Model - Phase II.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "import pickle as pkl\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "from keras.datasets import imdb\n",
    "\n",
    "\n",
    "from sklearn.metrics import confusion_matrix\n",
    "from sklearn.utils.multiclass import unique_labels\n",
    "from sklearn.metrics import f1_score\n",
    "\n",
    "from sklearn.preprocessing import LabelEncoder, OneHotEncoder\n",
    "from ast import literal_eval\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.metrics import confusion_matrix, f1_score\n",
    "from keras.models import Sequential, Model\n",
    "from keras.layers import Dense, Activation, Dropout, Conv1D, Flatten, MaxPooling1D,\\\n",
    "                        AveragePooling1D, Concatenate, LeakyReLU, Embedding, GlobalMaxPooling1D\n",
    "from keras.utils.vis_utils import plot_model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Get postive samples"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "drugbank = pkl.load(open('data/drugbank_embds.pkl', 'rb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "drugbank['target_gene_code'] = drugbank['target_gene'].apply(lambda x: ''.join(list(x.split('\\n')[1:])))\n",
    "drugbank['target_amino_code'] = drugbank['target_amino'].apply(lambda x: ''.join(list(x.split('\\n')[1:])))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open ('data/chem_cid_SMILE_fngr_vector_fngr.pickle', 'rb') as f:\n",
    "    smiles_df = pkl.load(f)\n",
    "    \n",
    "smiles_df['cid'] = smiles_df['cid'].astype(int)\n",
    "drugbank = pd.merge(smiles_df, drugbank, right_on = 'pubchem_id', left_on = 'cid', how = 'right')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "drugbank = drugbank[['drug_id', 'target_id', 'fingerprint', 'target_gene_code']]\n",
    "drugbank = drugbank[~ drugbank['fingerprint'].isna()]\n",
    "drugbank = drugbank.rename(columns = {'fingerprint': 'drug_fingerprint_code', 'target_id': 'gene_id'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>drug_id</th>\n",
       "      <th>gene_id</th>\n",
       "      <th>drug_fingerprint_code</th>\n",
       "      <th>target_gene_code</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>DB03313</td>\n",
       "      <td>BE0001349</td>\n",
       "      <td>2246728737 3545365497 2246699815 2664995851 86...</td>\n",
       "      <td>ATGGTCTCAGGAACGGTGGGCAGAGGTACGGCGCTGGGCGCGGTGC...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>DB03313</td>\n",
       "      <td>BE0004335</td>\n",
       "      <td>2246728737 3545365497 2246699815 2664995851 86...</td>\n",
       "      <td>ATGCAAAATCAATTAAATGAATTAAAACGAAAAATGCTGGAATTTT...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>DB03313</td>\n",
       "      <td>BE0004290</td>\n",
       "      <td>2246728737 3545365497 2246699815 2664995851 86...</td>\n",
       "      <td>ATGAACAAACCAACGATTCTGCGCCTAATCAAGTATCTGAGCATTA...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>DB03313</td>\n",
       "      <td>BE0003624</td>\n",
       "      <td>2246728737 3545365497 2246699815 2664995851 86...</td>\n",
       "      <td>ATGAGACTGATTTGTATGAGAAAATTTAACAGCCATTCGATTCCGA...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>DB03313</td>\n",
       "      <td>BE0004644</td>\n",
       "      <td>2246728737 3545365497 2246699815 2664995851 86...</td>\n",
       "      <td>ATGAAATTAGATAAATTATTTGAGAAATTTCTTTCTCTTTTTAAAA...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14071</th>\n",
       "      <td>DB02843</td>\n",
       "      <td>BE0001369</td>\n",
       "      <td>2245384272 3153477100 2976033787 1916236386 29...</td>\n",
       "      <td>ATGAGCACTGTAAAAGCACCGACGCTGCCCGCCAGCATCTTCCGCG...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14072</th>\n",
       "      <td>DB02843</td>\n",
       "      <td>BE0000916</td>\n",
       "      <td>2245384272 3153477100 2976033787 1916236386 29...</td>\n",
       "      <td>ATGTCCCGGCCCCTGTCAGACCAAGAGAAAAGAAAGCAAATCAGTG...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14073</th>\n",
       "      <td>DB02843</td>\n",
       "      <td>BE0002493</td>\n",
       "      <td>2245384272 3153477100 2976033787 1916236386 29...</td>\n",
       "      <td>ATGAAACGCAAGGGCATCATCCTCGCCGGAGGCTCGGGCACCCGCC...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14074</th>\n",
       "      <td>DB02843</td>\n",
       "      <td>BE0001698</td>\n",
       "      <td>2245384272 3153477100 2976033787 1916236386 29...</td>\n",
       "      <td>ATGTCACAACCTATTTTTAACGATAAGCAATTTCAGGAAGCGCTTT...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14075</th>\n",
       "      <td>DB02843</td>\n",
       "      <td>BE0001703</td>\n",
       "      <td>2245384272 3153477100 2976033787 1916236386 29...</td>\n",
       "      <td>ATGAACAAAACGCTAATCGCCGCAGCTGTGGCAGGGATAGTTTTAC...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>14076 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       drug_id    gene_id                              drug_fingerprint_code  \\\n",
       "0      DB03313  BE0001349  2246728737 3545365497 2246699815 2664995851 86...   \n",
       "1      DB03313  BE0004335  2246728737 3545365497 2246699815 2664995851 86...   \n",
       "2      DB03313  BE0004290  2246728737 3545365497 2246699815 2664995851 86...   \n",
       "3      DB03313  BE0003624  2246728737 3545365497 2246699815 2664995851 86...   \n",
       "4      DB03313  BE0004644  2246728737 3545365497 2246699815 2664995851 86...   \n",
       "...        ...        ...                                                ...   \n",
       "14071  DB02843  BE0001369  2245384272 3153477100 2976033787 1916236386 29...   \n",
       "14072  DB02843  BE0000916  2245384272 3153477100 2976033787 1916236386 29...   \n",
       "14073  DB02843  BE0002493  2245384272 3153477100 2976033787 1916236386 29...   \n",
       "14074  DB02843  BE0001698  2245384272 3153477100 2976033787 1916236386 29...   \n",
       "14075  DB02843  BE0001703  2245384272 3153477100 2976033787 1916236386 29...   \n",
       "\n",
       "                                        target_gene_code  \n",
       "0      ATGGTCTCAGGAACGGTGGGCAGAGGTACGGCGCTGGGCGCGGTGC...  \n",
       "1      ATGCAAAATCAATTAAATGAATTAAAACGAAAAATGCTGGAATTTT...  \n",
       "2      ATGAACAAACCAACGATTCTGCGCCTAATCAAGTATCTGAGCATTA...  \n",
       "3      ATGAGACTGATTTGTATGAGAAAATTTAACAGCCATTCGATTCCGA...  \n",
       "4      ATGAAATTAGATAAATTATTTGAGAAATTTCTTTCTCTTTTTAAAA...  \n",
       "...                                                  ...  \n",
       "14071  ATGAGCACTGTAAAAGCACCGACGCTGCCCGCCAGCATCTTCCGCG...  \n",
       "14072  ATGTCCCGGCCCCTGTCAGACCAAGAGAAAAGAAAGCAAATCAGTG...  \n",
       "14073  ATGAAACGCAAGGGCATCATCCTCGCCGGAGGCTCGGGCACCCGCC...  \n",
       "14074  ATGTCACAACCTATTTTTAACGATAAGCAATTTCAGGAAGCGCTTT...  \n",
       "14075  ATGAACAAAACGCTAATCGCCGCAGCTGTGGCAGGGATAGTTTTAC...  \n",
       "\n",
