examples/TorchANI.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Adversarial attacks using TorchANI models and data\n",
    "\n",
    "In this notebook, we use TorchANI ([paper](https://pubs.acs.org/doi/10.1021/acs.jcim.0c00451) and [source code](https://github.com/aiqm/torchani)) to show that adversarial attacks on uncertainty gives distinct configurations from the training data, as shown in the supporting information of [our paper](https://arxiv.org/abs/2101.11588).\n",
    "\n",
    "The notebook uses the models with the helper functions in the `robust` folder in this repo and requires installation of the [`torchani`](https://github.com/aiqm/torchani) package."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: torchani in /home/atan14/anaconda3/lib/python3.8/site-packages (2.2)\n",
      "Requirement already satisfied: lark-parser in /home/atan14/anaconda3/lib/python3.8/site-packages (from torchani) (0.11.3)\n",
      "Requirement already satisfied: requests in /home/atan14/anaconda3/lib/python3.8/site-packages (from torchani) (2.24.0)\n",
      "Requirement already satisfied: torch in /home/atan14/anaconda3/lib/python3.8/site-packages (from torchani) (1.8.1)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in /home/atan14/anaconda3/lib/python3.8/site-packages (from requests->torchani) (2020.6.20)\n",
      "Requirement already satisfied: urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1 in /home/atan14/anaconda3/lib/python3.8/site-packages (from requests->torchani) (1.25.9)\n",
      "Requirement already satisfied: idna<3,>=2.5 in /home/atan14/anaconda3/lib/python3.8/site-packages (from requests->torchani) (2.10)\n",
      "Requirement already satisfied: chardet<4,>=3.0.2 in /home/atan14/anaconda3/lib/python3.8/site-packages (from requests->torchani) (3.0.4)\n",
      "Requirement already satisfied: numpy in /home/atan14/anaconda3/lib/python3.8/site-packages (from torch->torchani) (1.18.5)\n",
      "Requirement already satisfied: typing-extensions in /home/atan14/anaconda3/lib/python3.8/site-packages (from torch->torchani) (3.7.4.2)\n",
      "--2021-06-02 21:19:29--  https://www.dropbox.com/sh/2c8zdqc1hrqsgwy/AAD6l24ngoiFa6DRapF6HPk5a/\n",
      "Resolving www.dropbox.com (www.dropbox.com)... 162.125.4.18, 2620:100:6019:18::a27d:412\n",
      "Connecting to www.dropbox.com (www.dropbox.com)|162.125.4.18|:443... connected.\n",
      "HTTP request sent, awaiting response... 301 Moved Permanently\n",
      "Location: /sh/raw/2c8zdqc1hrqsgwy/AAD6l24ngoiFa6DRapF6HPk5a [following]\n",
      "--2021-06-02 21:19:29--  https://www.dropbox.com/sh/raw/2c8zdqc1hrqsgwy/AAD6l24ngoiFa6DRapF6HPk5a\n",
      "Reusing existing connection to www.dropbox.com:443.\n",
      "HTTP request sent, awaiting response... 302 Found\n",
      "Location: https://ucb8cad6845c8493be5be45272e7.dl.dropboxusercontent.com/zip_download_get/AzE31PibD0ty4-zKoPBgcp4Vgm_lx1SOI6Ry0q5M1RIJTBFnQciy51KN2EOW8XGs81DAmqHs6aG8v52ce6Vh6zeSDSN3aoJW9Hy0WYVtfXh2Tw# [following]\n",
      "--2021-06-02 21:19:30--  https://ucb8cad6845c8493be5be45272e7.dl.dropboxusercontent.com/zip_download_get/AzE31PibD0ty4-zKoPBgcp4Vgm_lx1SOI6Ry0q5M1RIJTBFnQciy51KN2EOW8XGs81DAmqHs6aG8v52ce6Vh6zeSDSN3aoJW9Hy0WYVtfXh2Tw\n",
      "Resolving ucb8cad6845c8493be5be45272e7.dl.dropboxusercontent.com (ucb8cad6845c8493be5be45272e7.dl.dropboxusercontent.com)... 162.125.4.15, 2620:100:6019:15::a27d:40f\n",
      "Connecting to ucb8cad6845c8493be5be45272e7.dl.dropboxusercontent.com (ucb8cad6845c8493be5be45272e7.dl.dropboxusercontent.com)|162.125.4.15|:443... connected.\n",
      "HTTP request sent, awaiting response... 200 OK\n",
      "Length: 257749784 (246M) [application/zip]\n",
      "Saving to: ‘download.zip’\n",
      "\n",
      "download.zip        100%[===================>] 245.81M  23.0MB/s    in 19s     \n",
      "\n",
      "2021-06-02 21:19:49 (12.9 MB/s) - ‘download.zip’ saved [257749784/257749784]\n",
      "\n",
      "Archive:  download.zip\n",
      "warning:  stripped absolute path spec from /\n",
      "mapname:  conversion of  failed\n",
      "replace download/nist.json? [y]es, [n]o, [A]ll, [N]one, [r]ename: "
     ]
    }
   ],
   "source": [
    "! pip install torchani\n",
    "! wget https://www.dropbox.com/sh/2c8zdqc1hrqsgwy/AAD6l24ngoiFa6DRapF6HPk5a/ -O download.zip\n",
    "! unzip download.zip -d download || [[ $? == 2 ]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sys\n",
    "sys.path.append('..')\n",
    "\n",
    "import robust as rb\n",
    "import torch\n",
    "import torchani\n",
    "import ase\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib import cm\n",
    "import matplotlib as mpl\n",
    "from robust.geom import compute_rmsd"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Load ANI-1x data\n",
    "\n",
    "ANI-1x dataset contains organic molecules built from H, C, N, and O elements. In this specific file `ani_gdb_s01.h5`, there are only three unique molecules ($H_2O$, $CH_4$, and $NH_3$). \n",
    "\n",
    "Note that subtracting energies by the self energies of all atoms for each molecule makes the range of energies reasonable. For more information, refer to [TorchANI paper](https://pubs.acs.org/doi/10.1021/acs.jcim.0c00451)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "species_order = ['H', 'C', 'N', 'O']\n",
    "energy_shifter = torchani.utils.EnergyShifter(None)\n",
    "\n",
    "dspath = 'download/dataset/ani1-up_to_gdb4/ani_gdb_s01.h5'\n",
    "data = torchani.data.load(dspath)\n",
    "dset = list(data.subtract_self_energies(energy_shifter, species_order).species_to_indices())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Load torchANI.ANI1ccx ensemble models\n",
    "\n",
    "Loading ANI-1ccx models from `torchANI` package. ANI-1ccx was trained to the ANI-1x data set and then retrained to a 10% smaller data set of accurate coupled cluster calculations. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
    "models = torchani.models.ANI1x(periodic_table_index=False).to(device)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Perform adversarial attack \n",
    "\n",
    "Now we select the parameters for adversarial attacks to find configurations with high uncertainty. We randomly choose seed configurations to maximize adversarial loss on. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "num_attacks = 100\n",
    "lr = 5e-5\n",
    "kT = 0.7\n",
    "n_epochs = 100"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Get energy distribution for each molecule to build individual `AdvLoss` functions. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "ch_format = {\"CHHHH\": \"CH4\", 'NHHH': 'H3N', 'OHH': 'H2O'}\n",
    "atomic_number = {'H': 1, 'C': 6, 'N': 7, 'O': 8}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "energies = {}\n",
    "for d in data:\n",
    "    mol = ch_format[\"\".join(d['species'])]\n",
    "    energies[mol] = energies.get(mol, [])\n",
    "    energies[mol].append(torchani.units.hartree2kcalmol(d['energies']))\n",
    "    \n",
    "loss_fn = {}\n",
    "for mol in energies.keys():\n",
    "    potential_dset = rb.PotentialDataset(\n",
    "            x=np.zeros((len(energies[mol]))),\n",
    "            energy=np.array(energies[mol]) - np.array(energies[mol]).min(),\n",
    "            forces=np.zeros((len(energies[mol]), 3)),\n",
    "    )\n",
    "    loss_fn[mol] = rb.AdvLoss(train=potential_dset, temperature=kT)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Starting adversarial attack"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_force_std(forces, mode='max', modeldim=0, natomdim=1, xyzdim=2):\n",
    "    forces = (forces.squeeze() ** 2).sum(xyzdim)**0.5\n",
    "    fstd = forces.std(modeldim)\n",
    "    if mode == 'max':\n",
    "        return fstd.max()\n",
    "    elif mode == 'mean':\n",
    "        return fstd.mean()\n",
    "    \n",
    "def get_force_max(forces, modeldim=0, natomdim=1, xyzdim=2):\n",
    "    forces = (forces.squeeze() ** 2).sum(xyzdim)**0.5\n",
    "    fstd = forces.std(modeldim)\n",
    "    return forces[:, fstd.argmax()].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Step |     AdvLoss     | Delta  | Energy  |  Force  |  Fstd  | RMSD \n",
