Test&Viz_AFSE.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "os.environ[\"CUDA_VISIBLE_DEVICES\"] = \"3\"\n",
    "import torch\n",
    "import torch.autograd as autograd\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "import torch.optim as optim\n",
    "import torch.utils.data as Data\n",
    "import math\n",
    "torch.manual_seed(8)\n",
    "import time\n",
    "import numpy as np\n",
    "import gc\n",
    "import sys\n",
    "sys.setrecursionlimit(50000)\n",
    "import pickle\n",
    "torch.backends.cudnn.benchmark = True\n",
    "torch.set_default_tensor_type('torch.cuda.FloatTensor')\n",
    "# from tensorboardX import SummaryWriter\n",
    "torch.nn.Module.dump_patches = True\n",
    "import copy\n",
    "import pandas as pd\n",
    "#then import my own modules\n",
    "from AttentiveFP.AttentiveLayers import Fingerprint\n",
    "from AttentiveFP import save_smiles_dicts, get_smiles_dicts, get_smiles_array, moltosvg_highlight"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from rdkit import Chem\n",
    "# from rdkit.Chem import AllChem\n",
    "from rdkit.Chem import Draw\n",
    "from rdkit.Chem.Draw import IPythonConsole\n",
    "IPythonConsole.ipython_useSVG = True\n",
    "from rdkit.Chem import QED\n",
    "from rdkit.Chem import rdMolDescriptors, MolSurf\n",
    "from rdkit.Chem.Draw import SimilarityMaps\n",
    "from rdkit import Chem\n",
    "from rdkit.Chem import AllChem\n",
    "from rdkit.Chem import rdDepictor\n",
    "from rdkit.Chem.Draw import rdMolDraw2D\n",
    "%matplotlib inline\n",
    "from numpy.polynomial.polynomial import polyfit\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.cm as cm\n",
    "import matplotlib\n",
    "import seaborn as sns; sns.set()\n",
    "from IPython.display import SVG, display\n",
    "import sascorer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from AttentiveFP.utils import EarlyStopping\n",
    "from AttentiveFP.utils import Meter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "model_file/1_sim_EC50_P21554_theorem\n"
     ]
    }
   ],
   "source": [
    "raw_filename = \"../data/active_inactive/median_active/EC50/P21554.csv\" #目标域active\n",
    "test_active = 200\n",
    "train_rate = 1\n",
    "random_seed = 68\n",
    "inactive_filename = \"../data/active_inactive/median_inactive/EC50/P21554.csv\"#目标域inactive\n",
    "file_list1 = raw_filename.split('/')\n",
    "file1 = file_list1[-2]+'_'+file_list1[-1]\n",
    "file1 = file1[:-4]\n",
    "model_file = \"model_file/1_sim_\"+file1+\"_theorem\"\n",
    "print(model_file)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                              smiles     value\n",
      "0  CCC1=C(NC2=C1C=C(C=C2)Cl)C(=O)NCCC3=CC=C(C=C3)... -2.683047\n",
      "1  CCCC=CC1=CC=C(S1)C2=C(C(=NN2C3=C(C=C(C=C3)Cl)C... -2.164947\n",
      "2  C1CN(CCC1C2=C3C=C(C=CC3=NN2)C(C4=CC=C(C=C4)Cl)... -1.690196\n",
      "3  C1CC1C2=C(N=C(C=C2)C(=O)NC(C3=CC=CC=N3)C(F)(F)... -2.816241\n",
      "4  C1CN(CCC1NC2=NC=NC3=C2C=C(C=C3)C(C4=CC=C(C=C4)... -2.255273\n",
      "number of all smiles:  1426\n",
      "number of successfully processed smiles:  1426\n",
      "                                              smiles     value  \\\n",
      "0  CCC1=C(NC2=C1C=C(C=C2)Cl)C(=O)NCCC3=CC=C(C=C3)... -2.683047   \n",
      "1  CCCC=CC1=CC=C(S1)C2=C(C(=NN2C3=C(C=C(C=C3)Cl)C... -2.164947   \n",
      "2  C1CN(CCC1C2=C3C=C(C=CC3=NN2)C(C4=CC=C(C=C4)Cl)... -1.690196   \n",
      "3  C1CC1C2=C(N=C(C=C2)C(=O)NC(C3=CC=CC=N3)C(F)(F)... -2.816241   \n",
      "4  C1CN(CCC1NC2=NC=NC3=C2C=C(C=C3)C(C4=CC=C(C=C4)... -2.255273   \n",
      "\n",
      "                                         cano_smiles  \n",
      "0  CCc1c(C(=O)NCCc2ccc(N3CCCCC3)cc2)[nH]c2ccc(Cl)...  \n",
      "1  CCCC=Cc1ccc(-c2c(C)c(C(=O)NN3CCCCC3)nn2-c2ccc(...  \n",
      "2  O=S(=O)(c1ccccc1C(F)(F)F)N1CCC(c2[nH]nc3ccc(C(...  \n",
      "3  O=C(NC(c1ccccn1)C(F)(F)F)c1ccc(C2CC2)c(OCC2CCO...  \n",
      "4  O=C(NC(CO)(CO)CO)c1ccc(N2CCC(Nc3ncnc4ccc(C(c5c...  \n"
     ]
    }
   ],
   "source": [
    "tasks = ['value']\n",
    "feature_filename = raw_filename.replace('.csv','.pickle')\n",
    "filename = raw_filename.replace('.csv','')\n",
    "prefix_filename = raw_filename.split('/')[-1].replace('.csv','')\n",
    "smiles_tasks_df = pd.read_csv(raw_filename, header=0, names = [\"smiles\",\"value\"],usecols=[3,8])\n",
    "# smiles_tasks_df = smiles_tasks_df[1:]\n",
    "# smiles_tasks_df = smiles_tasks_df.drop(0,axis=1,inplace=False) \n",
    "print(smiles_tasks_df[:5])\n",
    "# print(smiles_tasks_df.iloc(1))\n",
    "def add_canonical_smiles(smiles_tasks_df):\n",
    "    smilesList = smiles_tasks_df.smiles.values\n",
    "    print(\"number of all smiles: \",len(smilesList))\n",
    "    atom_num_dist = []\n",
    "    remained_smiles = []\n",
    "    canonical_smiles_list = []\n",
    "    for smiles in smilesList:\n",
    "        try:        \n",
    "            mol = Chem.MolFromSmiles(smiles)\n",
    "            atom_num_dist.append(len(mol.GetAtoms()))\n",
    "            remained_smiles.append(smiles)\n",
    "            canonical_smiles_list.append(Chem.MolToSmiles(Chem.MolFromSmiles(smiles), isomericSmiles=True))\n",
    "        except:\n",
    "            print(smiles)\n",
    "            pass\n",
    "    print(\"number of successfully processed smiles: \", len(remained_smiles))\n",
    "    smiles_tasks_df = smiles_tasks_df[smiles_tasks_df[\"smiles\"].isin(remained_smiles)]\n",
    "    smiles_tasks_df['cano_smiles'] =canonical_smiles_list\n",
    "    return smiles_tasks_df\n",
    "# print(smiles_tasks_df)\n",
    "smiles_tasks_df = add_canonical_smiles(smiles_tasks_df)\n",
    "\n",
    "print(smiles_tasks_df.head())\n",
    "# plt.figure(figsize=(5, 3))\n",
    "# sns.set(font_scale=1.5)\n",
    "# ax = sns.distplot(atom_num_dist, bins=28, kde=False)\n",
    "# plt.tight_layout()\n",
    "# # plt.savefig(\"atom_num_dist_\"+prefix_filename+\".png\",dpi=200)\n",
    "# plt.show()\n",
    "# plt.close()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "random_seed = 68\n",
    "start_time = str(time.ctime()).replace(':','-').replace(' ','_')\n",
    "\n",
    "p_dropout= 0.03\n",
    "fingerprint_dim = 100\n",
    "\n",
    "weight_decay = 4.3 # also known as l2_regularization_lambda\n",
    "learning_rate = 4\n",
    "radius = 2\n",
    "T = 1\n",
    "per_task_output_units_num = 1 # for regression model\n",
    "output_units_num = len(tasks) * per_task_output_units_num"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1226, 3) (1426, 3) (200, 3)\n"
     ]
    }
   ],
   "source": [
    "# remained_df[tasks[0]] = remained_df[tasks[0]].astype(float)\n",
    "# test_df = remained_df.sample(frac=0.8,random_state=random_seed) #随机取0.8做test\n",
    "\n",
    "# smiles_tasks_df 为读取出来的数据\n",
    "# tmp = smiles_tasks_df.loc[smiles_tasks_df['value'].str.contains('-')]\n",
    "# print(tmp.shape)\n",
    "# smiles_tasks_df = smiles_tasks_df.drop(tmp.index)\n",
    "sample_df = smiles_tasks_df.sample(frac=train_rate,random_state=random_seed)\n",
    "\n",
    "# sample_df[\"value\"] = sample_df[\"value\"].astype(float) #astype()强制转换数据类型\n",
    "# sample_df[\"value\"] = sample_df[\"value\"].apply(lambda x:-math.log10(x))\n",
    "# print(sample_df[:10])\n",
    "\n",
    "test_df = sample_df.sample(test_active,random_state=random_seed)\n",
    "train_df = sample_df.drop(test_df.index) \n",
    "print(train_df.shape,sample_df.shape,test_df.shape)\n",
    "train_df = train_df.reset_index(drop=True)\n",
    "test_df = test_df.reset_index(drop=True)\n",
    "# print(train_df.head())\n",
    "# print(test_df.head())\n",
    "# print(len(test_df),sorted(test_df.cano_smiles.values))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "number of all smiles:  1730\n",
      "number of successfully processed smiles:  1730\n",
      "                                              smiles  \\\n",
      "0  CCC(CC)(C(=O)NC)NC(=O)C1=NC(=C(C=C1)C2CCCC2)OC...   \n",
      "1  CC1CCC2=C(C1)SC3=NC(=NC(=C23)N4CCN(CC4)C(=O)C5...   \n",
      "2  CC(C)(C)OC(=O)N(CC1CC1)NC(=O)C2=NC(=C(C=C2)C3C...   \n",
      "3  CC1=NC(=NO1)C(C)(C2CC2)NC(=O)C3=NC(=C(C=C3)C4C...   \n",
      "4  CNC(=O)CC1(COC1)NC(=O)C2=NC(=C(C=C2)C(F)(F)F)O...   \n",
      "\n",
      "                                         cano_smiles  value  \n",
      "0   CCC(CC)(NC(=O)c1ccc(C2CCCC2)c(OCC2CC2)n1)C(=O)NC   -4.1  \n",
      "1  Cc1nc(N2CCN(C(=O)c3ccccc3)CC2)c2c3c(sc2n1)CC(C...   -4.1  \n",
      "2  CC(C)(C)OC(=O)N(CC1CC1)NC(=O)c1ccc(C2CC2)c(OCC...   -4.1  \n",
      "3  Cc1nc(C(C)(NC(=O)c2ccc(C3CC3)c(OCC3CC3)n2)C2CC...   -4.1  \n",
      "4  CNC(=O)CC1(NC(=O)c2ccc(C(F)(F)F)c(OCC3CC3)n2)COC1   -4.1  \n",
      "(1930, 3)\n",
      "                                              smiles     value  \\\n",
      "0  C1CN(CCC1NC2=NC=NC3=C2C=C(C=C3)C(C4=CC=C(C=C4)... -1.204120   \n",
      "1  CC(C)(C1=CC=CC=C1)NC(=O)C2=NN(C3=C2CC4C3C4)C5=... -1.198657   \n",
      "2   CC(C)(C)C(C(=O)NC)NC(=O)C1=CC=CC(=N1)C2=CC=CC=C2 -2.867585   \n",
      "3  C1CC1N2C(=C3C=C(C=CC3=N2)C(C4=CC=C(C=C4)Cl)C5=... -1.230449   \n",
      "4  CC(C(CC1=CC=C(C=C1)Cl)C2=CC=CC(=C2)C#N)NC(=O)C... -0.278754   \n",
      "\n",
      "                                         cano_smiles  \n",
      "0  O=S(=O)(CCO)N1CCC(Nc2ncnc3ccc(C(c4ccc(Cl)cc4)c...  \n",
      "1  CC(C)(NC(=O)c1nn(-c2ccc(F)cc2F)c2c1CC1CC21)c1c...  \n",
      "2        CNC(=O)C(NC(=O)c1cccc(-c2ccccc2)n1)C(C)(C)C  \n",
      "3  NC(=O)c1ccc(S(=O)(=O)N2CCC(c3c4cc(C(c5ccc(Cl)c...  \n",
      "4  CC(NC(=O)C(C)(C)Nc1cc2ccccc2cn1)C(Cc1ccc(Cl)cc...  \n"
     ]
    }
   ],
   "source": [
    "# inactive_filename = \"../data/active_inactive/median_inactive/EC50/Q99500.csv\"\n",
    "inactive_df = pd.read_csv(inactive_filename,header=0,names=[\"smiles\"],usecols=[3])\n",
    "inactive_df = add_canonical_smiles(inactive_df)\n",
    "inactive_df['value']=-4.1\n",
    "print(inactive_df.head())\n",
    "for l in inactive_df[\"cano_smiles\"]:\n",
    "    if l in sample_df[\"cano_smiles\"]:\n",
    "        print(\"same smiles:\",l)\n",
    "        \n",
    "test_df = pd.concat([test_df,inactive_df],axis=0)\n",
    "test_df = test_df.reset_index(drop=True)\n",
    "print(test_df.shape)\n",
    "print(test_df.head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "../data/active_inactive/median_active/EC50/P21554.pickle\n",
      "../data/active_inactive/median_active/EC50/P21554\n",
      "3156\n",
      "feature dicts file saved as ../data/active_inactive/median_active/EC50/P21554.pickle\n"
     ]
    }
   ],
   "source": [
    "print(feature_filename)\n",
    "print(filename)\n",
    "total_df = pd.concat([train_df,test_df],axis=0)\n",
    "total_smilesList = total_df['smiles'].values\n",
    "print(len(total_smilesList))\n",
    "# if os.path.isfile(feature_filename):\n",
    "#     feature_dicts = pickle.load(open(feature_filename, \"rb\" ))\n",
    "# else:\n",
    "#     feature_dicts = save_smiles_dicts(smilesList,filename)\n",
    "feature_dicts = save_smiles_dicts(total_smilesList,filename)\n",
