384e+01] logz [-3.9471e+02] vol [1.12535e-07] ncalls [507783] scale [0.2990]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5520] min ESS [5.2415] max ESS [12.5958] average jump [0.0614]\n",
      "[nnest.sampler] [INFO] Step [16200] loglstar [-3.5656e+02] maxlogl [-6.3384e+01] logz [-3.7705e+02] vol [9.21360e-08] ncalls [515391] scale [0.1404]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5940] min ESS [7.1343] max ESS [51.0000] average jump [0.0518]\n",
      "[nnest.sampler] [INFO] Step [16400] loglstar [-3.3734e+02] maxlogl [-6.3384e+01] logz [-3.5809e+02] vol [7.54346e-08] ncalls [522977] scale [0.1995]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5980] min ESS [3.3950] max ESS [10.9107] average jump [0.0412]\n",
      "[nnest.sampler] [INFO] Step [16600] loglstar [-3.1865e+02] maxlogl [-6.3384e+01] logz [-3.3975e+02] vol [6.17606e-08] ncalls [530555] scale [0.0716]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5220] min ESS [4.9428] max ESS [23.1221] average jump [0.0577]\n",
      "[nnest.sampler] [INFO] Step [16800] loglstar [-3.0338e+02] maxlogl [-5.8804e+01] logz [-3.2470e+02] vol [5.05653e-08] ncalls [538177] scale [0.1287]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0184]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.0846] validation loss [-0.0892]\n",
      "[nnest.trainer] [INFO] Epoch [88] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [38] validation loss [-0.0911] train time (s) [38.6131]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5300] min ESS [3.1431] max ESS [51.0000] average jump [0.0602]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[nnest.sampler] [INFO] Step [17000] loglstar [-2.8817e+02] maxlogl [-4.5922e+01] logz [-3.0953e+02] vol [4.13994e-08] ncalls [545702] scale [0.2490]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4740] min ESS [2.2318] max ESS [18.8039] average jump [0.0535]\n",
      "[nnest.sampler] [INFO] Step [17200] loglstar [-2.7304e+02] maxlogl [-4.5922e+01] logz [-2.9438e+02] vol [3.38949e-08] ncalls [553316] scale [0.3748]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5580] min ESS [5.3318] max ESS [35.9814] average jump [0.0507]\n",
      "[nnest.sampler] [INFO] Step [17400] loglstar [-2.6012e+02] maxlogl [-4.5922e+01] logz [-2.8147e+02] vol [2.77508e-08] ncalls [560787] scale [0.2339]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5820] min ESS [4.2009] max ESS [12.0693] average jump [0.0493]\n",
      "[nnest.sampler] [INFO] Step [17600] loglstar [-2.4757e+02] maxlogl [-4.5922e+01] logz [-2.6934e+02] vol [2.27205e-08] ncalls [568476] scale [0.1816]\n",
      "[nnest.sampler] [INFO] Acceptance [0.6920] min ESS [5.6413] max ESS [22.9391] average jump [0.0284]\n",
      "[nnest.sampler] [INFO] Step [17800] loglstar [-2.3588e+02] maxlogl [-4.1310e+01] logz [-2.5772e+02] vol [1.86019e-08] ncalls [576068] scale [0.1771]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0167]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.0947] validation loss [-0.0981]\n",
      "[nnest.trainer] [INFO] Epoch [58] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [8] validation loss [-0.0988] train time (s) [22.8614]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5660] min ESS [6.9845] max ESS [26.2761] average jump [0.0470]\n",
      "[nnest.sampler] [INFO] Step [18000] loglstar [-2.2440e+02] maxlogl [-4.1310e+01] logz [-2.4621e+02] vol [1.52300e-08] ncalls [583612] scale [0.2358]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5260] min ESS [11.2800] max ESS [35.3137] average jump [0.0559]\n",
      "[nnest.sampler] [INFO] Step [18200] loglstar [-2.1426e+02] maxlogl [-4.0694e+01] logz [-2.3643e+02] vol [1.24693e-08] ncalls [591280] scale [0.2888]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4500] min ESS [11.3533] max ESS [38.6672] average jump [0.0599]\n",
      "[nnest.sampler] [INFO] Step [18400] loglstar [-2.0453e+02] maxlogl [-3.8744e+01] logz [-2.2689e+02] vol [1.02090e-08] ncalls [598835] scale [0.3249]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5520] min ESS [7.4008] max ESS [46.4055] average jump [0.0522]\n",
      "[nnest.sampler] [INFO] Step [18600] loglstar [-1.9525e+02] maxlogl [-3.8744e+01] logz [-2.1759e+02] vol [8.35839e-09] ncalls [606579] scale [0.6339]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5980] min ESS [5.9755] max ESS [20.2982] average jump [0.0421]\n",
      "[nnest.sampler] [INFO] Step [18800] loglstar [-1.8729e+02] maxlogl [-3.8744e+01] logz [-2.0961e+02] vol [6.84327e-09] ncalls [614360] scale [0.2111]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0151]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.1049] validation loss [-0.1093]\n",
      "[nnest.trainer] [INFO] Epoch [91] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [41] validation loss [-0.1109] train time (s) [40.6632]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5160] min ESS [1.6150] max ESS [17.6606] average jump [0.0463]\n",
      "[nnest.sampler] [INFO] Step [19000] loglstar [-1.7800e+02] maxlogl [-3.8744e+01] logz [-2.0075e+02] vol [5.60280e-09] ncalls [621949] scale [0.2133]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5460] min ESS [6.8332] max ESS [51.0000] average jump [0.0444]\n",
      "[nnest.sampler] [INFO] Step [19200] loglstar [-1.7038e+02] maxlogl [-3.8744e+01] logz [-1.9314e+02] vol [4.58718e-09] ncalls [629270] scale [0.3467]\n",
      "[nnest.sampler] [INFO] Acceptance [0.6160] min ESS [7.8460] max ESS [51.0000] average jump [0.0352]\n",
      "[nnest.sampler] [INFO] Step [19400] loglstar [-1.6243e+02] maxlogl [-3.8744e+01] logz [-1.8551e+02] vol [3.75567e-09] ncalls [636864] scale [0.4087]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5680] min ESS [7.4941] max ESS [39.2630] average jump [0.0402]\n",
      "[nnest.sampler] [INFO] Step [19600] loglstar [-1.5477e+02] maxlogl [-3.8744e+01] logz [-1.7772e+02] vol [3.07488e-09] ncalls [644840] scale [0.3172]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5820] min ESS [7.4994] max ESS [26.5971] average jump [0.0387]\n",
      "[nnest.sampler] [INFO] Step [19800] loglstar [-1.4875e+02] maxlogl [-3.4085e+01] logz [-1.7175e+02] vol [2.51750e-09] ncalls [652393] scale [0.2832]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0136]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.1160] validation loss [-0.1208]\n",