      "   0 | -1.09142593e-02 | 0.0007 | 12.3374 | 24.6929 | 2.7067 | 0.0000\n",
      "  20 | -7.23734660e-01 | 0.0144 | 9.1495 | 18.7695 | 2.2414 | 0.0150\n",
      "  40 | -4.39512400e+02 | 0.0217 | 4.0032 | 14.3481 | 2.0170 | 0.0555\n",
      "  60 | -2.56519986e+03 | 0.0270 | 2.4968 | 14.1121 | 2.7103 | 0.1138\n",
      "  80 | -7.67260869e-01 | 0.0256 | 9.7929 | 34.8672 | 3.8260 | 0.1732\n",
      " 100 | -1.51124181e-09 | 0.0266 | 24.1900 | 57.4413 | 5.0738 | 0.2350\n",
      "   0 | -2.59934495e+03 | 0.0008 | 2.7640 | 16.3041 | 2.0898 | 0.0000\n",
      "  20 | -3.95649164e+03 | 0.0153 | 1.1963 | 7.2082 | 1.5675 | 0.0165\n",
      "  40 | -7.91452183e+03 | 0.0155 | 2.5085 | 19.7004 | 2.3377 | 0.0538\n",
      "  60 | -1.11560335e+04 | 0.0184 | 1.6438 | 14.8119 | 2.1163 | 0.0807\n",
      "  80 | -7.44793421e+02 | 0.0236 | 4.1530 | 20.3624 | 2.1338 | 0.0990\n",
      " 100 | -1.68046824e-09 | 0.0224 | 24.2807 | 58.1733 | 3.5599 | 0.1239\n",
      "   0 | -4.91201832e-03 | 0.0007 | 12.2519 | 33.6516 | 4.1762 | 0.0000\n",
      "  20 | -9.77550616e-01 | 0.0144 | 8.2346 | 21.4700 | 2.4101 | 0.0153\n",
      "  40 | -1.62290240e+02 | 0.0239 | 4.2845 | 13.5518 | 1.9177 | 0.0575\n",
      "  60 | -1.36910082e+02 | 0.0246 | 5.2507 | 21.3129 | 2.0768 | 0.1145\n",
      "  80 | -2.74556483e+02 | 0.0238 | 4.7276 | 18.9512 | 2.4714 | 0.1650\n",
      " 100 | -7.80612540e-01 | 0.0237 | 9.0037 | 28.4021 | 2.8088 | 0.1994\n",
      "   0 | -2.44717569e+02 | 0.0003 | 5.2887 | 33.2207 | 3.9346 | 0.0000\n",
      "  20 | -4.68321889e+03 | 0.0084 | 2.7153 | 21.8459 | 2.8074 | 0.0134\n",
      "  40 | -1.35755809e+04 | 0.0128 | 1.9748 | 21.9284 | 3.0458 | 0.0528\n",
      "  60 | -5.00508822e-01 | 0.0113 | 10.6755 | 66.5670 | 5.7677 | 0.0898\n",
      "  80 | -1.08017487e-06 | 0.0111 | 20.3006 | 96.6330 | 13.6991 | 0.1185\n",
      " 100 | -4.29745208e-12 | 0.0111 | 29.3084 | 124.1502 | 21.2575 | 0.1431\n",
      "   0 | -2.04353284e-07 | 0.0007 | 20.7953 | 50.6364 | 5.1968 | 0.0000\n",
      "  20 | -9.10847002e-04 | 0.0136 | 14.5348 | 38.5188 | 5.0061 | 0.0141\n",
      "  40 | -1.00392251e+02 | 0.0238 | 5.2741 | 17.0051 | 2.4246 | 0.0518\n",
      "  60 | -1.15457458e+02 | 0.0259 | 4.8757 | 23.5946 | 2.2548 | 0.1118\n",
      "  80 | -2.09459960e-01 | 0.0256 | 9.9208 | 29.4189 | 2.2727 | 0.1752\n",
      " 100 | -5.90875422e-09 | 0.0258 | 22.5709 | 34.5111 | 3.4008 | 0.2418\n",
      "   0 | -1.24858205e+03 | 0.0008 | 2.4648 | 14.3223 | 2.1872 | 0.0000\n",
      "  20 | -4.72239251e+03 | 0.0152 | 0.9450 | 5.7060 | 1.5255 | 0.0170\n",
      "  40 | -2.07104611e+03 | 0.0195 | 1.6081 | 11.6468 | 1.6984 | 0.0611\n",
      "  60 | -4.45620551e+02 | 0.0178 | 3.6965 | 16.2929 | 1.7851 | 0.1020\n",
      "  80 | -5.06106197e+01 | 0.0184 | 5.6905 | 14.8960 | 2.0008 | 0.1329\n",
      " 100 | -4.27884216e-02 | 0.0178 | 10.8978 | 27.0305 | 2.6428 | 0.1628\n",
      "   0 | -8.91222562e+00 | 0.0008 | 6.4688 | 20.3209 | 3.2133 | 0.0000\n",
      "  20 | -2.67421176e+02 | 0.0147 | 4.1062 | 13.2914 | 2.3002 | 0.0158\n",
      "  40 | -2.79168863e+03 | 0.0247 | 1.7296 | 10.0201 | 1.8113 | 0.0577\n",
      "  60 | -1.53152022e+03 | 0.0296 | 2.4651 | 13.6455 | 1.7735 | 0.1151\n",
      "  80 | -1.02441938e+01 | 0.0298 | 6.6087 | 18.5317 | 2.2255 | 0.1733\n",
      " 100 | -2.44563451e-08 | 0.0301 | 20.9478 | 35.2241 | 4.0474 | 0.2333\n",
      "   0 | -1.06000568e+02 | 0.0003 | 5.3834 | 20.8497 | 3.7628 | 0.0000\n",
      "  20 | -5.57517242e+02 | 0.0067 | 3.7386 | 15.2710 | 3.3240 | 0.0123\n",
      "  40 | -6.49251212e+03 | 0.0114 | 2.3180 | 19.4988 | 2.7103 | 0.0456\n",
      "  60 | -3.53667284e+04 | 0.0139 | 1.5605 | 26.5851 | 3.3788 | 0.0817\n",
      "  80 | -3.70301460e+03 | 0.0214 | 2.8095 | 24.1878 | 3.1150 | 0.1256\n",
      " 100 | -1.94328708e-09 | 0.0223 | 23.7799 | 59.3624 | 8.4682 | 0.1872\n",
      "   0 | -5.23647255e+01 | 0.0008 | 5.8871 | 17.0279 | 2.5079 | 0.0000\n",
      "  20 | -3.18844100e+02 | 0.0152 | 3.9315 | 11.2878 | 2.0156 | 0.0151\n",
      "  40 | -7.59722111e+03 | 0.0235 | 1.9201 | 12.5427 | 1.9726 | 0.0555\n",
      "  60 | -1.18056967e+04 | 0.0312 | 0.9307 | 8.7012 | 1.7217 | 0.1092\n",
      "  80 | -3.56315814e-01 | 0.0343 | 10.1914 | 30.7504 | 3.4435 | 0.1714\n",
      " 100 | -2.56190023e-18 | 0.0346 | 39.0090 | 68.1247 | 8.2836 | 0.2363\n",
      "   0 | -1.44300116e-01 | 0.0006 | 10.6422 | 32.9357 | 3.8755 | 0.0000\n",
      "  20 | -1.03394016e+01 | 0.0117 | 7.2656 | 24.2295 | 4.4681 | 0.0152\n",
      "  40 | -9.63159421e+02 | 0.0180 | 3.4562 | 17.4463 | 4.1489 | 0.0557\n",
      "  60 | -1.92789030e+02 | 0.0137 | 5.9311 | 37.3434 | 4.8830 | 0.1058\n",
      "  80 | -1.55281339e+01 | 0.0172 | 7.9287 | 44.7674 | 5.9665 | 0.1547\n",
      " 100 | -2.82808601e-02 | 0.0179 | 11.9130 | 41.4848 | 4.8751 | 0.2073\n",
      "   0 | -1.80930948e+00 | 0.0006 | 9.2291 | 47.5527 | 5.3283 | 0.0000\n",
      "  20 | -6.04746177e+02 | 0.0117 | 4.5651 | 31.3616 | 4.8831 | 0.0160\n",
      "  40 | -5.97692085e+03 | 0.0215 | 2.0917 | 18.1261 | 3.4412 | 0.0632\n",
      "  60 | -8.60175114e-04 | 0.0174 | 14.6208 | 56.6687 | 4.8061 | 0.1176\n",
      "  80 | -1.62601292e-15 | 0.0167 | 33.8366 | 79.2731 | 7.3865 | 0.1687\n",
      " 100 | -1.49131558e-30 | 0.0166 | 58.3589 | 106.1542 | 13.8853 | 0.2198\n",
      "   0 | -4.22565295e-01 | 0.0006 | 10.0728 | 31.4651 | 5.2084 | 0.0000\n",
      "  20 | -1.66888908e+01 | 0.0108 | 7.0866 | 21.5787 | 4.0293 | 0.0147\n",
      "  40 | -7.73120382e+02 | 0.0165 | 3.8718 | 16.7582 | 4.0315 | 0.0506\n",
      "  60 | -1.32911443e+02 | 0.0168 | 6.0927 | 32.5779 | 4.8213 | 0.0895\n",
      "  80 | -3.50230014e-02 | 0.0186 | 12.5839 | 51.4692 | 6.3706 | 0.1263\n",
      " 100 | -3.29041887e-11 | 0.0188 | 27.5691 | 88.7474 | 12.5710 | 0.1685\n",
      "   0 | -1.84540631e-03 | 0.0003 | 13.9436 | 63.6964 | 8.4127 | 0.0000\n",
      "  20 | -1.22636633e+00 | 0.0077 | 9.0236 | 49.2371 | 6.4360 | 0.0126\n",
      "  40 | -6.01576870e+03 | 0.0170 | 1.3207 | 11.5941 | 2.4988 | 0.0501\n",
      "  60 | -1.78164279e+02 | 0.0198 | 5.0634 | 31.4059 | 4.0445 | 0.1125\n",
      "  80 | -3.26348549e-06 | 0.0187 | 18.3330 | 55.2812 | 3.6938 | 0.1727\n",
      " 100 | -9.35271537e-20 | 0.0185 | 40.7094 | 72.8566 | 7.2700 | 0.2335\n",
      "   0 | -7.65589538e-02 | 0.0003 | 11.6136 | 66.0890 | 7.8476 | 0.0000\n",
      "  20 | -4.59189075e+01 | 0.0079 | 6.7406 | 46.0077 | 4.8352 | 0.0130\n",
      "  40 | -2.10337521e+04 | 0.0158 | 0.7007 | 11.2371 | 2.5650 | 0.0507\n",
      "  60 | -1.27316332e+01 | 0.0166 | 7.4243 | 39.5830 | 4.5525 | 0.1067\n",
      "  80 | -3.38696913e-11 | 0.0159 | 26.7603 | 64.0998 | 6.6034 | 0.1629\n",
      " 100 | -2.58601614e-26 | 0.0158 | 51.2759 | 70.8780 | 7.9622 | 0.2210\n",
      "   0 | -3.61856163e+02 | 0.0006 | 3.3740 | 15.2642 | 4.3213 | 0.0000\n",
      "  20 | -1.49395586e+03 | 0.0115 | 2.2455 | 9.8921 | 3.7516 | 0.0162\n",
      "  40 | -9.16251235e+03 | 0.0105 | 1.5435 | 14.2262 | 3.6641 | 0.0482\n",
      "  60 | -1.26846803e+04 | 0.0167 | 1.2769 | 12.3375 | 2.7962 | 0.0839\n",
      "  80 | -4.47147894e+01 | 0.0172 | 6.3877 | 33.9216 | 4.8570 | 0.1216\n",
      " 100 | -3.82570876e-06 | 0.0172 | 18.4658 | 59.0012 | 8.7141 | 0.1621\n",
      "   0 | -2.11335182e+01 | 0.0006 | 5.3837 | 19.5657 | 4.0074 | 0.0000\n",
      "  20 | -2.86573024e+02 | 0.0117 | 3.3883 | 13.6866 | 3.9428 | 0.0141\n",
      "  40 | -1.22351765e+03 | 0.0178 | 2.5527 | 17.8588 | 3.7311 | 0.0539\n",
      "  60 | -1.99776475e+02 | 0.0196 | 4.6297 | 28.8948 | 4.5568 | 0.0947\n",
      "  80 | -3.18570394e+02 | 0.0207 | 4.6767 | 26.6075 | 4.6750 | 0.1287\n",
      " 100 | -2.85717310e+01 | 0.0197 | 5.9090 | 19.4068 | 4.1628 | 0.1537\n",
      "   0 | -1.75924838e-04 | 0.0006 | 15.1047 | 29.8201 | 6.4810 | 0.0000\n",
      "  20 | -1.16456364e-02 | 0.0118 | 11.9997 | 22.3892 | 5.4400 | 0.0154\n",
      "  40 | -1.67735318e+00 | 0.0183 | 7.7618 | 18.3012 | 4.6714 | 0.0539\n",
      "  60 | -7.73479597e+00 | 0.0215 | 7.8659 | 36.2348 | 5.1125 | 0.1028\n",
      "  80 | -9.29118446e+02 | 0.0285 | 4.6476 | 29.9825 | 5.1914 | 0.1555\n",
      " 100 | -3.77297433e+00 | 0.0323 | 7.6973 | 26.9351 | 4.6472 | 0.2197\n",
