6_test_textbook_problems.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from __future__ import absolute_import, division, print_function\n",
    "\n",
    "import cPickle, gzip, glob\n",
    "from six.moves import xrange\n",
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "\n",
    "from tensorflow.models.rnn.translate import data_utils\n",
    "import seq2seq_model\n",
    "\n",
    "from rdkit import Chem\n",
    "from rdkit.Chem import AllChem\n",
    "import parser.Smipar as Smipar\n",
    "\n",
    "from rdkit import DataStructs\n",
    "from rdkit.Chem.Fingerprints import FingerprintMols"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "flags = tf.app.flags\n",
    "\n",
    "flags.DEFINE_float(\"learning_rate\", 0.5, \"Learning rate.\")\n",
    "flags.DEFINE_float(\"learning_rate_decay_factor\", 0.99, \"Learning rate decays by this much.\")\n",
    "flags.DEFINE_float(\"max_gradient_norm\", 5.0, \"Clip gradients to this norm.\")\n",
    "flags.DEFINE_integer(\"batch_size\", 1, \"Batch size to use during training.\")\n",
    "flags.DEFINE_integer(\"size\", 600, \"Size of each model layer.\")\n",
    "flags.DEFINE_integer(\"num_layers\", 3, \"Number of layers in the model.\")\n",
    "flags.DEFINE_integer(\"reactant_vocab_size\", 311, \"Reactant vocabulary size.\")\n",
    "flags.DEFINE_integer(\"product_vocab_size\", 180, \"Product vocabulary size.\")\n",
    "flags.DEFINE_string(\"train_dir\", \"checkpoint/saved_models\", \"Training dir.\")\n",
    "FLAGS = flags.FLAGS\n",
    "\n",
    "_buckets = [(54, 54), (70, 60), (90, 65), (150, 80)]\n",
    "\n",
    "with gzip.open('data/vocab/vocab_list.pkl.gz', 'rb') as list_file:\n",
    "    reactants_token_list, products_token_list = cPickle.load(list_file)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def create_model(session, forward_only):\n",
    "    model = seq2seq_model.Seq2SeqModel( \\\n",
    "        FLAGS.reactant_vocab_size, FLAGS.product_vocab_size, _buckets, \\\n",
    "        FLAGS.size, FLAGS.num_layers, FLAGS.max_gradient_norm, FLAGS.batch_size, \\\n",
    "        FLAGS.learning_rate, FLAGS.learning_rate_decay_factor, forward_only=forward_only)\n",
    "    ckpt = tf.train.get_checkpoint_state(FLAGS.train_dir)\n",
    "    if ckpt and tf.gfile.Exists(ckpt.model_checkpoint_path):\n",
    "        print(\"Reading model parameters from %s\" % ckpt.model_checkpoint_path)\n",