    "remained_df = total_df[total_df[\"cano_smiles\"].isin(feature_dicts['smiles_to_atom_mask'].keys())]\n",
    "# uncovered_df = total_df.drop(remained_df.index)\n",
    "# uncovered_df\n",
    "train_df = train_df[train_df[\"cano_smiles\"].isin(feature_dicts['smiles_to_atom_mask'].keys())]\n",
    "test_df = test_df[test_df[\"cano_smiles\"].isin(feature_dicts['smiles_to_atom_mask'].keys())]\n",
    "train_df = train_df.reset_index(drop=True)\n",
    "test_df = test_df.reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "val_df = train_df.sample(frac=0.2,random_state=random_seed)\n",
    "train_df = train_df.drop(val_df.index)\n",
    "train_df = train_df.reset_index(drop=True)\n",
    "val_df = val_df.reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(981, 3) (1930, 3) (245, 3)\n"
     ]
    }
   ],
   "source": [
    "print(train_df.shape,test_df.shape,val_df.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "241904\n",
      "atom_fc.weight torch.Size([100, 39])\n",
      "atom_fc.bias torch.Size([100])\n",
      "neighbor_fc.weight torch.Size([100, 49])\n",
      "neighbor_fc.bias torch.Size([100])\n",
      "GRUCell.0.weight_ih torch.Size([300, 100])\n",
      "GRUCell.0.weight_hh torch.Size([300, 100])\n",
      "GRUCell.0.bias_ih torch.Size([300])\n",
      "GRUCell.0.bias_hh torch.Size([300])\n",
      "GRUCell.1.weight_ih torch.Size([300, 100])\n",
      "GRUCell.1.weight_hh torch.Size([300, 100])\n",
      "GRUCell.1.bias_ih torch.Size([300])\n",
      "GRUCell.1.bias_hh torch.Size([300])\n",
      "align.0.weight torch.Size([1, 200])\n",
      "align.0.bias torch.Size([1])\n",
      "align.1.weight torch.Size([1, 200])\n",
      "align.1.bias torch.Size([1])\n",
      "attend.0.weight torch.Size([100, 100])\n",
      "attend.0.bias torch.Size([100])\n",
      "attend.1.weight torch.Size([100, 100])\n",
      "attend.1.bias torch.Size([100])\n",
      "mol_GRUCell.weight_ih torch.Size([300, 100])\n",
      "mol_GRUCell.weight_hh torch.Size([300, 100])\n",
      "mol_GRUCell.bias_ih torch.Size([300])\n",
      "mol_GRUCell.bias_hh torch.Size([300])\n",
      "mol_align.weight torch.Size([1, 200])\n",
      "mol_align.bias torch.Size([1])\n",
      "mol_attend.weight torch.Size([100, 100])\n",
      "mol_attend.bias torch.Size([100])\n",
      "metric.weight torch.Size([100, 200])\n",
      "metric.bias torch.Size([100])\n",
      "output.weight torch.Size([1, 100])\n",
      "output.bias torch.Size([1])\n"
     ]
    }
   ],
   "source": [
    "x_atom, x_bonds, x_atom_index, x_bond_index, x_mask, smiles_to_rdkit_list = get_smiles_array([remained_df[\"cano_smiles\"].values[0]],feature_dicts)\n",
    "num_atom_features = x_atom.shape[-1]\n",
    "num_bond_features = x_bonds.shape[-1]\n",
    "loss_function = nn.MSELoss()\n",
    "model = Fingerprint_viz(radius, T, num_atom_features, num_bond_features,\n",
    "            fingerprint_dim, output_units_num, p_dropout)\n",
    "model.cuda()\n",
    "max_epoch = 1000\n",
    "batch_size = 10\n",
    "\n",
    "# optimizer = optim.Adam(model.parameters(), learning_rate, weight_decay=weight_decay)\n",
    "# optimizer = optim.Adam(model.parameters(), 10**-learning_rate, weight_decay=10**-weight_decay)\n",
    "optimizer = optim.Adam([\n",
    "{'params': model.parameters(), 'lr': 10**(-learning_rate), 'weight_decay ': 10**-weight_decay}, \n",
    "])\n",
    "# optimizer = optim.SGD(model.parameters(), 10**-learning_rate, weight_decay=10**-weight_decay)\n",
    "\n",
    "# tensorboard = SummaryWriter(log_dir=\"runs/\"+start_time+\"_\"+prefix_filename+\"_\"+str(fingerprint_dim)+\"_\"+str(p_dropout))\n",
    "\n",
    "model_parameters = filter(lambda p: p.requires_grad, model.parameters())\n",
    "params = sum([np.prod(p.size()) for p in model_parameters])\n",
    "print(params)\n",
    "for name, param in model.named_parameters():\n",
    "    if param.requires_grad:\n",
    "        print(name, param.data.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from matplotlib import pyplot as plt\n",
    "\n",
    "def plot_output(active,inactive,epoch):\n",
    "    plt.figure(figsize=(24,8))\n",
    "    plt.subplot(1,2,1)\n",
    "    x_a = np.arange(0,len(active),1)\n",
    "    plt.plot(x_a,active,marker='.')\n",
    "    plt.ylabel('binary_active')\n",
    "    plt.title(\"epoch: {}, mean_value: {}\".format(epoch,np.mean(active)))\n",
    "    plt.subplot(1,2,2)\n",
    "    x_ina = np.arange(0,len(inactive),1)\n",
    "    plt.plot(x_ina,inactive,marker='.')\n",
    "    plt.ylabel('binary_inactive')\n",
    "    plt.title(\"epoch: {}, mean_value: {}\".format(epoch,np.mean(inactive)))\n",
    "    plt.show()\n",
    "    \n",
    "\n",
    "def sorted_show_pik(dataset, p, k, k_predict, i, acc):\n",
    "    p_value = dataset[tasks[0]].astype(float).tolist()\n",
    "    x = np.arange(0,len(dataset),1)\n",
    "#     print('plt',dataset.head(),p[:10],k_predict,k)\n",
    "#     plt.figure()\n",
    "#     fig, ax1 = plt.subplots()\n",
    "#     ax1.grid(False)\n",
    "#     ax2 = ax1.twinx()\n",
    "#     plt.grid(False)\n",
    "    plt.scatter(x,p,marker='.',s=6,color='r',label='predict')\n",
    "#     plt.ylabel('predict')\n",
    "    plt.scatter(x,p_value,s=6,marker=',',color='blue',label='p_value')\n",
    "    plt.axvline(x=k-1,ls=\"-\",c=\"black\")#添加垂直直线\n",
    "    k_value = np.ones(len(dataset))\n",
    "# #     print(EC50[k-1])\n",
    "    k_value = k_value*k_predict\n",
    "    plt.plot(x,k_value,'-',color='black')\n",
    "    plt.ylabel('p_value')\n",
    "    plt.title(\"epoch: {},  top-k recall: {}\".format(i,acc))\n",
    "    plt.legend(loc=3,fontsize=5)\n",
    "    plt.show()\n",
    "    \n",
    "\n",
    "\n",
    "def topk_acc2(df, predict, k, show_flag, i):\n",
    "    df['predict'] = predict\n",
    "    df2 = df.sort_values(by='predict',ascending=False) # 拼接预测值后对预测值进行排序\n",
    "#     print('df2:\\n',df2)\n",
    "        \n",
    "    df3 = df2[:k]  #取按预测值排完序后的前k个\n",
    "    \n",
    "    true_sort = df.sort_values(by=tasks[0],ascending=False) #返回一个新的按真实值排序列表\n",
    "    k_true = true_sort[tasks[0]].values[k-1]  # 真实排第k个的活性值\n",
    "#     print('df3:\\n',df3['predict'])\n",
    "#     print('k_true: ',type(k_true),k_true)\n",
    "#     print('k_true: ',k_true,'min_predict: ',df3['predict'].values[-1],'index: ',df3['predict'].values>=k_true,'acc_num: ',len(df3[df3['predict'].values>=k_true]),\n",
    "#           'fp_num: ',len(df3[df3['predict'].values>=-4.1]),'k: ',k)\n",
    "    acc = len(df3[df3[tasks[0]].values>=k_true])/k #预测值前k个中真实排在前k个的个数/k\n",
    "    fp = len(df3[df3[tasks[0]].values==-4.1])/k  #预测值前k个中为-4.1的个数/k\n",
    "    \n",
    "    if(show_flag):\n",
    "        #进来的是按实际活性值排好序的\n",
    "        sorted_show_pik(true_sort,true_sort['predict'],k,k_predict,i,acc)\n",
    "    return acc,fp\n",
    "\n",
    "def topk_recall(df, predict, k, active_num, show_flag, i):\n",
    "    df['predict'] = predict\n",
    "    df2 = df.sort_values(by='predict',ascending=False) # 拼接预测值后对预测值进行排序\n",
    "#     print('df2:\\n',df2)\n",
    "        \n",
    "    df3 = df2[:k]  #取按预测值排完序后的前active_num个，因为后面的全是-4.1\n",
    "    \n",
    "    true_sort = df.sort_values(by=tasks[0],ascending=False) #返回一个新的按真实值排序列表\n",
    "    min_active = true_sort[tasks[0]].values[test_active-1]  # 真实排第k个的活性值\n",
    "#     print('df3:\\n',df3['predict'])\n",
    "#     print('min_active: ',type(min_active),min_active)\n",
    "#     print('min_active: ',min_active,'min_predict: ',df3['predict'].values[-1],'index: ',df3['predict'].values>=min_active,'acc_num: ',len(df3[df3['predict'].values>=min_active]),\n",
    "#           'fp_num: ',len(df3[df3['predict'].values>=-4.1]),'k: ',k,'test_active: ',test_active)\n",
    "    acc = len(df3[df3[tasks[0]].values>=min_active])/test_active #预测值前k个中真实排在前test_active个的个数/test_active\n",
    "    fp = len(df3[df3[tasks[0]].values==-4.1])/k  #预测值前k个中为-4.1的个数/test_active\n",
    "    \n",
    "    if(show_flag):\n",
    "        #进来的是按实际活性值排好序的\n",
    "        sorted_show_pik(true_sort,true_sort['predict'],k,k_predict,i,acc)\n",
    "    return acc,fp\n",
    "\n",
    "    \n",
    "def topk_acc_recall(df, predict, k, active_num, show_flag, i):\n",
    "    if k>active_num:\n",
    "        return topk_recall(df, predict, k, active_num, show_flag, i)\n",
    "    return topk_acc2(df,predict,k,show_flag,i)\n",
    "# test_acc_df = pd.DataFrame(np.random.rand(20).reshape(10,2),index=range(0,10),columns=['p_value','smiles'])\n",
    "# # print(test_acc_df)\n",
    "# # df2 = test_acc_df.sort_values(by=tasks[0],ascending=False)\n",
    "# df2 = test_acc_df\n",
    "\n",
    "# # print('按活性值排序后：\\n',df2)\n",
    "# # print('原df：\\n',test_acc_df)\n",
    "# # print(df2.iloc[2,0])\n",
    "# # print(df2.loc[2])\n",
    "# pik = [0.6,0.8,0.5,0.2,0.5,0.7,0.2,0.8,0.9,0.2]\n",
    "# topk_acc2(df2,pik,5,True,1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "def caculate_r2(y,predict):\n",
    "#     print(y)\n",
    "#     print(predict)\n",
    "    y = torch.FloatTensor(y).reshape(-1,1)\n",
    "    predict = torch.FloatTensor(predict).reshape(-1,1)\n",
    "    y_mean = torch.mean(y)\n",
    "    predict_mean = torch.mean(predict)\n",
    "    \n",
    "    y1 = torch.pow(torch.mm((y-y_mean).t(),(predict-predict_mean)),2)\n",
    "    y2 = torch.mm((y-y_mean).t(),(y-y_mean))*torch.mm((predict-predict_mean).t(),(predict-predict_mean))\n",
    "#     print(y1,y2)\n",
    "    return y1/y2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "def topk_acc_weighted(df, predict):\n",
    "    df['predict'] = predict\n",
    "    df2 = df.sort_values(by='predict',ascending=False) # 拼接预测值后按预测值进行排序\n",
    "#     print('df2:\\n',df2)\n",
    "    weighted_acc_list = []\n",
    "    for k in np.arange(1,len(df),1):\n",
    "#         k = int(k*len(df))\n",
    "#         print(k)\n",
    "        df3 = df2[:k]  #按预测值进行排序后的前k个\n",
    "\n",
    "        val_sort = df.sort_values(by=tasks[0],ascending=False) #返回一个新的按真实值排序的列表\n",
    "        k_val = val_sort[tasks[0]].values[k-1] # 真实排在第k个的活性值\n",
    "    #     print('df3:\\n',df3['predict'])\n",
    "    #     print(type(k_predict),k_predict)\n",
    "    #     print('df3:\\n',df3)\n",
    "        acc = len(df3[df3[tasks[0]].values>=k_val])/k\n",
    "        weight = (len(df)-k)/len(df)\n",
    "#         print('weight=',weight,'acc=',acc)\n",
    "        weighted_acc_list.append(acc*weight)#\n",
    "    weighted_acc_list = np.array(weighted_acc_list)\n",
    "#     print('weighted_acc_list=',weighted_acc_list)\n",
    "    return np.sum(weighted_acc_list)/weighted_acc_list.shape[0]\n",
    "#     num = len(df[(df[tasks[0]].values==-4.1) & (df['predict'].values>=k_predict)])#预测值前k个中为-4.1的个数\n",
    "    \n",
    "#     if(show_flag):\n",
    "#         #进来的是按实际活性值排好序的\n",
    "#         sorted_show_pik(df2,df2['predict'],k,k_predict,i,acc)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "def plot_hist(test_binary_output_a,test_binary_output_ina,flag):\n",
    "    if not flag:\n",
    "        return\n",
    "    test_binary_output_a = np.array(test_binary_output_a)\n",