       "[14076 rows x 4 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "drugbank"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Get DrugID -> Finderprint mapping from positive samples"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "drug_map = drugbank[['drug_id', 'drug_fingerprint_code']].drop_duplicates('drug_id').set_index('drug_id')\n",
    "target_map = drugbank[['gene_id', 'target_gene_code']].drop_duplicates('gene_id').set_index('gene_id')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "negative_df = pd.read_csv('negative_samples/negative_samples_thrsh:12')[['node_0','node_1','shortest_path']] \n",
    "negative_df = negative_df.rename(columns = {'node_0': 'drug_id', 'node_1': 'gene_id'})\n",
    "negative_df = negative_df[negative_df['drug_id'].str.contains('DB') & negative_df['gene_id'].str.contains('BE')]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "negative_df = pd.merge(negative_df, drug_map, left_on = 'drug_id', right_index = True, how = 'left')\n",
    "negative_df = pd.merge(negative_df, target_map, left_on = 'gene_id', right_index = True, how = 'left')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>drug_id</th>\n",
       "      <th>gene_id</th>\n",
       "      <th>shortest_path</th>\n",
       "      <th>drug_fingerprint_code</th>\n",
       "      <th>target_gene_code</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>DB04481</td>\n",
       "      <td>BE0002266</td>\n",
       "      <td>12.0</td>\n",
       "      <td>2245384272 4022716898 2245273601 1735424888 22...</td>\n",
       "      <td>ATGGCGACCTCCACGGGTCGCTGGCTTCTCCTCCGGCTTGCACTAT...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>DB04481</td>\n",
       "      <td>BE0000206</td>\n",
       "      <td>13.0</td>\n",
       "      <td>2245384272 4022716898 2245273601 1735424888 22...</td>\n",
       "      <td>ATGTTTGACTGTATGGATGTTCTGTCAGTGAGTCCTGGGCAAATCC...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>DB04481</td>\n",
       "      <td>BE0002401</td>\n",
       "      <td>13.0</td>\n",
       "      <td>2245384272 4022716898 2245273601 1735424888 22...</td>\n",
       "      <td>ATGGCAGCAGACCTGGGCCCCTGGAATGACACCATCAATGGCACCT...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>DB04481</td>\n",
       "      <td>BE0004180</td>\n",
       "      <td>13.0</td>\n",
       "      <td>2245384272 4022716898 2245273601 1735424888 22...</td>\n",
       "      <td>ATGACGGAAAACAAAGGATCTTCTCAGCCTAAGAAAAACGGTAATA...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>DB04481</td>\n",
       "      <td>BE0002840</td>\n",
       "      <td>12.0</td>\n",
       "      <td>2245384272 4022716898 2245273601 1735424888 22...</td>\n",
       "      <td>ATGAAATTTCTTCTGGACATCCTCCTGCTTCTCCCGTTACTGATCG...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    drug_id    gene_id  shortest_path  \\\n",
       "1   DB04481  BE0002266           12.0   \n",
       "3   DB04481  BE0000206           13.0   \n",
       "4   DB04481  BE0002401           13.0   \n",
       "6   DB04481  BE0004180           13.0   \n",
       "10  DB04481  BE0002840           12.0   \n",
       "\n",
       "                                drug_fingerprint_code  \\\n",
       "1   2245384272 4022716898 2245273601 1735424888 22...   \n",
       "3   2245384272 4022716898 2245273601 1735424888 22...   \n",
       "4   2245384272 4022716898 2245273601 1735424888 22...   \n",
       "6   2245384272 4022716898 2245273601 1735424888 22...   \n",
       "10  2245384272 4022716898 2245273601 1735424888 22...   \n",
       "\n",
       "                                     target_gene_code  \n",
       "1   ATGGCGACCTCCACGGGTCGCTGGCTTCTCCTCCGGCTTGCACTAT...  \n",
       "3   ATGTTTGACTGTATGGATGTTCTGTCAGTGAGTCCTGGGCAAATCC...  \n",
       "4   ATGGCAGCAGACCTGGGCCCCTGGAATGACACCATCAATGGCACCT...  \n",
       "6   ATGACGGAAAACAAAGGATCTTCTCAGCCTAAGAAAAACGGTAATA...  \n",
       "10  ATGAAATTTCTTCTGGACATCCTCCTGCTTCTCCCGTTACTGATCG...  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "negative_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Note: Around 13% of negative samples don't match a drug id. Since the number is small, we ignore it for now"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "len(negative_df[(negative_df['target_gene_code'].isna()) |\n",
    "                (negative_df['drug_fingerprint_code'].isna())])/len(negative_df)\n",
    "\n",
    "negative_df = negative_df[(~ negative_df['target_gene_code'].isna()) &\n",
    "                (~ negative_df['drug_fingerprint_code'].isna())]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/omai.r/.local/lib/python3.5/site-packages/pandas/core/frame.py:7138: FutureWarning: Sorting because non-concatenation axis is not aligned. A future version\n",
      "of pandas will change to not sort by default.\n",
      "\n",
      "To accept the future behavior, pass 'sort=False'.\n",
      "\n",
      "To retain the current behavior and silence the warning, pass 'sort=True'.\n",
      "\n",
      "  sort=sort,\n"
     ]
    }
   ],
   "source": [
    "negative_df['Y'] = 0\n",
    "drugbank['Y'] = 1\n",
    "drugbank = drugbank.append(negative_df.sample(len(drugbank)))\n",
    "drugbank = drugbank.drop('shortest_path', axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Y</th>\n",
       "      <th>drug_fingerprint_code</th>\n",
       "      <th>drug_id</th>\n",
       "      <th>gene_id</th>\n",
       "      <th>target_gene_code</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [Y, drug_fingerprint_code, drug_id, gene_id, target_gene_code]\n",
       "Index: []"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "drugbank[drugbank['target_gene_code'].str.contains('S').isna()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "drugbank = drugbank[~ drugbank['drug_fingerprint_code'].str.contains('n')]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### The dataset is clean and balanced"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f7dd6d73198>"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "drugbank['Y'].hist()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Preprocessing fingerprints to feed to CNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LabelEncoder()"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "label_encoder = LabelEncoder()\n",
    "label_encoder.fit(list(drugbank['target_gene_code'].iloc[0]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "gene_chars = set(list(drugbank['target_gene_code'].iloc[0]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "drugbank['target_gene_encoded'] = drugbank['target_gene_code'].apply(\\\n",
    "                                   lambda x: label_encoder.transform([i for i in list(x) if i in gene_chars]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "drugbank['drug_fingerprint_encoded'] = drugbank['drug_fingerprint_code'].apply(lambda x: [int(i) for i in x.replace(' ', '')])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Y</th>\n",
       "      <th>drug_fingerprint_code</th>\n",
       "      <th>drug_id</th>\n",
       "      <th>gene_id</th>\n",
       "      <th>target_gene_code</th>\n",
       "      <th>target_gene_encoded</th>\n",