    "        model.saver.restore(session, ckpt.model_checkpoint_path)\n",
    "    else:\n",
    "        print(\"Created model with fresh parameters.\")\n",
    "        session.run(tf.initialize_all_variables())\n",
    "    return model\n",
    "\n",
    "def cano(smiles): # canonicalize smiles by MolToSmiles function\n",
    "    return Chem.MolToSmiles(Chem.MolFromSmiles(smiles)) if (smiles != '') else ''\n",
    "\n",
    "def fp_sim(smiles1, smiles2): # fingerprint similarity (Tanimoto)\n",
    "    try:\n",
    "        fp1 = FingerprintMols.FingerprintMol(Chem.MolFromSmiles(smiles1))\n",
    "        fp2 = FingerprintMols.FingerprintMol(Chem.MolFromSmiles(smiles2))\n",
    "        return DataStructs.FingerprintSimilarity(fp1, fp2)\n",
    "    except:\n",
    "        return 0\n",
    "    \n",
    "def avg(l):\n",
    "    return sum(l)/float(len(l))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Reading model parameters from checkpoint/saved_models/gen.ckpt-8\n",
      "CC=C(C)C.Cl>>CCC(C)(C)Cl CCC(C)(C)Cl 1.0\n",
      "CC1=CCCC1.BrBr>ClC(Cl)(Cl)Cl>CC1(Br)CCCC1Br CC1(Br)CCCC1Br 1.0\n",
      "CC1=CCCC1.O>[H]B([H])[H].C1CCOC1.OO.[Na+].[OH-]>CC1CCCC1O CC1CCCC1O 1.0\n",
      "CCC(C)=CC.[O-][O+]=O>CSC>CC=O.CCC(C)=O CC=O.CCC(C)=O 1.0\n",
      "CC=C1CCCCC1.Br>COOC>CC(Br)C1CCCCC1 CC(Br)C1CCCCC1 1.0\n",
      "CC=C1CCCCC1.Cl>COOC>CCC1(Cl)CCCCC1 CCC1(O)CCCCC1 0.4625\n",
      "CCC=CC>OOC(=O)c1ccccc1>CCC1OC1C CCC1OC1C 1.0\n",
      "CC1=CCCC1>O=[Os](=O)(=O)=O.OO>CC1(O)CCCC1O CC1(O)CCCC1O 1.0\n",
      "CC1=CCCCC1>[K+].[O-][Mn](=O)(=O)=O.[OH-]>CC1(O)CCCCC1O CC1(O)CCCCC1O 1.0\n",
      "CC1=CCCCC1>CC(=O)OO.OS(O)(=O)=O>CC(=O)CCCCC(O)=O CC1=C(C2CCCC)OCC1 0\n",
      "CC1=CCCCC1.[O-][O+]=O>CSC>CC(=O)CCCCC=O CC(=O)CCCCC=O 1.0\n",
      "CC1=C(C)CCCC1.[H][H]>[Pt]>CC1CCCCC1C CC1CCCCC1C 1.0\n",
      "C1CCC2=CCCCC2C1.O>OS(O)(=O)=O>OC12CCCCC1CCCC2 OC12CCCCC1CCCC2 1.0\n",
      "C1CCC2=CCCCC2C1.O>CC(=O)O[Hg]OC(C)=O.[Na+].[H][B-]([H])([H])[H]>OC12CCCCC1CCCC2 OC12CCCCC1CCCC2 1.0\n",
      "C1CCC2=CCCCC2C1.ClCl.O>>OC12CCCCC1CCCC2Cl OC12CCCCC1CCCC2Cl 1.0\n",
      "data/problems/8-47.txt\t1.07091126624\t0.8975\t13\t1\n",
      "---------------\n",
      "CCC(C)(Br)Br>[Na+].[NH2-]>CCC#C CCC(C)(C)Br 0.285714285714\n",
      "CCC(C)(Br)Br>[K+].[OH-]>CC#CC CCC(C)Br 0.294117647059\n",
      "CCC#C.CCCCBr>[Na+].[NH2-]>CCCCC#CCC CCCC=CC(C)C 0.4\n",
      "C1CCCCC#CCCC1.[H][H]>c1ccc2ncccc2c1.[Pd].[Ba++].[O-]S([O-])(=O)=O>C1CCCCC=CCCC1 C1=CCCCCCCCC1 1.0\n",