    "    test_binary_output_ina = np.array(test_binary_output_ina)\n",
    "    weights_a = np.ones_like(test_binary_output_a) / len(test_binary_output_a)\n",
    "    weights_ina = np.ones_like(test_binary_output_ina) / len(test_binary_output_ina)\n",
    "\n",
    "    ax = plt.subplot(1,2,1)\n",
    "    plt.hist(test_binary_output_a,bins=np.arange(0,1.01,0.1),weights=weights_a)\n",
    "    plt.xlim(0,1)\n",
    "    ax.set_title(\"test_binary_output_a\")\n",
    "\n",
    "    ax = plt.subplot(1,2,2)\n",
    "    # plt.hist(test_binary_output_ina,bins=20,density =True)\n",
    "    plt.hist(test_binary_output_ina,bins=np.arange(0,1.01,0.1),weights=weights_ina)\n",
    "    plt.xlim(0,1)\n",
    "    ax.set_title(\"test_binary_output_ina\")\n",
    "    plt.show()\n",
    "\n",
    "def caculate_binary_acc(test_binary_output_a,test_binary_output_ina,test_binary_output):\n",
    "#     test_binary_output_a = np.array(test_binary_output_a)\n",
    "#     test_binary_output_ina = np.array(test_binary_output_ina)\n",
    "    acc = (len(test_binary_output_a[test_binary_output_a>0.5])+len(test_binary_output_ina[test_binary_output_ina<0.5]))/len(test_binary_output)\n",
    "#     print(acc)\n",
    "#     print(len(test_binary_output_a),len(test_binary_output_ina),len(test_binary_output))\n",
    "    return acc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "def mol_with_atom_index( mol ):\n",
    "    atoms = mol.GetNumAtoms()\n",
    "    for idx in range( atoms ):\n",
    "        mol.GetAtomWithIdx( idx ).SetProp( 'molAtomMapNumber', str( mol.GetAtomWithIdx( idx ).GetIdx() ) )\n",
    "    return mol\n",
    "\n",
    "def min_max_norm(dataset):\n",
    "    if isinstance(dataset, list):\n",
    "        norm_list = list()\n",
    "        min_value = min(dataset)\n",
    "        max_value = max(dataset)\n",
    "\n",
    "        for value in dataset:\n",
    "            tmp = (value - min_value) / (max_value - min_value)\n",
    "            norm_list.append(tmp)\n",
    "    return norm_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "def eval(model, dataset, test_viz=False, k=50):\n",
    "    model.eval()\n",
    "    predict_list = []\n",
    "    test_MSE_list = []\n",
    "    r2_list = []\n",
    "    valList = np.arange(0,dataset.shape[0])\n",
    "    batch_list = []\n",
    "    feature_list = []\n",
    "    if test_viz:\n",
    "        false_positive_num = 0\n",
    "        dataset_predict = dataset.sort_values(by='predict',ascending=False) # 对预测值进行排序   \n",
    "        k_predict = dataset_predict['predict'].values[k-1]\n",
    "        true_sort = dataset.sort_values(by=tasks[0],ascending=False) #返回一个新的按真实值排序列表\n",
    "        k_true = true_sort[tasks[0]].values[k-1]\n",
    "#         print('k_predict =',k_predict,'k_true =',k_true)\n",
    "    \n",
    "# #     取dataset中排序后的第k个\n",
    "#     sorted_dataset = dataset.sort_values(by=tasks[0],ascending=False)\n",
    "#     k_df = sorted_dataset.iloc[[k-1]]\n",
    "#     k_smiles = k_df['cano_smiles'].values\n",
    "#     k_value = k_df[tasks[0]].values.astype(float)    \n",
    "    \n",
    "    for i in range(0, dataset.shape[0], batch_size):\n",
    "        batch = valList[i:i+batch_size]\n",
    "        batch_list.append(batch) \n",
    "#     print(batch_list)\n",
    "    for counter, batch in enumerate(batch_list):\n",
    "#         print(type(batch))\n",
    "        batch_df = dataset.loc[batch,:]\n",
    "        smiles_list = batch_df.cano_smiles.values\n",
    "#         print(batch_df)\n",
    "        y_val = batch_df[tasks[0]].values\n",
    "#         print(type(y_val))\n",
    "        \n",
    "    \n",
    "        x_atom, x_bonds, x_atom_index, x_bond_index, x_mask, smiles_to_rdkit_list = get_smiles_array(smiles_list,feature_dicts)\n",
    "        mol_feature, mol_prediction = model(torch.Tensor(x_atom),torch.Tensor(x_bonds),torch.cuda.LongTensor(x_atom_index),torch.cuda.LongTensor(x_bond_index),torch.Tensor(x_mask),output_feature=True)\n",
    "        atom_feature_viz, atom_attention_weight_viz, mol_feature_viz, mol_feature_unbounded_viz, \\\n",
    "        mol_attention_weight_viz, mol_prediction = model(torch.Tensor(x_atom),torch.Tensor(x_bonds),torch.cuda.LongTensor(x_atom_index),torch.cuda.LongTensor(x_bond_index),torch.Tensor(x_mask),viz=True)\n",
    "        mol_feature_output = mol_feature.cpu().detach().numpy()\n",
    "        feature_list.extend(mol_feature_output)\n",
    "        \n",
    "        MSE = F.mse_loss(mol_prediction, torch.Tensor(y_val).view(-1,1), reduction='none')\n",
    "#         r2 = caculate_r2(mol_prediction, torch.Tensor(y_val).view(-1,1))\n",
    "# #         r2_list.extend(r2.cpu().detach().numpy())\n",
    "#         if r2!=r2:\n",
    "#             r2 = torch.tensor(0)\n",
    "#         r2_list.append(r2.item())\n",
    "#         predict_list.extend(mol_prediction.cpu().detach().numpy())\n",
    "#         print(x_mask[:2],atoms_prediction.shape, mol_prediction,MSE)\n",
    "        predict_list.extend(mol_prediction.cpu().detach().numpy())\n",
    "#         test_MAE_list.extend(MAE.data.squeeze().cpu().numpy())\n",
    "        test_MSE_list.extend(MSE.data.view(-1,1).cpu().numpy())\n",
    "#     print(r2_list)\n",
    "        if test_viz:\n",
    "            out_feature_sorted = []\n",
    "            out_weight_sorted = []\n",
    "            mol_feature_sorted = []\n",
    "            mol_pred = np.array(mol_prediction.data.squeeze().cpu().numpy())\n",
    "            atom_feature = np.stack([atom_feature_viz[L].cpu().detach().numpy() for L in range(radius+1)])\n",
    "            atom_weight = np.stack([mol_attention_weight_viz[t].cpu().detach().numpy() for t in range(T)])\n",
    "            mol_feature = np.stack([mol_feature_viz[t].cpu().detach().numpy() for t in range(T)])\n",
    "            \n",
    "            mol_feature_sorted.extend([mol_feature[:,i,:] for i in range(mol_feature.shape[1])])\n",
    "            for i, smiles in enumerate(smiles_list):\n",
    "                atom_num = i\n",
    "                ind_mask = x_mask[i]\n",
    "                ind_atom = smiles_to_rdkit_list[smiles]\n",
    "                ind_feature = atom_feature[:, i]\n",
    "                ind_weight = atom_weight[:, i]\n",
    "                out_feature = []\n",
    "                out_weight = []\n",
    "                for j, one_or_zero in enumerate(list(ind_mask)):\n",
    "                    if one_or_zero == 1.0:\n",
    "                        out_feature.append(ind_feature[:,j])\n",
    "                        out_weight.append(ind_weight[:,j])\n",
    "                out_feature_sorted.extend([out_feature[m] for m in np.argsort(ind_atom)])\n",
    "                out_weight_sorted.extend([out_weight[m] for m in np.argsort(ind_atom)])        \n",
    "\n",
    "                mol = Chem.MolFromSmiles(smiles)\n",
    "\n",
    "                aromatic_boolean = [int(mol.GetAtomWithIdx(i).GetIsAromatic()) for i in range(mol.GetNumAtoms())]\n",
    "    #             if len(aromatic_boolean)>30 and np.sum(aromatic_boolean)>3:\n",
    "#                 if mol_pred[atom_num] >= k_predict and dataset[tasks[0]].values[counter*batch_size+atom_num] == -4.1:\n",
    "#                 if smiles == true_sort.cano_smiles.values[k-1]:\n",
    "                if dataset['smiles'].values[counter*batch_size+atom_num] == \"C1CC1N2C(=C3C=C(C=CC3=N2)C(C4=CC=C(C=C4)Cl)C5=CC=C(C=C5)Cl)C6CCN(CC6)S(=O)(=O)C7=CC=CC=C7C(=O)N\" \\\n",
    "    or dataset['smiles'].values[counter*batch_size+atom_num] ==\"C1CC1N2C(=C3C=C(C=CC3=N2)C(C4=CC=C(C=C4)Cl)C5=CC=C(C=C5)Cl)C6CCN(CC6)S(=O)(=O)C7=CC=CC=C7C(=O)O\":\n",
    "                    false_positive_num += 1\n",
    "                    weight_norm = min_max_norm([out_weight[m][0] for m in np.argsort(ind_atom)])\n",
    "\n",
    "                    print('init_smiles:',dataset['smiles'].values[counter*batch_size+atom_num]+'\\n'\n",
    "                          'cano_smiles: '+smiles+'\\n'\n",
    "                          'prediction: '+str(mol_pred[atom_num])+'\\n'\n",
    "                         'prediction_rank: '+str(len(dataset['predict'].values[dataset['predict'].values>mol_pred[atom_num]])+1)+'\\n'\n",
    "                         'prediction_k: '+str(k_predict)+'\\n'\n",
    "                         'ture_k: '+str(k_true)+'\\n'\n",
    "#                           'ture_rank: '+str(len(dataset[tasks[0]].values[dataset[tasks[0]].values>-2.79588])+1)+'\\n'\n",
    "                         'true_activity: '+str(dataset[tasks[0]].values[dataset.cano_smiles.values==smiles])+'\\n')\n",
    "                    norm = matplotlib.colors.Normalize(vmin=0,vmax=1.28)\n",
    "                    cmap = cm.get_cmap('Oranges')\n",
    "                    plt_colors = cm.ScalarMappable(norm=norm, cmap=cmap)\n",
    "                    atom_colors = {}\n",
    "                    weight_norm = np.array(weight_norm).flatten()\n",
    "                    threshold = weight_norm[np.argsort(weight_norm)[6]]\n",
    "                    weight_norm = np.where(weight_norm < threshold, 0, weight_norm)\n",
    "\n",
    "                    for i in range(len(ind_atom)):\n",
    "                        atom_colors[i] = plt_colors.to_rgba(float(weight_norm[i]))\n",
    "                    rdDepictor.Compute2DCoords(mol)\n",
    "\n",
    "                    drawer = rdMolDraw2D.MolDraw2DSVG(800,400)\n",
    "                    drawer.SetFontSize(1)\n",
    "                    op = drawer.drawOptions()\n",
    "\n",
    "                    mol = rdMolDraw2D.PrepareMolForDrawing(mol)\n",
    "                    drawer.DrawMolecule(mol,highlightAtoms=range(0,len(ind_atom)),highlightBonds=[],\n",
    "                        highlightAtomColors=atom_colors)\n",
    "                    drawer.FinishDrawing()\n",
    "                    svg = drawer.GetDrawingText()\n",
    "                    svg2 = svg.replace('svg:','')\n",
    "                    svg3 = SVG(svg2)\n",
    "                    display(svg3)\n",
    "    if test_viz:\n",
    "        print('false_positive_num =',false_positive_num) \n",
    "    \n",
    "    return feature_list,caculate_r2(predict_list,dataset[tasks[0]].values.astype(float).tolist()),np.array(test_MSE_list).mean(),predict_list\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "init_smiles: C1CC1N2C(=C3C=C(C=CC3=N2)C(C4=CC=C(C=C4)Cl)C5=CC=C(C=C5)Cl)C6CCN(CC6)S(=O)(=O)C7=CC=CC=C7C(=O)N\n",
      "cano_smiles: NC(=O)c1ccccc1S(=O)(=O)N1CCC(c2c3cc(C(c4ccc(Cl)cc4)c4ccc(Cl)cc4)ccc3nn2C2CC2)CC1\n",
      "prediction: -1.1775482\n",
      "prediction_rank: 30\n",
      "prediction_k: [-0.08818804]\n",
      "ture_k: 0.3010299956639812\n",
      "true_activity: [-2.30535137]\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "RDKit WARNING: [19:13:15] The new font size 1 is below the current minimum (6).\n"
     ]
    },
    {
     "data": {
      "image/svg+xml": [