       "      <th>drug_fingerprint_encoded</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>2246728737 3545365497 2246699815 2664995851 86...</td>\n",
       "      <td>DB03313</td>\n",
       "      <td>BE0001349</td>\n",
       "      <td>ATGGTCTCAGGAACGGTGGGCAGAGGTACGGCGCTGGGCGCGGTGC...</td>\n",
       "      <td>[0, 3, 2, 2, 3, 1, 3, 1, 0, 2, 2, 0, 0, 1, 2, ...</td>\n",
       "      <td>[2, 2, 4, 6, 7, 2, 8, 7, 3, 7, 3, 5, 4, 5, 3, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>2246728737 3545365497 2246699815 2664995851 86...</td>\n",
       "      <td>DB03313</td>\n",
       "      <td>BE0004335</td>\n",
       "      <td>ATGCAAAATCAATTAAATGAATTAAAACGAAAAATGCTGGAATTTT...</td>\n",
       "      <td>[0, 3, 2, 1, 0, 0, 0, 0, 3, 1, 0, 0, 3, 3, 0, ...</td>\n",
       "      <td>[2, 2, 4, 6, 7, 2, 8, 7, 3, 7, 3, 5, 4, 5, 3, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>2246728737 3545365497 2246699815 2664995851 86...</td>\n",
       "      <td>DB03313</td>\n",
       "      <td>BE0004290</td>\n",
       "      <td>ATGAACAAACCAACGATTCTGCGCCTAATCAAGTATCTGAGCATTA...</td>\n",
       "      <td>[0, 3, 2, 0, 0, 1, 0, 0, 0, 1, 1, 0, 0, 1, 2, ...</td>\n",
       "      <td>[2, 2, 4, 6, 7, 2, 8, 7, 3, 7, 3, 5, 4, 5, 3, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>2246728737 3545365497 2246699815 2664995851 86...</td>\n",
       "      <td>DB03313</td>\n",
       "      <td>BE0003624</td>\n",
       "      <td>ATGAGACTGATTTGTATGAGAAAATTTAACAGCCATTCGATTCCGA...</td>\n",
       "      <td>[0, 3, 2, 0, 2, 0, 1, 3, 2, 0, 3, 3, 3, 2, 3, ...</td>\n",
       "      <td>[2, 2, 4, 6, 7, 2, 8, 7, 3, 7, 3, 5, 4, 5, 3, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>2246728737 3545365497 2246699815 2664995851 86...</td>\n",
       "      <td>DB03313</td>\n",
       "      <td>BE0004644</td>\n",
       "      <td>ATGAAATTAGATAAATTATTTGAGAAATTTCTTTCTCTTTTTAAAA...</td>\n",
       "      <td>[0, 3, 2, 0, 0, 0, 3, 3, 0, 2, 0, 3, 0, 0, 0, ...</td>\n",
       "      <td>[2, 2, 4, 6, 7, 2, 8, 7, 3, 7, 3, 5, 4, 5, 3, ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Y                              drug_fingerprint_code  drug_id    gene_id  \\\n",
       "0  1  2246728737 3545365497 2246699815 2664995851 86...  DB03313  BE0001349   \n",
       "1  1  2246728737 3545365497 2246699815 2664995851 86...  DB03313  BE0004335   \n",
       "2  1  2246728737 3545365497 2246699815 2664995851 86...  DB03313  BE0004290   \n",
       "3  1  2246728737 3545365497 2246699815 2664995851 86...  DB03313  BE0003624   \n",
       "4  1  2246728737 3545365497 2246699815 2664995851 86...  DB03313  BE0004644   \n",
       "\n",
       "                                    target_gene_code  \\\n",
       "0  ATGGTCTCAGGAACGGTGGGCAGAGGTACGGCGCTGGGCGCGGTGC...   \n",
       "1  ATGCAAAATCAATTAAATGAATTAAAACGAAAAATGCTGGAATTTT...   \n",
       "2  ATGAACAAACCAACGATTCTGCGCCTAATCAAGTATCTGAGCATTA...   \n",
       "3  ATGAGACTGATTTGTATGAGAAAATTTAACAGCCATTCGATTCCGA...   \n",
       "4  ATGAAATTAGATAAATTATTTGAGAAATTTCTTTCTCTTTTTAAAA...   \n",
       "\n",
       "                                 target_gene_encoded  \\\n",
       "0  [0, 3, 2, 2, 3, 1, 3, 1, 0, 2, 2, 0, 0, 1, 2, ...   \n",
       "1  [0, 3, 2, 1, 0, 0, 0, 0, 3, 1, 0, 0, 3, 3, 0, ...   \n",
       "2  [0, 3, 2, 0, 0, 1, 0, 0, 0, 1, 1, 0, 0, 1, 2, ...   \n",
       "3  [0, 3, 2, 0, 2, 0, 1, 3, 2, 0, 3, 3, 3, 2, 3, ...   \n",
       "4  [0, 3, 2, 0, 0, 0, 3, 3, 0, 2, 0, 3, 0, 0, 0, ...   \n",
       "\n",
       "                            drug_fingerprint_encoded  \n",
       "0  [2, 2, 4, 6, 7, 2, 8, 7, 3, 7, 3, 5, 4, 5, 3, ...  \n",
       "1  [2, 2, 4, 6, 7, 2, 8, 7, 3, 7, 3, 5, 4, 5, 3, ...  \n",
       "2  [2, 2, 4, 6, 7, 2, 8, 7, 3, 7, 3, 5, 4, 5, 3, ...  \n",
       "3  [2, 2, 4, 6, 7, 2, 8, 7, 3, 7, 3, 5, 4, 5, 3, ...  \n",
       "4  [2, 2, 4, 6, 7, 2, 8, 7, 3, 7, 3, 5, 4, 5, 3, ...  "
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "drugbank.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Calculate length mismatches and pad"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "drugbank['target_gene_encoded_len'] = drugbank['target_gene_encoded']\\\n",
    "                                    .apply(lambda x: len(x))\n",
    "drugbank['drug_fingerprint_encoded_len'] = drugbank['drug_fingerprint_encoded']\\\n",
    "                                    .apply(lambda x: len(x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "drugbank['mismatch'] = drugbank[['target_gene_encoded', 'drug_fingerprint_encoded']]\\\n",
    "                        .apply(lambda x: len(x['target_gene_encoded']) - len(x['drug_fingerprint_encoded']), axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f7dd67bbc18>"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "drugbank['mismatch'].hist(bins = 100, figsize = (10, 5))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "95 percent of the mismatches are of the range 22.998257 to 35.001743\n"
     ]
    }
   ],
   "source": [
    "print('95 percent of the mismatches are of the range %f to %f' % (max(drugbank['mismatch'].value_counts()) - np.std(drugbank['mismatch'].value_counts()),\n",
    "                                                           max(drugbank['mismatch'].value_counts()) + np.std(drugbank['mismatch'].value_counts())))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "max_target_gene_encoded_len = max(drugbank['target_gene_encoded_len'])\n",
    "max_drug_fingerprint_encoded_len = max(drugbank['drug_fingerprint_encoded_len'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "drugbank['target_gene_encoded_mismatch'] = max_target_gene_encoded_len - drugbank['target_gene_encoded_len']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "drugbank['drug_fingerprint_encoded_mismatch'] = max_drug_fingerprint_encoded_len - drugbank['drug_fingerprint_encoded_len'] "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f7dd69075c0>"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlkAAAEyCAYAAADJI8VDAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAE3FJREFUeJzt3X+snfd9F/D3pwkZXVxuxzZZkEQ4yGnBJEJdr0ijScjepsXFcztN1UhUjWYKsYqWMVAl5gBSK35oRmKMFbJOVpt1Q1VMFKYtTiJCGfMfSKUkXhFpFqJFXbbY6ppu6QwOg5Lx4Y/7pNylvvWJ7/nec8+9r5cU2c9zn3Oe7zkfP+e+83y/5/ut7g4AAPP1pkU3AABgJxKyAAAGELIAAAYQsgAABhCyAAAGELIAAAYQsgAABhCyAAAGELIAAAa4epEnr6qjSY6+5S1vuedtb3vb0HO98sorufbaa4eeg/lQq+WhVstDrZaHWm1/Z8+e/d3u/vbLHVfbYVmd1dXVfuqpp4ae48yZMzl48ODQczAfarU81Gp5qNXyUKvtr6rOdvfq5Y7TXQgAMICQBQAwwEJDVlUdraqTFy5cWGQzAADmbqEhq7tPd/exlZWVRTYDAGDudBcCAAwgZAEADGBMFgDAAMZkAQAMoLsQAGAAIQsAYIBtsXbh/v37F9kMAGCwfccf+7p9L5w4soCWbB1jsgAABtBdCAAwgJAFADCAkAUAMIDJSAEABjDwHQBggIVO4QAA7DyXmq5hNzImCwBgACELAGAA3YUAwELs9Fng3ckCABjAFA4AAAOYwgEAYABjsgCAbWOj6R+WcayWMVkAAAMIWQAAA+guBACumNndN+ZOFgDAAEIWAMAA5skCABjAPFkAAAPoLgQAGEDIAgAYQMgCABhAyAIAGEDIAgAYQMgCABjAsjoAwLZ3qeV7XjhxZAEtmZ07WQAAAwhZAAADWFYHAGAAy+oAAAyguxAAYAAhCwBgACELAGAAIQsAYAAhCwBgACELAGAAIQsAYAAhCwBgACELAGAAIQsAYAAhCwBgACELAGAAIQsAYAAhCwBgACELAGCAuYesqvrzVfWzVfVwVf2NeT8/AMAyuHqWg6rqgSTfl+Sl7r553f7DSX46yVVJPt7dJ7r72SQfrKo3JfmFJB+bf7MBgK227/hji27CUpn1TtYnkxxev6Oqrkpyf5J3JzmQ5M6qOjD97D1JHkvy+NxaCgCwRKq7Zzuwal+SR1+7k1VVtyX5SHffPm3flyTd/RPrHvNYdx/Z