      "C1CCCCC#CCCC1>N.[Na]>C1CCCCC=CCCC1 C1=CCCCCCCCC1 1.0\n",
      "CCCCC#C>CC(C)C(C)BC(C)C(C)C.OO.[Na+].[OH-]>CCCCCC=O CCCCCC(C)C=O 0.636363636364\n",
      "CCCC#C.[H][H]>c1ccc2ncccc2c1.[Pd].[Ba++].[O-]S([O-])(=O)=O>CCCC=C C=CCCC 1.0\n",
      "data/problems/9-34.txt\t8.38281387964\t0.659456509877\t3\t0\n",
      "---------------\n",
      "c1ccccc1.CC(C)(C)Br>[Al+3].[Cl-].[Cl-].[Cl-]>CC(C)(C)c1ccccc1 CC(C)(C)c1ccccc1 1.0\n",
      "c1ccccc1.CCCCCl>[Al+3].[Cl-].[Cl-].[Cl-]>CCC(C)c1ccccc1 CCCCc1ccccc1 0.735294117647\n",
      "c1ccccc1.CC(C)CO>FB(F)F>CC(C)(C)c1ccccc1 CC(C)COc1ccccc1 0.31884057971\n",
      "c1ccccc1.BrBr>[Fe]>Brc1ccccc1 Brc1ccccc1 1.0\n",
      "c1ccccc1.CC(C)=C>F>CC(C)(C)c1ccccc1 CC(C)(C)c1ccccc1C 0.743362831858\n",
      "c1ccccc1.O=S(=O)=O>OS(O)(=O)=O>OS(=O)(=O)c1ccccc1 O=C(Cl)c1ccccc1 0.276315789474\n",
      "c1ccccc1.CC(C)(C)CCl>[Al+3].[Cl-].[Cl-].[Cl-]>CCC(C)(C)c1ccccc1 CC(C)(C)Cc1ccccc1 0.666666666667\n",
      "c1ccccc1.ClC(=O)c1ccccc1>[Al+3].[Cl-].[Cl-].[Cl-]>O=C(c1ccccc1)c1ccccc1 OCc1ccccc1 0.505617977528\n",
      "c1ccccc1.II>O[N+]([O-])=O>Ic1ccccc1 Oc1ccccc1 0.512820512821\n",
      "c1ccccc1.O[N+]([O-])=O>OS(O)(=O)=O>[O-][N+](=O)c1ccccc1 O=C(O)c1ccccc1 0.272727272727\n",
      "c1ccccc1.[C-]#[O+]>[Al+3].[Cu+].[Cl-].[Cl-].[Cl-].Cl.[Cl-]>O=Cc1ccccc1 BrC1 0\n",
      "data/problems/17-44.txt\t10.7509029969\t0.548331431676\t2\t1\n",
      "---------------\n",
      "CCC(C)O.Cc1ccc(cc1)S(Cl)(=O)=O>c1ccncc1>CCC(C)OS(=O)(=O)c1ccc(C)cc1 CCC(C)OS(=O)(=O)c1ccc(C)cc1 1.0\n",
      "CCC(C)OS(=O)(=O)c1ccc(C)cc1.[Na+].[Br-]>>CCC(C)Br CCC(C)(=O)O.CC(O)c1ccccc1 0\n",
      "OC1CCCCCCC1>OS(O)(=O)=O.O=[Cr](=O)=O>O=C1CCCCCCC1 O=C1CCCCCCC1 1.0\n",
      "OCC1CCCC1>[O-][Cr](Cl)(=O)=O.c1cc[nH+]cc1>O=CC1CCCC1 O=C(O)C1CCCC1 0.596153846154\n",
      "OCC1CCCC1>OS(O)(=O)=O.[O-][Cr](=O)(=O)O[Cr]([O-])(=O)=O.[Na+].[Na+]>OC(=O)C1CCCC1 O=C(O)C1CCCC1 1.0\n",
      "OC1CCCC1.Cl>[Zn++].[Cl-].[Cl-]>ClC1CCCC1 OC1CCCC1 0.4375\n",
      "CCCCO.Br>>CCCCBr CCCCBr 1.0\n",
      "CC(C)(C)[O-].[K+].CI>>COC(C)(C)C CC(C)(C)OC 1.0\n",
      "[Na+].C[O-].CC(C)(C)I>>CO.CC(C)=C COC(C)(C)C 0.428571428571\n",
      "OC1CCCC1>OS(O)(=O)=O>C1CC=CC1 ClC1CCCC1 0.305555555556\n",
      "C1CC=CC1>O=[Os](=O)(=O)=O.O[I](=O)(=O)=O.O>[H]C(=O)CCCC([H])=O OC1CCCC1O 0.305555555556\n",
      "CC[O-].[Na+].CCCCBr>CCO>CCCCOCC CCCCOC1C 0\n",
      "data/problems/11-41.txt\t8.87203523499\t0.58944469882\t5\t2\n",