       "<svg baseProfile=\"full\" height=\"400px\" version=\"1.1\" viewBox=\"0 0 800 400\" width=\"800px\" xml:space=\"preserve\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:rdkit=\"http://www.rdkit.org/xml\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
       "<!-- END OF HEADER -->\n",
       "<rect height=\"400\" style=\"opacity:1.0;fill:#FFFFFF;stroke:none\" width=\"800\" x=\"0\" y=\"0\"> </rect>\n",
       "<ellipse cx=\"252.326\" cy=\"355.067\" rx=\"18.3942\" ry=\"12.8195\" style=\"fill:#FDA25A;fill-rule:evenodd;stroke:#FDA25A;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"243.763\" cy=\"309.368\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDBE85;fill-rule:evenodd;stroke:#FDBE85;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"280.654\" cy=\"288.002\" rx=\"8.53558\" ry=\"8.77571\" style=\"fill:#FC8C3B;fill-rule:evenodd;stroke:#FC8C3B;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"206.734\" cy=\"288.15\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDA55E;fill-rule:evenodd;stroke:#FDA55E;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"169.843\" cy=\"309.609\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDA863;fill-rule:evenodd;stroke:#FDA863;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"132.814\" cy=\"288.391\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDAC68;fill-rule:evenodd;stroke:#FDAC68;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"132.675\" cy=\"245.713\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDAC68;fill-rule:evenodd;stroke:#FDAC68;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"169.565\" cy=\"224.254\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDA761;fill-rule:evenodd;stroke:#FDA761;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"206.595\" cy=\"245.472\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDA761;fill-rule:evenodd;stroke:#FDA761;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"243.485\" cy=\"224.175\" rx=\"8.53558\" ry=\"8.75157\" style=\"fill:#F67A26;fill-rule:evenodd;stroke:#F67A26;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"264.944\" cy=\"260.997\" rx=\"8.53558\" ry=\"8.77571\" style=\"fill:#FD8E3E;fill-rule:evenodd;stroke:#FD8E3E;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"222.026\" cy=\"187.216\" rx=\"8.53558\" ry=\"8.77571\" style=\"fill:#FD8E3E;fill-rule:evenodd;stroke:#FD8E3E;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"280.375\" cy=\"202.553\" rx=\"8.53558\" ry=\"8.54636\" style=\"fill:#FDA761;fill-rule:evenodd;stroke:#FDA761;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"317.405\" cy=\"223.772\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FFF5EB;fill-rule:evenodd;stroke:#FFF5EB;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"354.295\" cy=\"202.312\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FFF5EB;fill-rule:evenodd;stroke:#FFF5EB;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"354.156\" cy=\"159.635\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FC8A38;fill-rule:evenodd;stroke:#FC8A38;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"391.047\" cy=\"138.175\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FD9D52;fill-rule:evenodd;stroke:#FD9D52;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"430.091\" cy=\"155.407\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FD9C51;fill-rule:evenodd;stroke:#FD9C51;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"443.412\" cy=\"195.953\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FD9F55;fill-rule:evenodd;stroke:#FD9F55;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"485.186\" cy=\"204.69\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDA761;fill-rule:evenodd;stroke:#FDA761;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"498.506\" cy=\"245.236\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FD9C51;fill-rule:evenodd;stroke:#FD9C51;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"470.052\" cy=\"277.044\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDA660;fill-rule:evenodd;stroke:#FDA660;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"428.279\" cy=\"268.307\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDA55E;fill-rule:evenodd;stroke:#FDA55E;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"399.825\" cy=\"300.116\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDA863;fill-rule:evenodd;stroke:#FDA863;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"413.145\" cy=\"340.662\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDA660;fill-rule:evenodd;stroke:#FDA660;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"379.297\" cy=\"372.548\" rx=\"8.7395\" ry=\"9.27063\" style=\"fill:#FD9F55;fill-rule:evenodd;stroke:#FD9F55;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"454.919\" cy=\"349.399\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDA863;fill-rule:evenodd;stroke:#FDA863;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"483.373\" cy=\"317.59\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDA55E;fill-rule:evenodd;stroke:#FDA55E;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"540.28\" cy=\"253.973\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDA660;fill-rule:evenodd;stroke:#FDA660;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"553.6\" cy=\"294.519\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDA55E;fill-rule:evenodd;stroke:#FDA55E;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"595.374\" cy=\"303.256\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDA863;fill-rule:evenodd;stroke:#FDA863;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"623.828\" cy=\"271.447\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDA660;fill-rule:evenodd;stroke:#FDA660;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"667.121\" cy=\"280.261\" rx=\"8.7395\" ry=\"9.27063\" style=\"fill:#FD9F55;fill-rule:evenodd;stroke:#FD9F55;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"610.508\" cy=\"230.901\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDA863;fill-rule:evenodd;stroke:#FDA863;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"568.734\" cy=\"222.164\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDA55E;fill-rule:evenodd;stroke:#FDA55E;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"513.639\" cy=\"172.881\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDA55E;fill-rule:evenodd;stroke:#FDA55E;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"500.319\" cy=\"132.335\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDA964;fill-rule:evenodd;stroke:#FDA964;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"458.545\" cy=\"123.598\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FD9244;fill-rule:evenodd;stroke:#FD9244;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"437.085\" cy=\"86.7075\" rx=\"8.53558\" ry=\"8.54636\" style=\"fill:#CB4301;fill-rule:evenodd;stroke:#CB4301;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"395.369\" cy=\"95.7168\" rx=\"8.53558\" ry=\"8.54636\" style=\"fill:#DE4F05;fill-rule:evenodd;stroke:#DE4F05;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"363.56\" cy=\"67.2633\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FD8D3C;fill-rule:evenodd;stroke:#FD8D3C;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"321.786\" cy=\"58.5262\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FFF5EB;fill-rule:evenodd;stroke:#FFF5EB;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"350.24\" cy=\"26.7174\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FFF5EB;fill-rule:evenodd;stroke:#FFF5EB;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"317.127\" cy=\"138.416\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FFF5EB;fill-rule:evenodd;stroke:#FFF5EB;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"280.236\" cy=\"159.876\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FFF5EB;fill-rule:evenodd;stroke:#FFF5EB;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-0 atom-0 atom-1\" d=\"M 243.879,344.935 L 243.821,327.152\" style=\"fill:none;fill-rule:evenodd;stroke:#0000FF;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-0 atom-0 atom-1\" d=\"M 243.821,327.152 L 243.763,309.368\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-1 atom-1 atom-2\" d=\"M 245.909,313.057 L 261.047,304.251\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-1 atom-1 atom-2\" d=\"M 261.047,304.251 L 276.185,295.446\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-1 atom-1 atom-2\" d=\"M 241.617,305.679 L 256.755,296.873\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-1 atom-1 atom-2\" d=\"M 256.755,296.873 L 271.893,288.068\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-2 atom-1 atom-3\" d=\"M 243.763,309.368 L 206.734,288.15\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-3 atom-3 atom-4\" d=\"M 206.734,288.15 L 169.843,309.609\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-3 atom-3 atom-4\" d=\"M 196.908,283.99 L 171.085,299.012\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-44 atom-8 atom-3\" d=\"M 206.595,245.472 L 206.734,288.15\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-4 atom-4 atom-5\" d=\"M 169.843,309.609 L 132.814,288.391\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-5 atom-5 atom-6\" d=\"M 132.814,288.391 L 132.675,245.713\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-5 atom-5 atom-6\" d=\"M 141.329,281.961 L 141.231,252.087\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-6 atom-6 atom-7\" d=\"M 132.675,245.713 L 169.565,224.254\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-7 atom-7 atom-8\" d=\"M 169.565,224.254 L 206.595,245.472\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-7 atom-7 atom-8\" d=\"M 170.876,234.842 L 196.797,249.695\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-8 atom-8 atom-9\" d=\"M 206.595,245.472 L 222.569,236.18\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-8 atom-8 atom-9\" d=\"M 222.569,236.18 L 238.543,226.887\" style=\"fill:none;fill-rule:evenodd;stroke:#CCCC00;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-9 atom-9 atom-10\" d=\"M 244.111,233.576 L 250.658,244.83\" style=\"fill:none;fill-rule:evenodd;stroke:#CCCC00;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-9 atom-9 atom-10\" d=\"M 250.658,244.83 L 257.204,256.085\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-9 atom-9 atom-10\" d=\"M 251.489,229.284 L 258.036,240.539\" style=\"fill:none;fill-rule:evenodd;stroke:#CCCC00;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-9 atom-9 atom-10\" d=\"M 258.036,240.539 L 264.582,251.793\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-10 atom-9 atom-11\" d=\"M 243.048,214.774 L 236.464,203.455\" style=\"fill:none;fill-rule:evenodd;stroke:#CCCC00;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-10 atom-9 atom-11\" d=\"M 236.464,203.455 L 229.88,192.137\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-10 atom-9 atom-11\" d=\"M 235.67,219.065 L 229.086,207.747\" style=\"fill:none;fill-rule:evenodd;stroke:#CCCC00;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-10 atom-9 atom-11\" d=\"M 229.086,207.747 L 222.502,196.429\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-11 atom-9 