4PmOJTmWJHv37n3nqVOnrvxVzODixYvZs2fP0HMwH2q1PNRqeajV8tjOtXr6/IVFN+GPuOW6lYWc99ChQ2e7e/Vyx83UXbiB65K8uG77XJJbq+pgkh9I8k35BneyuvtkkpNJsrq62gcPHtxEUy7vzJkzGX0O5kOtlodaLQ+1Wh7buVZ3bbPuwhfef3DRTfiGNhOyLqm7zyQ5M+/nBQBYJpv5duH5JDes275+2jezqjpaVScvXNhetx8BADZrMyHrySQ3VdWNVXVNkjuSPPJGnqC7T3f3sZWVxfSpAgCMMusUDg8mOZjk26rqXJIPd/cnqureJE9kbQqHB7r7mWEtBQC2jOkaNm+mkNXdd26w//FsYpqGqjqa5Oj+/fuv9CkAALalhS6ro7sQANiprF0IADCAkAUAMMBCQ5YpHACAncqYLACAAXQXAgAMIGQBAAxgTBYAwADGZAEADKC7EABggJmW1QEA2G4utb7iCyeOLKAll+ZOFgDAAEIWAMAAvl0IADCAbxcCAAyguxAAYAAhCwBgACELAGAA82QBwC53qfmm2DzfLgQAGGChd7K6+3SS06urq/cssh0AsFu4a7V1jMkCABhAyAIAGEDIAgAYQMgCABhAyAIAGEDIAgAYwDxZAAADLDRkdffp7j62srKyyGYAAMyd7kIAgAGELACAAYQsAIABhCwAgAGELACAAYQsAIABhCwAgAGELACAAYQsAIABhCwAgAGsXQgAMIC1CwEABrh60Q0AAMbYd/yxRTdhVzMmCwBgACELAGAAIQsAYAAhCwBgACELAGAA3y4EgB3ANwm3H3eyAAAGELIAAAYQsgAABhCyAAAGELIAAAYQsgAABhCyAAAGGDJPVlV9f5IjSf5Ekk90978bcR4AgO1q5jtZVfVAVb1UVZ9/3f7DVfVcVT1fVceTpLt/qbvvSfLBJH91vk0GANj+3kh34SeTHF6/o6quSnJ/kncnOZDkzqo6sO6Qvz/9HABgV6nunv3gqn1JHu3um6ft25J8pLtvn7bvmw49Mf336e7+9xs817Ekx5Jk79697zx16tQVvoTZXLx4MXv27Bl6DuZDrZaHWi0PtVoes9Tq6fMXtqg1y+eW61aGn+PQoUNnu3v1csdtdkzWdUleXLd9LsmtSX40yfckWamq/d39s69/YHefTHIySVZXV/vgwYObbMo3dubMmYw+B/OhVstDrZaHWi2PWWp1l3UKN/TC+w8uuglfM2Tge3d/NMlHRzw3AMAy2OwUDueT3LBu+/pp30yq6mhVnbxwwW1PAGBn2WzIejLJTVV1Y1Vdk+SOJI/M+uDuPt3dx1ZWxvefAgBspTcyhcODST6T5O1Vda6q7u7uV5Pcm+SJJM8meai7nxnTVACA5THzmKzuvnOD/Y8neXxuLQIA2AEWuqyOMVkAwE610JBlTBYAsFMNmcIBAHajfZeYv+qFE0cW0BK2g4XeyQIA2KmMyQIAGMCYLACAAXQXAgAMIGQBAAxgTBYAwADGZAEADKC7EABgACELAGAAIQsAYICFLqtTVUeTHN2/f/8imwEA29alluphORj4DgAwgO5CAIABFtpdCAA73aW6+144cWQBLWGruZMFADCAkAUAMICQBQAwgLULAQAGMIUDAMAAugsBAAYQsgAABhCyAAAGELIAAAYQsgAABhCyAAAGME8WAMAAC10gurtPJzm9urp6zyLbAcDOtwwLNe87/lg+dMuruesSbWX56C4EABhAyAIAGGCh3YUAsAyWoauR7cedLACAAYQsAIABhCwAgAGELACAAYQsAIABhCwAgAEsqwMAMMBCQ1Z3n+7uYysrK4tsBgDA3OkuBAAYQMgCABhAyAIAGEDIAgAYQMgCABhAyAIAGODqRTcAAHabfccfW3QT2ALuZAEADCBkAQAMIGQBAAwgZAEADCBkAQAMIGQBAAwgZAEADCBkAQAMMPfJSKvqzyb5e0lWuvt9835+tsZGE+W9cOLIFrcEYHsyoSiXM9OdrKp6oKpeqqrPv27/4ap6rqqer6rjSdLdX+juu0c0FgBgWczaXfjJJIfX76iqq5Lcn+TdSQ4kubOqDsy1dQAAS6q6e7YDq/YlebS7b562b0vyke6+fdq+L0m6+yem7Ye/UXdhVR1LcixJ9u7d+85Tp05d+auYwcWLF7Nnz56h59hJnj5/4ZL7b7luZfi51Wp5qNXyUKtLf67N+pm20WfiCHvfnHzpD7bsdDvOVvyeOnTo0NnuXr3ccZsZk3VdkhfXbZ9LcmtVfWuSf5zkHVV132uh6/W6+2SSk0myurraBw8e3ERTLu/MmTMZfY6d5K6NxmS9/+Dwc6vV8lCr5aFWl/5cm/UzbaPPxBE+dMur+cmn5z5ketfYit9Ts5p7Fbv795J8cN7PCwCwTDYzhcP5JDes275+2jezqjpaVScvXNi627AAAFthMyHrySQ3VdWNVXVNkjuSPPJGnqC7T3f3sZWV8f2nAABbadYpHB5M8pkkb6+qc1V1d3e/muTeJE8keTbJQ939zLimAgAsj5nGZHX3nRvsfzzJ41d68qo6muTo/v37r/QpAAC2pYUuq6O7EADYqaxdCAAwgJAFADDAQkOWKRwAgJ3KmCwAgAF0FwIADCBkAQAMsNAVKM2TBewm+y61QPGJIwtoyXLwfrHsjMkCABhAdyEAwABCFgDAAEIWAMAAQhYAwAC75tuFT5+/kLte900V31IBWC5b8Y3DS50DroRvFwIADKC7EABgACELAGAAIQsAYAAhCwBggF3z7ULWWAtseagVW82/OZgv3y4EABhAdyEAwABCFgDAAEIWAMAAQhYAwABCFgDAAEIWAMAA5slakJ20kry5dZaX2rET+HfMdmWeLACAAXQXAgAMIGQBAAwgZAEADCBkAQAMIGQBAAwgZAEADCBkAQAMIGQBAAwgZAEADCBkAQAMsKvXLtxobT9rXsF4i1q/0/UNbBVrFwIADKC7EABgACELAGAAIQsAYAAhCwBgACELAGAAIQsAYAAhCwBgACELAGAAIQsAYAAhCwBgACELAGAAIQsAYAAhCwBgACELAGCAq+f9hFV1bZKfSfLVJGe6+1PzPgcAwHY3052sqnqgql6qqs+/bv/hqnquqp6vquPT7h9I8nB335PkPXNuLwDAUpi1u/CTSQ6v31FVVyW5P8m7kxxIcmdVHUhyfZIXp8P+cD7NBABYLtXdsx1YtS/Jo91987R9W5KPdPft0/Z906Hnknylux+tqlPdfccGz3csybEk2bt37ztPnTq1mddxWS+9fCFf+oPZjr3lupW5nvvp8xeW7rwbPXbWYzfzWi5evJg9e/Zc8eNnsdmabOY1z/rYeb+vs7blUjZq39435+uuq3m/D7Oa9bXMat7vdbI1Nd3ILNfVVvy7fiOP360udV0xu624pg4dOnS2u1cvd9xmQtb7khzu7r8+bf9QkluT/HiSf5nkfyX5j7OMyVpdXe2nnnpqpnZcqX/xqV/OTz492xC0F04cmeu59x1/bOnOu9FjZz12M6/lzJkzOXjw4BU/fhabrclmXvOsj533+zprWy5lo/Z96JZXv+66mvf7MKtZX8us5v1eJ1tT043Mcl1txb/rN/L43epS1xWz24prqqpmCllzr2J3v5Lkh+f9vAAAy2QzUzicT3LDuu3rp30zq6qjVXXywgW3iQGAnWUzIevJJDdV1Y1VdU2SO5I88kaeoLtPd/exlZWtGZMAALBVZp3C4cEkn0ny9qo6V1V3d/erSe5N8kSSZ5M81N3PjGsqAMDymGlMVnffucH+x5M8PtcWAQDsAAtdVseYLABgp1poyDImCwDYqSwQDQAwgJAFADCAMVkAAAMYkwUAMMDMaxcObUTVl5P81uDTfFuS3x18DuZDrZaHWi0PtVoearX9/Znu/vbLHbQtQtZWqKqnZlnMkcVTq+WhVstDrZaHWu0cBr4DAAwgZAEADLCbQtbJRTeAmanV8lCr5aFWy0OtdohdMyYLAGAr7aY7WQAAW0bIAgAYYFeErKo6XFXPVdXzVXV80e3Zbarqhqr61ar69ap6pqp+bNr/J6vq01X1G9Of3zLtr6r66FSv/1pV37HuuT4wHf8bVfWBRb2mna6qrqqqz1XVo9P2jVX12akm/7qqrpn2f9O0/fz0833rnuO+af9zVXX7Yl7JzlZVb62qh6vqv1XVs1V1m+tqe6qqvz19/n2+qh6sqj/uutr5dnzIqqqrktyf5N1JDiS5s6oOLLZVu86rST7U3QeSvCvJj0w1OJ7kV7r7piS/Mm0na7W6afrvWJKPJWuhLMmHk9ya5C8l+fBrv0CYux9L8uy67X+S5Ke6e3+SryS5e9p/d5KvTPt/ajouU33vSPIXkhxO8jPTtch8/XSSf9vdfy7JX8xazVxX20xVXZfkbyZZ7e6bk1yVtevDdbXD7fiQlbUPjee7+wvd/dUkp5K8d8Ft2lW6+4vd/WvT3/9H1n4RXJe1Ovz8dNjPJ/n+6e/vTfILveY/JXlrVf2pJLcn+XR3v9zdX0ny6ax90DBHVXV9kiNJPj5tV5LvSvLwdMjra/VaDR9O8t3T8e9Ncqq7/3d3/2aS57N2LTInVbWS5C8n+USSdPdXu/v347rarq5O8uaqujrJNyf5YlxXO95uCFnXJXlx3fa5aR8LMN32fkeSzybZ291fnH70O0n2Tn/fqGZquTX+eZK/k+T/TtvfmuT3u/vVaXv9+/61mkw/vzAdr1bj3Zjky0l+bura/XhVXRvX1bbT3eeT/NMkv521cHUhydm4rna83RCy2Caqak+Sf5Pkb3X3f1//s16bS8R8IgtWVd+X5KXuPrvotnBZVyf5jiQf6+53JHkl/79rMInraruYul/fm7Vg/KeTXBt3C3eF3RCyzie5Yd329dM+tlBV/bGsBaxPdfcvTru/NHVXZPrzpWn/RjVTy/G+M8l7quqFrHWtf1fWxv28dermSP7o+/61mkw/X0nye1GrrXAuybnu/uy0/XDWQpfravv5niS/2d1f7u7/k+QXs3atua52uN0Qsp5MctP0LY5rsjZo8JEFt2lXmcYSfCLJs939z9b96JEkr32T6QNJfnnd/r82fRvqXUkuTN0fTyT53qr6lun/DL932secdPd93X19d+/L2rXyH7r7/Ul+Ncn7psNeX6vXavi+6fie9t8xfUvqxqwNtv7PW/QydoXu/p0kL1bV26dd353k1+O62o5+O8m7quqbp8/D12rlutrhrr78Icutu1+tqnuz9qFxVZIHuvuZBTdrt/nOJD+U5Omq+i/Tvr+b5ESSh6rq7iS/leQHp589nuSvZG1Q5/9M8sNJ0t0vV9U/zFpwTpJ/0N0vb81L2PV+PMmpqvpHST6XabD19Oe/qqrnk7yctWCW7n6mqh7K2i+SV5P8SHf/4dY3e8f70SSfmv4H8gtZu1beFNfVttLdn62qh5P8Wtauh89lbemcx+K62tEsqwMAMMBu6C4EANhyQhYAwABCFgDAAEIWAMAAQhYAwABCFgDAAEIWAMAA/w8jdHraqnrY2QAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "drugbank['drug_fingerprint_encoded_mismatch'].hist(bins = 100, figsize = (10, 5), log = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f7dd632add8>"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "drugbank['target_gene_encoded_mismatch'].hist(bins = 100, figsize = (10, 5), log = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### There appears to be quite large intracolumn mismatches in length. Unsure how to deal with it now. Current strategy is to pad the inputs and model it"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "drugbank['target_gene_encoded_padded'] = drugbank[['target_gene_encoded','target_gene_encoded_mismatch']]\\\n",
    "                                         .apply(lambda x: ['-1' for i in range(int(x['target_gene_encoded_mismatch']/2))] +   \n",
    "                                                        list(x['target_gene_encoded']) +\n",
    "                                                        ['-1' for i in range (int(x['target_gene_encoded_mismatch']/2))]\n",
    "                                                        , axis = 1)\n",
    "drugbank['target_gene_encoded_padded'] = drugbank['target_gene_encoded_padded'].apply(lambda x: x + ['-1'] if len(x) < max_target_gene_encoded_len else x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "drugbank['drug_fingerprint_encoded_padded'] = drugbank[['drug_fingerprint_encoded','drug_fingerprint_encoded_mismatch']]\\\n",
    "                                         .apply(lambda x: ['-1' for i in range(int(x['drug_fingerprint_encoded_mismatch']/2))] +   \n",