      "---------------\n",
      "O=C1CCCCC1.CN>Cl>CN=C1CCCCC1 CN=C1CCCCC1 1.0\n",
      "O=C1CCCCC1.CO>Cl>COC1(CCCCC1)OC COC1(O)CCCCC1 0.907216494845\n",
      "O=C1CCCCC1.NO>Cl>ON=C1CCCCC1 ON=C1CCCCC1 1.0\n",
      "O=C1CCCCC1.OCCO>Cc1ccc(cc1)S(O)(=O)=O>C1COC2(CCCCC2)O1 C1CCC2(CC1)OCCO2 1.0\n",
      "O=C1CCCCC1.NNc1ccccc1>Cl>C1CCC(CC1)=NNc1ccccc1 c1ccc(NN=C2CCCCC2)cc1 1.0\n",
      "O=C1CCCCC1.[Mg+]c1ccccc1.[Br-]>>OC1(CCCCC1)c1ccccc1 OC1(c2ccccc2)CCCCC1 1.0\n",
      "O=C1CCCCC1.[Na+].C#[C-]>>OC1(CCCCC1)C#C NC1(O)CCCCC1 0.516853932584\n",
      "O=C1CCCCC1>[K+].NN.[OH-]>C1CCCCC1 C1CCCCC1 1.0\n",
      "O=C1CCCCC1.C=P(c1ccccc1)(c1ccccc1)c1ccccc1>>C=C1CCCCC1 C=C(Cc1ccccc1)=C1CCCCC1 0\n",
      "O=C1CCCCC1.[Na+].[C-]#N>>OC1(CCCCC1)C#N N#CC1(O)CCCCC1 1.0\n",
      "OC1(CCCCC1)C#N.O>Cl>OC(=O)C1(O)CCCCC1 N=C(O)C1(CCCCC2CO2)C 0\n",
      "data/problems/18-56.txt\t5.64648598101\t0.765824584312\t7\t2\n",
      "---------------\n",
      "ClC(=O)c1ccccc1.OC1CCCCC1>>O=C(OC1CCCCC1)c1ccccc1 O=C(OCC1CCCCC1)c1ccccc1 0.782426778243\n",
      "COC(=O)C1CCCCC1.CN>>CNC(=O)C1CCCCC1 COC(=O)C1CCCCC1 0.423312883436\n",
      "ClC(=O)c1ccccc1.C1CCNC1>>O=C(N1CCCC1)c1ccccc1 C1CCNC(=O)c1ccccc1 0\n",
      "O=C1CCCC(=O)O1.NC1CCCCC1>>OC(=O)CCCC(=O)NC1CCCCC1 O=C(CCCCCO)NC1CCCC1 0.802816901408\n",
      "CCOC(=O)c1ccccc1>[Al+3].[H-].[H-].[H-].[H-].[Li+]>OCc1ccccc1 CCO.OCc1ccccc1 0.944444444444\n",
      "O=C1CCCCN1>[Al+3].[H-].[H-].[H-].[H-].[Li+]>C1CCNCC1 C1CCCNC1 1.0\n",
      "O=C1CCCCCO1.[Na+].C[O-]>CO>OCCCCCC(O)=O COC(=O)CCCCCO 0.733333333333\n",
      "O=C1CCCCO1.[Mg+]c1ccccc1.[Br-]>>OCCCCC(O)(c1ccccc1)c1ccccc1 OCCCCC(O)(c1ccccc1)c1ccccc1 1.0\n",
      "N#CC1CCCC1.C[Mg+].[I-]>>CC(=O)C1CCCC1 CC(=O)C1CCCC1 1.0\n",
      "O=C1CCCCCN1.O>[OH-].[Na+]>NCCCCCC(O)=O NCCCCCC(=O)O 1.0\n",
      "CC(CC(N)=O)c1ccccc1>[Na+].BrBr.[OH-]>CC(CN)c1ccccc1 CC(CN1)Cc1ccccc1 0\n",
      "O=C1CCCCC1.OCCO>Cl>C1COC2(CCCCC2)O1 C1CCC2(CC1)OCCO2 1.0\n",
      "O=C1CCCCO1.OCCO>Cl>OCCCCC(=O)OCCO COCCCC(=O)OC1CC1 0.550847457627\n",
      "data/problems/21-50.txt\t8.0618322567\t0.710552446038\t5\t2\n",
      "---------------\n",
      "C=CC=C.OC(=O)C=C>>OC(=O)C1CCC=CC1 C=C(C)OCC 0.344262295082\n",
      "C1CC=CC=C1.COC(=O)C#C>>COC(=O)C1=CC2CCC1C=C2 CC(=O)OC(=O)CC1CCC1 0.493212669683\n",
      "C1C=CC=C1.OC(=O)C=C>>OC(=O)C1CC2CC1C=C2 CC(C)=OCC1CC1 0\n",
      "CO\\C=C\\C=C.[H]C(=O)C=C>>[H]C(=O)C1CCC=CC1OC C=C(C)OCC 0.322580645161\n",