atom-12\" d=\"M 248.427,221.138 L 261.647,213.447\" style=\"fill:none;fill-rule:evenodd;stroke:#CCCC00;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-11 atom-9 atom-12\" d=\"M 261.647,213.447 L 274.868,205.757\" style=\"fill:none;fill-rule:evenodd;stroke:#0000FF;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-12 atom-12 atom-13\" d=\"M 286.026,205.791 L 301.715,214.781\" style=\"fill:none;fill-rule:evenodd;stroke:#0000FF;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-12 atom-12 atom-13\" d=\"M 301.715,214.781 L 317.405,223.772\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-45 atom-44 atom-12\" d=\"M 280.236,159.876 L 280.294,177.659\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-45 atom-44 atom-12\" d=\"M 280.294,177.659 L 280.352,195.443\" style=\"fill:none;fill-rule:evenodd;stroke:#0000FF;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-13 atom-13 atom-14\" d=\"M 317.405,223.772 L 354.295,202.312\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-14 atom-14 atom-15\" d=\"M 354.295,202.312 L 354.156,159.635\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-15 atom-15 atom-16\" d=\"M 354.156,159.635 L 391.047,138.175\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-42 atom-15 atom-43\" d=\"M 354.156,159.635 L 317.127,138.416\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-16 atom-16 atom-17\" d=\"M 391.047,138.175 L 430.091,155.407\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-16 atom-16 atom-17\" d=\"M 400.35,132.951 L 427.681,145.013\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-46 atom-39 atom-16\" d=\"M 394.645,102.827 L 392.846,120.501\" style=\"fill:none;fill-rule:evenodd;stroke:#0000FF;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-46 atom-39 atom-16\" d=\"M 392.846,120.501 L 391.047,138.175\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-17 atom-17 atom-18\" d=\"M 430.091,155.407 L 443.412,195.953\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-48 atom-37 atom-17\" d=\"M 458.545,123.598 L 430.091,155.407\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-18 atom-18 atom-19\" d=\"M 443.412,195.953 L 485.186,204.69\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-18 atom-18 atom-19\" d=\"M 451.425,188.908 L 480.667,195.024\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-19 atom-19 atom-20\" d=\"M 485.186,204.69 L 498.506,245.236\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-34 atom-19 atom-35\" d=\"M 485.186,204.69 L 513.639,172.881\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-20 atom-20 atom-21\" d=\"M 498.506,245.236 L 470.052,277.044\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-27 atom-20 atom-28\" d=\"M 498.506,245.236 L 540.28,253.973\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-21 atom-21 atom-22\" d=\"M 470.052,277.044 L 428.279,268.307\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-21 atom-21 atom-22\" d=\"M 462.039,284.089 L 432.797,277.973\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-49 atom-27 atom-21\" d=\"M 483.373,317.59 L 470.052,277.044\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-22 atom-22 atom-23\" d=\"M 428.279,268.307 L 399.825,300.116\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-23 atom-23 atom-24\" d=\"M 399.825,300.116 L 413.145,340.662\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-23 atom-23 atom-24\" d=\"M 409.932,303.534 L 419.257,331.916\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-24 atom-24 atom-25\" d=\"M 413.145,340.662 L 399.845,355.531\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-24 atom-24 atom-25\" d=\"M 399.845,355.531 L 386.544,370.4\" style=\"fill:none;fill-rule:evenodd;stroke:#00CC00;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-25 atom-24 atom-26\" d=\"M 413.145,340.662 L 454.919,349.399\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-26 atom-26 atom-27\" d=\"M 454.919,349.399 L 483.373,317.59\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-26 atom-26 atom-27\" d=\"M 452.826,338.937 L 472.743,316.671\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-28 atom-28 atom-29\" d=\"M 540.28,253.973 L 553.6,294.519\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-28 atom-28 atom-29\" d=\"M 550.387,257.39 L 559.712,285.773\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-50 atom-34 atom-28\" d=\"M 568.734,222.164 L 540.28,253.973\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-29 atom-29 atom-30\" d=\"M 553.6,294.519 L 595.374,303.256\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-30 atom-30 atom-31\" d=\"M 595.374,303.256 L 623.828,271.447\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-30 atom-30 atom-31\" d=\"M 593.281,292.794 L 613.198,270.527\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-31 atom-31 atom-32\" d=\"M 623.828,271.447 L 641.851,275.216\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-31 atom-31 atom-32\" d=\"M 641.851,275.216 L 659.875,278.986\" style=\"fill:none;fill-rule:evenodd;stroke:#00CC00;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-32 atom-31 atom-33\" d=\"M 623.828,271.447 L 610.508,230.901\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-33 atom-33 atom-34\" d=\"M 610.508,230.901 L 568.734,222.164\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-33 atom-33 atom-34\" d=\"M 602.494,237.945 L 573.252,231.829\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-35 atom-35 atom-36\" d=\"M 513.639,172.881 L 500.319,132.335\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-35 atom-35 atom-36\" d=\"M 503.532,169.463 L 494.208,141.081\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-36 atom-36 atom-37\" d=\"M 500.319,132.335 L 458.545,123.598\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-37 atom-37 atom-38\" d=\"M 458.545,123.598 L 449.883,108.708\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-37 atom-37 atom-38\" d=\"M 449.883,108.708 L 441.221,93.8177\" style=\"fill:none;fill-rule:evenodd;stroke:#0000FF;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-37 atom-37 atom-38\" d=\"M 448.568,123.423 L 442.505,113\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-37 atom-37 atom-38\" d=\"M 442.505,113 L 436.442,102.577\" style=\"fill:none;fill-rule:evenodd;stroke:#0000FF;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-38 atom-38 atom-39\" d=\"M 431.435,87.9278 L 400.875,94.5277\" style=\"fill:none;fill-rule:evenodd;stroke:#0000FF;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-39 atom-39 atom-40\" d=\"M 389.719,90.6623 L 376.639,78.9628\" style=\"fill:none;fill-rule:evenodd;stroke:#0000FF;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-39 atom-39 atom-40\" d=\"M 376.639,78.9628 L 363.56,67.2633\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-40 atom-40 atom-41\" d=\"M 363.56,67.2633 L 321.786,58.5262\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-47 atom-42 atom-40\" d=\"M 350.24,26.7174 L 363.56,67.2633\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-41 atom-41 atom-42\" d=\"M 321.786,58.5262 L 350.24,26.7174\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-43 atom-43 atom-44\" d=\"M 317.127,138.416 L 280.236,159.876\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"atom-0\" d=\"M 241.231 346.002 L 245.191 352.404 Q 245.584 353.036, 246.216 354.18 Q 246.847 355.323, 246.881 355.392 L 246.881 346.002 L 248.486 346.002 L 248.486 358.089 L 246.83 358.089 L 242.579 351.09 Q 242.084 350.27, 241.555 349.331 Q 241.043 348.392, 240.889 348.102 L 240.889 358.089 L 239.319 358.089 L 239.319 346.002 L 241.231 346.002 \" fill=\"#0000FF\"/>\n",
       "<path class=\"atom-0\" d=\"M 249.937 346.002 L 251.576 346.002 L 251.576 351.141 L 257.756 351.141 L 257.756 346.002 L 259.395 346.002 L 259.395 358.089 L 257.756 358.089 L 257.756 352.507 L 251.576 352.507 L 251.576 358.089 L 249.937 358.089 L 249.937 346.002 \" fill=\"#0000FF\"/>\n",
       "<path class=\"atom-0\" d=\"M 259.98 357.665 Q 260.273 356.91, 260.972 356.493 Q 261.67 356.065, 262.639 356.065 Q 263.845 356.065, 264.521 356.718 Q 265.197 357.372, 265.197 358.532 Q 265.197 359.715, 264.318 360.82 Q 263.451 361.924, 261.648 363.231 L 265.332 363.231 L 265.332 364.132 L 259.958 364.132 L 259.958 363.377 Q 261.445 362.318, 262.324 361.529 Q 263.214 360.741, 263.642 360.031 Q 264.07 359.321, 264.07 358.589 Q 264.07 357.822, 263.687 357.394 Q 263.304 356.966, 262.639 356.966 Q 261.997 356.966, 261.569 357.225 Q 261.141 357.484, 260.837 358.059 L 259.98 357.665 \" fill=\"#0000FF\"/>\n",
       "<path class=\"atom-2\" d=\"M 275.106 287.943 Q 275.106 285.041, 276.54 283.419 Q 277.974 281.797, 280.654 281.797 Q 283.334 281.797, 284.768 283.419 Q 286.202 285.041, 286.202 287.943 Q 286.202 290.879, 284.751 292.552 Q 283.3 294.208, 280.654 294.208 Q 277.991 294.208, 276.54 292.552 Q 275.106 290.896, 275.106 287.943 M 280.654 292.842 Q 282.497 292.842, 283.487 291.613 Q 284.495 290.367, 284.495 287.943 Q 284.495 285.57, 283.487 284.375 Q 282.497 283.163, 280.654 283.163 Q 278.81 283.163, 277.803 284.358 Q 276.813 285.553, 276.813 287.943 Q 276.813 290.384, 277.803 291.613 Q 278.81 292.842, 280.654 292.842 \" fill=\"#FF0000\"/>\n",
       "<path class=\"atom-9\" d=\"M 240.071 228.161 Q 240.207 228.212, 240.771 228.451 Q 241.334 228.69, 241.949 228.844 Q 242.58 228.98, 243.195 228.98 Q 244.339 228.98, 245.004 228.434 Q 245.67 227.871, 245.67 226.898 Q 245.67 226.232, 245.329 225.822 Q 245.004 225.412, 244.492 225.191 Q 243.98 224.969, 243.127 224.713 Q 242.051 224.388, 241.402 224.081 Q 240.771 223.774, 240.31 223.125 Q 239.866 222.476, 239.866 221.384 Q 239.866 219.864, 240.89 218.925 Q 241.932 217.986, 243.98 217.986 Q 245.38 217.986, 246.968 218.652 L 246.575 219.967 Q 245.124 219.369, 244.031 219.369 Q 242.853 219.369, 242.205 219.864 Q 241.556 220.342, 241.573 221.179 Q 241.573 221.827, 241.897 222.22 Q 242.239 222.613, 242.717 222.835 Q 243.212 223.057, 244.031 223.313 Q 245.124 223.654, 245.773 223.996 Q 246.421 224.337, 246.882 225.037 Q 247.36 225.72, 247.36 226.898 Q 247.36 228.571, 246.233 229.475 Q 245.124 230.363, 243.263 230.363 Q 242.188 230.363, 241.368 230.124 Q 240.566 229.902, 239.61 229.51 L 240.071 228.161 \" fill=\"#CCCC00\"/>\n",
       "<path class=\"atom-10\" d=\"M 259.396 260.937 Q 259.396 258.035, 260.83 256.413 Q 262.264 254.791, 264.944 254.791 Q 267.625 254.791, 269.059 256.413 Q 270.492 258.035, 270.492 260.937 Q 270.492 263.873, 269.041 265.546 Q 267.59 267.202, 264.944 267.202 Q 262.281 267.202, 260.83 265.546 Q 259.396 263.89, 259.396 260.937 M 264.944 265.837 Q 266.788 265.837, 267.778 264.607 Q 268.785 263.361, 268.785 260.937 Q 268.785 258.564, 267.778 257.369 Q 266.788 256.157, 264.944 256.157 Q 263.101 256.157, 262.093 257.352 Q 261.103 258.547, 261.103 260.937 Q 261.103 263.378, 262.093 264.607 Q 263.101 265.837, 264.944 265.837 \" fill=\"#FF0000\"/>\n",