      "[nnest.trainer] [INFO] Epoch [58] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [8] validation loss [-0.1209] train time (s) [22.4831]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4640] min ESS [6.3590] max ESS [51.0000] average jump [0.0426]\n",
      "[nnest.sampler] [INFO] Step [20000] loglstar [-1.4324e+02] maxlogl [-3.4085e+01] logz [-1.6652e+02] vol [2.06115e-09] ncalls [659840] scale [0.1886]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4320] min ESS [7.5492] max ESS [51.0000] average jump [0.0434]\n",
      "[nnest.sampler] [INFO] Step [20200] loglstar [-1.3645e+02] maxlogl [-3.4085e+01] logz [-1.6025e+02] vol [1.68753e-09] ncalls [667232] scale [0.3177]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5280] min ESS [3.9532] max ESS [29.5403] average jump [0.0377]\n",
      "[nnest.sampler] [INFO] Step [20400] loglstar [-1.3022e+02] maxlogl [-3.4085e+01] logz [-1.5397e+02] vol [1.38163e-09] ncalls [674619] scale [0.1551]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5020] min ESS [9.0865] max ESS [31.7376] average jump [0.0365]\n",
      "[nnest.sampler] [INFO] Step [20600] loglstar [-1.2459e+02] maxlogl [-3.4085e+01] logz [-1.4853e+02] vol [1.13119e-09] ncalls [682169] scale [0.2740]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5520] min ESS [6.6986] max ESS [39.1368] average jump [0.0389]\n",
      "[nnest.sampler] [INFO] Step [20800] loglstar [-1.2001e+02] maxlogl [-3.0399e+01] logz [-1.4385e+02] vol [9.26136e-10] ncalls [689753] scale [0.2444]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0124]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.1237] validation loss [-0.1297]\n",
      "[nnest.trainer] [INFO] Epoch [59] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [9] validation loss [-0.1319] train time (s) [23.4032]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5640] min ESS [6.2308] max ESS [42.5554] average jump [0.0293]\n",
      "[nnest.sampler] [INFO] Step [21000] loglstar [-1.1459e+02] maxlogl [-3.0399e+01] logz [-1.3875e+02] vol [7.58256e-10] ncalls [697378] scale [0.1552]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5780] min ESS [6.5639] max ESS [46.2607] average jump [0.0430]\n",
      "[nnest.sampler] [INFO] Step [21200] loglstar [-1.1048e+02] maxlogl [-3.0399e+01] logz [-1.3465e+02] vol [6.20808e-10] ncalls [705068] scale [0.2714]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5240] min ESS [4.2081] max ESS [21.4926] average jump [0.0343]\n",
      "[nnest.sampler] [INFO] Step [21400] loglstar [-1.0620e+02] maxlogl [-3.0399e+01] logz [-1.3062e+02] vol [5.08274e-10] ncalls [713129] scale [0.1097]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5640] min ESS [11.0278] max ESS [51.0000] average jump [0.0350]\n",
      "[nnest.sampler] [INFO] Step [21600] loglstar [-1.0164e+02] maxlogl [-3.0399e+01] logz [-1.2633e+02] vol [4.16140e-10] ncalls [720711] scale [0.1913]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5380] min ESS [7.8830] max ESS [51.0000] average jump [0.0376]\n",
      "[nnest.sampler] [INFO] Step [21800] loglstar [-9.7860e+01] maxlogl [-2.9627e+01] logz [-1.2238e+02] vol [3.40706e-10] ncalls [728444] scale [0.3136]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0114]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.1343] validation loss [-0.1381]\n",
      "[nnest.trainer] [INFO] Epoch [88] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [38] validation loss [-0.1393] train time (s) [39.5034]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4800] min ESS [13.1726] max ESS [51.0000] average jump [0.0343]\n",
      "[nnest.sampler] [INFO] Step [22000] loglstar [-9.3861e+01] maxlogl [-2.9627e+01] logz [-1.1869e+02] vol [2.78947e-10] ncalls [736112] scale [0.2694]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5160] min ESS [6.1873] max ESS [51.0000] average jump [0.0307]\n",
      "[nnest.sampler] [INFO] Step [22200] loglstar [-8.9914e+01] maxlogl [-2.9627e+01] logz [-1.1513e+02] vol [2.28382e-10] ncalls [743571] scale [0.1207]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[nnest.sampler] [INFO] Acceptance [0.5460] min ESS [1.0790] max ESS [24.0346] average jump [0.0299]\n",
      "[nnest.sampler] [INFO] Step [22400] loglstar [-8.6209e+01] maxlogl [-2.9627e+01] logz [-1.1140e+02] vol [1.86984e-10] ncalls [751018] scale [0.1139]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5920] min ESS [8.0975] max ESS [51.0000] average jump [0.0300]\n",
      "[nnest.sampler] [INFO] Step [22600] loglstar [-8.2718e+01] maxlogl [-2.9627e+01] logz [-1.0817e+02] vol [1.53089e-10] ncalls [758466] scale [0.1656]\n",
      "[nnest.sampler] [INFO] Acceptance [0.6340] min ESS [6.7245] max ESS [51.0000] average jump [0.0179]\n",
      "[nnest.sampler] [INFO] Step [22800] loglstar [-7.9212e+01] maxlogl [-2.2025e+01] logz [-1.0479e+02] vol [1.25339e-10] ncalls [765996] scale [0.2136]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0101]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.1453] validation loss [-0.1479]\n",
      "[nnest.trainer] [INFO] Epoch [52] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [2] validation loss [-0.1480] train time (s) [19.3312]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4360] min ESS [10.2446] max ESS [51.0000] average jump [0.0405]\n",
      "[nnest.sampler] [INFO] Step [23000] loglstar [-7.6657e+01] maxlogl [-2.0044e+01] logz [-1.0214e+02] vol [1.02619e-10] ncalls [773628] scale [0.3463]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5160] min ESS [15.1674] max ESS [51.0000] average jump [0.0330]\n",
      "[nnest.sampler] [INFO] Step [23200] loglstar [-7.3800e+01] maxlogl [-2.0044e+01] logz [-9.9480e+01] vol [8.40172e-11] ncalls [781713] scale [0.1309]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5240] min ESS [11.2964] max ESS [51.0000] average jump [0.0257]\n",
      "[nnest.sampler] [INFO] Step [23400] loglstar [-7.0973e+01] maxlogl [-2.0044e+01] logz [-9.6913e+01] vol [6.87874e-11] ncalls [789268] scale [0.1560]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5160] min ESS [7.1949] max ESS [20.2729] average jump [0.0343]\n",