      "   0 | -4.24013756e+04 | 0.0003 | 0.6636 | 13.9614 | 2.9434 | 0.0000\n",
      "  20 | -6.59974760e+03 | 0.0086 | -0.0127 | 4.5391 | 2.5378 | 0.0127\n",
      "  40 | -3.57312191e+04 | 0.0082 | 0.9940 | 16.2158 | 3.1879 | 0.0383\n",
      "  60 | -2.06401553e+04 | 0.0119 | 0.6297 | 10.0887 | 2.5262 | 0.0520\n",
      "  80 | -1.05434506e+04 | 0.0121 | 2.4828 | 28.1090 | 3.5384 | 0.0569\n",
      " 100 | -1.75342678e+02 | 0.0131 | 6.1794 | 52.2945 | 6.4496 | 0.0483\n",
      "   0 | -2.07221021e+01 | 0.0003 | 6.7939 | 42.6188 | 4.1994 | 0.0000\n",
      "  20 | -1.06434487e+03 | 0.0086 | 3.5498 | 29.0455 | 3.0706 | 0.0132\n",
      "  40 | -1.69880326e+04 | 0.0154 | 0.9422 | 12.8313 | 2.6746 | 0.0522\n",
      "  60 | -1.24383849e+01 | 0.0131 | 7.2793 | 41.7764 | 3.7951 | 0.0994\n",
      "  80 | -3.20438040e-08 | 0.0124 | 21.9247 | 64.8590 | 4.6161 | 0.1454\n",
      " 100 | -5.55038386e-21 | 0.0123 | 42.7056 | 72.2474 | 7.0119 | 0.1935\n",
      "   0 | -8.77605396e+03 | 0.0006 | 0.3120 | 4.7639 | 2.8972 | 0.0000\n",
      "  20 | -1.17723033e+04 | 0.0054 | 0.3841 | 5.6668 | 2.3797 | 0.0129\n",
      "  40 | -1.34806697e+04 | 0.0067 | 0.4009 | 7.1294 | 2.7451 | 0.0135\n",
      "  60 | -1.80890193e+04 | 0.0071 | 1.6821 | 19.5118 | 3.6314 | 0.0186\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  80 | -1.59153567e+04 | 0.0099 | 1.7416 | 18.1827 | 3.6438 | 0.0228\n",
      " 100 | -6.32797910e+03 | 0.0166 | 2.3257 | 23.3859 | 4.2591 | 0.0305\n",
      "   0 | -3.24670972e-08 | 0.0006 | 21.6039 | 45.0585 | 5.7034 | 0.0000\n",
      "  20 | -2.35574855e-05 | 0.0114 | 16.7372 | 38.4188 | 5.2609 | 0.0162\n",
      "  40 | -1.04154918e+00 | 0.0203 | 8.2176 | 23.4601 | 5.0670 | 0.0609\n",
      "  60 | -9.90393382e+00 | 0.0210 | 7.7534 | 37.9332 | 4.7690 | 0.1244\n",
      "  80 | -1.29450560e-02 | 0.0192 | 12.4227 | 57.1334 | 6.3970 | 0.1830\n",
      " 100 | -5.48365685e-09 | 0.0190 | 22.8925 | 70.9798 | 7.1623 | 0.2420\n",
      "   0 | -3.71735819e-05 | 0.0006 | 15.8397 | 34.5416 | 6.1090 | 0.0000\n",
      "  20 | -7.89588967e-03 | 0.0118 | 12.0095 | 30.3243 | 5.1695 | 0.0153\n",
      "  40 | -5.69697826e+02 | 0.0211 | 4.0036 | 19.9028 | 4.4123 | 0.0574\n",
      "  60 | -4.38396022e+01 | 0.0267 | 5.1482 | 32.9078 | 5.0668 | 0.1217\n",
      "  80 | -6.62939085e-15 | 0.0273 | 33.0429 | 111.7670 | 18.8349 | 0.1940\n",
      " 100 | -2.23046346e-62 | 0.0274 | 111.0937 | 274.8762 | 85.7572 | 0.2690\n",
      "   0 | -5.49446463e-04 | 0.0006 | 14.6674 | 40.6761 | 4.4018 | 0.0000\n",
      "  20 | -9.93286745e-02 | 0.0119 | 10.5307 | 28.8986 | 4.6500 | 0.0158\n",
      "  40 | -2.60324400e+01 | 0.0192 | 5.8922 | 19.4348 | 3.4952 | 0.0595\n",
      "  60 | -3.08116486e+00 | 0.0149 | 8.5546 | 40.4207 | 4.5567 | 0.1116\n",
      "  80 | -5.59036509e-03 | 0.0154 | 13.2778 | 54.0589 | 6.7889 | 0.1580\n",
      " 100 | -8.85519682e-10 | 0.0156 | 24.6767 | 81.3052 | 9.8344 | 0.2070\n",
      "   0 | -3.59544768e-05 | 0.0007 | 16.7054 | 29.3452 | 3.1992 | 0.0000\n",
      "  20 | -9.82482950e-03 | 0.0141 | 12.6430 | 26.0215 | 2.4678 | 0.0146\n",
      "  40 | -3.45499483e+02 | 0.0237 | 4.4106 | 14.3454 | 2.1583 | 0.0536\n",
      "  60 | -5.10826169e+02 | 0.0302 | 4.9755 | 24.2674 | 3.1229 | 0.1151\n",
      "  80 | -1.10123917e-07 | 0.0297 | 21.7646 | 61.1256 | 7.0255 | 0.1792\n",
      " 100 | -7.78063052e-29 | 0.0299 | 56.4709 | 105.2116 | 9.8488 | 0.2469\n",
      "   0 | -3.05714405e-04 | 0.0006 | 15.1616 | 41.9761 | 5.2154 | 0.0000\n",
      "  20 | -5.83606968e-02 | 0.0119 | 10.9282 | 29.6314 | 4.8397 | 0.0151\n",
      "  40 | -9.05791482e+01 | 0.0152 | 5.4002 | 28.6956 | 4.4678 | 0.0542\n",
      "  60 | -4.69013568e+02 | 0.0196 | 4.9807 | 36.2454 | 5.1848 | 0.1048\n",
      "  80 | -2.57406294e-04 | 0.0222 | 15.0664 | 64.3611 | 8.2312 | 0.1562\n",
      " 100 | -6.66564284e-21 | 0.0226 | 42.5715 | 123.0009 | 18.2879 | 0.2135\n",
      "   0 | -2.11021671e-11 | 0.0001 | 27.7016 | 51.5525 | 6.2639 | 0.0000\n",
      "  20 | -5.86927293e-09 | 0.0079 | 23.5064 | 43.0878 | 5.4133 | 0.0072\n",
      "  40 | -1.99126467e-03 | 0.0199 | 13.5591 | 22.0767 | 3.8279 | 0.0376\n",
      "  60 | -2.53962235e+00 | 0.0280 | 8.2775 | 19.5628 | 2.7204 | 0.0912\n",
      "  80 | -3.50964958e+00 | 0.0361 | 8.1940 | 21.7720 | 2.2148 | 0.1588\n",
      " 100 | -1.03771113e-04 | 0.0385 | 15.9054 | 31.1931 | 2.6866 | 0.2296\n",
      "   0 | -3.02881057e-08 | 0.0006 | 21.6892 | 40.2299 | 7.4477 | 0.0000\n",
      "  20 | -7.10772808e-06 | 0.0109 | 17.5302 | 31.3347 | 6.6475 | 0.0145\n",
      "  40 | -3.01531792e-02 | 0.0195 | 10.7802 | 19.0268 | 4.3380 | 0.0533\n",
      "  60 | -2.00769483e+00 | 0.0232 | 8.5233 | 29.3459 | 4.7847 | 0.1142\n",
      "  80 | -1.99035921e+02 | 0.0250 | 5.2619 | 24.4047 | 3.5240 | 0.1717\n",
      " 100 | -2.02436507e+01 | 0.0278 | 6.8242 | 23.0343 | 4.6806 | 0.2237\n",
      "   0 | -7.66952412e+03 | 0.0007 | 0.2986 | 4.6789 | 1.4643 | 0.0000\n",
      "  20 | -9.93748545e+03 | 0.0118 | 0.1174 | 4.3530 | 1.4974 | 0.0152\n",
      "  40 | -1.12476698e+04 | 0.0110 | 0.3534 | 5.9967 | 1.6224 | 0.0380\n",
      "  60 | -6.68369656e+03 | 0.0118 | 1.4902 | 12.7800 | 2.1904 | 0.0518\n",
      "  80 | -2.36095657e+03 | 0.0134 | 2.6558 | 17.6814 | 2.4493 | 0.0580\n",
      " 100 | -1.36076630e+03 | 0.0154 | 3.6916 | 18.6275 | 3.1660 | 0.0670\n",
      "   0 | -1.63696733e-06 | 0.0003 | 19.5556 | 69.1012 | 13.8147 | 0.0000\n",
      "  20 | -1.22906108e-03 | 0.0068 | 14.4694 | 48.6697 | 9.1978 | 0.0128\n",
      "  40 | -7.80049800e-01 | 0.0133 | 8.5134 | 21.4763 | 3.1639 | 0.0486\n",
      "  60 | -8.31529940e-01 | 0.0137 | 9.4594 | 36.3017 | 3.4501 | 0.0967\n",
      "  80 | -2.21024960e+01 | 0.0191 | 7.1902 | 36.6343 | 3.0422 | 0.1399\n",
      " 100 | -9.59710611e+03 | 0.0223 | 1.6693 | 17.9265 | 2.6060 | 0.1701\n",
      "   0 | -2.78193516e+03 | 0.0008 | -0.0248 | 1.9480 | 1.3368 | 0.0000\n",
      "  20 | -5.61972919e+03 | 0.0069 | 0.0010 | 2.7539 | 1.3882 | 0.0114\n",
      "  40 | -3.40732335e+03 | 0.0099 | 0.0732 | 3.0937 | 1.3980 | 0.0187\n",
      "  60 | -1.31358986e+04 | 0.0072 | 0.7244 | 8.7793 | 2.0095 | 0.0266\n",
      "  80 | -1.61972821e+04 | 0.0168 | 0.4537 | 6.1554 | 1.6876 | 0.0288\n",
      " 100 | -2.20980059e+02 | 0.0169 | 4.6490 | 24.0272 | 2.1242 | 0.0490\n",
      "   0 | -2.19658156e-08 | 0.0006 | 22.0672 | 54.4546 | 6.8684 | 0.0000\n",
      "  20 | -4.14020266e-05 | 0.0112 | 16.5337 | 41.9184 | 6.3989 | 0.0143\n",
      "  40 | -8.74915908e-01 | 0.0191 | 8.9209 | 25.0264 | 4.7947 | 0.0537\n",
      "  60 | -1.50912310e+00 | 0.0191 | 9.2761 | 44.4842 | 5.4564 | 0.1051\n",
      "  80 | -9.89678016e+01 | 0.0227 | 6.2686 | 38.9326 | 5.3091 | 0.1515\n",
      " 100 | -1.80237957e+02 | 0.0260 | 5.1148 | 21.4494 | 5.0399 | 0.2003\n",
      "   0 | -7.64747243e-09 | 0.0007 | 22.2565 | 25.0289 | 3.5793 | 0.0000\n",
      "  20 | -7.24524325e-07 | 0.0144 | 18.8842 | 20.7723 | 2.9974 | 0.0143\n",
      "  40 | -4.77211330e-03 | 0.0208 | 12.2874 | 18.6980 | 2.6423 | 0.0520\n",
      "  60 | -1.57870668e+01 | 0.0289 | 5.3837 | 15.0965 | 1.9175 | 0.1092\n",
      "  80 | -5.13752897e+02 | 0.0348 | 4.5023 | 23.4864 | 3.6439 | 0.1849\n",
      " 100 | -3.25714183e+00 | 0.0337 | 7.9144 | 30.3267 | 2.9646 | 0.2630\n",
      "   0 | -1.06045888e-04 | 0.0007 | 15.1993 | 30.3452 | 2.7198 | 0.0000\n",
      "  20 | -1.93724758e-02 | 0.0135 | 11.4684 | 22.0979 | 2.3707 | 0.0134\n",
      "  40 | -1.79411382e+01 | 0.0207 | 6.3722 | 17.1428 | 2.3210 | 0.0488\n",
      "  60 | -2.98013306e+01 | 0.0251 | 6.8973 | 28.0709 | 3.3761 | 0.0960\n",
      "  80 | -6.10014837e-02 | 0.0258 | 12.1234 | 48.3031 | 3.5375 | 0.1429\n",
      " 100 | -8.80127597e-07 | 0.0263 | 20.1578 | 56.5288 | 2.9122 | 0.1942\n",
      "   0 | -1.29731074e-03 | 0.0006 | 13.0330 | 38.5878 | 6.4193 | 0.0000\n",
      "  20 | -1.43417569e-01 | 0.0113 | 9.4230 | 24.0294 | 5.0293 | 0.0141\n",
      "  40 | -1.54687078e+01 | 0.0174 | 5.8902 | 18.5902 | 4.3176 | 0.0520\n",
      "  60 | -1.52831940e+02 | 0.0211 | 5.1779 | 26.8618 | 4.7432 | 0.1007\n",
      "  80 | -2.99104751e+03 | 0.0286 | 1.4710 | 12.1774 | 3.4883 | 0.1486\n",
      " 100 | -8.81498758e-02 | 0.0288 | 11.0816 | 37.9805 | 5.0482 | 0.1959\n",