      "C\\C=C\\C=C\\C.N#C\\C=C/C#N>>CC1C=CC(C)C(C#N)C1C#N CCC(C)=NCCC(C)=O 0.259259259259\n",
      "C\\C=C/C=C\\C.N#C\\C=C/C#N>>CC1C=CC(C)C(C#N)C1C#N CCC(C)=NCCC(C)=O 0.259259259259\n",
      "data/problems/15-30.txt\t21.4003211657\t0.279762354741\t0\t1\n",
      "---------------\n",
      "CCCC#C.Cl>>CCCC(C)(Cl)Cl CCCC(C)(Cl)Cl 1.0\n",
      "CCCC#C.[H][H]>[Ni]>CCCCC CCCCC 1.0\n",
      "CCCC#C.BrBr>>CCCC(Br)(Br)C(Br)Br CCCC(Br)(Br)C(Br)Br 1.0\n",
      "CCCC#C>[K+].[O-][Mn](=O)(=O)=O>CCCC(=O)C(O)=O CCCC(=O)C=O 0.595238095238\n",
      "CCCC#C>[Na+].[K+].[OH-].[O-][Mn](=O)(=O)=O>CCCC(O)=O CCCC(=O)O 1.0\n",
      "CCCC#C>N.[Na]>C=CCCC C=CCCC 1.0\n",
      "CCCC#C.O>[Hg++].OS(O)(=O)=O.[O-]S([O-])(=O)=O>CCCC(C)=O CCCC(C)=O 1.0\n",
      "CCCC#C.O>CC(C)C(C)BC(C)C(C)C.[OH-].[Na+].OO>CCCCC=O CCCCC(C)=O 0.708333333333\n",
      "data/problems/9-33.txt\t2.40399896926\t0.912946428571\t6\t0\n",
      "---------------\n",
      "[Mg+]C1CCCCC1.[Br-].C=O>>OCC1CCCCC1 CC1(O)CCCCC1 0.616666666667\n",
      "[Mg+]C1CCCC1.[Cl-].CCCC(C)=O>>CCCC(C)(O)C1CCCC1 CCCC(C)(O)C1CCCC1 1.0\n",
      "CC(C)[Mg+].[I-].[H]C(=O)c1ccccc1>>CC(C)C(O)c1ccccc1 CC(C)c1ccccc1.CC(C)O 0.627659574468\n",
      "C[Mg+].[I-].OC1CCCC(=O)C1>>CC1(O)CCCC(O)C1 CC1(O)CCCC(C)C1 0.737373737374\n",
      "[H]C(=O)C1CCCC1>[Na+].[BH4-]>OCC1CCCC1 OCC1CCCC1 1.0\n",
      "[Mg+]c1ccccc1.[Br-].O=C(c1ccccc1)c1ccccc1>>OC(c1ccccc1)(c1ccccc1)c1ccccc1 OC(c1ccccc1)c1ccccc1 0.801418439716\n",
      "[Mg+]c1ccccc1.[Br-].COC(=O)C1CCCC1>>OC(C1CCCC1)(c1ccccc1)c1ccccc1 CO.OC(c1ccccc1)(c1ccccc1)C1CCCC1 1.0\n",
      "[Mg+]CC1CCCC1.[Cl-].ClC(=O)c1ccccc1>>OC(CC1CCCC1)(CC1CCCC1)c1ccccc1 OC(Cc1ccccc1)(c1ccccc1)C1CCCC1 0.574948665298\n",
      "COC(=O)CC1CCCC(=O)C1>[Na+].[BH4-]>COC(=O)CC1CCCC(O)C1 CO.O=CCC1CCCC1 0.422680412371\n",
      "COC(=O)CC1CCCC(=O)C1>[Al+3].[H-].[H-].[H-].[H-].[Li+]>OCCC1CCCC(O)C1 CO.OCCC1CCCC1 0.575757575758\n",
      "C1CCC2=CCCCC2C1.O>CC(=O)O[Hg]OC(C)=O.[BH4-].[Na+]>OC12CCCCC1CCCC2 OC12CCCCC1CCCC2 1.0\n",
      "C1CCC2=CCCCC2C1.O>B.C1CCOC1.OO.[Na+].[OH-]>OC1CCCC2CCCCC12 OC1CCCC2CCCCC12 1.0\n",
      "CCC\\C=C\\C>[O-][Mn](=O)(=O)=O.[K+].[Na+].[OH-]>CCCC(O)C(C)O CCCC(O)C(C)O 1.0\n",
      "CCC\\C=C\\C>[H]C(=O)OO.Cl>CCCC(O)C(C)O CCCC=OC1C 0\n",
      "CC[Cu-]CC.[Li+].CC\\C=C\\CBr>>CC\\C=C\\CC=C CCCC(CC)=CCC 0.375\n",
      "data/problems/10-38.txt\t10.7982503688\t0.715433671443\t6\t1\n",
      "---------------\n",