      "[nnest.sampler] [INFO] Step [23600] loglstar [-6.7822e+01] maxlogl [-1.5331e+01] logz [-9.4184e+01] vol [5.63184e-11] ncalls [797002] scale [0.2807]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5600] min ESS [9.1467] max ESS [51.0000] average jump [0.0235]\n",
      "[nnest.sampler] [INFO] Step [23800] loglstar [-6.5280e+01] maxlogl [-1.5331e+01] logz [-9.1506e+01] vol [4.61096e-11] ncalls [804680] scale [0.4973]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0091]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.1542] validation loss [-0.1578]\n",
      "[nnest.trainer] [INFO] Epoch [87] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [37] validation loss [-0.1612] train time (s) [38.0458]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5300] min ESS [10.1578] max ESS [51.0000] average jump [0.0324]\n",
      "[nnest.sampler] [INFO] Step [24000] loglstar [-6.2875e+01] maxlogl [-1.5331e+01] logz [-8.9272e+01] vol [3.77513e-11] ncalls [812399] scale [0.4739]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5520] min ESS [10.0772] max ESS [51.0000] average jump [0.0206]\n",
      "[nnest.sampler] [INFO] Step [24200] loglstar [-6.0447e+01] maxlogl [-1.5331e+01] logz [-8.7066e+01] vol [3.09082e-11] ncalls [819716] scale [0.1635]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5260] min ESS [6.0514] max ESS [51.0000] average jump [0.0277]\n",
      "[nnest.sampler] [INFO] Step [24400] loglstar [-5.8228e+01] maxlogl [-1.5331e+01] logz [-8.4921e+01] vol [2.53055e-11] ncalls [827350] scale [0.2961]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4900] min ESS [11.8674] max ESS [51.0000] average jump [0.0286]\n",
      "[nnest.sampler] [INFO] Step [24600] loglstar [-5.6021e+01] maxlogl [-1.5331e+01] logz [-8.2881e+01] vol [2.07184e-11] ncalls [834839] scale [0.1947]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5080] min ESS [14.0294] max ESS [51.0000] average jump [0.0296]\n",
      "[nnest.sampler] [INFO] Step [24800] loglstar [-5.3967e+01] maxlogl [-1.5331e+01] logz [-8.0955e+01] vol [1.69628e-11] ncalls [842269] scale [0.2656]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0083]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.1648] validation loss [-0.1695]\n",
      "[nnest.trainer] [INFO] Epoch [53] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [3] validation loss [-0.1695] train time (s) [20.2553]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5900] min ESS [9.5187] max ESS [51.0000] average jump [0.0234]\n",
      "[nnest.sampler] [INFO] Step [25000] loglstar [-5.2072e+01] maxlogl [-1.5331e+01] logz [-7.9137e+01] vol [1.38879e-11] ncalls [849873] scale [0.1745]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5080] min ESS [11.6536] max ESS [51.0000] average jump [0.0242]\n",
      "[nnest.sampler] [INFO] Step [25200] loglstar [-5.0342e+01] maxlogl [-1.5331e+01] logz [-7.7565e+01] vol [1.13705e-11] ncalls [857504] scale [0.2515]\n",
      "[nnest.sampler] [INFO] Acceptance [0.6700] min ESS [9.6195] max ESS [51.0000] average jump [0.0157]\n",
      "[nnest.sampler] [INFO] Step [25400] loglstar [-4.8527e+01] maxlogl [-1.5331e+01] logz [-7.6030e+01] vol [9.30937e-12] ncalls [865049] scale [0.2215]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5280] min ESS [8.0846] max ESS [51.0000] average jump [0.0230]\n",
      "[nnest.sampler] [INFO] Step [25600] loglstar [-4.6827e+01] maxlogl [-1.5331e+01] logz [-7.4455e+01] vol [7.62187e-12] ncalls [872798] scale [0.2002]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5500] min ESS [8.3907] max ESS [51.0000] average jump [0.0225]\n",
      "[nnest.sampler] [INFO] Step [25800] loglstar [-4.4998e+01] maxlogl [-1.5331e+01] logz [-7.2842e+01] vol [6.24026e-12] ncalls [880626] scale [0.2029]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0075]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.1742] validation loss [-0.1789]\n",
      "[nnest.trainer] [INFO] Epoch [79] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [29] validation loss [-0.1794] train time (s) [31.4687]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5340] min ESS [8.1914] max ESS [51.0000] average jump [0.0220]\n",
      "[nnest.sampler] [INFO] Step [26000] loglstar [-4.3291e+01] maxlogl [-1.5331e+01] logz [-7.1311e+01] vol [5.10909e-12] ncalls [888530] scale [0.2771]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5100] min ESS [10.3956] max ESS [51.0000] average jump [0.0247]\n",
      "[nnest.sampler] [INFO] Step [26200] loglstar [-4.1646e+01] maxlogl [-1.5331e+01] logz [-6.9830e+01] vol [4.18297e-12] ncalls [896482] scale [0.3290]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5620] min ESS [18.9410] max ESS [51.0000] average jump [0.0249]\n",
      "[nnest.sampler] [INFO] Step [26400] loglstar [-4.0199e+01] maxlogl [-1.5331e+01] logz [-6.8486e+01] vol [3.42472e-12] ncalls [904256] scale [0.3331]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5120] min ESS [8.0865] max ESS [51.0000] average jump [0.0228]\n",
      "[nnest.sampler] [INFO] Step [26600] loglstar [-3.8563e+01] maxlogl [-1.5331e+01] logz [-6.7106e+01] vol [2.80393e-12] ncalls [911862] scale [0.2344]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5880] min ESS [9.0177] max ESS [51.0000] average jump [0.0251]\n",
      "[nnest.sampler] [INFO] Step [26800] loglstar [-3.7308e+01] maxlogl [-1.0435e+01] logz [-6.5872e+01] vol [2.29566e-12] ncalls [919576] scale [0.4445]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0067]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.1851] validation loss [-0.1891]\n",
      "[nnest.trainer] [INFO] Epoch [54] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [4] validation loss [-0.1895] train time (s) [20.5405]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5540] min ESS [6.7126] max ESS [51.0000] average jump [0.0214]\n",
      "[nnest.sampler] [INFO] Step [27000] loglstar [-3.6035e+01] maxlogl [-1.0435e+01] logz [-6.4739e+01] vol [1.87953e-12] ncalls [927225] scale [0.3536]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5300] min ESS [6.7020] max ESS [51.0000] average jump [0.0256]\n",
      "[nnest.sampler] [INFO] Step [27200] loglstar [-3.4764e+01] maxlogl [-1.0435e+01] logz [-6.3646e+01] vol [1.53883e-12] ncalls [934943] scale [0.2426]\n",
      "[nnest.sampler] [INFO] Acceptance [0.6220] min ESS [6.4022] max ESS [51.0000] average jump [0.0195]\n",