      "   0 | -8.64828970e-10 | 0.0005 | 23.8041 | 39.7994 | 11.1737 | 0.0000\n",
      "  20 | -2.58703516e-07 | 0.0112 | 19.6082 | 35.0311 | 8.0973 | 0.0143\n",
      "  40 | -1.67046402e-02 | 0.0212 | 11.1528 | 21.9460 | 5.7383 | 0.0565\n",
      "  60 | -7.70134621e+02 | 0.0258 | 3.4682 | 15.5355 | 4.7537 | 0.1227\n",
      "  80 | -6.98221562e+03 | 0.0282 | 1.7203 | 13.5532 | 3.6415 | 0.2054\n",
      " 100 | -2.60237893e-03 | 0.0291 | 13.1944 | 32.5139 | 4.9948 | 0.2922\n",
      "   0 | -7.27255248e-04 | 0.0007 | 14.6430 | 31.8859 | 4.6061 | 0.0000\n",
      "  20 | -1.38435954e-01 | 0.0141 | 10.5172 | 23.8554 | 3.6419 | 0.0147\n",
      "  40 | -1.67508140e+02 | 0.0283 | 4.1878 | 12.5499 | 1.8864 | 0.0569\n",
      "  60 | -8.32947576e+02 | 0.0323 | 3.6787 | 16.4567 | 1.5751 | 0.1242\n",
      "  80 | -1.90930546e+00 | 0.0285 | 8.6574 | 25.7237 | 2.4362 | 0.1877\n",
      " 100 | -1.89025607e-06 | 0.0279 | 18.5822 | 31.9773 | 3.2310 | 0.2492\n",
      "   0 | -2.77996141e+04 | 0.0003 | 1.3350 | 18.7029 | 3.2853 | 0.0000\n",
      "  20 | -1.03093663e+04 | 0.0090 | 0.2750 | 5.8635 | 2.7008 | 0.0129\n",
      "  40 | -2.91212788e+04 | 0.0094 | 1.3042 | 19.0877 | 3.3913 | 0.0422\n",
      "  60 | -2.11592684e+04 | 0.0136 | 1.3031 | 17.0377 | 2.7726 | 0.0619\n",
      "  80 | -8.48731803e+03 | 0.0166 | 2.0533 | 18.6589 | 3.0677 | 0.0777\n",
      " 100 | -2.62894747e-01 | 0.0156 | 10.9717 | 61.2757 | 8.8093 | 0.0963\n",
      "   0 | -1.29965855e-09 | 0.0006 | 23.0341 | 29.0910 | 6.6007 | 0.0000\n",
      "  20 | -2.45876642e-07 | 0.0143 | 19.2431 | 24.6039 | 5.2885 | 0.0158\n",
      "  40 | -1.05458431e-02 | 0.0256 | 11.3923 | 18.2914 | 2.7802 | 0.0614\n",
      "  60 | -4.57786654e+02 | 0.0332 | 3.4909 | 12.4334 | 2.1644 | 0.1316\n",
      "  80 | -1.77712230e+03 | 0.0367 | 2.8206 | 14.3211 | 2.0433 | 0.2192\n",
      " 100 | -2.12648073e-05 | 0.0365 | 16.2777 | 29.1856 | 3.4666 | 0.3110\n",
      "   0 | -1.42566152e-06 | 0.0007 | 18.8710 | 48.7337 | 5.9791 | 0.0000\n",
      "  20 | -4.35155355e-03 | 0.0130 | 13.0681 | 34.7258 | 5.1141 | 0.0149\n",
      "  40 | -2.01128605e+02 | 0.0216 | 5.0970 | 18.7604 | 2.9686 | 0.0541\n",
      "  60 | -4.81169122e+02 | 0.0255 | 4.0273 | 18.9371 | 1.7622 | 0.1108\n",
      "  80 | -3.53376552e-03 | 0.0255 | 13.2571 | 36.6861 | 3.1060 | 0.1640\n",
      " 100 | -1.39833462e-17 | 0.0255 | 37.4591 | 64.1806 | 5.3237 | 0.2188\n",
      "   0 | -7.63707053e+00 | 0.0008 | 7.5870 | 22.1907 | 2.6634 | 0.0000\n",
      "  20 | -2.58834411e+02 | 0.0148 | 4.7909 | 15.5649 | 2.2136 | 0.0152\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  40 | -4.79326955e+03 | 0.0213 | 1.9977 | 13.4744 | 1.9520 | 0.0558\n",
      "  60 | -1.57621275e+03 | 0.0234 | 4.1401 | 25.5160 | 3.4914 | 0.1065\n",
      "  80 | -3.13265364e-01 | 0.0234 | 10.6418 | 36.2534 | 3.0642 | 0.1606\n",
      " 100 | -6.68148993e-13 | 0.0237 | 30.3632 | 62.9813 | 8.4233 | 0.2212\n",
      "   0 | -2.02135549e-12 | 0.0000 | 28.9224 | 45.3814 | 7.5144 | 0.0000\n",
      "  20 | -5.05674454e-12 | 0.0015 | 28.2363 | 44.0748 | 7.5266 | 0.0014\n",
      "  40 | -8.86049879e-09 | 0.0121 | 22.6198 | 33.4744 | 6.9612 | 0.0148\n",
      "  60 | -8.61625285e-04 | 0.0242 | 13.6876 | 19.3243 | 3.8577 | 0.0577\n",
      "  80 | -9.28234507e-01 | 0.0320 | 8.9399 | 19.9686 | 2.7073 | 0.1240\n",
      " 100 | -5.58114110e+01 | 0.0363 | 6.1369 | 21.6985 | 2.1788 | 0.2017\n",
      "   0 | -3.08159060e-04 | 0.0007 | 14.1305 | 27.8047 | 3.6457 | 0.0000\n",
      "  20 | -1.02219302e-02 | 0.0147 | 10.7955 | 17.9631 | 3.4275 | 0.0164\n",
      "  40 | -8.22652059e+00 | 0.0228 | 7.3272 | 21.3646 | 2.5794 | 0.0596\n",
      "  60 | -7.73248029e+02 | 0.0295 | 3.8872 | 17.1006 | 1.9980 | 0.1166\n",
      "  80 | -1.95455211e+02 | 0.0281 | 4.9324 | 19.6649 | 1.8828 | 0.1744\n",
      " 100 | -1.01756163e-04 | 0.0275 | 15.8698 | 31.2352 | 3.1139 | 0.2240\n",
      "   0 | -1.27231844e+02 | 0.0006 | 4.5281 | 14.5556 | 4.1669 | 0.0000\n",
      "  20 | -7.70416898e+02 | 0.0119 | 3.0928 | 9.6525 | 3.8032 | 0.0160\n",
      "  40 | -4.55948234e+03 | 0.0108 | 1.8862 | 14.4737 | 3.8484 | 0.0543\n",
      "  60 | -1.64156523e+04 | 0.0154 | 0.5510 | 7.0133 | 2.8490 | 0.0885\n",
      "  80 | -7.06981355e+01 | 0.0175 | 6.0965 | 29.8192 | 4.3460 | 0.1280\n",
      " 100 | -7.39885956e-08 | 0.0171 | 21.7265 | 58.4952 | 5.4308 | 0.1725\n",
      "   0 | -3.04111080e-05 | 0.0006 | 16.4561 | 34.4146 | 5.1618 | 0.0000\n",
      "  20 | -5.26983741e-03 | 0.0113 | 12.7276 | 29.0966 | 4.7372 | 0.0142\n",
      "  40 | -7.43837179e+01 | 0.0194 | 5.5006 | 20.6040 | 4.8691 | 0.0541\n",
      "  60 | -6.89967137e+03 | 0.0239 | 2.3643 | 20.4951 | 3.9819 | 0.1170\n",
      "  80 | -7.91170918e+01 | 0.0196 | 6.3650 | 35.9383 | 5.5471 | 0.1790\n",
      " 100 | -2.21303672e-03 | 0.0197 | 13.6667 | 36.9402 | 4.2521 | 0.2417\n",
      "   0 | -1.09450451e-05 | 0.0006 | 18.3098 | 60.4148 | 5.3176 | 0.0000\n",
      "  20 | -1.75565185e-02 | 0.0103 | 12.6199 | 41.0632 | 5.1475 | 0.0142\n",
      "  40 | -4.94568288e+01 | 0.0181 | 5.5570 | 15.7155 | 4.8293 | 0.0506\n",
      "  60 | -2.25976929e+00 | 0.0205 | 9.0341 | 35.0003 | 4.1185 | 0.0999\n",
      "  80 | -6.45249367e-06 | 0.0202 | 18.8694 | 63.1208 | 7.0181 | 0.1474\n",
      " 100 | -5.04204918e-15 | 0.0202 | 33.9005 | 80.4371 | 6.9958 | 0.1988\n",
      "   0 | -2.88550655e-03 | 0.0006 | 13.2490 | 38.6347 | 4.3788 | 0.0000\n",
      "  20 | -6.45759387e-01 | 0.0113 | 9.1536 | 31.4886 | 4.2780 | 0.0150\n",
      "  40 | -5.06929173e+02 | 0.0210 | 3.5449 | 12.3136 | 4.0319 | 0.0576\n",
      "  60 | -3.23853362e-02 | 0.0172 | 12.5307 | 62.5617 | 7.4389 | 0.1093\n",
      "  80 | -2.58320385e-14 | 0.0163 | 32.6554 | 104.2788 | 14.9141 | 0.1554\n",
      " 100 | -2.76690746e-32 | 0.0162 | 62.2891 | 160.4377 | 27.4903 | 0.2022\n",
      "   0 | -8.75063066e-03 | 0.0003 | 13.6701 | 72.0564 | 8.2087 | 0.0000\n",
      "  20 | -1.35299204e+01 | 0.0077 | 8.1789 | 54.9694 | 6.2112 | 0.0128\n",
      "  40 | -2.83661964e+04 | 0.0155 | 0.6211 | 12.6483 | 2.5455 | 0.0494\n",
      "  60 | -6.46853215e+01 | 0.0143 | 6.3343 | 36.9383 | 3.5047 | 0.1051\n",
      "  80 | -5.89066026e-08 | 0.0130 | 21.3745 | 49.4403 | 3.4293 | 0.1572\n",
      " 100 | -2.11867659e-19 | 0.0128 | 39.9280 | 53.6231 | 5.3721 | 0.2097\n",
      "   0 | -7.87887234e+01 | 0.0008 | 5.4625 | 23.8225 | 3.5943 | 0.0000\n",
      "  20 | -2.02550552e+03 | 0.0148 | 2.7420 | 13.8885 | 1.9381 | 0.0153\n",
      "  40 | -6.68428327e+03 | 0.0226 | 1.6195 | 14.1709 | 2.2287 | 0.0578\n",
      "  60 | -5.02216835e+02 | 0.0194 | 4.5304 | 20.6128 | 1.9363 | 0.1058\n",
      "  80 | -1.59680611e+00 | 0.0190 | 8.7142 | 24.3038 | 2.7378 | 0.1556\n",
      " 100 | -6.52888307e-07 | 0.0193 | 19.2645 | 27.6659 | 3.2275 | 0.2106\n",
      "   0 | -1.20214102e+03 | 0.0008 | 3.2829 | 17.1526 | 2.0568 | 0.0000\n",
      "  20 | -5.87323447e+03 | 0.0146 | 1.4219 | 8.5757 | 1.7624 | 0.0148\n",
      "  40 | -6.07968807e+03 | 0.0156 | 2.4261 | 18.7337 | 2.3011 | 0.0496\n",
      "  60 | -2.88985316e+02 | 0.0146 | 5.2227 | 29.0764 | 3.2828 | 0.0752\n",
      "  80 | -1.06639148e+02 | 0.0156 | 5.8705 | 24.7301 | 3.5824 | 0.0993\n",
      " 100 | -3.71860867e+00 | 0.0156 | 8.0785 | 21.5294 | 3.4874 | 0.1275\n",
      "   0 | -5.13773439e+02 | 0.0006 | 4.4865 | 26.8059 | 4.2336 | 0.0000\n",
      "  20 | -3.02697263e+03 | 0.0128 | 2.1204 | 12.5914 | 3.4944 | 0.0151\n",
      "  40 | -5.09471257e+03 | 0.0149 | 2.6148 | 20.5241 | 3.9951 | 0.0549\n",
      "  60 | -2.57295653e+01 | 0.0131 | 6.3403 | 31.9922 | 4.6700 | 0.0918\n",
      "  80 | -1.47349152e-03 | 0.0131 | 13.7613 | 41.6845 | 4.2758 | 0.1271\n",
      " 100 | -9.08104311e-11 | 0.0131 | 25.7833 | 50.4212 | 6.3835 | 0.1651\n",
      "   0 | -9.96363095e-08 | 0.0007 | 21.0885 | 36.7361 | 6.8099 | 0.0000\n",
      "  20 | -4.15378733e-05 | 0.0130 | 16.4532 | 28.1021 | 4.9395 | 0.0143\n",
      "  40 | -2.52458806e-01 | 0.0242 | 9.0423 | 14.4135 | 2.7624 | 0.0528\n",
      "  60 | -4.24145278e+01 | 0.0263 | 6.1966 | 21.9991 | 2.3727 | 0.1102\n",
      "  80 | -4.10054833e+02 | 0.0313 | 4.5330 | 18.4692 | 2.0304 | 0.1700\n",
      " 100 | -3.17210179e-01 | 0.0336 | 8.9969 | 21.3015 | 2.4758 | 0.2276\n",