       "<path class=\"atom-11\" d=\"M 216.478 187.156 Q 216.478 184.254, 217.912 182.633 Q 219.346 181.011, 222.026 181.011 Q 224.706 181.011, 226.14 182.633 Q 227.574 184.254, 227.574 187.156 Q 227.574 190.093, 226.123 191.766 Q 224.672 193.421, 222.026 193.421 Q 219.363 193.421, 217.912 191.766 Q 216.478 190.11, 216.478 187.156 M 222.026 192.056 Q 223.869 192.056, 224.859 190.827 Q 225.867 189.58, 225.867 187.156 Q 225.867 184.783, 224.859 183.588 Q 223.869 182.376, 222.026 182.376 Q 220.182 182.376, 219.175 183.571 Q 218.185 184.766, 218.185 187.156 Q 218.185 189.598, 219.175 190.827 Q 220.182 192.056, 222.026 192.056 \" fill=\"#FF0000\"/>\n",
       "<path class=\"atom-12\" d=\"M 277.704 196.51 L 281.664 202.912 Q 282.057 203.543, 282.689 204.687 Q 283.32 205.831, 283.354 205.899 L 283.354 196.51 L 284.959 196.51 L 284.959 208.596 L 283.303 208.596 L 279.052 201.597 Q 278.557 200.778, 278.028 199.839 Q 277.516 198.9, 277.362 198.61 L 277.362 208.596 L 275.792 208.596 L 275.792 196.51 L 277.704 196.51 \" fill=\"#0000FF\"/>\n",
       "<path class=\"atom-25\" d=\"M 373.118 372.889 Q 373.118 369.884, 374.517 368.314 Q 375.934 366.726, 378.615 366.726 Q 381.107 366.726, 382.439 368.485 L 381.312 369.406 Q 380.339 368.126, 378.615 368.126 Q 376.788 368.126, 375.815 369.355 Q 374.859 370.567, 374.859 372.889 Q 374.859 375.279, 375.849 376.508 Q 376.856 377.737, 378.802 377.737 Q 380.134 377.737, 381.687 376.935 L 382.165 378.215 Q 381.534 378.625, 380.578 378.864 Q 379.622 379.103, 378.563 379.103 Q 375.934 379.103, 374.517 377.498 Q 373.118 375.893, 373.118 372.889 \" fill=\"#00CC00\"/>\n",
       "<path class=\"atom-25\" d=\"M 383.907 365.992 L 385.477 365.992 L 385.477 378.949 L 383.907 378.949 L 383.907 365.992 \" fill=\"#00CC00\"/>\n",
       "<path class=\"atom-32\" d=\"M 660.942 280.602 Q 660.942 277.598, 662.341 276.027 Q 663.758 274.439, 666.438 274.439 Q 668.931 274.439, 670.262 276.198 L 669.136 277.12 Q 668.163 275.839, 666.438 275.839 Q 664.612 275.839, 663.639 277.068 Q 662.683 278.28, 662.683 280.602 Q 662.683 282.992, 663.673 284.221 Q 664.68 285.45, 666.626 285.45 Q 667.958 285.45, 669.511 284.648 L 669.989 285.928 Q 669.358 286.338, 668.402 286.577 Q 667.446 286.816, 666.387 286.816 Q 663.758 286.816, 662.341 285.211 Q 660.942 283.607, 660.942 280.602 \" fill=\"#00CC00\"/>\n",
       "<path class=\"atom-32\" d=\"M 671.731 273.705 L 673.301 273.705 L 673.301 286.662 L 671.731 286.662 L 671.731 273.705 \" fill=\"#00CC00\"/>\n",
       "<path class=\"atom-38\" d=\"M 434.414 80.6643 L 438.374 87.066 Q 438.767 87.6976, 439.398 88.8414 Q 440.03 89.9852, 440.064 90.0535 L 440.064 80.6643 L 441.669 80.6643 L 441.669 92.7507 L 440.013 92.7507 L 435.762 85.7515 Q 435.267 84.9321, 434.738 83.9932 Q 434.226 83.0543, 434.072 82.7641 L 434.072 92.7507 L 432.502 92.7507 L 432.502 80.6643 L 434.414 80.6643 \" fill=\"#0000FF\"/>\n",
       "<path class=\"atom-39\" d=\"M 392.698 89.6736 L 396.658 96.0753 Q 397.051 96.7069, 397.682 97.8507 Q 398.314 98.9945, 398.348 99.0628 L 398.348 89.6736 L 399.953 89.6736 L 399.953 101.76 L 398.297 101.76 L 394.046 94.7608 Q 393.551 93.9414, 393.022 93.0025 Q 392.51 92.0636, 392.356 91.7734 L 392.356 101.76 L 390.786 101.76 L 390.786 89.6736 L 392.698 89.6736 \" fill=\"#0000FF\"/>\n",
       "</svg>"
      ],
      "text/plain": [
       "<IPython.core.display.SVG object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "init_smiles: C1CC1N2C(=C3C=C(C=CC3=N2)C(C4=CC=C(C=C4)Cl)C5=CC=C(C=C5)Cl)C6CCN(CC6)S(=O)(=O)C7=CC=CC=C7C(=O)O\n",
      "cano_smiles: O=C(O)c1ccccc1S(=O)(=O)N1CCC(c2c3cc(C(c4ccc(Cl)cc4)c4ccc(Cl)cc4)ccc3nn2C2CC2)CC1\n",
      "prediction: -1.615232\n",
      "prediction_rank: 80\n",
      "prediction_k: [-0.08818804]\n",
      "ture_k: 0.3010299956639812\n",
      "true_activity: [-3.76342799]\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "RDKit WARNING: [19:13:15] The new font size 1 is below the current minimum (6).\n"
     ]
    },
    {
     "data": {
      "image/svg+xml": [
       "<svg baseProfile=\"full\" height=\"400px\" version=\"1.1\" viewBox=\"0 0 800 400\" width=\"800px\" xml:space=\"preserve\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:rdkit=\"http://www.rdkit.org/xml\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
       "<!-- END OF HEADER -->\n",
       "<rect height=\"400\" style=\"opacity:1.0;fill:#FFFFFF;stroke:none\" width=\"800\" x=\"0\" y=\"0\"> </rect>\n",
       "<ellipse cx=\"280.654\" cy=\"288.002\" rx=\"8.53558\" ry=\"8.77571\" style=\"fill:#F98331;fill-rule:evenodd;stroke:#F98331;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"243.763\" cy=\"309.368\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDAA66;fill-rule:evenodd;stroke:#FDAA66;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"249.357\" cy=\"352.14\" rx=\"15.5597\" ry=\"8.77571\" style=\"fill:#CB4301;fill-rule:evenodd;stroke:#CB4301;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"206.734\" cy=\"288.15\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDA761;fill-rule:evenodd;stroke:#FDA761;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"169.843\" cy=\"309.609\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDB06F;fill-rule:evenodd;stroke:#FDB06F;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"132.814\" cy=\"288.391\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDB474;fill-rule:evenodd;stroke:#FDB474;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"132.675\" cy=\"245.713\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDB474;fill-rule:evenodd;stroke:#FDB474;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"169.565\" cy=\"224.254\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDAF6D;fill-rule:evenodd;stroke:#FDAF6D;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"206.595\" cy=\"245.472\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDAE6C;fill-rule:evenodd;stroke:#FDAE6C;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"243.485\" cy=\"224.175\" rx=\"8.53558\" ry=\"8.75157\" style=\"fill:#FA8533;fill-rule:evenodd;stroke:#FA8533;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"264.944\" cy=\"260.997\" rx=\"8.53558\" ry=\"8.77571\" style=\"fill:#FD974B;fill-rule:evenodd;stroke:#FD974B;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"222.026\" cy=\"187.216\" rx=\"8.53558\" ry=\"8.77571\" style=\"fill:#FD974B;fill-rule:evenodd;stroke:#FD974B;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"280.375\" cy=\"202.553\" rx=\"8.53558\" ry=\"8.54636\" style=\"fill:#FDAE6C;fill-rule:evenodd;stroke:#FDAE6C;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"317.405\" cy=\"223.772\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FFF5EB;fill-rule:evenodd;stroke:#FFF5EB;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"354.295\" cy=\"202.312\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FFF5EB;fill-rule:evenodd;stroke:#FFF5EB;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"354.156\" cy=\"159.635\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FD9345;fill-rule:evenodd;stroke:#FD9345;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"391.047\" cy=\"138.175\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDA55E;fill-rule:evenodd;stroke:#FDA55E;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"430.091\" cy=\"155.407\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDA45D;fill-rule:evenodd;stroke:#FDA45D;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"443.412\" cy=\"195.953\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDA761;fill-rule:evenodd;stroke:#FDA761;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"485.186\" cy=\"204.69\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDAE6C;fill-rule:evenodd;stroke:#FDAE6C;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"498.506\" cy=\"245.236\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDA45D;fill-rule:evenodd;stroke:#FDA45D;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"470.052\" cy=\"277.044\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDAD6A;fill-rule:evenodd;stroke:#FDAD6A;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"428.279\" cy=\"268.307\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDAC68;fill-rule:evenodd;stroke:#FDAC68;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"399.825\" cy=\"300.116\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDAF6D;fill-rule:evenodd;stroke:#FDAF6D;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"413.145\" cy=\"340.662\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDAE6C;fill-rule:evenodd;stroke:#FDAE6C;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"379.297\" cy=\"372.548\" rx=\"8.7395\" ry=\"9.27063\" style=\"fill:#FDA761;fill-rule:evenodd;stroke:#FDA761;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"454.919\" cy=\"349.399\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDAF6D;fill-rule:evenodd;stroke:#FDAF6D;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"483.373\" cy=\"317.59\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDAC68;fill-rule:evenodd;stroke:#FDAC68;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"540.28\" cy=\"253.973\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDAD6A;fill-rule:evenodd;stroke:#FDAD6A;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"553.6\" cy=\"294.519\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDAC68;fill-rule:evenodd;stroke:#FDAC68;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"595.374\" cy=\"303.256\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDAF6D;fill-rule:evenodd;stroke:#FDAF6D;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"623.828\" cy=\"271.447\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDAE6C;fill-rule:evenodd;stroke:#FDAE6C;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"667.121\" cy=\"280.261\" rx=\"8.7395\" ry=\"9.27063\" style=\"fill:#FDA761;fill-rule:evenodd;stroke:#FDA761;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"610.508\" cy=\"230.901\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDAF6D;fill-rule:evenodd;stroke:#FDAF6D;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"568.734\" cy=\"222.164\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDAC68;fill-rule:evenodd;stroke:#FDAC68;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"513.639\" cy=\"172.881\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDAC68;fill-rule:evenodd;stroke:#FDAC68;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"500.319\" cy=\"132.335\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FDB171;fill-rule:evenodd;stroke:#FDB171;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"458.545\" cy=\"123.598\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FD9A4F;fill-rule:evenodd;stroke:#FD9A4F;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"437.085\" cy=\"86.7075\" rx=\"8.53558\" ry=\"8.54636\" style=\"fill:#DD4E04;fill-rule:evenodd;stroke:#DD4E04;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"395.369\" cy=\"95.7168\" rx=\"8.53558\" ry=\"8.54636\" style=\"fill:#E85C0C;fill-rule:evenodd;stroke:#E85C0C;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"363.56\" cy=\"67.2633\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FD964A;fill-rule:evenodd;stroke:#FD964A;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"321.786\" cy=\"58.5262\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FFF5EB;fill-rule:evenodd;stroke:#FFF5EB;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"350.24\" cy=\"26.7174\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FFF5EB;fill-rule:evenodd;stroke:#FFF5EB;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"317.127\" cy=\"138.416\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FFF5EB;fill-rule:evenodd;stroke:#FFF5EB;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<ellipse