      "[nnest.sampler] [INFO] Step [27400] loglstar [-3.3597e+01] maxlogl [-1.0435e+01] logz [-6.2602e+01] vol [1.25989e-12] ncalls [942765] scale [0.3604]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4940] min ESS [8.1121] max ESS [30.7468] average jump [0.0208]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[nnest.sampler] [INFO] Step [27600] loglstar [-3.2526e+01] maxlogl [-1.0435e+01] logz [-6.1644e+01] vol [1.03151e-12] ncalls [950455] scale [0.4392]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4860] min ESS [6.7622] max ESS [51.0000] average jump [0.0191]\n",
      "[nnest.sampler] [INFO] Step [27800] loglstar [-3.1513e+01] maxlogl [-1.0435e+01] logz [-6.0763e+01] vol [8.44527e-13] ncalls [958392] scale [0.2211]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0061]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.1936] validation loss [-0.1968]\n",
      "[nnest.trainer] [INFO] Epoch [100] train loss [-0.2034] validation loss [-0.1988]\n",
      "[nnest.trainer] [INFO] Epoch [145] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [95] validation loss [-0.1995] train time (s) [70.7922]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4860] min ESS [7.6183] max ESS [51.0000] average jump [0.0235]\n",
      "[nnest.sampler] [INFO] Step [28000] loglstar [-3.0428e+01] maxlogl [-1.0435e+01] logz [-5.9897e+01] vol [6.91440e-13] ncalls [966669] scale [0.1752]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5400] min ESS [12.8082] max ESS [51.0000] average jump [0.0249]\n",
      "[nnest.sampler] [INFO] Step [28200] loglstar [-2.9516e+01] maxlogl [-1.0435e+01] logz [-5.9079e+01] vol [5.66103e-13] ncalls [973955] scale [0.3217]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5760] min ESS [6.5674] max ESS [51.0000] average jump [0.0221]\n",
      "[nnest.sampler] [INFO] Step [28400] loglstar [-2.8550e+01] maxlogl [-1.0435e+01] logz [-5.8287e+01] vol [4.63486e-13] ncalls [981363] scale [0.3022]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5680] min ESS [9.2527] max ESS [51.0000] average jump [0.0228]\n",
      "[nnest.sampler] [INFO] Step [28600] loglstar [-2.7628e+01] maxlogl [-8.3657e+00] logz [-5.7546e+01] vol [3.79470e-13] ncalls [988761] scale [0.1830]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5940] min ESS [9.8594] max ESS [51.0000] average jump [0.0181]\n",
      "[nnest.sampler] [INFO] Step [28800] loglstar [-2.6766e+01] maxlogl [-8.3657e+00] logz [-5.6811e+01] vol [3.10684e-13] ncalls [996373] scale [0.1994]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0056]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.2052] validation loss [-0.2101]\n",
      "[nnest.trainer] [INFO] Epoch [57] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [7] validation loss [-0.2114] train time (s) [22.2061]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5020] min ESS [5.5643] max ESS [51.0000] average jump [0.0183]\n",
      "[nnest.sampler] [INFO] Step [29000] loglstar [-2.5946e+01] maxlogl [-8.3657e+00] logz [-5.6126e+01] vol [2.54367e-13] ncalls [1003868] scale [0.2541]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5540] min ESS [7.6653] max ESS [51.0000] average jump [0.0144]\n",
      "[nnest.sampler] [INFO] Step [29200] loglstar [-2.5199e+01] maxlogl [-8.3657e+00] logz [-5.5503e+01] vol [2.08258e-13] ncalls [1011462] scale [0.1885]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5360] min ESS [6.9273] max ESS [51.0000] average jump [0.0196]\n",
      "[nnest.sampler] [INFO] Step [29400] loglstar [-2.4464e+01] maxlogl [-8.3657e+00] logz [-5.4912e+01] vol [1.70507e-13] ncalls [1019494] scale [0.2375]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5500] min ESS [7.6630] max ESS [51.0000] average jump [0.0200]\n",
      "[nnest.sampler] [INFO] Step [29600] loglstar [-2.3810e+01] maxlogl [-8.3657e+00] logz [-5.4368e+01] vol [1.39599e-13] ncalls [1027116] scale [0.2805]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4560] min ESS [9.0675] max ESS [51.0000] average jump [0.0180]\n",
      "[nnest.sampler] [INFO] Step [29800] loglstar [-2.3082e+01] maxlogl [-8.3657e+00] logz [-5.3844e+01] vol [1.14294e-13] ncalls [1034625] scale [0.1879]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0050]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.2133] validation loss [-0.2167]\n",
      "[nnest.trainer] [INFO] Epoch [88] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [38] validation loss [-0.2196] train time (s) [38.4525]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5400] min ESS [6.2211] max ESS [51.0000] average jump [0.0162]\n",
      "[nnest.sampler] [INFO] Step [30000] loglstar [-2.2451e+01] maxlogl [-8.3657e+00] logz [-5.3339e+01] vol [9.35762e-14] ncalls [1042364] scale [0.3261]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5380] min ESS [5.3889] max ESS [51.0000] average jump [0.0186]\n",
      "[nnest.sampler] [INFO] Step [30200] loglstar [-2.1790e+01] maxlogl [-8.3657e+00] logz [-5.2870e+01] vol [7.66137e-14] ncalls [1049862] scale [0.4485]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5700] min ESS [7.7441] max ESS [51.0000] average jump [0.0244]\n",
      "[nnest.sampler] [INFO] Step [30400] loglstar [-2.1111e+01] maxlogl [-8.3657e+00] logz [-5.2384e+01] vol [6.27260e-14] ncalls [1057514] scale [0.2119]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5220] min ESS [9.0879] max ESS [51.0000] average jump [0.0207]\n",
      "[nnest.sampler] [INFO] Step [30600] loglstar [-2.0509e+01] maxlogl [-8.3657e+00] logz [-5.1923e+01] vol [5.13557e-14] ncalls [1065174] scale [0.1255]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5240] min ESS [10.0002] max ESS [51.0000] average jump [0.0195]\n",
      "[nnest.sampler] [INFO] Step [30800] loglstar [-1.9945e+01] maxlogl [-8.3657e+00] logz [-5.1495e+01] vol [4.20465e-14] ncalls [1072911] scale [0.1209]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0046]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.2249] validation loss [-0.2276]\n",
      "[nnest.trainer] [INFO] Epoch [60] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [10] validation loss [-0.2288] train time (s) [23.3031]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5440] min ESS [8.2108] max ESS [13.0809] average jump [0.0193]\n",