    "                                                        list(x['drug_fingerprint_encoded']) +\n",
    "                                                        ['-1' for i in range (int(x['drug_fingerprint_encoded_mismatch']/2))]\n",
    "                                                        , axis = 1)\n",
    "\n",
    "drugbank['drug_fingerprint_encoded_padded'] = drugbank['drug_fingerprint_encoded_padded'].apply(lambda x: x + ['-1'] if len(x) < max_drug_fingerprint_encoded_len else x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "drugbank.to_csv('data/intermediate/interactions_preprocessed.csv', index = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "del smiles_df, negative_df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Building a CNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "drugbank_reader = pd.read_csv('data/intermediate/interactions_preprocessed.csv', chunksize = 7500, iterator = True, converters={\"target_gene_encoded_padded\": \\\n",
    "                                                                                     lambda x: np.array(x.strip(\"[]\").replace(\"'\",\"\").split(\", \")),\n",
    "                                                                                    \"drug_fingerprint_encoded_padded\": \\\n",
    "                                                                                    lambda x: np.array(x.strip(\"[]\").replace(\"'\",\"\").split(\", \"))})\n",
    "\n",
    "#drugbank = drugbank.sample(7500)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "drugbank = drugbank_reader.__next__().drop('Unnamed: 0', axis = 1)\n",
    "drugbank_test = drugbank_reader.__next__().drop('Unnamed: 0', axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "drugbank_reader = pd.read_csv('data/intermediate/interactions_preprocessed.csv', converters={\"target_gene_encoded_padded\": \\\n",
    "                                                                                     lambda x: np.array(x.strip(\"[]\").replace(\"'\",\"\").split(\", \")),\n",
    "                                                                                    \"drug_fingerprint_encoded_padded\": \\\n",
    "                                                                                    lambda x: np.array(x.strip(\"[]\").replace(\"'\",\"\").split(\", \"))})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "drugbank = drugbank_reader.sample(7500, replace = False)\n",
    "drugbank_test = drugbank_reader.sample(7500, replace = False)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['-1', '-1', '-1', ..., '-1', '-1', '-1'], dtype='<U2')"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "drugbank['target_gene_encoded_padded'][0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_0 = np.array(list(drugbank['target_gene_encoded_padded'])).astype(int)\n",
    "X_1 = np.array(list(drugbank['drug_fingerprint_encoded_padded'])).astype(int)\n",
    "Y   = np.array(list(drugbank['Y'])).astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_0_test = np.array(list(drugbank_test['target_gene_encoded_padded'])).astype(int)\n",
    "X_1_test = np.array(list(drugbank_test['drug_fingerprint_encoded_padded'])).astype(int)\n",
    "Y_test   = np.array(list(drugbank_test['Y'])).astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(7500, 21222)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_0.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(7500, 21222)"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_0_test.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Trimming the feature martrix to minimize padding. This trimming may result in loss of data from the largest SMILE and fingerprints, but are neccesary to reduce the amount of padding"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Trim the X of -1's. Might loose actual sequences\n",
    "padding_fractions = []\n",
    "padding_fraction = 0.9\n",
    "\n",
    "while padding_fraction > 0.5:\n",
    "    \n",
    "    padding_fraction = len(X_0[np.where(X_0 == -1)]) / (X_0.shape[0] * X_0.shape[1])\n",
    "    padding_fractions.append(padding_fraction)\n",
    "    trimming_factor = 0.1\n",
    "\n",
    "    X_0 = X_0[:,int(X_0.shape[1]*trimming_factor) : X_0.shape[1] - int(X_0.shape[1]*trimming_factor)]\n",
    "    X_1 = X_1[:,int(X_1.shape[1]*trimming_factor) : X_1.shape[1] - int(X_1.shape[1]*trimming_factor)]\n",
    "\n",
    "    X_0_test = X_0_test[:,int(X_0_test.shape[1]*trimming_factor) : X_0_test.shape[1] - int(X_0_test.shape[1]*trimming_factor)]\n",
    "    X_1_test = X_1_test[:,int(X_1_test.shape[1]*trimming_factor) : X_1_test.shape[1] - int(X_1_test.shape[1]*trimming_factor)]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f2f50d94eb8>]"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(padding_fractions)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(7500, 1828)"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_0.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(7500, 1828)"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_0_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_0 = np.hstack((X_0, X_1))\n",
    "X_0_test = np.hstack((X_0_test, X_1_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(7500, 2649)"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_0.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(7500, 2649)"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_0_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "del drugbank, drugbank_test, drugbank_reader"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 7500 samples, validate on 7500 samples\n",
      "Epoch 1/80\n",
      "7500/7500 [==============================] - 50s 7ms/step - loss: 2.9786 - acc: 0.5155 - val_loss: 0.7661 - val_acc: 0.5240\n",
      "Epoch 2/80\n",
      "7500/7500 [==============================] - 49s 6ms/step - loss: 0.9479 - acc: 0.5297 - val_loss: 0.6774 - val_acc: 0.5808\n",
      "Epoch 3/80\n",
      "7500/7500 [==============================] - 49s 6ms/step - loss: 0.7027 - acc: 0.5476 - val_loss: 0.6713 - val_acc: 0.5844\n",
      "Epoch 4/80\n",
      "7500/7500 [==============================] - 49s 6ms/step - loss: 0.6806 - acc: 0.5685 - val_loss: 0.6590 - val_acc: 0.6093\n",
      "Epoch 5/80\n",
      "7500/7500 [==============================] - 49s 7ms/step - loss: 0.6665 - acc: 0.5945 - val_loss: 0.6390 - val_acc: 0.6432\n",
      "Epoch 6/80\n",
      "7500/7500 [==============================] - 48s 6ms/step - loss: 0.6305 - acc: 0.6433 - val_loss: 0.5885 - val_acc: 0.7029\n",
      "Epoch 7/80\n",
      "7500/7500 [==============================] - 49s 6ms/step - loss: 0.5766 - acc: 0.6939 - val_loss: 0.5256 - val_acc: 0.7347\n",
      "Epoch 8/80\n",
      "7500/7500 [==============================] - 48s 6ms/step - loss: 0.4775 - acc: 0.7688 - val_loss: 0.5305 - val_acc: 0.7400\n",
      "Epoch 9/80\n",
      "7500/7500 [==============================] - 48s 6ms/step - loss: 0.4156 - acc: 0.8139 - val_loss: 0.4231 - val_acc: 0.8071\n",