cx=\"280.236\" cy=\"159.876\" rx=\"8.53558\" ry=\"8.53558\" style=\"fill:#FFF5EB;fill-rule:evenodd;stroke:#FFF5EB;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-0 atom-0 atom-1\" d=\"M 271.893,288.068 L 256.755,296.873\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-0 atom-0 atom-1\" d=\"M 256.755,296.873 L 241.617,305.679\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-0 atom-0 atom-1\" d=\"M 276.185,295.446 L 261.047,304.251\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-0 atom-0 atom-1\" d=\"M 261.047,304.251 L 245.909,313.057\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-1 atom-1 atom-2\" d=\"M 243.763,309.368 L 243.821,327.118\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-1 atom-1 atom-2\" d=\"M 243.821,327.118 L 243.879,344.867\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-2 atom-1 atom-3\" d=\"M 243.763,309.368 L 206.734,288.15\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-3 atom-3 atom-4\" d=\"M 206.734,288.15 L 169.843,309.609\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-3 atom-3 atom-4\" d=\"M 196.908,283.99 L 171.085,299.012\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-44 atom-8 atom-3\" d=\"M 206.595,245.472 L 206.734,288.15\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-4 atom-4 atom-5\" d=\"M 169.843,309.609 L 132.814,288.391\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-5 atom-5 atom-6\" d=\"M 132.814,288.391 L 132.675,245.713\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-5 atom-5 atom-6\" d=\"M 141.329,281.961 L 141.231,252.087\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-6 atom-6 atom-7\" d=\"M 132.675,245.713 L 169.565,224.254\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-7 atom-7 atom-8\" d=\"M 169.565,224.254 L 206.595,245.472\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-7 atom-7 atom-8\" d=\"M 170.876,234.842 L 196.797,249.695\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-8 atom-8 atom-9\" d=\"M 206.595,245.472 L 222.569,236.18\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-8 atom-8 atom-9\" d=\"M 222.569,236.18 L 238.543,226.887\" style=\"fill:none;fill-rule:evenodd;stroke:#CCCC00;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-9 atom-9 atom-10\" d=\"M 244.111,233.576 L 250.658,244.83\" style=\"fill:none;fill-rule:evenodd;stroke:#CCCC00;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-9 atom-9 atom-10\" d=\"M 250.658,244.83 L 257.204,256.085\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-9 atom-9 atom-10\" d=\"M 251.489,229.284 L 258.036,240.539\" style=\"fill:none;fill-rule:evenodd;stroke:#CCCC00;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-9 atom-9 atom-10\" d=\"M 258.036,240.539 L 264.582,251.793\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-10 atom-9 atom-11\" d=\"M 243.048,214.774 L 236.464,203.455\" style=\"fill:none;fill-rule:evenodd;stroke:#CCCC00;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-10 atom-9 atom-11\" d=\"M 236.464,203.455 L 229.88,192.137\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-10 atom-9 atom-11\" d=\"M 235.67,219.065 L 229.086,207.747\" style=\"fill:none;fill-rule:evenodd;stroke:#CCCC00;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-10 atom-9 atom-11\" d=\"M 229.086,207.747 L 222.502,196.429\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-11 atom-9 atom-12\" d=\"M 248.427,221.138 L 261.647,213.447\" style=\"fill:none;fill-rule:evenodd;stroke:#CCCC00;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-11 atom-9 atom-12\" d=\"M 261.647,213.447 L 274.868,205.757\" style=\"fill:none;fill-rule:evenodd;stroke:#0000FF;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-12 atom-12 atom-13\" d=\"M 286.026,205.791 L 301.715,214.781\" style=\"fill:none;fill-rule:evenodd;stroke:#0000FF;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-12 atom-12 atom-13\" d=\"M 301.715,214.781 L 317.405,223.772\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-45 atom-44 atom-12\" d=\"M 280.236,159.876 L 280.294,177.659\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-45 atom-44 atom-12\" d=\"M 280.294,177.659 L 280.352,195.443\" style=\"fill:none;fill-rule:evenodd;stroke:#0000FF;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-13 atom-13 atom-14\" d=\"M 317.405,223.772 L 354.295,202.312\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-14 atom-14 atom-15\" d=\"M 354.295,202.312 L 354.156,159.635\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-15 atom-15 atom-16\" d=\"M 354.156,159.635 L 391.047,138.175\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-42 atom-15 atom-43\" d=\"M 354.156,159.635 L 317.127,138.416\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-16 atom-16 atom-17\" d=\"M 391.047,138.175 L 430.091,155.407\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-16 atom-16 atom-17\" d=\"M 400.35,132.951 L 427.681,145.013\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-46 atom-39 atom-16\" d=\"M 394.645,102.827 L 392.846,120.501\" style=\"fill:none;fill-rule:evenodd;stroke:#0000FF;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-46 atom-39 atom-16\" d=\"M 392.846,120.501 L 391.047,138.175\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-17 atom-17 atom-18\" d=\"M 430.091,155.407 L 443.412,195.953\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-48 atom-37 atom-17\" d=\"M 458.545,123.598 L 430.091,155.407\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-18 atom-18 atom-19\" d=\"M 443.412,195.953 L 485.186,204.69\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-18 atom-18 atom-19\" d=\"M 451.425,188.908 L 480.667,195.024\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-19 atom-19 atom-20\" d=\"M 485.186,204.69 L 498.506,245.236\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-34 atom-19 atom-35\" d=\"M 485.186,204.69 L 513.639,172.881\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-20 atom-20 atom-21\" d=\"M 498.506,245.236 L 470.052,277.044\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-27 atom-20 atom-28\" d=\"M 498.506,245.236 L 540.28,253.973\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-21 atom-21 atom-22\" d=\"M 470.052,277.044 L 428.279,268.307\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-21 atom-21 atom-22\" d=\"M 462.039,284.089 L 432.797,277.973\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-49 atom-27 atom-21\" d=\"M 483.373,317.59 L 470.052,277.044\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-22 atom-22 atom-23\" d=\"M 428.279,268.307 L 399.825,300.116\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-23 atom-23 atom-24\" d=\"M 399.825,300.116 L 413.145,340.662\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-23 atom-23 atom-24\" d=\"M 409.932,303.534 L 419.257,331.916\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-24 atom-24 atom-25\" d=\"M 413.145,340.662 L 399.845,355.531\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-24 atom-24 atom-25\" d=\"M 399.845,355.531 L 386.544,370.4\" style=\"fill:none;fill-rule:evenodd;stroke:#00CC00;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-25 atom-24 atom-26\" d=\"M 413.145,340.662 L 454.919,349.399\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-26 atom-26 atom-27\" d=\"M 454.919,349.399 L 483.373,317.59\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-26 atom-26 atom-27\" d=\"M 452.826,338.937 L 472.743,316.671\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-28 atom-28 atom-29\" d=\"M 540.28,253.973 L 553.6,294.519\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-28 atom-28 atom-29\" d=\"M 550.387,257.39 L 559.712,285.773\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-50 atom-34 atom-28\" d=\"M 568.734,222.164 L 540.28,253.973\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-29 atom-29 atom-30\" d=\"M 553.6,294.519 L 595.374,303.256\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-30 atom-30 atom-31\" d=\"M 595.374,303.256 L 623.828,271.447\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-30 atom-30 atom-31\" d=\"M 593.281,292.794 L 613.198,270.527\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-31 atom-31 atom-32\" d=\"M 623.828,271.447 L 641.851,275.216\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-31 atom-31 atom-32\" d=\"M 641.851,275.216 L 659.875,278.986\" style=\"fill:none;fill-rule:evenodd;stroke:#00CC00;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-32 atom-31 atom-33\" d=\"M 623.828,271.447 L 610.508,230.901\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-33 atom-33 atom-34\" d=\"M 610.508,230.901 L 568.734,222.164\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-33 atom-33 atom-34\" d=\"M 602.494,237.945 L 573.252,231.829\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-35 atom-35 atom-36\" d=\"M 513.639,172.881 L 500.319,132.335\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-35 atom-35 atom-36\" d=\"M 503.532,169.463 L 494.208,141.081\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-36 atom-36 atom-37\" d=\"M 500.319,132.335 L 458.545,123.598\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-37 atom-37 atom-38\" d=\"M 458.545,123.598 L 449.883,108.708\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-37 atom-37 atom-38\" d=\"M 449.883,108.708 L 441.221,93.8177\" style=\"fill:none;fill-rule:evenodd;stroke:#0000FF;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-37 atom-37 atom-38\" d=\"M 448.568,123.423 L 442.505,113\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-37 atom-37 atom-38\" d=\"M 442.505,113 L 436.442,102.577\" style=\"fill:none;fill-rule:evenodd;stroke:#0000FF;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-38 atom-38 atom-39\" d=\"M 431.435,87.9278 L 400.875,94.5277\" style=\"fill:none;fill-rule:evenodd;stroke:#0000FF;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-39 atom-39 atom-40\" d=\"M 389.719,90.6623 L 376.639,78.9628\" style=\"fill:none;fill-rule:evenodd;stroke:#0000FF;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-39 atom-39 atom-40\" d=\"M 376.639,78.9628 L 363.56,67.2633\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-40 atom-40 atom-41\" d=\"M 363.56,67.2633 L 321.786,58.5262\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-47 atom-42 atom-40\" d=\"M 350.24,26.7174 L 363.56,67.2633\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-41 atom-41 atom-42\" d=\"M 321.786,58.5262 L 350.24,26.7174\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"bond-43 atom-43 atom-44\" d=\"M 317.127,138.416 L 280.236,159.876\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n",