      "[nnest.sampler] [INFO] Step [31000] loglstar [-1.9432e+01] maxlogl [-5.8216e+00] logz [-5.1089e+01] vol [3.44248e-14] ncalls [1080664] scale [0.2087]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5380] min ESS [8.1538] max ESS [51.0000] average jump [0.0183]\n",
      "[nnest.sampler] [INFO] Step [31200] loglstar [-1.8883e+01] maxlogl [-5.8216e+00] logz [-5.0707e+01] vol [2.81846e-14] ncalls [1088384] scale [0.2496]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5240] min ESS [12.8148] max ESS [51.0000] average jump [0.0191]\n",
      "[nnest.sampler] [INFO] Step [31400] loglstar [-1.8378e+01] maxlogl [-5.8216e+00] logz [-5.0345e+01] vol [2.30756e-14] ncalls [1096140] scale [0.5395]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5260] min ESS [5.7850] max ESS [10.6553] average jump [0.0214]\n",
      "[nnest.sampler] [INFO] Step [31600] loglstar [-1.7908e+01] maxlogl [-5.8216e+00] logz [-5.0007e+01] vol [1.88927e-14] ncalls [1104004] scale [0.2818]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5440] min ESS [13.2152] max ESS [51.0000] average jump [0.0172]\n",
      "[nnest.sampler] [INFO] Step [31800] loglstar [-1.7468e+01] maxlogl [-5.8216e+00] logz [-4.9692e+01] vol [1.54680e-14] ncalls [1111862] scale [0.4195]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0042]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.2316] validation loss [-0.2339]\n",
      "[nnest.trainer] [INFO] Epoch [65] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [15] validation loss [-0.2371] train time (s) [26.2663]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5100] min ESS [4.8432] max ESS [51.0000] average jump [0.0169]\n",
      "[nnest.sampler] [INFO] Step [32000] loglstar [-1.7005e+01] maxlogl [-5.8216e+00] logz [-4.9397e+01] vol [1.26642e-14] ncalls [1120152] scale [0.1572]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5000] min ESS [9.2019] max ESS [51.0000] average jump [0.0201]\n",
      "[nnest.sampler] [INFO] Step [32200] loglstar [-1.6566e+01] maxlogl [-5.8216e+00] logz [-4.9110e+01] vol [1.03685e-14] ncalls [1127911] scale [0.4569]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5720] min ESS [4.7889] max ESS [51.0000] average jump [0.0122]\n",
      "[nnest.sampler] [INFO] Step [32400] loglstar [-1.6144e+01] maxlogl [-5.8216e+00] logz [-4.8833e+01] vol [8.48904e-15] ncalls [1135803] scale [0.3087]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5340] min ESS [7.4325] max ESS [51.0000] average jump [0.0161]\n",
      "[nnest.sampler] [INFO] Step [32600] loglstar [-1.5723e+01] maxlogl [-5.8216e+00] logz [-4.8570e+01] vol [6.95024e-15] ncalls [1143518] scale [0.0716]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4780] min ESS [6.5413] max ESS [51.0000] average jump [0.0146]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[nnest.sampler] [INFO] Step [32800] loglstar [-1.5288e+01] maxlogl [-4.8657e+00] logz [-4.8316e+01] vol [5.69038e-15] ncalls [1151401] scale [0.2095]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0038]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.2413] validation loss [-0.2455]\n",
      "[nnest.trainer] [INFO] Epoch [85] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [35] validation loss [-0.2488] train time (s) [36.9057]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.6700] min ESS [3.8790] max ESS [20.3432] average jump [0.0133]\n",
      "[nnest.sampler] [INFO] Step [33000] loglstar [-1.4921e+01] maxlogl [-4.8657e+00] logz [-4.8074e+01] vol [4.65889e-15] ncalls [1159352] scale [0.2054]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5420] min ESS [6.1998] max ESS [16.3988] average jump [0.0197]\n",
      "[nnest.sampler] [INFO] Step [33200] loglstar [-1.4542e+01] maxlogl [-4.8657e+00] logz [-4.7847e+01] vol [3.81437e-15] ncalls [1167153] scale [0.2387]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5200] min ESS [8.7000] max ESS [51.0000] average jump [0.0188]\n",
      "[nnest.sampler] [INFO] Step [33400] loglstar [-1.4241e+01] maxlogl [-4.8657e+00] logz [-4.7638e+01] vol [3.12294e-15] ncalls [1174942] scale [0.1453]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5500] min ESS [4.9222] max ESS [51.0000] average jump [0.0211]\n",
      "[nnest.sampler] [INFO] Step [33600] loglstar [-1.3893e+01] maxlogl [-4.8657e+00] logz [-4.7442e+01] vol [2.55685e-15] ncalls [1182783] scale [0.1620]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5320] min ESS [7.3335] max ESS [51.0000] average jump [0.0206]\n",
      "[nnest.sampler] [INFO] Step [33800] loglstar [-1.3557e+01] maxlogl [-4.8657e+00] logz [-4.7259e+01] vol [2.09337e-15] ncalls [1190594] scale [0.1247]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0034]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.2516] validation loss [-0.2555]\n",
      "[nnest.trainer] [INFO] Epoch [72] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [22] validation loss [-0.2581] train time (s) [29.9348]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4580] min ESS [11.1996] max ESS [33.5564] average jump [0.0282]\n",
      "[nnest.sampler] [INFO] Step [34000] loglstar [-1.3224e+01] maxlogl [-4.8657e+00] logz [-4.7083e+01] vol [1.71391e-15] ncalls [1198911] scale [0.2759]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5480] min ESS [9.1485] max ESS [51.0000] average jump [0.0213]\n",
      "[nnest.sampler] [INFO] Step [34200] loglstar [-1.2879e+01] maxlogl [-4.8657e+00] logz [-4.6912e+01] vol [1.40323e-15] ncalls [1206861] scale [0.2948]\n",
      "[nnest.sampler] [INFO] Acceptance [0.6440] min ESS [3.9386] max ESS [51.0000] average jump [0.0174]\n",
      "[nnest.sampler] [INFO] Step [34400] loglstar [-1.2532e+01] maxlogl [-4.8657e+00] logz [-4.6745e+01] vol [1.14887e-15] ncalls [1214867] scale [0.1967]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5320] min ESS [13.2801] max ESS [51.0000] average jump [0.0224]\n",
      "[nnest.sampler] [INFO] Step [34600] loglstar [-1.2198e+01] maxlogl [-2.7055e+00] logz [-4.6583e+01] vol [9.40613e-16] ncalls [1222837] scale [0.2492]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5460] min ESS [3.7308] max ESS [51.0000] average jump [0.0199]\n",
      "[nnest.sampler] [INFO] Step [34800] loglstar [-1.1911e+01] maxlogl [-2.7055e+00] logz [-4.6427e+01] vol [7.70109e-16] ncalls [1230734] scale [0.1884]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0032]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.2628] validation loss [-0.2667]\n",