      "CC=C(C)C.Br>COOC>CC(C)C(C)Br CC(C)C(C)Br 1.0\n",
      "CC=C1CCCC1.O>OS(O)(=O)=O>CCC1(O)CCCC1 CCC1(O)CCCC1 1.0\n",
      "CC(C)C=C.Br>>CC(C)C(C)Br CC(C)C(C)Br 1.0\n",
      "C=Cc1ccccc1.BrC(Br)Br>[Na+].[OH-]>BrC1(Br)CC1c1ccccc1 BrC1(Br)CC1c1ccccc1 1.0\n",
      "CC=CC.CO>Cl>CCC(C)OC CCC(C)OC 1.0\n",
      "CC=CC.BrBr>CO.[Li+].[Cl-]>CC(Br)C(C)Br CC(Br)C(C)Br 1.0\n",
      "c1ccccc1C=C.c1ccccc1C=C>Cl>CC(c1ccccc1)C=Cc2ccccc2 C=Cc1ccccc1=CC(C)c1ccccc1 0\n",
      "data/problems/8-48.txt\t2.38319761784\t0.857142857143\t6\t1\n",
      "---------------\n",
      "data/problems/8-47.txt\t1.07091126624\t0.8975\t13\t1\n",
      "data/problems/9-34.txt\t8.38281387964\t0.659456509877\t3\t0\n",
      "data/problems/17-44.txt\t10.7509029969\t0.548331431676\t2\t1\n",
      "data/problems/11-41.txt\t8.87203523499\t0.58944469882\t5\t2\n",
      "data/problems/18-56.txt\t5.64648598101\t0.765824584312\t7\t2\n",
      "data/problems/21-50.txt\t8.0618322567\t0.710552446038\t5\t2\n",
      "data/problems/15-30.txt\t21.4003211657\t0.279762354741\t0\t1\n",
      "data/problems/9-33.txt\t2.40399896926\t0.912946428571\t6\t0\n",
      "data/problems/10-38.txt\t10.7982503688\t0.715433671443\t6\t1\n",
      "data/problems/8-48.txt\t2.38319761784\t0.857142857143\t6\t1\n"
     ]
    }
   ],
   "source": [
    "loss_list = [[] for _ in range(10)]\n",
    "sim_list = [[] for _ in range(10)]\n",
    "products_list = [[] for _ in range(10)]\n",
    "\n",
    "with tf.Session() as sess:\n",
    "    model = create_model(sess, True)\n",
    "    \n",
    "    for i, rxn_file in enumerate(glob.iglob('data/problems/*.txt')):\n",
    "        with open(rxn_file, 'r') as f:\n",
    "            rsmi_list = f.read().splitlines()\n",
    "        \n",
    "        for rsmi in rsmi_list: # list of RSMIs to be tested\n",
    "\n",
    "            reactant_list = []\n",
    "            agent_list = []\n",
    "            product_list = []\n",
    "\n",
    "            split_rsmi = rsmi.split('>')\n",
    "            reactants = cano(split_rsmi[0]).split('.')\n",
    "            agents = cano(split_rsmi[1]).split('.')\n",
    "            products = cano(split_rsmi[2]).split('.')\n",
    "\n",
    "            for reactant in reactants:\n",
    "                reactant_list += Smipar.parser_list(reactant)\n",
    "                reactant_list += '.'\n",
    "            for agent in agents:\n",
    "                agent_list += Smipar.parser_list(agent)\n",