      "   0 | -7.45802644e+02 | 0.0006 | 3.6459 | 16.8143 | 4.1820 | 0.0000\n",
      "  20 | -1.22249033e+03 | 0.0112 | 2.2121 | 10.3981 | 3.4497 | 0.0159\n",
      "  40 | -1.27747438e+04 | 0.0146 | 2.0289 | 18.5582 | 3.8925 | 0.0545\n",
      "  60 | -1.10966394e+04 | 0.0157 | 1.1460 | 11.5559 | 3.6173 | 0.0890\n",
      "  80 | -3.15011588e+01 | 0.0179 | 6.7537 | 31.4071 | 4.6014 | 0.1214\n",
      " 100 | -1.49468430e-07 | 0.0181 | 20.7170 | 43.1510 | 5.3533 | 0.1620\n",
      "   0 | -7.01943364e-02 | 0.0006 | 10.6483 | 21.4459 | 4.2582 | 0.0000\n",
      "  20 | -9.87502929e-01 | 0.0116 | 8.5147 | 17.2061 | 4.6894 | 0.0162\n",
      "  40 | -7.78616229e+01 | 0.0173 | 5.1959 | 15.3247 | 4.2560 | 0.0564\n",
      "  60 | -2.26501463e+02 | 0.0186 | 4.0450 | 21.7198 | 3.7676 | 0.1079\n",
      "  80 | -7.46358374e+02 | 0.0246 | 3.6134 | 16.2349 | 4.8262 | 0.1652\n",
      " 100 | -4.36439079e-04 | 0.0290 | 13.1241 | 32.5930 | 5.6214 | 0.2201\n",
      "   0 | -2.06152548e-02 | 0.0003 | 13.0160 | 66.7854 | 9.2491 | 0.0000\n",
      "  20 | -1.98380564e+01 | 0.0075 | 7.8512 | 52.1094 | 6.7742 | 0.0123\n",
      "  40 | -2.17078295e+04 | 0.0149 | 0.5623 | 10.6796 | 2.1976 | 0.0482\n",
      "  60 | -1.01964751e+02 | 0.0170 | 6.0468 | 37.0220 | 4.3352 | 0.1027\n",
      "  80 | -1.18006820e-07 | 0.0166 | 20.9103 | 52.8325 | 3.5834 | 0.1564\n",
      " 100 | -7.86318237e-22 | 0.0165 | 44.0901 | 71.1451 | 5.8960 | 0.2135\n",
      "   0 | -1.87278507e-07 | 0.0006 | 19.5772 | 30.6230 | 7.1112 | 0.0000\n",
      "  20 | -2.01156296e-05 | 0.0117 | 16.1517 | 27.8197 | 6.2462 | 0.0147\n",
      "  40 | -1.88298756e-01 | 0.0185 | 9.1126 | 14.5868 | 4.9246 | 0.0526\n",
      "  60 | -6.61681596e-02 | 0.0252 | 10.5831 | 36.1988 | 5.9940 | 0.1095\n",
      "  80 | -9.68701210e-08 | 0.0256 | 20.9397 | 68.8828 | 9.2855 | 0.1677\n",
      " 100 | -1.35737410e-18 | 0.0257 | 39.1824 | 119.6358 | 15.6797 | 0.2289\n",
      "   0 | -6.39964066e+01 | 0.0006 | 5.8136 | 20.4580 | 4.1912 | 0.0000\n",
      "  20 | -3.42950388e+02 | 0.0114 | 3.9134 | 12.7372 | 4.2680 | 0.0149\n",
      "  40 | -3.51467939e+03 | 0.0145 | 2.8579 | 18.9726 | 3.9212 | 0.0516\n",
      "  60 | -2.54175038e+04 | 0.0192 | 1.4096 | 15.9761 | 3.7123 | 0.0888\n",
      "  80 | -8.74461802e+01 | 0.0272 | 6.1350 | 31.4850 | 4.7676 | 0.1382\n",
      " 100 | -1.97617640e-15 | 0.0276 | 34.3665 | 79.8486 | 7.6532 | 0.2039\n",
      "   0 | -4.66704332e+00 | 0.0006 | 7.8450 | 25.7923 | 3.2974 | 0.0000\n",
      "  20 | -1.79904904e+02 | 0.0124 | 5.0555 | 19.8670 | 3.8365 | 0.0147\n",
      "  40 | -5.08333870e+03 | 0.0212 | 1.6576 | 10.6802 | 3.4992 | 0.0565\n",
      "  60 | -1.18582577e+02 | 0.0202 | 6.2870 | 31.8597 | 4.8857 | 0.1113\n",
      "  80 | -3.37175763e-05 | 0.0198 | 17.5594 | 66.3024 | 6.7003 | 0.1668\n",
      " 100 | -8.93308899e-18 | 0.0198 | 38.0743 | 96.8710 | 13.1063 | 0.2261\n",
      "   0 | -2.76835157e-07 | 0.0007 | 20.9840 | 57.4397 | 4.3390 | 0.0000\n",
      "  20 | -4.07644412e-03 | 0.0133 | 13.8776 | 44.0510 | 3.8201 | 0.0137\n",
      "  40 | -2.03237351e+03 | 0.0271 | 3.0398 | 14.0278 | 2.0782 | 0.0520\n",
      "  60 | -1.02463486e+02 | 0.0307 | 6.0830 | 25.6195 | 2.3294 | 0.1151\n",
      "  80 | -1.50419115e-07 | 0.0281 | 21.1712 | 47.1339 | 4.7489 | 0.1757\n",
      " 100 | -1.28371541e-19 | 0.0278 | 40.8981 | 61.8161 | 6.2089 | 0.2366\n",
      "   0 | -1.11624102e-03 | 0.0006 | 13.8491 | 40.2278 | 4.5698 | 0.0000\n",
      "  20 | -3.03115564e-01 | 0.0117 | 9.5446 | 32.0621 | 4.3719 | 0.0161\n",
      "  40 | -5.37382817e+02 | 0.0220 | 3.4766 | 12.6673 | 3.8344 | 0.0626\n",
      "  60 | -7.03525438e-01 | 0.0185 | 9.1598 | 50.6788 | 6.7134 | 0.1253\n",
      "  80 | -2.14907002e-08 | 0.0177 | 21.8215 | 83.0929 | 8.2646 | 0.1839\n",
      " 100 | -8.09459623e-18 | 0.0176 | 37.2468 | 96.6646 | 12.8269 | 0.2435\n",
      "   0 | -5.93839211e-01 | 0.0006 | 8.6514 | 27.5058 | 4.3875 | 0.0000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  20 | -1.50829300e+01 | 0.0121 | 6.0539 | 17.3745 | 4.4260 | 0.0140\n",
      "  40 | -1.85363306e+02 | 0.0194 | 3.9194 | 17.4489 | 4.1724 | 0.0528\n",
      "  60 | -3.78972487e+03 | 0.0225 | 2.5459 | 19.9082 | 4.5791 | 0.1010\n",
      "  80 | -1.71634073e+03 | 0.0245 | 2.6537 | 14.0316 | 3.2205 | 0.1491\n",
      " 100 | -7.72155035e-10 | 0.0241 | 24.5597 | 64.7610 | 11.4879 | 0.2019\n",
      "   0 | -2.13866728e-08 | 0.0007 | 21.8415 | 41.0647 | 3.9573 | 0.0000\n",
      "  20 | -2.18860331e-05 | 0.0140 | 16.6557 | 30.5488 | 3.4047 | 0.0153\n",
      "  40 | -2.32695144e-01 | 0.0224 | 9.6513 | 18.1568 | 2.4819 | 0.0578\n",
      "  60 | -1.64256401e-01 | 0.0223 | 10.1077 | 32.6268 | 3.5300 | 0.1141\n",
      "  80 | -1.50290205e-02 | 0.0219 | 12.8526 | 46.1539 | 4.4561 | 0.1626\n",
      " 100 | -4.70822917e-07 | 0.0229 | 20.7985 | 57.5924 | 7.7779 | 0.2078\n",
      "   0 | -5.64143473e+02 | 0.0006 | 4.0811 | 26.9668 | 4.8251 | 0.0000\n",
      "  20 | -4.71542476e+03 | 0.0118 | 1.8414 | 11.0147 | 3.5773 | 0.0158\n",
      "  40 | -1.03332558e+03 | 0.0144 | 3.2339 | 23.8879 | 4.7071 | 0.0549\n",
      "  60 | -1.22695663e+01 | 0.0120 | 7.1881 | 34.9751 | 4.5968 | 0.0894\n",
      "  80 | -6.06137672e-02 | 0.0121 | 11.1220 | 37.0366 | 5.6274 | 0.1240\n",
      " 100 | -3.77096926e-05 | 0.0122 | 16.4182 | 37.7648 | 5.1514 | 0.1614\n",
      "   0 | -7.52130262e-12 | 0.0000 | 26.8749 | 24.1668 | 3.8547 | 0.0000\n",
      "  20 | -2.22881087e-11 | 0.0023 | 26.0711 | 23.8316 | 3.7743 | 0.0026\n",
      "  40 | -2.38930207e-08 | 0.0135 | 20.9634 | 21.6564 | 3.3910 | 0.0208\n",
      "  60 | -1.32228119e-02 | 0.0272 | 11.1844 | 15.6977 | 2.7381 | 0.0677\n",
      "  80 | -1.55664601e+03 | 0.0403 | 2.9587 | 12.3728 | 1.8058 | 0.1448\n",
      " 100 | -2.60118951e+03 | 0.0415 | 2.1352 | 9.9739 | 1.8138 | 0.2405\n",
      "   0 | -6.74320131e-08 | 0.0006 | 21.4452 | 59.6279 | 6.3183 | 0.0000\n",
      "  20 | -1.86309694e-04 | 0.0110 | 15.5575 | 43.8579 | 4.6515 | 0.0145\n",
      "  40 | -1.38174241e+01 | 0.0204 | 6.8870 | 21.1735 | 4.3088 | 0.0542\n",
      "  60 | -7.97075524e+00 | 0.0218 | 7.8907 | 37.6883 | 5.5473 | 0.1112\n",
      "  80 | -4.09347186e-04 | 0.0213 | 15.6981 | 61.9115 | 6.7302 | 0.1657\n",
      " 100 | -1.09221289e-11 | 0.0217 | 28.3384 | 80.8398 | 6.4550 | 0.2231\n",
      "   0 | -1.23658192e-06 | 0.0007 | 18.6218 | 29.2686 | 3.2842 | 0.0000\n",
      "  20 | -2.44026864e-04 | 0.0144 | 14.6086 | 25.0349 | 3.0533 | 0.0162\n",
      "  40 | -2.76182723e+00 | 0.0268 | 6.5886 | 15.3398 | 2.2648 | 0.0604\n",
      "  60 | -8.78139364e+02 | 0.0324 | 4.1848 | 20.8626 | 2.4425 | 0.1295\n",
      "  80 | -1.18216507e+02 | 0.0292 | 5.2860 | 20.0448 | 3.1535 | 0.1982\n",
      " 100 | -2.12632829e-03 | 0.0286 | 13.5217 | 29.1966 | 4.7780 | 0.2626\n",
      "   0 | -3.27959676e-02 | 0.0008 | 11.8011 | 34.4252 | 3.0341 | 0.0000\n",
      "  20 | -5.80366609e+00 | 0.0144 | 7.6119 | 22.6430 | 2.4990 | 0.0167\n",
      "  40 | -4.53342971e+02 | 0.0211 | 4.0964 | 15.4403 | 1.8209 | 0.0610\n",
      "  60 | -2.27470652e+00 | 0.0178 | 8.9180 | 38.2226 | 4.0964 | 0.1069\n",
      "  80 | -3.26812073e-06 | 0.0182 | 19.0784 | 58.5939 | 5.1696 | 0.1493\n",
      " 100 | -6.22542355e-17 | 0.0184 | 36.8209 | 79.8068 | 10.8818 | 0.1959\n",
      "   0 | -4.99349791e-02 | 0.0008 | 11.1710 | 31.3831 | 2.7939 | 0.0000\n",
      "  20 | -5.55125625e+00 | 0.0143 | 7.2232 | 21.9473 | 2.5902 | 0.0164\n",
      "  40 | -2.04168331e+03 | 0.0259 | 3.0306 | 13.3278 | 1.9007 | 0.0630\n",
      "  60 | -8.01626148e+01 | 0.0227 | 6.1381 | 32.0907 | 3.6815 | 0.1208\n",
      "  80 | -7.64478771e-02 | 0.0217 | 11.1664 | 38.8297 | 4.3119 | 0.1707\n",
      " 100 | -1.04794693e-07 | 0.0215 | 20.8714 | 51.9878 | 4.9072 | 0.2220\n",
      "   0 | -1.24758538e-04 | 0.0006 | 14.5335 | 37.8296 | 5.1265 | 0.0000\n",
      "  20 | -2.70361882e-02 | 0.0116 | 10.4708 | 31.7235 | 4.6313 | 0.0144\n",
      "  40 | -6.04478204e+02 | 0.0195 | 3.6137 | 15.3255 | 4.2195 | 0.0541\n",
      "  60 | -1.90424931e+00 | 0.0229 | 7.8896 | 48.3833 | 5.5724 | 0.1096\n",
      "  80 | -1.01228782e-09 | 0.0230 | 23.6343 | 88.9891 | 12.5129 | 0.1659\n",