      "[nnest.trainer] [INFO] Epoch [64] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [14] validation loss [-0.2671] train time (s) [25.2992]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.6180] min ESS [21.8876] max ESS [51.0000] average jump [0.0209]\n",
      "[nnest.sampler] [INFO] Step [35000] loglstar [-1.1618e+01] maxlogl [-2.7055e+00] logz [-4.6281e+01] vol [6.30512e-16] ncalls [1238707] scale [0.2388]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5120] min ESS [12.1436] max ESS [51.0000] average jump [0.0185]\n",
      "[nnest.sampler] [INFO] Step [35200] loglstar [-1.1313e+01] maxlogl [-2.7055e+00] logz [-4.6140e+01] vol [5.16219e-16] ncalls [1246845] scale [0.6317]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5240] min ESS [15.2613] max ESS [51.0000] average jump [0.0252]\n",
      "[nnest.sampler] [INFO] Step [35400] loglstar [-1.0995e+01] maxlogl [-2.4175e+00] logz [-4.6004e+01] vol [4.22645e-16] ncalls [1255431] scale [0.1624]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5380] min ESS [12.1264] max ESS [51.0000] average jump [0.0205]\n",
      "[nnest.sampler] [INFO] Step [35600] loglstar [-1.0733e+01] maxlogl [-2.4175e+00] logz [-4.5874e+01] vol [3.46032e-16] ncalls [1263639] scale [0.5073]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4980] min ESS [4.0687] max ESS [11.7892] average jump [0.0177]\n",
      "[nnest.sampler] [INFO] Step [35800] loglstar [-1.0446e+01] maxlogl [-2.4175e+00] logz [-4.5750e+01] vol [2.83307e-16] ncalls [1271975] scale [0.1401]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0029]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.2707] validation loss [-0.2768]\n",
      "[nnest.trainer] [INFO] Epoch [73] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [23] validation loss [-0.2808] train time (s) [30.5056]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5400] min ESS [17.0321] max ESS [51.0000] average jump [0.0235]\n",
      "[nnest.sampler] [INFO] Step [36000] loglstar [-1.0157e+01] maxlogl [-2.4175e+00] logz [-4.5630e+01] vol [2.31952e-16] ncalls [1280134] scale [0.1283]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5320] min ESS [42.2662] max ESS [51.0000] average jump [0.0283]\n",
      "[nnest.sampler] [INFO] Step [36200] loglstar [-9.8969e+00] maxlogl [-2.4175e+00] logz [-4.5515e+01] vol [1.89906e-16] ncalls [1288192] scale [0.2481]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5800] min ESS [18.5342] max ESS [51.0000] average jump [0.0230]\n",
      "[nnest.sampler] [INFO] Step [36400] loglstar [-9.6218e+00] maxlogl [-2.4175e+00] logz [-4.5405e+01] vol [1.55482e-16] ncalls [1296204] scale [0.3693]\n",
      "[nnest.sampler] [INFO] Acceptance [0.6140] min ESS [12.7801] max ESS [51.0000] average jump [0.0187]\n",
      "[nnest.sampler] [INFO] Step [36600] loglstar [-9.3481e+00] maxlogl [-2.4175e+00] logz [-4.5298e+01] vol [1.27298e-16] ncalls [1304334] scale [0.2086]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4680] min ESS [28.5778] max ESS [51.0000] average jump [0.0238]\n",
      "[nnest.sampler] [INFO] Step [36800] loglstar [-9.0641e+00] maxlogl [-2.4175e+00] logz [-4.5194e+01] vol [1.04223e-16] ncalls [1312377] scale [0.2007]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0026]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.2819] validation loss [-0.2860]\n",
      "[nnest.trainer] [INFO] Epoch [100] train loss [-0.2884] validation loss [-0.2863]\n",
      "[nnest.trainer] [INFO] Epoch [126] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [76] validation loss [-0.2882] train time (s) [59.2459]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5000] min ESS [10.2875] max ESS [51.0000] average jump [0.0156]\n",
      "[nnest.sampler] [INFO] Step [37000] loglstar [-8.7706e+00] maxlogl [-2.4175e+00] logz [-4.5092e+01] vol [8.53305e-17] ncalls [1320528] scale [0.1892]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5600] min ESS [4.4462] max ESS [51.0000] average jump [0.0261]\n",
      "[nnest.sampler] [INFO] Step [37200] loglstar [-8.5143e+00] maxlogl [-2.4175e+00] logz [-4.4993e+01] vol [6.98627e-17] ncalls [1328461] scale [0.3026]\n",
      "[nnest.sampler] [INFO] Acceptance [0.6140] min ESS [20.4975] max ESS [51.0000] average jump [0.0188]\n",
      "[nnest.sampler] [INFO] Step [37400] loglstar [-8.2727e+00] maxlogl [-1.1575e+00] logz [-4.4898e+01] vol [5.71987e-17] ncalls [1336500] scale [0.1843]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5980] min ESS [4.5755] max ESS [51.0000] average jump [0.0205]\n",
      "[nnest.sampler] [INFO] Step [37600] loglstar [-8.0287e+00] maxlogl [-1.1575e+00] logz [-4.4808e+01] vol [4.68304e-17] ncalls [1344575] scale [0.3171]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5180] min ESS [11.9828] max ESS [51.0000] average jump [0.0194]\n",
      "[nnest.sampler] [INFO] Step [37800] loglstar [-7.7657e+00] maxlogl [-1.1575e+00] logz [-4.4721e+01] vol [3.83415e-17] ncalls [1352651] scale [0.2187]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[nnest.trainer] [INFO] Training jitter [0.0024]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.2917] validation loss [-0.2963]\n",
      "[nnest.trainer] [INFO] Epoch [89] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [39] validation loss [-0.3004] train time (s) [39.1734]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5100] min ESS [7.6825] max ESS [51.0000] average jump [0.0232]\n",
      "[nnest.sampler] [INFO] Step [38000] loglstar [-7.5116e+00] maxlogl [-1.1575e+00] logz [-4.4636e+01] vol [3.13913e-17] ncalls [1360708] scale [0.1739]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4620] min ESS [16.7946] max ESS [51.0000] average jump [0.0222]\n",
      "[nnest.sampler] [INFO] Step [38200] loglstar [-7.2819e+00] maxlogl [-1.1575e+00] logz [-4.4555e+01] vol [2.57010e-17] ncalls [1368611] scale [0.2465]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4980] min ESS [33.2600] max ESS [51.0000] average jump [0.0218]\n",
      "[nnest.sampler] [INFO] Step [38400] loglstar [-7.0425e+00] maxlogl [-1.1575e+00] logz [-4.4478e+01] vol [2.10422e-17] ncalls [1376673] scale [0.2137]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5000] min ESS [5.7141] max ESS [51.0000] average jump [0.0286]\n",