    "                agent_list += '.'\n",
    "            for product in products:\n",
    "                product_list += Smipar.parser_list(product)\n",
    "                product_list += '.'\n",
    "\n",
    "            reactant_list.pop() # to pop last '.'\n",
    "            agent_list.pop()\n",
    "            product_list.pop()\n",
    "\n",
    "            products.append(data_utils.EOS_ID)\n",
    "\n",
    "            reactant_list += '>'\n",
    "            reactant_list += agent_list\n",
    "\n",
    "            token_ids = [reactants_token_list.index(r) for r in reactant_list]\n",
    "            product_ids = [products_token_list.index(p) for p in product_list]\n",
    "\n",
    "            bucket_id = min([b for b in xrange(len(_buckets)) if _buckets[b][0] > len(token_ids)])\n",
    "            encoder_inputs, decoder_inputs, target_weights = model.get_batch( \\\n",
    "                                       {bucket_id: [(token_ids, product_ids)]}, bucket_id)\n",
    "            _, loss, output_logits = model.step(sess, encoder_inputs, decoder_inputs,\n",
    "                                           target_weights, bucket_id, True)\n",
    "\n",
    "            outputs = [int(np.argmax(logit, axis=1)) for logit in output_logits]\n",
    "            if data_utils.EOS_ID in outputs:\n",
    "                outputs = outputs[:outputs.index(data_utils.EOS_ID)]\n",
    "            products_ = ''.join([tf.compat.as_str(products_token_list[output]) for output in outputs])\n",
    "\n",
    "            loss_list[i].append(loss)\n",
    "            sim_list[i].append(fp_sim(split_rsmi[2], products_))\n",
    "            products_list[i].append(products_)\n",
    "            \n",
    "            print(rsmi, products_list[i][-1], sim_list[i][-1])\n",
    "            \n",
    "        print(rxn_file, avg(loss_list[i]), avg(sim_list[i]), sim_list[i].count(1), sim_list[i].count(0), sep='\\t')\n",
    "        print('-' * 15)\n",
    "        \n",
    "    for i, rxn_file in enumerate(glob.iglob('data/problems/*.txt')):\n",
    "        print(rxn_file, avg(loss_list[i]), avg(sim_list[i]), sim_list[i].count(1), sim_list[i].count(0), sep='\\t')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "number, average loss(cross-entropy), average Tanimoto similarity, right, error"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}