      " 100 | -2.97961353e-26 | 0.0230 | 51.5644 | 157.5443 | 34.9260 | 0.2252\n",
      "   0 | -6.28005560e+03 | 0.0006 | 0.8537 | 7.1478 | 3.2354 | 0.0000\n",
      "  20 | -1.14099292e+04 | 0.0093 | 0.5642 | 6.4510 | 3.0879 | 0.0144\n",
      "  40 | -7.23402945e+03 | 0.0096 | 0.3552 | 4.9843 | 2.9629 | 0.0369\n",
      "  60 | -1.08285257e+04 | 0.0062 | 2.1752 | 20.0667 | 3.8599 | 0.0591\n",
      "  80 | -6.69196614e+03 | 0.0105 | 2.0895 | 12.7370 | 3.6750 | 0.0729\n",
      " 100 | -6.13509159e+02 | 0.0135 | 4.3966 | 23.3668 | 4.2155 | 0.0851\n",
      "   0 | -8.25304550e+00 | 0.0006 | 7.8671 | 31.1656 | 4.5680 | 0.0000\n",
      "  20 | -5.22076286e+02 | 0.0122 | 4.5585 | 22.9454 | 4.7230 | 0.0147\n",
      "  40 | -5.84312522e+03 | 0.0191 | 2.3560 | 17.1215 | 3.8414 | 0.0572\n",
      "  60 | -2.38172234e-01 | 0.0149 | 11.1947 | 58.6385 | 7.1644 | 0.0992\n",
      "  80 | -9.21514090e-10 | 0.0144 | 25.3428 | 90.5562 | 10.6492 | 0.1360\n",
      " 100 | -3.78223892e-22 | 0.0144 | 45.8281 | 125.2535 | 15.5657 | 0.1741\n",
      "   0 | -3.85663788e-10 | 0.0004 | 25.2530 | 46.1099 | 3.6497 | 0.0000\n",
      "  20 | -7.64814761e-07 | 0.0116 | 19.6620 | 38.2081 | 4.3287 | 0.0123\n",
      "  40 | -2.91953973e-01 | 0.0225 | 9.5236 | 18.8078 | 2.7716 | 0.0487\n",
      "  60 | -1.80558270e+01 | 0.0232 | 7.3974 | 28.1916 | 2.5015 | 0.1022\n",
      "  80 | -8.26975135e-02 | 0.0263 | 11.6451 | 44.9244 | 4.1201 | 0.1561\n",
      " 100 | -4.33753617e-09 | 0.0282 | 23.8290 | 55.7630 | 6.6045 | 0.2151\n",
      "   0 | -5.52018895e+03 | 0.0006 | 2.4004 | 17.6285 | 3.9503 | 0.0000\n",
      "  20 | -3.20946949e+03 | 0.0123 | 1.0896 | 6.5320 | 3.0709 | 0.0162\n",
      "  40 | -1.20841804e+04 | 0.0139 | 1.9381 | 16.3505 | 4.1726 | 0.0535\n",
      "  60 | -7.40722855e+03 | 0.0159 | 2.1258 | 16.0585 | 3.5339 | 0.0853\n",
      "  80 | -1.30867568e+03 | 0.0171 | 3.5248 | 20.7497 | 3.9568 | 0.1070\n",
      " 100 | -2.33659074e-06 | 0.0171 | 19.7743 | 73.8486 | 8.7271 | 0.1315\n",
      "   0 | -5.00177170e-08 | 0.0007 | 21.0616 | 30.0458 | 6.4905 | 0.0000\n",
      "  20 | -8.56039810e-06 | 0.0138 | 17.2895 | 23.2024 | 5.1891 | 0.0155\n",
      "  40 | -1.03126330e-02 | 0.0205 | 11.6430 | 18.6190 | 3.5024 | 0.0530\n",
      "  60 | -6.46800492e+00 | 0.0256 | 7.2462 | 20.5416 | 2.4980 | 0.1051\n",
      "  80 | -4.06150668e+02 | 0.0300 | 4.1546 | 17.2152 | 1.9779 | 0.1662\n",
      " 100 | -2.61513423e+00 | 0.0352 | 7.8434 | 24.1887 | 2.4569 | 0.2344\n",
      "   0 | -1.23700994e+02 | 0.0008 | 5.0075 | 16.2908 | 1.9121 | 0.0000\n",
      "  20 | -5.69438339e+02 | 0.0154 | 3.1474 | 8.7997 | 1.5389 | 0.0155\n",
      "  40 | -5.43858287e+03 | 0.0202 | 2.1463 | 12.9586 | 1.8513 | 0.0548\n",
      "  60 | -4.74245465e+03 | 0.0233 | 1.7988 | 12.2578 | 1.7450 | 0.1003\n",
      "  80 | -1.21242782e+02 | 0.0227 | 5.5292 | 22.2722 | 2.2757 | 0.1421\n",
      " 100 | -1.81781050e-06 | 0.0232 | 19.3208 | 51.1965 | 3.1613 | 0.1821\n",
      "   0 | -1.69295276e-04 | 0.0006 | 15.3168 | 30.4795 | 6.2827 | 0.0000\n",
      "  20 | -1.36251054e-02 | 0.0119 | 11.9881 | 25.4433 | 5.2656 | 0.0146\n",
      "  40 | -1.23611107e+01 | 0.0225 | 6.6823 | 16.6043 | 3.7799 | 0.0573\n",
      "  60 | -9.61413004e+01 | 0.0206 | 6.3537 | 36.8080 | 4.6227 | 0.1144\n",
      "  80 | -5.58171610e+03 | 0.0306 | 2.0749 | 18.9919 | 4.6190 | 0.1736\n",
      " 100 | -1.42840580e-01 | 0.0358 | 10.6778 | 30.3180 | 4.9717 | 0.2389\n",
      "   0 | -5.33670812e-08 | 0.0006 | 21.8389 | 53.1307 | 7.6659 | 0.0000\n",
      "  20 | -6.95192118e-05 | 0.0112 | 16.3612 | 41.9055 | 6.5624 | 0.0143\n",
      "  40 | -4.79762862e+00 | 0.0214 | 6.9563 | 19.3465 | 4.1138 | 0.0545\n",
      "  60 | -3.21347998e+02 | 0.0242 | 4.7289 | 25.3863 | 4.2973 | 0.1174\n",
      "  80 | -1.50160215e+02 | 0.0263 | 5.5266 | 22.6988 | 3.6025 | 0.1863\n",
      " 100 | -8.47179853e-03 | 0.0288 | 12.2210 | 29.8870 | 4.8218 | 0.2578\n",
      "   0 | -8.65087438e+01 | 0.0006 | 4.9297 | 25.8435 | 5.0636 | 0.0000\n",
      "  20 | -1.49831438e+03 | 0.0123 | 2.3531 | 16.6372 | 3.9981 | 0.0149\n",
      "  40 | -2.94106855e+03 | 0.0180 | 2.1647 | 22.3437 | 4.0906 | 0.0575\n",
      "  60 | -4.21984328e-02 | 0.0145 | 11.0500 | 59.3249 | 7.1574 | 0.0985\n",
      "  80 | -5.45169248e-11 | 0.0140 | 26.1967 | 95.1448 | 14.0110 | 0.1376\n",
      " 100 | -6.90303039e-27 | 0.0140 | 52.8897 | 171.1621 | 41.4167 | 0.1780\n",
      "   0 | -1.27516480e-04 | 0.0006 | 14.8402 | 35.9511 | 6.0347 | 0.0000\n",
      "  20 | -3.01888789e-02 | 0.0117 | 10.8301 | 31.7099 | 5.3630 | 0.0153\n",
      "  40 | -1.44592997e+03 | 0.0233 | 2.8267 | 14.5474 | 3.7489 | 0.0593\n",
      "  60 | -1.46056446e+01 | 0.0253 | 6.4220 | 41.5123 | 5.8255 | 0.1260\n",
      "  80 | -2.26456882e-12 | 0.0243 | 28.5893 | 101.8331 | 15.1466 | 0.1933\n",
      " 100 | -1.35138443e-44 | 0.0241 | 82.0778 | 236.2707 | 59.7287 | 0.2620\n",
      "   0 | -1.77688160e-09 | 0.0006 | 23.4621 | 32.0428 | 3.1102 | 0.0000\n",
      "  20 | -5.37376462e-07 | 0.0143 | 19.2930 | 26.1267 | 2.7969 | 0.0154\n",
      "  40 | -3.64363854e-02 | 0.0222 | 11.1238 | 22.5989 | 2.9054 | 0.0554\n",
      "  60 | -9.02330821e+01 | 0.0274 | 5.7353 | 18.2109 | 2.2602 | 0.1094\n",
      "  80 | -7.49730567e-02 | 0.0312 | 11.4374 | 39.2569 | 5.1943 | 0.1719\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " 100 | -1.97597612e-16 | 0.0324 | 35.9652 | 80.7107 | 10.7474 | 0.2395\n",
      "   0 | -3.74348725e+00 | 0.0006 | 8.6277 | 38.0359 | 6.0730 | 0.0000\n",
      "  20 | -2.18827391e+02 | 0.0122 | 4.9697 | 23.1539 | 4.3853 | 0.0157\n",
      "  40 | -6.00652299e+03 | 0.0213 | 2.4351 | 17.6007 | 3.6943 | 0.0617\n",
      "  60 | -1.71865667e+02 | 0.0197 | 5.2712 | 25.6453 | 3.8318 | 0.1176\n",
      "  80 | -5.96097903e-04 | 0.0196 | 14.6012 | 38.2050 | 4.3942 | 0.1756\n",
      " 100 | -8.16449349e-16 | 0.0197 | 34.2197 | 56.5501 | 8.5481 | 0.2378\n",
      "   0 | -1.91692162e+02 | 0.0008 | 4.1659 | 16.8433 | 2.0551 | 0.0000\n",
      "  20 | -2.68884553e+03 | 0.0148 | 2.2268 | 9.7608 | 1.7156 | 0.0167\n",
      "  40 | -3.91746534e+03 | 0.0189 | 1.7904 | 13.6890 | 2.0648 | 0.0573\n",
      "  60 | -1.17522692e+02 | 0.0213 | 4.9444 | 25.4321 | 3.4389 | 0.1012\n",
      "  80 | -2.02371097e-01 | 0.0220 | 10.3117 | 30.7878 | 4.7515 | 0.1464\n",
      " 100 | -2.20377732e-06 | 0.0221 | 18.7658 | 36.7887 | 6.0394 | 0.1933\n",
      "   0 | -9.21904651e-01 | 0.0003 | 9.2056 | 38.3316 | 6.9145 | 0.0000\n",
      "  20 | -2.99841083e+01 | 0.0076 | 6.4162 | 27.5472 | 4.2329 | 0.0121\n",
      "  40 | -1.96841609e+03 | 0.0156 | 2.5613 | 15.6179 | 2.9422 | 0.0484\n",
      "  60 | -6.84096128e+03 | 0.0162 | 2.5281 | 29.5404 | 4.0740 | 0.1012\n",
      "  80 | -3.03553839e+03 | 0.0218 | 1.8561 | 11.2059 | 2.3912 | 0.1481\n",
      " 100 | -2.77003028e-05 | 0.0242 | 16.9531 | 59.0314 | 6.6805 | 0.1943\n",
      "   0 | -3.76728561e-04 | 0.0006 | 14.9919 | 34.7525 | 4.8517 | 0.0000\n",
      "  20 | -7.16534871e-02 | 0.0118 | 11.1605 | 29.6483 | 4.8256 | 0.0157\n",
      "  40 | -8.61413888e+02 | 0.0210 | 3.7791 | 18.3887 | 3.7808 | 0.0595\n",
      "  60 | -3.06795237e+03 | 0.0222 | 3.7536 | 29.9480 | 4.0435 | 0.1196\n",
      "  80 | -3.32527235e-01 | 0.0202 | 10.7975 | 48.5963 | 5.7353 | 0.1731\n",
      " 100 | -6.83651692e-08 | 0.0201 | 21.3791 | 44.1579 | 6.4845 | 0.2277\n",
      "   0 | -2.66466308e-03 | 0.0003 | 13.4566 | 42.8164 | 9.8140 | 0.0000\n",
      "  20 | -8.22179950e-02 | 0.0084 | 10.4691 | 22.5069 | 4.0978 | 0.0134\n",
      "  40 | -6.57754566e-01 | 0.0103 | 8.8820 | 26.2941 | 2.9175 | 0.0473\n",
      "  60 | -2.05914168e+01 | 0.0155 | 6.7878 | 29.3420 | 3.4710 | 0.0872\n",
      "  80 | -8.25697324e+03 | 0.0148 | 2.4870 | 26.0915 | 3.4612 | 0.1255\n",
      " 100 | -2.56565132e+03 | 0.0183 | 2.8431 | 28.1078 | 3.3590 | 0.1657\n",
      "   0 | -1.36219525e-04 | 0.0006 | 16.3617 | 54.3175 | 7.1921 | 0.0000\n",
      "  20 | -1.91962971e-01 | 0.0113 | 10.8902 | 40.8321 | 4.9272 | 0.0146\n",
      "  40 | -4.91998712e+02 | 0.0208 | 3.6438 | 14.3936 | 3.1474 | 0.0547\n",
      "  60 | -4.29808856e+00 | 0.0205 | 8.4619 | 35.3722 | 4.5841 | 0.1093\n",
      "  80 | -8.77429473e-12 | 0.0195 | 28.3610 | 72.8215 | 6.4411 | 0.1607\n",