      "[nnest.sampler] [INFO] Step [38600] loglstar [-6.8354e+00] maxlogl [-1.1575e+00] logz [-4.4405e+01] vol [1.72279e-17] ncalls [1384595] scale [0.1882]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5040] min ESS [10.3978] max ESS [51.0000] average jump [0.0203]\n",
      "[nnest.sampler] [INFO] Step [38800] loglstar [-6.6128e+00] maxlogl [-1.1575e+00] logz [-4.4335e+01] vol [1.41050e-17] ncalls [1392894] scale [0.1798]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0021]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.3022] validation loss [-0.3076]\n",
      "[nnest.trainer] [INFO] Epoch [100] train loss [-0.3072] validation loss [-0.3071]\n",
      "[nnest.trainer] [INFO] Epoch [133] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [83] validation loss [-0.3083] train time (s) [64.2255]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5400] min ESS [10.7225] max ESS [51.0000] average jump [0.0253]\n",
      "[nnest.sampler] [INFO] Step [39000] loglstar [-6.3901e+00] maxlogl [-1.1575e+00] logz [-4.4268e+01] vol [1.15482e-17] ncalls [1400866] scale [0.2029]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5520] min ESS [19.3879] max ESS [51.0000] average jump [0.0234]\n",
      "[nnest.sampler] [INFO] Step [39200] loglstar [-6.1587e+00] maxlogl [-1.1575e+00] logz [-4.4204e+01] vol [9.45489e-18] ncalls [1408786] scale [0.3764]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5940] min ESS [19.8221] max ESS [51.0000] average jump [0.0233]\n",
      "[nnest.sampler] [INFO] Step [39400] loglstar [-5.9553e+00] maxlogl [-1.1575e+00] logz [-4.4143e+01] vol [7.74101e-18] ncalls [1416720] scale [0.4486]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5200] min ESS [31.0662] max ESS [51.0000] average jump [0.0258]\n",
      "[nnest.sampler] [INFO] Step [39600] loglstar [-5.7510e+00] maxlogl [-1.1575e+00] logz [-4.4085e+01] vol [6.33780e-18] ncalls [1424639] scale [0.3569]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5580] min ESS [10.1870] max ESS [51.0000] average jump [0.0180]\n",
      "[nnest.sampler] [INFO] Step [39800] loglstar [-5.5602e+00] maxlogl [-1.1575e+00] logz [-4.4031e+01] vol [5.18895e-18] ncalls [1432529] scale [0.2200]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0019]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.3132] validation loss [-0.3189]\n",
      "[nnest.trainer] [INFO] Epoch [65] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [15] validation loss [-0.3197] train time (s) [26.2290]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4820] min ESS [25.1355] max ESS [51.0000] average jump [0.0271]\n",
      "[nnest.sampler] [INFO] Step [40000] loglstar [-5.3758e+00] maxlogl [-1.1575e+00] logz [-4.3979e+01] vol [4.24835e-18] ncalls [1440336] scale [0.2353]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5560] min ESS [11.6145] max ESS [51.0000] average jump [0.0181]\n",
      "[nnest.sampler] [INFO] Step [40200] loglstar [-5.1536e+00] maxlogl [-1.1575e+00] logz [-4.3931e+01] vol [3.47826e-18] ncalls [1448411] scale [0.1459]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5940] min ESS [12.3563] max ESS [51.0000] average jump [0.0207]\n",
      "[nnest.sampler] [INFO] Step [40400] loglstar [-4.9735e+00] maxlogl [-1.1575e+00] logz [-4.3884e+01] vol [2.84776e-18] ncalls [1456592] scale [0.5452]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5680] min ESS [6.7322] max ESS [51.0000] average jump [0.0236]\n",
      "[nnest.sampler] [INFO] Step [40600] loglstar [-4.7847e+00] maxlogl [-1.1575e+00] logz [-4.3840e+01] vol [2.33155e-18] ncalls [1464641] scale [0.3431]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5340] min ESS [9.3036] max ESS [51.0000] average jump [0.0240]\n",
      "[nnest.sampler] [INFO] Step [40800] loglstar [-4.6307e+00] maxlogl [-1.1575e+00] logz [-4.3799e+01] vol [1.90891e-18] ncalls [1472707] scale [0.4104]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0017]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.3229] validation loss [-0.3284]\n",
      "[nnest.trainer] [INFO] Epoch [75] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [25] validation loss [-0.3300] train time (s) [30.8702]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5220] min ESS [24.7310] max ESS [51.0000] average jump [0.0238]\n",
      "[nnest.sampler] [INFO] Step [41000] loglstar [-4.4348e+00] maxlogl [-7.4324e-01] logz [-4.3761e+01] vol [1.56288e-18] ncalls [1480918] scale [0.1450]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5000] min ESS [11.1324] max ESS [51.0000] average jump [0.0231]\n",
      "[nnest.sampler] [INFO] Step [41200] loglstar [-4.2551e+00] maxlogl [-7.4324e-01] logz [-4.3724e+01] vol [1.27958e-18] ncalls [1489060] scale [0.1925]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5620] min ESS [10.0554] max ESS [51.0000] average jump [0.0215]\n",
      "[nnest.sampler] [INFO] Step [41400] loglstar [-4.1058e+00] maxlogl [-7.4324e-01] logz [-4.3690e+01] vol [1.04763e-18] ncalls [1497133] scale [0.2560]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5220] min ESS [15.7624] max ESS [51.0000] average jump [0.0235]\n",
      "[nnest.sampler] [INFO] Step [41600] loglstar [-3.9599e+00] maxlogl [-7.4324e-01] logz [-4.3659e+01] vol [8.57728e-19] ncalls [1505386] scale [0.2504]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4480] min ESS [16.9699] max ESS [51.0000] average jump [0.0177]\n",
      "[nnest.sampler] [INFO] Step [41800] loglstar [-3.8290e+00] maxlogl [-7.4324e-01] logz [-4.3631e+01] vol [7.02248e-19] ncalls [1513885] scale [0.3119]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0016]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.3331] validation loss [-0.3389]\n",
      "[nnest.trainer] [INFO] Epoch [100] train loss [-0.3374] validation loss [-0.3394]\n",
      "[nnest.trainer] [INFO] Epoch [125] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [75] validation loss [-0.3411] train time (s) [58.8937]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.6080] min ESS [5.6898] max ESS [51.0000] average jump [0.0219]\n",
      "[nnest.sampler] [INFO] Step [42000] loglstar [-3.6787e+00] maxlogl [-7.4324e-01] logz [-4.3604e+01] vol [5.74952e-19] ncalls [1522135] scale [0.2165]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5420] min ESS [4.8369] max ESS [51.0000] average jump [0.0238]\n",