       "<path class=\"atom-0\" d=\"M 275.106 287.943 Q 275.106 285.041, 276.54 283.419 Q 277.974 281.797, 280.654 281.797 Q 283.334 281.797, 284.768 283.419 Q 286.202 285.041, 286.202 287.943 Q 286.202 290.879, 284.751 292.552 Q 283.3 294.208, 280.654 294.208 Q 277.991 294.208, 276.54 292.552 Q 275.106 290.896, 275.106 287.943 M 280.654 292.842 Q 282.497 292.842, 283.487 291.613 Q 284.495 290.367, 284.495 287.943 Q 284.495 285.57, 283.487 284.375 Q 282.497 283.163, 280.654 283.163 Q 278.81 283.163, 277.803 284.358 Q 276.813 285.553, 276.813 287.943 Q 276.813 290.384, 277.803 291.613 Q 278.81 292.842, 280.654 292.842 \" fill=\"#FF0000\"/>\n",
       "<path class=\"atom-2\" d=\"M 238.354 352.08 Q 238.354 349.178, 239.788 347.556 Q 241.222 345.934, 243.902 345.934 Q 246.583 345.934, 248.017 347.556 Q 249.451 349.178, 249.451 352.08 Q 249.451 355.016, 248 356.689 Q 246.548 358.345, 243.902 358.345 Q 241.239 358.345, 239.788 356.689 Q 238.354 355.033, 238.354 352.08 M 243.902 356.979 Q 245.746 356.979, 246.736 355.75 Q 247.743 354.504, 247.743 352.08 Q 247.743 349.707, 246.736 348.512 Q 245.746 347.3, 243.902 347.3 Q 242.059 347.3, 241.052 348.495 Q 240.061 349.69, 240.061 352.08 Q 240.061 354.521, 241.052 355.75 Q 242.059 356.979, 243.902 356.979 \" fill=\"#FF0000\"/>\n",
       "<path class=\"atom-2\" d=\"M 250.902 346.071 L 252.54 346.071 L 252.54 351.209 L 258.72 351.209 L 258.72 346.071 L 260.359 346.071 L 260.359 358.157 L 258.72 358.157 L 258.72 352.575 L 252.54 352.575 L 252.54 358.157 L 250.902 358.157 L 250.902 346.071 \" fill=\"#FF0000\"/>\n",
       "<path class=\"atom-9\" d=\"M 240.071 228.161 Q 240.207 228.212, 240.771 228.451 Q 241.334 228.69, 241.949 228.844 Q 242.58 228.98, 243.195 228.98 Q 244.339 228.98, 245.004 228.434 Q 245.67 227.871, 245.67 226.898 Q 245.67 226.232, 245.329 225.822 Q 245.004 225.412, 244.492 225.191 Q 243.98 224.969, 243.127 224.713 Q 242.051 224.388, 241.402 224.081 Q 240.771 223.774, 240.31 223.125 Q 239.866 222.476, 239.866 221.384 Q 239.866 219.864, 240.89 218.925 Q 241.932 217.986, 243.98 217.986 Q 245.38 217.986, 246.968 218.652 L 246.575 219.967 Q 245.124 219.369, 244.031 219.369 Q 242.853 219.369, 242.205 219.864 Q 241.556 220.342, 241.573 221.179 Q 241.573 221.827, 241.897 222.22 Q 242.239 222.613, 242.717 222.835 Q 243.212 223.057, 244.031 223.313 Q 245.124 223.654, 245.773 223.996 Q 246.421 224.337, 246.882 225.037 Q 247.36 225.72, 247.36 226.898 Q 247.36 228.571, 246.233 229.475 Q 245.124 230.363, 243.263 230.363 Q 242.188 230.363, 241.368 230.124 Q 240.566 229.902, 239.61 229.51 L 240.071 228.161 \" fill=\"#CCCC00\"/>\n",
       "<path class=\"atom-10\" d=\"M 259.396 260.937 Q 259.396 258.035, 260.83 256.413 Q 262.264 254.791, 264.944 254.791 Q 267.625 254.791, 269.059 256.413 Q 270.492 258.035, 270.492 260.937 Q 270.492 263.873, 269.041 265.546 Q 267.59 267.202, 264.944 267.202 Q 262.281 267.202, 260.83 265.546 Q 259.396 263.89, 259.396 260.937 M 264.944 265.837 Q 266.788 265.837, 267.778 264.607 Q 268.785 263.361, 268.785 260.937 Q 268.785 258.564, 267.778 257.369 Q 266.788 256.157, 264.944 256.157 Q 263.101 256.157, 262.093 257.352 Q 261.103 258.547, 261.103 260.937 Q 261.103 263.378, 262.093 264.607 Q 263.101 265.837, 264.944 265.837 \" fill=\"#FF0000\"/>\n",
       "<path class=\"atom-11\" d=\"M 216.478 187.156 Q 216.478 184.254, 217.912 182.633 Q 219.346 181.011, 222.026 181.011 Q 224.706 181.011, 226.14 182.633 Q 227.574 184.254, 227.574 187.156 Q 227.574 190.093, 226.123 191.766 Q 224.672 193.421, 222.026 193.421 Q 219.363 193.421, 217.912 191.766 Q 216.478 190.11, 216.478 187.156 M 222.026 192.056 Q 223.869 192.056, 224.859 190.827 Q 225.867 189.58, 225.867 187.156 Q 225.867 184.783, 224.859 183.588 Q 223.869 182.376, 222.026 182.376 Q 220.182 182.376, 219.175 183.571 Q 218.185 184.766, 218.185 187.156 Q 218.185 189.598, 219.175 190.827 Q 220.182 192.056, 222.026 192.056 \" fill=\"#FF0000\"/>\n",
       "<path class=\"atom-12\" d=\"M 277.704 196.51 L 281.664 202.912 Q 282.057 203.543, 282.689 204.687 Q 283.32 205.831, 283.354 205.899 L 283.354 196.51 L 284.959 196.51 L 284.959 208.596 L 283.303 208.596 L 279.052 201.597 Q 278.557 200.778, 278.028 199.839 Q 277.516 198.9, 277.362 198.61 L 277.362 208.596 L 275.792 208.596 L 275.792 196.51 L 277.704 196.51 \" fill=\"#0000FF\"/>\n",
       "<path class=\"atom-25\" d=\"M 373.118 372.889 Q 373.118 369.884, 374.517 368.314 Q 375.934 366.726, 378.615 366.726 Q 381.107 366.726, 382.439 368.485 L 381.312 369.406 Q 380.339 368.126, 378.615 368.126 Q 376.788 368.126, 375.815 369.355 Q 374.859 370.567, 374.859 372.889 Q 374.859 375.279, 375.849 376.508 Q 376.856 377.737, 378.802 377.737 Q 380.134 377.737, 381.687 376.935 L 382.165 378.215 Q 381.534 378.625, 380.578 378.864 Q 379.622 379.103, 378.563 379.103 Q 375.934 379.103, 374.517 377.498 Q 373.118 375.893, 373.118 372.889 \" fill=\"#00CC00\"/>\n",
       "<path class=\"atom-25\" d=\"M 383.907 365.992 L 385.477 365.992 L 385.477 378.949 L 383.907 378.949 L 383.907 365.992 \" fill=\"#00CC00\"/>\n",
       "<path class=\"atom-32\" d=\"M 660.942 280.602 Q 660.942 277.598, 662.341 276.027 Q 663.758 274.439, 666.438 274.439 Q 668.931 274.439, 670.262 276.198 L 669.136 277.12 Q 668.163 275.839, 666.438 275.839 Q 664.612 275.839, 663.639 277.068 Q 662.683 278.28, 662.683 280.602 Q 662.683 282.992, 663.673 284.221 Q 664.68 285.45, 666.626 285.45 Q 667.958 285.45, 669.511 284.648 L 669.989 285.928 Q 669.358 286.338, 668.402 286.577 Q 667.446 286.816, 666.387 286.816 Q 663.758 286.816, 662.341 285.211 Q 660.942 283.607, 660.942 280.602 \" fill=\"#00CC00\"/>\n",
       "<path class=\"atom-32\" d=\"M 671.731 273.705 L 673.301 273.705 L 673.301 286.662 L 671.731 286.662 L 671.731 273.705 \" fill=\"#00CC00\"/>\n",
       "<path class=\"atom-38\" d=\"M 434.414 80.6643 L 438.374 87.066 Q 438.767 87.6976, 439.398 88.8414 Q 440.03 89.9852, 440.064 90.0535 L 440.064 80.6643 L 441.669 80.6643 L 441.669 92.7507 L 440.013 92.7507 L 435.762 85.7515 Q 435.267 84.9321, 434.738 83.9932 Q 434.226 83.0543, 434.072 82.7641 L 434.072 92.7507 L 432.502 92.7507 L 432.502 80.6643 L 434.414 80.6643 \" fill=\"#0000FF\"/>\n",
       "<path class=\"atom-39\" d=\"M 392.698 89.6736 L 396.658 96.0753 Q 397.051 96.7069, 397.682 97.8507 Q 398.314 98.9945, 398.348 99.0628 L 398.348 89.6736 L 399.953 89.6736 L 399.953 101.76 L 398.297 101.76 L 394.046 94.7608 Q 393.551 93.9414, 393.022 93.0025 Q 392.51 92.0636, 392.356 91.7734 L 392.356 101.76 L 390.786 101.76 L 390.786 89.6736 L 392.698 89.6736 \" fill=\"#0000FF\"/>\n",
       "</svg>"
      ],
      "text/plain": [
       "<IPython.core.display.SVG object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "false_positive_num = 2\n",
      "----------\n"
     ]
    }
   ],
   "source": [
    "train_mse_list=[]\n",
    "train_r2_list=[]\n",
    "test_mse_list=[]\n",
    "test_r2_list=[]\n",
    "val_mse_list=[]\n",
    "val_r2_list=[]\n",
    "\n",
    "batch_size = 10\n",
    "epoch = 0\n",
    "\n",
    "total_epoch = 1000\n",
    "k_list = [int(len(test_df)*0.01),int(len(test_df)*0.1),50,100,150,200,250,300]\n",
    "stopper = EarlyStopping(mode='higher', patience=30, filename=model_file + '_model.pth')\n",
    "\n",
    "stopper.load_checkpoint(model)\n",
    "\n",
    "test_feature_list, test_r2, test_MSE, test_predict = eval(model, test_df)\n",
    "test_df['predict']=test_predict\n",
    "train_feature_list, train_r2, train_MSE, train_predict = eval(model, train_df)\n",
    "test_feature_list, test_r2, test_MSE, test_predict = eval(model, test_df, test_viz=True,k=1)\n",
    "test_feature_list_active, test_r2_a, test_MSE_a, test_predict_a = eval(model, test_df[:test_active])\n",
    "    \n",
    "test_predict = np.array(test_predict)\n",
    "\n",
    "k_list = [int(len(test_df)*0.01),int(len(test_df)*0.1),50,100,150,200,250,300]\n",
    "topk_list =[]\n",
    "false_positive_rate_list = []\n",
    "for k in k_list:\n",
    "    a,b = topk_acc_recall(test_df, test_predict, k, test_active, False, epoch)\n",
    "    topk_list.append(a)\n",
    "    false_positive_rate_list.append(b)\n",
    "\n",
    "print(\"----------\")\n",
    "# print(epoch, 'R2:%.4f'%train_r2,'%.4f'%val_r2,'%.4f'%test_r2, 'RMSE:%.4f'%train_MSE**0.5, '%.4f'%val_MSE**0.5, '%.4f'%test_MSE**0.5)\n",
    "# print(test_r2_a, test_MSE_a)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 r2:0.5025 RMSE:0.4878 \n",
      " top_precent1:0.5263 top_precent1-fp:0.2105 \n",
      " top_precent10:0.5855 top_precent10-fp:0.4145 \n",
      " Top50:0.5400 Top50-Ina:0.1800 \n",
      " Top100:0.7000 Top100-Ina:0.1900 \n",
      " Top150:0.6467 Top150-Ina:0.3267 \n",
      " Top200:0.5750 Top200-Ina:0.4250 \n",
      " Top250:0.6200 Top250-Ina:0.5040 \n",
      " Top300:0.6400 Top300-Ina:0.5733 \n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(epoch,'r2:%.4f'%test_r2_a,'RMSE:%.4f'%test_MSE_a,'\\n',\n",
    "      'top_precent1:%.4f'%topk_list[0],'top_precent1-fp:%.4f'%false_positive_rate_list[0],'\\n',\n",
    "      'top_precent10:%.4f'%topk_list[1],'top_precent10-fp:%.4f'%false_positive_rate_list[1],'\\n',\n",
    "      'Top50:%.4f'%topk_list[2],'Top50-Ina:%.4f'%false_positive_rate_list[2],'\\n',\n",
    "      'Top100:%.4f'%topk_list[3],'Top100-Ina:%.4f'%false_positive_rate_list[3],'\\n',\n",
    "      'Top150:%.4f'%topk_list[4],'Top150-Ina:%.4f'%false_positive_rate_list[4],'\\n',\n",
    "      'Top200:%.4f'%topk_list[5],'Top200-Ina:%.4f'%false_positive_rate_list[5],'\\n',\n",
    "      'Top250:%.4f'%topk_list[6],'Top250-Ina:%.4f'%false_positive_rate_list[6],'\\n',\n",
    "      'Top300:%.4f'%topk_list[7],'Top300-Ina:%.4f'%false_positive_rate_list[7],'\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "../data/active_inactive/median_active/EC50/P21554.csv\n",
      "../data/active_inactive/median_inactive/EC50/P21554.csv\n"
     ]
    }
   ],
   "source": [
    "print(raw_filename)#目标域的active\n",
    "print(inactive_filename)#目标域的inactive"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                                 smiles  value  \\\n",
      "1925  COC1=CC=CC(=C1)NC2=NC=C(C(=C2)C3CC3)C(=O)NCC4C...   -4.1   \n",
      "1926  CC(C)CSC1=C(C=CC(=N1)C(=O)NC2(CS(=O)(=O)C2)CC(...   -4.1   \n",
      "1927  CC1=NON=C1CN2C=NC3=C2N=C(N=C3N4CC(C(C4)(F)F)(F...   -4.1   \n",
      "1928   C1CC1COC2=C(C=CC(=N2)C(=O)NC3(COC3)CC(=O)N)C4CC4   -4.1   \n",
      "1929  CC(C)(C)C1=NC2=C(C=NN2CC3=C(C=C(C=C3)F)Cl)C(=N...   -4.1   \n",
      "\n",
      "                                            cano_smiles   predict  \n",
      "1925      COc1cccc(Nc2cc(C3CC3)c(C(=O)NCC3CCOCC3)cn2)c1 -2.558483  \n",
      "1926  CC(C)CSc1nc(C(=O)NC2(CC(N)=O)CS(=O)(=O)C2)ccc1... -3.644027  \n",
      "1927  Cc1nonc1Cn1cnc2c(N3CC(F)(F)C(F)(F)C3)nc(C(C)(C... -3.133777  \n",
      "1928      NC(=O)CC1(NC(=O)c2ccc(C3CC3)c(OCC3CC3)n2)COC1 -3.860179  \n",
      "1929  CC(C)(C)c1nc(N2CCC(F)(F)C2)c2cnn(Cc3ccc(F)cc3C... -2.461061  \n"
     ]
    }
   ],
   "source": [
    "test_df.to_csv(path_or_buf='/data/yinmingyue/Sim/code/result/Sim+/Sim_'+file1+'.csv')\n",
    "print(test_df[-5:].head())"
   ]
  },
  {
   "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": "3.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}