      " 100 | -9.12809080e-36 | 0.0194 | 68.0585 | 141.1905 | 24.4077 | 0.2137\n",
      "   0 | -3.32368914e-05 | 0.0006 | 16.7238 | 35.7765 | 7.5340 | 0.0000\n",
      "  20 | -1.93312560e-03 | 0.0109 | 13.2710 | 26.8038 | 7.0186 | 0.0145\n",
      "  40 | -1.00760686e+00 | 0.0191 | 7.9353 | 20.5779 | 4.6068 | 0.0531\n",
      "  60 | -2.39068597e+01 | 0.0238 | 6.5429 | 23.5800 | 3.4999 | 0.1131\n",
      "  80 | -1.47148572e-01 | 0.0203 | 10.0153 | 36.0462 | 5.0146 | 0.1668\n",
      " 100 | -1.32673339e-06 | 0.0200 | 18.0104 | 41.0845 | 5.9216 | 0.2187\n",
      "   0 | -1.24386873e-05 | 0.0006 | 17.5777 | 43.6907 | 5.1691 | 0.0000\n",
      "  20 | -6.59287623e-03 | 0.0112 | 13.0037 | 36.0455 | 4.3394 | 0.0149\n",
      "  40 | -1.65043743e+02 | 0.0198 | 4.8955 | 18.9677 | 4.1274 | 0.0556\n",
      "  60 | -2.02368458e+01 | 0.0227 | 7.5577 | 43.0267 | 5.6752 | 0.1154\n",
      "  80 | -2.04727644e-07 | 0.0214 | 21.4904 | 75.6152 | 10.6752 | 0.1726\n",
      " 100 | -2.79776947e-21 | 0.0214 | 44.5384 | 110.0772 | 20.7263 | 0.2325\n",
      "   0 | -8.17538750e+03 | 0.0008 | 1.5748 | 11.0021 | 1.8208 | 0.0000\n",
      "  20 | -7.82547108e+03 | 0.0138 | 0.6904 | 5.3011 | 1.5508 | 0.0159\n",
      "  40 | -2.11122976e+04 | 0.0119 | 1.4949 | 14.7141 | 2.1581 | 0.0483\n",
      "  60 | -1.99558027e+04 | 0.0175 | 0.6526 | 8.0077 | 1.8638 | 0.0650\n",
      "  80 | -2.45779040e+01 | 0.0182 | 7.2846 | 33.0129 | 3.8164 | 0.0805\n",
      " 100 | -8.51459350e-12 | 0.0181 | 28.3781 | 64.1280 | 4.5366 | 0.1087\n",
      "   0 | -1.00361103e-02 | 0.0006 | 12.9507 | 39.9928 | 6.3533 | 0.0000\n",
      "  20 | -1.10316405e+00 | 0.0120 | 8.9987 | 26.1113 | 4.6984 | 0.0149\n",
      "  40 | -1.31200745e+02 | 0.0223 | 5.3272 | 18.9116 | 4.1328 | 0.0585\n",
      "  60 | -1.41004727e+02 | 0.0235 | 5.4786 | 24.5022 | 3.2122 | 0.1151\n",
      "  80 | -1.20206024e+00 | 0.0226 | 9.1396 | 35.5672 | 4.3657 | 0.1656\n",
      " 100 | -3.33994147e-07 | 0.0228 | 20.0739 | 41.5510 | 4.7135 | 0.2134\n",
      "   0 | -8.55275344e-07 | 0.0006 | 19.0137 | 55.9564 | 5.2577 | 0.0000\n",
      "  20 | -9.81237199e-04 | 0.0111 | 13.8507 | 34.8629 | 5.8398 | 0.0137\n",
      "  40 | -1.88629509e+00 | 0.0168 | 8.2485 | 23.7899 | 5.1231 | 0.0475\n",
      "  60 | -7.05435718e+00 | 0.0223 | 6.6703 | 28.1551 | 4.4872 | 0.0937\n",
      "  80 | -2.86835936e+02 | 0.0221 | 5.1078 | 33.2531 | 5.0481 | 0.1403\n",
      " 100 | -1.15523865e+02 | 0.0275 | 5.2456 | 22.0621 | 4.3780 | 0.1913\n",
      "   0 | -6.43870526e+03 | 0.0007 | 1.1140 | 7.9727 | 1.5561 | 0.0000\n",
      "  20 | -8.18771196e+03 | 0.0137 | 0.5456 | 4.7268 | 1.4584 | 0.0161\n",
      "  40 | -1.15807387e+04 | 0.0115 | 0.7244 | 10.4290 | 1.9575 | 0.0454\n",
      "  60 | -1.34984309e+04 | 0.0173 | 1.0864 | 10.6165 | 2.0328 | 0.0696\n",
      "  80 | -2.51366791e+01 | 0.0150 | 6.3861 | 25.8551 | 2.5416 | 0.0904\n",
      " 100 | -1.42324040e-06 | 0.0149 | 19.4052 | 51.5703 | 4.0080 | 0.1129\n",
      "   0 | -1.26702825e+01 | 0.0003 | 7.6207 | 35.5250 | 3.5238 | 0.0000\n",
      "  20 | -5.82502688e+02 | 0.0068 | 4.6570 | 28.9041 | 3.5264 | 0.0128\n",
      "  40 | -1.19443463e+04 | 0.0138 | 0.2280 | 8.9334 | 2.6696 | 0.0516\n",
      "  60 | -3.65736670e+00 | 0.0132 | 9.2419 | 60.0730 | 5.7432 | 0.0994\n",
      "  80 | -2.02423482e-11 | 0.0133 | 28.2145 | 122.0748 | 27.2776 | 0.1358\n",
      " 100 | -9.91811980e-28 | 0.0133 | 54.9618 | 186.4110 | 64.6371 | 0.1666\n",
      "   0 | -1.09893512e+01 | 0.0008 | 7.8607 | 32.8618 | 2.9131 | 0.0000\n",
      "  20 | -1.11306452e+03 | 0.0151 | 3.9104 | 20.0162 | 2.3857 | 0.0156\n",
      "  40 | -5.13166231e+03 | 0.0242 | 2.8047 | 17.2724 | 2.2331 | 0.0607\n",
      "  60 | -5.34680636e-02 | 0.0199 | 12.0725 | 44.1659 | 3.2185 | 0.1086\n",
      "  80 | -3.56016244e-10 | 0.0193 | 25.7539 | 66.4519 | 4.5742 | 0.1550\n",
      " 100 | -2.05707356e-21 | 0.0192 | 44.0466 | 78.5947 | 5.3643 | 0.2027\n",
      "   0 | -1.44229297e-03 | 0.0007 | 14.0009 | 36.9849 | 4.1798 | 0.0000\n",
      "  20 | -2.87277253e-01 | 0.0133 | 9.5602 | 26.5974 | 2.9578 | 0.0148\n",
      "  40 | -5.27554777e+02 | 0.0224 | 3.7677 | 13.0709 | 1.8549 | 0.0524\n",
      "  60 | -2.93231099e+00 | 0.0240 | 7.8808 | 32.6214 | 2.8629 | 0.1028\n",
      "  80 | -3.44479620e-10 | 0.0245 | 24.6901 | 69.4528 | 5.1559 | 0.1550\n",
      " 100 | -1.14802995e-29 | 0.0245 | 56.6783 | 115.6086 | 8.1504 | 0.2122\n",
      "   0 | -1.93097191e-11 | 0.0001 | 26.7829 | 34.1615 | 3.5453 | 0.0000\n",
      "  20 | -2.75975170e-10 | 0.0070 | 24.7990 | 30.8683 | 3.3869 | 0.0058\n",
      "  40 | -5.28679743e-06 | 0.0209 | 17.3557 | 20.7717 | 2.9699 | 0.0370\n",
      "  60 | -1.40723932e-01 | 0.0280 | 10.3030 | 21.7483 | 2.5036 | 0.0942\n",
      "  80 | -2.84030222e+01 | 0.0350 | 7.1595 | 30.9841 | 3.4797 | 0.1689\n",
      " 100 | -2.24872930e+02 | 0.0342 | 5.3412 | 23.6200 | 3.6705 | 0.2370\n",
      "   0 | -3.60036989e+03 | 0.0007 | 1.5257 | 9.3981 | 1.8980 | 0.0000\n",
      "  20 | -2.48341285e+03 | 0.0145 | 0.5766 | 4.4241 | 1.5333 | 0.0164\n",
      "  40 | -1.55746946e+04 | 0.0157 | 0.6215 | 9.5902 | 1.9713 | 0.0531\n",
      "  60 | -1.36876912e+03 | 0.0189 | 2.4937 | 13.0528 | 2.1970 | 0.0920\n",
      "  80 | -2.07917648e-03 | 0.0171 | 13.3375 | 40.2053 | 3.0998 | 0.1315\n",
      " 100 | -2.32777020e-17 | 0.0170 | 36.5130 | 79.1170 | 8.5855 | 0.1729\n",
      "   0 | -9.32751914e-02 | 0.0006 | 10.5689 | 21.2026 | 4.7765 | 0.0000\n",
      "  20 | -1.85544497e+00 | 0.0124 | 8.2381 | 17.7103 | 4.0274 | 0.0142\n",
      "  40 | -3.73147612e+02 | 0.0195 | 4.2342 | 15.5412 | 3.8290 | 0.0554\n",
      "  60 | -1.23908008e+04 | 0.0235 | 2.0895 | 20.3147 | 4.2267 | 0.1165\n",
      "  80 | -3.27990084e+02 | 0.0278 | 4.4557 | 19.5620 | 4.3290 | 0.1872\n",
      " 100 | -8.44742765e-12 | 0.0284 | 27.4060 | 47.7109 | 7.1586 | 0.2672\n",
      "   0 | -2.35868500e-03 | 0.0003 | 13.5030 | 32.6856 | 2.9991 | 0.0000\n",
      "  20 | -9.88933723e-02 | 0.0068 | 10.6920 | 28.4521 | 3.0535 | 0.0121\n",
      "  40 | -6.57088463e+01 | 0.0130 | 5.0861 | 13.7430 | 3.7383 | 0.0483\n",
      "  60 | -2.38098202e+01 | 0.0160 | 7.4757 | 45.0539 | 5.6395 | 0.0993\n",
      "  80 | -2.64165022e-03 | 0.0170 | 14.5532 | 74.6020 | 9.7981 | 0.1421\n",
      " 100 | -1.53699354e-08 | 0.0172 | 23.3665 | 102.8810 | 25.1921 | 0.1864\n",
      "   0 | -1.67976149e-02 | 0.0003 | 12.1920 | 48.2616 | 7.0099 | 0.0000\n",
      "  20 | -3.15101518e+00 | 0.0078 | 8.1970 | 40.6457 | 5.1250 | 0.0130\n",
      "  40 | -5.52433353e+03 | 0.0158 | 1.5251 | 12.5427 | 2.2592 | 0.0514\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  60 | -2.94331006e+00 | 0.0138 | 8.7730 | 58.8258 | 6.0713 | 0.1048\n",
      "  80 | -1.35015950e-12 | 0.0132 | 29.7607 | 123.6637 | 33.0783 | 0.1487\n",
      " 100 | -9.62130759e-32 | 0.0132 | 61.3325 | 185.8180 | 62.9600 | 0.1886\n",
      "   0 | -1.04899747e-01 | 0.0008 | 11.0614 | 34.2099 | 3.1865 | 0.0000\n",
      "  20 | -1.53996616e+01 | 0.0143 | 6.9128 | 22.0782 | 2.7753 | 0.0161\n",
      "  40 | -8.24987967e+02 | 0.0250 | 3.6436 | 16.5590 | 2.1574 | 0.0625\n",
      "  60 | -4.78442204e+02 | 0.0209 | 4.7281 | 25.3157 | 2.1988 | 0.1161\n",
      "  80 | -1.20162302e+03 | 0.0260 | 3.3150 | 14.9378 | 1.8091 | 0.1528\n",
      " 100 | -8.00225831e-04 | 0.0279 | 14.1884 | 32.8971 | 2.6170 | 0.1747\n"
     ]
    }
   ],
   "source": [
    "batch_size = 1\n",
    "seed_data = [dset[i] for i in torch.randperm(len(dset))[:num_attacks]]\n",
    "loader = torchani.data.TransformableIterable(seed_data).collate(batch_size)\n",
    "\n",
    "print(\"Step |     AdvLoss     | Delta  | Energy  |  Force  |  Fstd  | RMSD \")\n",
    "\n",
    "molecules, energy, force_std, force_max, rmsd = [], [], [], [], []\n",
    "for i, batch in enumerate(loader):\n",
    "    species = batch['species'].to(device)\n",
    "    coordinates = batch['coordinates'].to(device)\n",
    "    coordinates.requires_grad = True\n",
    "    \n",
    "    delta = torch.zeros_like(coordinates, requires_grad=True, device=device)\n",
    "    optimizer = torch.optim.Adam([delta], lr=lr)\n",
    "    \n",
    "    # Get atomic numbers and coordinates of seed configuration\n",