      "[nnest.sampler] [INFO] Step [42200] loglstar [-3.5432e+00] maxlogl [-7.4324e-01] logz [-4.3581e+01] vol [4.70731e-19] ncalls [1530174] scale [0.2320]\n",
      "[nnest.sampler] [INFO] Acceptance [0.6080] min ESS [7.4021] max ESS [51.0000] average jump [0.0205]\n",
      "[nnest.sampler] [INFO] Step [42400] loglstar [-3.4051e+00] maxlogl [-7.4324e-01] logz [-4.3559e+01] vol [3.85402e-19] ncalls [1538104] scale [0.1847]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5240] min ESS [13.6470] max ESS [51.0000] average jump [0.0204]\n",
      "[nnest.sampler] [INFO] Step [42600] loglstar [-3.2717e+00] maxlogl [-7.4324e-01] logz [-4.3538e+01] vol [3.15540e-19] ncalls [1546100] scale [0.4117]\n",
      "[nnest.sampler] [INFO] Acceptance [0.4900] min ESS [12.4423] max ESS [51.0000] average jump [0.0200]\n",
      "[nnest.sampler] [INFO] Step [42800] loglstar [-3.1415e+00] maxlogl [-7.4324e-01] logz [-4.3520e+01] vol [2.58343e-19] ncalls [1554079] scale [0.2264]\n",
      "[nnest.trainer] [INFO] Number of training samples [1000]\n",
      "[nnest.trainer] [INFO] Training jitter [0.0014]\n",
      "[nnest.trainer] [INFO] Epoch [1] train loss [-0.3431] validation loss [-0.3483]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[nnest.trainer] [INFO] Epoch [100] train loss [-0.3468] validation loss [-0.3482]\n",
      "[nnest.trainer] [INFO] Epoch [200] train loss [-0.3484] validation loss [-0.3486]\n",
      "[nnest.trainer] [INFO] Epoch [212] ran out of patience\n",
      "[nnest.trainer] [INFO] Best epoch [162] validation loss [-0.3504] train time (s) [106.6622]]\n",
      "[nnest.sampler] [INFO] Acceptance [0.5680] min ESS [9.1330] max ESS [51.0000] average jump [0.0181]\n",
      "[nnest.sampler] [INFO] Step [43000] loglstar [-3.0162e+00] maxlogl [-7.4324e-01] logz [-4.3503e+01] vol [2.11513e-19] ncalls [1562091] scale [0.2786]\n",
      "[nnest.sampler] [INFO] niter: 43180\n",
      " ncall: 1569099\n",
      " nsamples: 44180\n",
      " logz: -43.364 +/-  0.193\n",
      " h: 37.226\n"
     ]
    }
   ],
   "source": [
    "sampler.run(strategy=['rejection_prior', 'mcmc'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-43.363593444541415\n"
     ]
    }
   ],
   "source": [
    "print(sampler.logz)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Removed no burn in\n"
     ]
    }
   ],
   "source": [
    "mc = MCSamples(samples=sampler.samples, weights=sampler.weights, loglikes=-sampler.loglikes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "18714.95808885336\n",
      "Marginalized limits: 0.68; 0.95; 0.99\n",
      "\n",
      "parameter   mean           sddev          lower1         upper1         limit1 lower2         upper2         limit2 lower3         upper3         limit3 \n",
      "param1      9.2942498E-01  2.6244553E-01  9.3568182E-01  1.0400492E+00  two    7.1014115E-01  1.1045941E+00  two   -9.8165318E-01  1.0898117E+00  two     p_{1}\n",
      "param2      9.3502268E-01  1.5582903E-01  9.2683009E-01  1.0415066E+00  two    5.9880925E-01  1.1003050E+00  two    5.5421009E-02  1.0885581E+00  two     p_{2}\n",
      "param3      8.9987233E-01  1.9995277E-01  9.0634676E-01  1.0434677E+00  two    8.6521282E-02  1.0999472E+00  two   -8.0397026E-04  1.1028576E+00  two     p_{3}\n",
      "param4      8.5213000E-01  2.4753901E-01  8.7008819E-01  1.0486420E+00  two   -2.1320781E-02  1.0921656E+00  two   -5.7217502E-02  1.0812706E+00  two     p_{4}\n",
      "param5      7.9267314E-01  2.9228390E-01  7.9540090E-01  1.0626390E+00  two   -3.6084414E-02  1.0922112E+00  two   -5.5718727E-02  1.0969761E+00  two     p_{5}\n",
      "param6      7.1834758E-01  3.3388961E-01  4.1295118E-02  1.0827183E+00  two   -6.3201019E-02  1.1045531E+00  two   -8.4568017E-02  1.1139771E+00  two     p_{6}\n",
      "param7      6.3178150E-01  3.6713910E-01 -1.3810443E-02  1.0978257E+00  two   -6.8638169E-02  1.1328175E+00  two   -1.1264044E-01  1.1530223E+00  two     p_{7}\n",
      "param8      5.3924032E-01  4.0182256E-01 -5.3170958E-02  1.1010693E+00  two   -8.2530704E-02  1.1896453E+00  two   -1.3447013E-01  1.2832371E+00  two     p_{8}\n",
      "param9      4.5649445E-01  4.5835841E-01 -9.4213990E-02  7.8864405E-01  two   -1.2660924E-01  1.3610048E+00  two   -1.6346511E-01  1.5926827E+00  two     p_{9}\n",
      "param10     4.1985883E-01  6.0913529E-01 -1.3111641E-01  4.6856271E-01  two   -1.7486429E-01  1.7893812E+00  two   -2.0956134E-01  2.4358404E+00  two     p_{10}\n",
      "\n",
      "Best fit sample -log(Like) = 0.743243\n",
      "mean(-Ln(like)) = 6.137791\n",
      "-Ln(mean like)  = 3.802688\n",
      "\n",
      "parameter   bestfit        lower1         upper1         lower2         upper2\n",
      "param1      1.0058366E+00 -1.0425877E+00  1.1084473E+00 -1.0744886E+00  1.1291890E+00   p_{1}\n",
      "param2      1.0188127E+00  6.6150710E-01  1.1130077E+00 -4.9577132E-02  1.1373989E+00   p_{2}\n",
      "param3      1.0280814E+00  4.6566062E-01  1.1046867E+00 -1.1210345E-01  1.1308499E+00   p_{3}\n",
      "param4      1.0290408E+00  1.9650081E-01  1.1096407E+00 -1.2310747E-01  1.1347923E+00   p_{4}\n",
      "param5      1.0114628E+00  2.1871137E-02  1.1184296E+00 -1.5401597E-01  1.1195874E+00   p_{5}\n",
      "param6      1.0388850E+00 -7.9114586E-02  1.1427660E+00 -1.5007887E-01  1.1427660E+00   p_{6}\n",
      "param7      1.0494729E+00 -7.6537896E-02  1.1846037E+00 -1.9067520E-01  1.1846037E+00   p_{7}\n",
      "param8      1.1332552E+00 -1.0247632E-01  1.3798316E+00 -2.3023030E-01  1.3798316E+00   p_{8}\n",
      "param9      1.3082969E+00 -1.3567181E-01  1.8124911E+00 -2.1502918E-01  1.8124911E+00   p_{9}\n",
      "param10     1.7298469E+00 -1.7362583E-01  3.2589066E+00 -2.4905037E-01  3.2589066E+00   p_{10}\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(mc.getEffectiveSamples())\n",
    "print(mc.getMargeStats())\n",
    "print(mc.likeStats)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 55 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 565,
       "width": 565
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "g = plots.getSubplotPlotter(width_inch=8)\n",
    "g.triangle_plot(mc, filled=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}