    "    z = torch.tensor(list(map(lambda x: atomic_number[species_order[x.item()]], batch['species'].squeeze())))\n",
    "    orig_mol = ase.Atoms(numbers=z, \n",
    "                         positions=coordinates[:len(z)].squeeze().cpu().detach().numpy())\n",
    "    v = np.concatenate([orig_mol.get_atomic_numbers().reshape(-1, 1), orig_mol.get_positions()], axis=1)\n",
    "    formula = orig_mol.get_chemical_formula()\n",
    "\n",
    "    molecules.append(formula)\n",
    "    energy.append([])\n",
    "    force_std.append([])\n",
    "    force_max.append([])\n",
    "    rmsd.append([])\n",
    "    \n",
    "    for epoch in range(n_epochs+1):\n",
    "        optimizer.zero_grad()\n",
    "\n",
    "        coordinates = coordinates + delta\n",
    "\n",
    "        results = [model((species, coordinates)) for model in models]\n",
    "\n",
    "        pred_energies = torch.stack([torchani.units.hartree2kcalmol(r.energies) - np.array(energies[formula]).min() for r in results], dim=1)\n",
    "        pred_forces = torch.stack(\n",
    "            [-torch.autograd.grad(\n",
    "                torchani.units.hartree2kcalmol(r.energies), coordinates, allow_unused=True, retain_graph=True\n",
    "            )[0] for r in results], dim=1).squeeze()\n",
    "\n",
    "        loss = loss_fn[formula].loss_fn(x=None, e=pred_energies, f=pred_forces).sum()\n",
    "\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "        \n",
    "        # calculate rmsd from seed configuration\n",
    "        disturbed_mol = ase.Atoms(numbers=np.array(z), \n",
    "                                      positions=coordinates[:len(z)].squeeze().cpu().detach().numpy())\n",
    "        w = np.concatenate([disturbed_mol.get_atomic_numbers().reshape(-1, 1), \n",
    "                            disturbed_mol.get_positions()], axis=1)\n",
    "        rmsd_vw = compute_rmsd(v, w)\n",
    "        \n",
    "        # save values \n",
    "        energy[i].append(pred_energies.mean(-1).item())\n",
    "        force_std[i].append(get_force_std(pred_forces, mode='max').detach().cpu().numpy())\n",
    "        force_max[i].append(get_force_max(pred_forces).detach().cpu().numpy())\n",
    "        rmsd[i].append(rmsd_vw)\n",
    "\n",
    "        if epoch % 20 == 0:\n",
    "            print(\"{:>4d} | {:>10.8e} | {:>5.4f} | {:>5.4f} | {:>5.4f} | {:>5.4f} | {:.4f}\".format(epoch, \n",
    "                                                                                        loss.item(), \n",
    "                                                                                        delta.abs().sum().item(), \n",
    "                                                                                        pred_energies.mean(-1).item(),\n",
    "                                                                                        pred_forces.abs().mean(),\n",
    "                                                                                        pred_forces.std(0).mean(),\n",
    "                                                                                        rmsd_vw))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "energy = np.array(energy)\n",
    "force_std = np.array(force_std)\n",
    "force_max = np.array(force_max)\n",
    "rmsd = np.array(rmsd)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Plot evolution of mean force uncertainty and mean rmsd as a function of steps for each molecule\n",
    "\n",
    "This plot reproduces figures S17 and S19 of the paper, up to differences in the random initialization. It essentially shows distortion of molecules to configurations that corresponds to high uncertainty (high rmsd) of the models. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "mpl.rcParams['pdf.fonttype'] = 42\n",
    "mpl.rcParams['ps.fonttype'] = 42\n",
    "mpl.rcParams.update({'font.size': 14})\n",
    "mpl.rc('lines', linewidth=2, color='g')\n",
    "mpl.rcParams['axes.linewidth'] = 1.5 \n",
    "mpl.rcParams['axes.linewidth'] = 1.5 \n",
    "mpl.rcParams[\"xtick.major.size\"] = 6 \n",
    "mpl.rcParams[\"ytick.major.size\"] = 6 \n",
    "mpl.rcParams[\"ytick.major.width\"] = 2 \n",
    "mpl.rcParams[\"xtick.major.width\"] = 2 "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "cmap = cm.inferno\n",
    "COLORS = {\n",
    "    \"CH4\": cmap(64),\n",
    "    \"H3N\": cmap(128),\n",
    "    \"H2O\": cmap(192),\n",
    "}\n",
    "\n",
    "MARKERS = {\n",
    "    \"CH4\": \"o\",\n",
    "    \"H3N\": \"^\",\n",
    "    \"H2O\": \"s\",\n",
    "}\n",
    "\n",
    "ce_format = {\n",
    "    \"CH4\": \"$CH_4$\",\n",
    "    \"H3N\": \"$NH_3$\",\n",
    "    \"H2O\": \"$H_2O$\"\n",
    "}\n",
    "\n",
    "natoms = {\"CH4\": 5, \"H2O\": 3, \"H3N\": 4}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(1, 2, figsize=(10, 4))\n",
    "\n",
    "max_epoch = 71\n",
    "interval = 10\n",
    "\n",
    "mol_indices = {mol: [i for i, x in enumerate(molecules) if x == mol] for mol in set(molecules)}\n",
    "for mol, idx in mol_indices.items():\n",
    "    fstd_mean = force_std[idx].mean(0)\n",
    "    fstd_std = force_std[idx].std(0)\n",
    "    rmsd_mean = rmsd[idx].mean(0)\n",
    "    rmsd_std = rmsd[idx].std(0)\n",
    "    \n",
    "    axes[0].errorbar(range(len(fstd_mean))[:max_epoch:interval], fstd_mean[:max_epoch:interval], yerr=fstd_std[:max_epoch:interval], \n",
    "                     label=ce_format[mol], color=COLORS[mol], marker=MARKERS[mol], capsize=3, linewidth=1)\n",
    "    axes[1].errorbar(range(len(rmsd_mean))[:max_epoch:interval], rmsd_mean[:max_epoch:interval], yerr=rmsd_std[:max_epoch:interval],\n",
    "                     label=ce_format[mol], color=COLORS[mol], marker=MARKERS[mol], capsize=3, linewidth=1)\n",
    "\n",
    "axes[0].set_ylabel(\"Max force std (kcal/mol/Å)\")\n",
    "axes[1].set_ylabel(\"RMSD\")\n",
    "\n",
    "axes[0].set_yscale(\"log\")\n",
    "\n",
    "for ax in axes:\n",
    "    ax.legend(frameon=False)\n",
    "    ax.set_xlabel(\"Step\")\n",
    "    ax.spines['right'].set_visible(False)\n",
    "    ax.spines['top'].set_visible(False)\n",
    "    \n",
    "fig.subplots_adjust(wspace=0.30)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This figure reproduces the one in S18 and S20 in the paper (without alanine dipeptide)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 936x936 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(3, 3, figsize=(13, 13))\n",
    "\n",
    "mol_indices = {mol: [i for i, x in enumerate(molecules) if x == mol] for mol in set(molecules)}\n",
    "for i, (mol, idx) in enumerate(mol_indices.items()):\n",
    "    axes[0, i].set_title(ce_format[mol])\n",
    "    energy_mol = energy[idx][:, :max_epoch].flatten() + np.array(energies[mol]).min()\n",
    "    energy_mol -= energy_mol.min() \n",
    "    energy_mol_peratom = energy_mol / natoms[mol]\n",
    "    fstd = force_std[idx][:, :max_epoch]\n",
    "    fmax = force_max[idx][:, :max_epoch]\n",
    "    axes[0, i].scatter(energy_mol_peratom, fstd.flatten()/fmax.flatten(), color=COLORS[mol])\n",
    "    axes[1, i].scatter(energy_mol_peratom, fstd, color=COLORS[mol])\n",
    "    axes[2, i].scatter(energy_mol, rmsd[idx][:, :max_epoch].flatten(), color=COLORS[mol])\n",
    "\n",
    "axes[0, 0].set_ylabel(\"Relative force std\")\n",
    "axes[1, 0].set_ylabel(\"Max force std (kcal/mol/Å)\")\n",
    "axes[2, 0].set_ylabel(\"RMSD\")\n",
    "\n",
    "for i in range(3):\n",
    "    axes[0, i].set_xlabel(\"Energy per atom (kcal/mol)\")\n",
    "    axes[1, i].set_xlabel(\"Energy per atom (kcal/mol)\")\n",
    "    axes[2, i].set_xlabel(\"Energy (kcal/mol)\")\n",
    "    \n",
    "    axes[0, i].xaxis.set_tick_params(which='both', labelbottom=True)\n",
    "\n",
    "    for j in range(3):\n",
    "        axes[j, i].spines['right'].set_visible(False)\n",
    "        axes[j, i].spines['top'].set_visible(False)\n",
    "\n",
    "\n",
    "fig.subplots_adjust(wspace=0.25, hspace=0.4)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "advsampling",
   "language": "python",
   "name": "advsampling"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}