      "Epoch 10/80\n",
      "7500/7500 [==============================] - 48s 6ms/step - loss: 0.3495 - acc: 0.8500 - val_loss: 0.5096 - val_acc: 0.7735\n",
      "Epoch 11/80\n",
      "7500/7500 [==============================] - 48s 6ms/step - loss: 0.3633 - acc: 0.8432 - val_loss: 0.4357 - val_acc: 0.8200\n",
      "Epoch 12/80\n",
      "7500/7500 [==============================] - 47s 6ms/step - loss: 0.3389 - acc: 0.8553 - val_loss: 0.4164 - val_acc: 0.8255\n",
      "Epoch 13/80\n",
      "7500/7500 [==============================] - 47s 6ms/step - loss: 0.2881 - acc: 0.8813 - val_loss: 0.3990 - val_acc: 0.8284\n",
      "Epoch 14/80\n",
      "7500/7500 [==============================] - 45s 6ms/step - loss: 0.2175 - acc: 0.9187 - val_loss: 0.4109 - val_acc: 0.8436\n",
      "Epoch 15/80\n",
      "7500/7500 [==============================] - 45s 6ms/step - loss: 0.2056 - acc: 0.9165 - val_loss: 0.4123 - val_acc: 0.8421\n",
      "Epoch 16/80\n",
      "7500/7500 [==============================] - 44s 6ms/step - loss: 0.2174 - acc: 0.9189 - val_loss: 0.4241 - val_acc: 0.8267\n",
      "Epoch 17/80\n",
      "7500/7500 [==============================] - 45s 6ms/step - loss: 0.2143 - acc: 0.9197 - val_loss: 0.4782 - val_acc: 0.8468\n",
      "Epoch 18/80\n",
      "7500/7500 [==============================] - 45s 6ms/step - loss: 0.1474 - acc: 0.9464 - val_loss: 0.4108 - val_acc: 0.8628\n",
      "Epoch 19/80\n",
      "7500/7500 [==============================] - 44s 6ms/step - loss: 0.1136 - acc: 0.9595 - val_loss: 0.5138 - val_acc: 0.8539\n",
      "Epoch 20/80\n",
      "7500/7500 [==============================] - 45s 6ms/step - loss: 0.1239 - acc: 0.9555 - val_loss: 0.4819 - val_acc: 0.8357\n",
      "Epoch 21/80\n",
      "7500/7500 [==============================] - 44s 6ms/step - loss: 0.1464 - acc: 0.9443 - val_loss: 0.4959 - val_acc: 0.8473\n",
      "Epoch 22/80\n",
      "7500/7500 [==============================] - 45s 6ms/step - loss: 0.1034 - acc: 0.9623 - val_loss: 0.5317 - val_acc: 0.8380\n",
      "Epoch 23/80\n",
      "7500/7500 [==============================] - 45s 6ms/step - loss: 0.1638 - acc: 0.9448 - val_loss: 0.5111 - val_acc: 0.8341\n",
      "Epoch 24/80\n",
      "7500/7500 [==============================] - 45s 6ms/step - loss: 0.1391 - acc: 0.9491 - val_loss: 0.4863 - val_acc: 0.8564\n",
      "Epoch 25/80\n",
      "7500/7500 [==============================] - 45s 6ms/step - loss: 0.0903 - acc: 0.9673 - val_loss: 0.5202 - val_acc: 0.8556\n",
      "Epoch 26/80\n",
      "7500/7500 [==============================] - 45s 6ms/step - loss: 0.0845 - acc: 0.9701 - val_loss: 0.5643 - val_acc: 0.8648\n",
      "Epoch 27/80\n",
      "2000/7500 [=======>......................] - ETA: 26s - loss: 0.0653 - acc: 0.9795"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-77-bf47bb47e120>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m     42\u001b[0m           \u001b[0mbatch_size\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mbatch_size\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     43\u001b[0m           \u001b[0mepochs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mepochs\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 44\u001b[0;31m           validation_data=(X_0_test[..., np.newaxis], Y_test))\n\u001b[0m",
      "\u001b[0;32m/usr/local/lib/python3.5/dist-packages/keras/engine/training.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_freq, max_queue_size, workers, use_multiprocessing, **kwargs)\u001b[0m\n\u001b[1;32m   1176\u001b[0m                                         \u001b[0msteps_per_epoch\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0msteps_per_epoch\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1177\u001b[0m                                         \u001b[0mvalidation_steps\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mvalidation_steps\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1178\u001b[0;31m                                         validation_freq=validation_freq)\n\u001b[0m\u001b[1;32m   1179\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1180\u001b[0m     def evaluate(self,\n",
      "\u001b[0;32m/usr/local/lib/python3.5/dist-packages/keras/engine/training_arrays.py\u001b[0m in \u001b[0;36mfit_loop\u001b[0;34m(model, fit_function, fit_inputs, out_labels, batch_size, epochs, verbose, callbacks, val_function, val_inputs, shuffle, callback_metrics, initial_epoch, steps_per_epoch, validation_steps, validation_freq)\u001b[0m\n\u001b[1;32m    202\u001b[0m                     \u001b[0mins_batch\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mins_batch\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtoarray\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    203\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 204\u001b[0;31m                 \u001b[0mouts\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfit_function\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mins_batch\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    205\u001b[0m                 \u001b[0mouts\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mto_list\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mouts\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    206\u001b[0m                 \u001b[0;32mfor\u001b[0m \u001b[0ml\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mo\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mzip\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mout_labels\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mouts\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.5/dist-packages/keras/backend/tensorflow_backend.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, inputs)\u001b[0m\n\u001b[1;32m   2977\u001b[0m                     \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_legacy_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minputs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2978\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2979\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minputs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   2980\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2981\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mpy_any\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mis_tensor\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mx\u001b[0m \u001b[0;32min\u001b[0m \u001b[0minputs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.5/dist-packages/keras/backend/tensorflow_backend.py\u001b[0m in \u001b[0;36m_call\u001b[0;34m(self, inputs)\u001b[0m\n\u001b[1;32m   2935\u001b[0m             \u001b[0mfetched\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_callable_fn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0marray_vals\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mrun_metadata\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrun_metadata\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2936\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2937\u001b[0;31m             \u001b[0mfetched\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_callable_fn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0marray_vals\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   2938\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mfetched\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2939\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.5/dist-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1456\u001b[0m         ret = tf_session.TF_SessionRunCallable(self._session._session,\n\u001b[1;32m   1457\u001b[0m                                                \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_handle\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1458\u001b[0;31m                                                run_metadata_ptr)\n\u001b[0m\u001b[1;32m   1459\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mrun_metadata\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1460\u001b[0m           \u001b[0mproto_data\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtf_session\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mTF_GetBuffer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrun_metadata_ptr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "#embedding_dims = 16\n",
    "hidden_dims = 512\n",
    "hidden_dims_2 = 256\n",
    "hidden_dims_3 = 128\n",
    "batch_size = 500\n",
    "epochs = 80\n",
    "#maxlen = \n",
    "\n",
    "model = Sequential()\n",
    "#model.add(Embedding(X_0.shape[1],\n",
    "#                    embedding_dims,\n",
    "#                    input_length=maxlen))\n",
    "\n",
    "#model.add(Dropout(0.2))\n",
    "model.add(Conv1D(X_0.shape[0],\n",
    "                 X_0.shape[1],\n",
    "                 activation='relu',\n",
    "                 strides=10))\n",
    "\n",
    "model.add(GlobalMaxPooling1D())\n",
    "\n",
    "model.add(Dense(hidden_dims))\n",
    "model.add(Dropout(0.2))\n",
    "model.add(Activation('relu'))\n",
    "\n",
    "model.add(Dense(hidden_dims_2))\n",
    "model.add(Dropout(0.2))\n",
    "model.add(Activation('relu'))\n",
    "\n",
    "model.add(Dense(hidden_dims_3))\n",
    "model.add(Dropout(0.2))\n",
    "model.add(Activation('relu'))\n",
    "\n",
    "model.add(Dense(1))\n",
    "model.add(Activation('sigmoid'))\n",
    "\n",
    "model.compile(loss='binary_crossentropy',\n",
    "              optimizer='adam',\n",
    "              metrics=['accuracy'])\n",
    "\n",
    "model.fit(X_0[..., np.newaxis], Y,\n",
    "          batch_size=batch_size,\n",
    "          epochs=epochs,\n",
    "          validation_data=(X_0_test[..., np.newaxis], Y_test))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [],
   "source": [
    "Y_pred = model.predict(X_0_test[..., np.newaxis])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [],
   "source": [
    "Y_pred = [1 if i > 0.5 else 0 for i in Y_pred]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [],
   "source": [
    "tn, fp, fn, tp = confusion_matrix(Y_test, Y_pred).ravel()\n",
    "cm = confusion_matrix(Y_test, Y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_confusion_matrix(y_true, y_pred, classes,\n",
    "                          normalize=False,\n",
    "                          title=None,\n",
    "                          cmap=plt.cm.Blues):\n",
    "    \"\"\"\n",
    "    This function prints and plots the confusion matrix.\n",
    "    Normalization can be applied by setting `normalize=True`.\n",
    "    \n",
    "    NOTE: Taken from sklearn example\n",
    "    \"\"\"\n",
    "    if not title:\n",
    "        if normalize:\n",
    "            title = 'Normalized confusion matrix'\n",
    "        else:\n",
    "            title = 'Confusion matrix, without normalization'\n",
    "\n",
    "    # Compute confusion matrix\n",
    "    cm = confusion_matrix(y_true, y_pred)\n",
    "    # Only use the labels that appear in the data\n",
    "    classes = [0,1]\n",
    "    if normalize:\n",
    "        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
    "        print(\"Normalized confusion matrix\")\n",
    "    else:\n",
    "        print('Confusion matrix, without normalization')\n",
    "\n",
    "    print(cm)\n",
    "\n",
    "    fig, ax = plt.subplots()\n",
    "    im = ax.imshow(cm, interpolation='nearest', cmap=cmap)\n",
    "    ax.figure.colorbar(im, ax=ax)\n",
    "    # We want to show all ticks...\n",
    "    ax.set(xticks=np.arange(cm.shape[1]),\n",
    "           yticks=np.arange(cm.shape[0]),\n",
    "           # ... and label them with the respective list entries\n",
    "           xticklabels=classes, yticklabels=classes,\n",
    "           title=title,\n",
    "           ylabel='True label',\n",
    "           xlabel='Predicted label')\n",
    "\n",
    "    # Rotate the tick labels and set their alignment.\n",
    "    plt.setp(ax.get_xticklabels(), rotation=45, ha=\"right\",\n",
    "             rotation_mode=\"anchor\")\n",
    "\n",
    "    # Loop over data dimensions and create text annotations.\n",
    "    fmt = '.2f' if normalize else 'd'\n",
    "    thresh = cm.max() / 2.\n",
    "    for i in range(cm.shape[0]):\n",
    "        for j in range(cm.shape[1]):\n",
    "            ax.text(j, i, format(cm[i, j], fmt),\n",
    "                    ha=\"center\", va=\"center\",\n",
    "                    color=\"white\" if cm[i, j] > thresh else \"black\")\n",
    "    fig.tight_layout()\n",
    "    return ax"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion matrix, without normalization\n",
      "[[3328  390]\n",
      " [ 667 3115]]\n",
      "Normalized confusion matrix\n",
      "[[0.8951049  0.1048951 ]\n",
      " [0.17636171 0.82363829]]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUMAAAEYCAYAAADGepQzAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzt3XmcXuP9//HXe2aSSCQE0VQSRAmKfmUTW9UeodqEUkGtqaBUtbRfW39oS3WxtKWIytdWW4tKNURQu0SCWGJLEJKILEgQ2fP5/XGuiTuRmbkn933PzD3zfuZxHrnPdZbrOnPu+cx1nes65ygiMDNr6SoauwBmZk2Bg6GZGQ6GZmaAg6GZGeBgaGYGOBiamQFlHAwltZX0b0nzJP2jgP0cKenBYpatsUjaTdIbTSU/Sd0lhaSqhipTuZA0RdI+6fM5kv5WgjyukfTLYu+3uVKpxxlKOgL4GbA18CkwAbgoIp4scL9HAT8GdomIpQUXtImTFECPiJjc2GWpiaQpwA8j4qE03x14B2hV7HMk6QZgWkScV8z9NpRVf1ZF2N+xaX/fLMb+WqKS1gwl/Qy4ArgY6AxsAvwVGFiE3W8KvNkSAmE+XPsqHf9sW4iIKMkErAt8BhxayzptyILl+2m6AmiTlu0BTAPOAGYBM4Dj0rILgcXAkpTHEOAC4JacfXcHAqhK88cCb5PVTt8BjsxJfzJnu12AccC89P8uOcseBX4NPJX28yDQqYZjqy7/L3LKPwg4AHgT+Ag4J2f9fsAzwNy07pVA67Ts8XQs89PxHpaz//8FPgBurk5L22ye8uid5rsAs4E98jh3NwJnpM9dU96nrLLfilXyuxlYDixIZfxFzjk4BngPmAOcm+f5X+m8pLQAtgCGpnO/OOX17xqOI4CTgEnp53oVX7SGKoDzgHfT+bkJWHeV786QVO7Hc9KOA6YCH6d97wC8lPZ/ZU7emwOPAB+m4/470DFn+RRgn/T5AtJ3N533z3KmpcAFadlZwFtk371XgYNS+teBhcCytM3clH4D8JucPE8AJqfzNwLoks/PqqVMpQyGA9KJrKplnV8BY4CvABsCTwO/Tsv2SNv/CmhFFkQ+B9Zb9QtUw3z1l7cKWBv4BNgqLdsI2HbVXzpg/fQlPyptd3ia3yAtfzR9GbcE2qb5S2o4tury/79U/hPIgtGtQAdgW7LAsVlavw+wU8q3O/AacPqqgWA1+/8dWVBpS05wyvnyvwq0A0YBf8zz3B1PCjDAEemY78hZdm9OGXLzm0L6BV/lHFyXyrc9sAj4eh7nf8V5Wd3PgFV+0Ws4jgDuAzqStUpmAwNyjmMy8DWgPXA3cPMq5b6J7LvTNiftGmAtoD9ZAPpXKn9XsqC6e9rHFsC+6dxsSBZQr1jdz4pVvrs56/RMZe6V5g8l+6NWQfYHcT6wUS0/rxU/I2AvsqDcO5XpL8Dj+fysWspUymbyBsCcqL0ZeyTwq4iYFRGzyWp8R+UsX5KWL4mIkWR/9bZaw/IsB7aT1DYiZkTExNWs821gUkTcHBFLI+I24HXgOznr/F9EvBkRC4A7yb6wNVlCdn10CXA70An4U0R8mvJ/lSxAEBHPRcSYlO8U4Fpg9zyO6fyIWJTKs5KIuI7sF34s2R+Ac+vYX7XHgG9KqgC+Bfwe2DUt2z0tr48LI2JBRLwIvEg6Zuo+/8VwSUTMjYj3gP/yxfk6ErgsIt6OiM+As4HBqzSJL4iI+av8bH8dEQsj4kGyYHRbKv904AmgF0BETI6I0enczAYuo+7zuYKkDckC7Y8j4oW0z39ExPsRsTwi7iCrxfXLc5dHAsMj4vmIWJSOd+d0XbdaTT+rFqGUwfBDoFMd11u6kDVTqr2b0lbsY5Vg+jnZX/F6iYj5ZH9JTwJmSPqPpK3zKE91mbrmzH9Qj/J8GBHL0ufqX6iZOcsXVG8vaUtJ90n6QNInZNdZO9Wyb4DZEbGwjnWuA7YD/pJ+CeoUEW+R/aL3BHYjqzG8L2kr1iwY1vQzq+v8F0N98q4iu7Zdbepq9rfq+avpfHaWdLuk6el83kLd55O0bSvgn8CtEXF7TvrRkiZImitpLtl5zWufrHK86Q/Ah6z5d7vZKWUwfIasSTSolnXeJ+sIqbZJSlsT88mag9W+mrswIkZFxL5kNaTXyYJEXeWpLtP0NSxTfVxNVq4eEbEOcA6gOrapdSiApPZk1+GuBy6QtH49yvMYcAjZdcvpaf4YYD2yEQH1Ls9q1Hb+VzqfklY6n2uQVz55L2Xl4FZIHhen7b+RzucPqPt8VvsL2WWdFT3lkjYl+86eSnbZpiPwSs4+6yrrSscraW2y1ltDfLfLQsmCYUTMI7tedpWkQZLaSWolaX9Jv0+r3QacJ2lDSZ3S+resYZYTgG9J2kTSumTNAGDFX+mB6QuwiKy5vXw1+xgJbCnpCElVkg4DtiGrGZVaB7JfgM9SrfXkVZbPJLu+VR9/AsZHxA+B/5Bd7wJA0gWSHq1l28fIfvEeT/OPpvknc2q7q6pvGWs7/y8C20rqKWktsutqheS1urx/Kmmz9EfjYrLrosUandCB7Hs2T1JX4Of5bCTpRLLa95ERkfsdXZss4M1O6x1HVjOsNhPoJql1Dbu+DTgu/TzbkB3v2HRJxijx0JqIuJRsjOF5ZCdxKtkv1L/SKr8BxpP1xr0MPJ/S1iSv0cAdaV/PsXIAq0jleJ+sJ213vhxsiIgPgQPJerA/JOsRPTAi5qxJmerpTLLOik/JagB3rLL8AuDG1ET6fl07kzSQrBOr+jh/BvSWdGSa35isV7wmj5H9QlcHwyfJamqP17gF/JYsuM2VdGZdZaSW8x8Rb5J1sDxEdm1s1XGp1wPbpLz+Rf0NJ+sBf5xsdMFCsnGrxXIhWWfFPLI/RHfnud3hZEH+fUmfpemciHgVuJSsxTUT+AYrn79HgInAB5K+9H2NbDzjL4G7yEYrbA4MXpMDa65KPujamiZJE4C90x8AsxbPwdDMjDK+N9nMrJgcDM3McDA0MwOyQaZNhqrahlp3aOxiWD30+vomjV0Eq4d3353CnDlz8h3vmJfKdTaNWPqlG6BWKxbMHhURA4qZf7E0rWDYugNttqpz1Ig1IU+NvbKxi2D1sOuOfYu+z1i6IO/f24UTrsr3jpkG16SCoZmVI4HK/4qbg6GZFUZARWVjl6JgDoZmVjgV9TJko3AwNLMCuZlsZpZxzdDMWjzhmqGZWdZMds3QzMy9yWZm7kAxM4N0zdDNZDMz1wzNzNxMNjOrVuFmspm1dL432cwM3Ew2M6vm3mQzM1wzNDNDvh3PzCzjDhQzs+bRgVL+R2Bmja+6qVzXVOdutJakZyW9KGmipAtT+maSxkqaLOkOSa1Teps0Pzkt756zr7NT+huS9qsrbwdDMytM9fMM85nqtgjYKyK2B3oCAyTtBPwOuDwitgA+Boak9YcAH6f0y9N6SNoGGAxsCwwA/iqp1ra8g6GZFUhFC4aR+SzNtkpTAHsB/0zpNwKD0ueBaZ60fG9JSum3R8SiiHgHmAz0qy1vB0MzK1z+zeROksbnTEO/vCtVSpoAzAJGA28BcyNiaVplGtA1fe4KTAVIy+cBG+Smr2ab1XIHipkVLv/e5DkRUeub7CNiGdBTUkfgHmDrAkuXF9cMzawwKl4zOVdEzAX+C+wMdJRUXXnrBkxPn6cDG2fFUBWwLvBhbvpqtlktB0MzK1zxepM3TDVCJLUF9gVeIwuKh6TVjgHuTZ9HpHnS8kciIlL64NTbvBnQA3i2trzdTDazgql4d6BsBNyYen4rgDsj4j5JrwK3S/oN8AJwfVr/euBmSZOBj8h6kImIiZLuBF4FlgKnpOZ3jRwMzawg2VP/ixMMI+IloNdq0t9mNb3BEbEQOLSGfV0EXJRv3g6GZlYYpanMORiaWYFERUX5dz84GJpZwYp4zbDROBiaWcEcDM3MfM3QzAyEXDM0MwPcgWJmBr5maGbma4ZmZtVcMzSzFs8dKGZmiYOhmZlAFQ6GZmauGZqZgYOhmZk7UMzMVij/WOhgaGYFkpvJZmaA7002M8uUf8XQwXBNtGldxUPXn07r1lVUVVZyz0Mv8JtrRnL1+UfQe5tNEGLye7M44f/dzPwFizntB3tx7EE7s3TpcuZ8/BknXXgL7834GICLfjKQAbttR4XEI2Nf54zf/7ORj675W7hwIfvs+S0WL1rE0mVLOejgQ/jl+Rfy6H8f4exfnMniJYvp1asP11x3PVVVVUQEZ/z0J4x6YCTt2rZj2PU30Kt378Y+jCalOTSTS1q3lTRA0huSJks6q5R5NaRFi5cyYOif2fGwS9hx8G/pv8s29PtGd37xx7vZ8bBL6HfYb5n6wcecPHh3ACa8PpVdj/w9/Q77Lfc8/AIX/WQQADttvxk79/waO3z/YvocehF9tt2U3fr0aMxDaxHatGnDA6Mf4dnnX2Ts+Ak8OOoBnnn6aX54/DHc9PfbeW7CK2yy6abcctONAIx64H7emjyJV16bxJVXD+O0U09u5CNoWiTlPTVlJQuG6b2nVwH7A9sAh0vaplT5NbT5CxYD0KqqkqqqSiKCT+cvXLF8rTatyN5lDY+Pn8SChUsAePalKXTt3BGACGjTuhWtW1XRpnUVVVWVzProkwY+kpZHEu3btwdgyZIlLF2yhMrKSlq3bk2PLbcEYK999uVf99wFwH0j7uWIHxyNJHbcaSfmzZvLjBkzGq38TZGDYe36AZMj4u2IWAzcDgwsYX4NqqJCjLn9LN57+BIeGfM64155F4BrL/gBUx66mK26d+avtz/2pe2OHbQzo556FYCxL73D4+Mn8c7oi3jnwYt56OnXeOOdmQ16HC3VsmXL2LFPTzbp8hX22mdfdujXj6VLl/Lc+PEA3HPXP5k2dSoA778/nW7dNl6xbdeu3Xh/+vRGKXdT5WBYu67A1Jz5aSltJZKGShovaXwsXVDC4hTX8uXBToMvYYv9zqPvdpuyzeYbAXDiBbfwtf7n8vo7H3BI/z4rbTP4gB3ovc0mXH7jwwB8beNObLVZZ7bY7zw23+9c9ui3Jbv22rzBj6UlqqysZOxzE5g8ZRrjxz3LqxMnctMtt/OLM3/KN3fuR4cOHaisrGzsYpYNVSivqSlr9P7wiBgWEX0joq+q2jZ2cept3mcLeGz8m/Tf5YsrAMuXB/8Y9RyD9u65Im3PHbfif4fsxyGnX8viJUsBGLjn9jz78hTmL1jM/AWLGfXURHb8n80a/Bhaso4dO7L7Hnvy4IMPsNPOO/Pwo0/w5DPP8s3dvsUWqcncpUtXpk374u/69OnT6NL1S3/XWy65ZliX6cDGOfPdUlrZ67Ree9ZtnwXutdq0Yu8dt+bNd2fytY07rVjnwN3/hzenZE3e7bfqxpXnDuaQn17L7I8/W7HO1A8+Zrc+W1BZWUFVVQW79e7B6+980LAH0wLNnj2buXPnArBgwQIefmg0W221NbNmzQJg0aJFXPqH33HC0JMA+PZ3vsutt9xERDB2zBjWWWddNtpoo0Yrf1MjQMpvaspKObRmHNBD0mZkQXAwcEQJ82swX+20Dtf96igqKyqoqBB3jX6e+5+YyMPDT6fD2m2R4OU3p3PaxXcAcPFPB7F2uzb8/fdDgCwIHnr6tdz90AvsvsOWjL/zHIJg9NOvMfLxVxrz0FqED2bM4ITjj2HZsmUsj+V875Dvc8C3D+Ts//0594+8j+XLl3PC0JPZY8+9ABiw/wGMun8k2269Be3atuPav/1fIx9BU9P0a335UHWPZ0l2Lh0AXAFUAsMj4qLa1q9o95Vos9X3S1YeK76Px13Z2EWweth1x74899z4okautb66ZWxy9J/zWnfSH/Z/LiL61rRc0sbATUBnIIBhEfEnSRcAJwCz06rnRMTItM3ZwBBgGXBaRIxK6QOAP5HFn79FxCW1la2kg65TYUeWMg8za2TKRlcUyVLgjIh4XlIH4DlJo9OyyyPijytlnQ3XGwxsC3QBHpK0ZVp8FbAvWeftOEkjIuLVmjL2HShmVhBRvGAYETOAGenzp5JeYzWjUHIMBG6PiEXAO5Imkw3rgzS0D0BS9dC+GoNho/cmm1n5q0cHSqfqoXRpGlrzPtUd6AWMTUmnSnpJ0nBJ66W0mobw5TW0L5eDoZkVrB5Da+ZUD6VL07Aa9tceuAs4PSI+Aa4GNgd6ktUcLy32MbiZbGaFKfKwGUmtyALh3yPiboCImJmz/DrgvjRb2xC+eg3tc83QzAqSjTMszqBrZStdD7wWEZflpOcO7DwIqB6DNgIYLKlNGsbXA3iWnKF9klqTdbKMqC1v1wzNrEAqZm/yrsBRwMuSJqS0c8ge9NKTbLjNFOBEgIiYKOlOso6RpcApEbEMQNKpwCi+GNo3sbaMHQzNrGDFGnQdEU+y+kfF1jhEL41f/tIY5voO7XMwNLPClMGtdvlwMDSzglRfMyx3DoZmVrBmEAsdDM2scK4ZmpkV997kRuNgaGYFqX6eYblzMDSzAjWP5xk6GJpZwZpBLHQwNLPCuWZoZi2e3IFiZpZxzdDMDF8zNDMDXDM0M/ODGszMAORxhmZmmUr3JpuZuZlsZpZeA1r+0bDGYChpndo2TK/vMzOjGbSSa60ZTiR7+UruYVbPB7BJCctlZmWkWdcMI2LjmpaZmeVqBrEwv/cmSxos6Zz0uZukPqUtlpmVCwGVUl5TU1ZnMJR0JbAn2btMAT4HrillocysjOT5Avmm3pTOpzd5l4joLekFgIj4KL2h3swMaB7N5HyC4RJJFWSdJkjaAFhe0lKZWdkQUNEMomE+1wyvAu4CNpR0IfAk8LuSlsrMyoqU39SU1VkzjIibJD0H7JOSDo2IV0pbLDMrFy3t4a6VwBKypnJePdBm1nK0iGaypHOB24AuQDfgVklnl7pgZlY+lOdU536kjSX9V9KrkiZK+klKX1/SaEmT0v/rpXRJ+rOkyZJektQ7Z1/HpPUnSTqmrrzzqeUdDewQEedFxLlAP+DYPLYzsxaiiENrlgJnRMQ2wE7AKZK2Ac4CHo6IHsDDaR5gf6BHmoYCV6fyrA+cD+xIFrPOrw6gNcknGM5g5eZ0VUozM0u9yflNdYmIGRHxfPr8KfAa0BUYCNyYVrsRGJQ+DwRuiswYoKOkjYD9gNER8VFEfAyMBgbUlndtD2q4nOwa4UfAREmj0nx/YFzdh2VmLUL9BlR3kjQ+Z35YRAxb/W7VHegFjAU6R0R1JewDoHP63BWYmrPZtJRWU3qNautAqe4xngj8Jyd9TG07NLOWpx69yXMiom9dK0lqTzak7/SI+CQ32EZESIo1KmgtantQw/XFzszMmp/qZnLR9ie1IguEf4+Iu1PyTEkbRcSM1AyeldKnA7kPlemW0qYDe6yS/mht+ebTm7y5pNtTT82b1VM+B2VmLUOxOlCUrXQ98FpEXJazaARQ3SN8DHBvTvrRqVd5J2Beak6PAvpLWi91nPRPaTXKZ5zhDcBvgD+S9dwcR7o1z8wM8hs2k6ddyR4K87KkCSntHOAS4E5JQ4B3ge+nZSOBA4DJZA+ROQ5WPEPh13zRv/GriPiotozzCYbtImKUpD9GxFvAeekC6C/zPjwza7ak4g26jognqTm27r2a9QM4pYZ9DQeG55t3PsFwUXpQw1uSTiJri3fINwMza/6awQ0oeQXDnwJrA6cBFwHrAseXslBmVl5axL3JETE2ffyULx7wamYGZC+Rbw73Jtc26PoeaukoiYiDS1IiMysvZfB4rnzUVjO8ssFKkXxjq40Z+cilDZ2tFWC9/X7b2EWwelg06YOS7LepP9I/H7UNun64IQtiZuWrOTzXL9/nGZqZrZZo5jVDM7N8VTWDqmHewVBSm4hYVMrCmFn5yd5vUv41w3zuTe4n6WVgUprfXtJfSl4yMysbxXqeYWPKp3L7Z+BA4EOAiHiR7KXyZmZAC3k7HlAREe+uUg1eVqLymFmZaS7vTc4nGE6V1A8ISZXAjwE/wsvMVqgs/1iYVzA8maypvAkwE3gopZmZITXz2/GqRcQsYHADlMXMylQziIV1B0NJ17Gae5QjYmhJSmRmZaep9xTnI59m8kM5n9cCDmLlt06ZWQvWYjpQIuKO3HlJNwNPlqxEZlZ2mkEsXKPb8Tbji3eWmllLJ6hsBtEwn2uGH/PFNcMKspfKn1XKQplZ+Sj2q0IbS63BML22b3uy954ALE8vYDEzW6E5BMNab8dLgW9kRCxLkwOhmX1Jsd6b3JjyuTd5gqReJS+JmZWl6mZyuT+oobZ3oFRFxFKgFzBO0lvAfLJjj4jo3UBlNLOmrAwewpCP2q4ZPgv0Br7bQGUxszIkoKqpV/vyUFswFEBEvNVAZTGzMtXca4YbSvpZTQsj4rISlMfMyo6ooPyjYW3BsBJoD83gKM2sZLIXQjV2KQpXWzCcERG/arCSmFl5KmJPsaThZE/WnxUR26W0C4ATgNlptXMiYmRadjYwhOyB06dFxKiUPgD4E1ml7m8RcUldedd5zdDMrDYCKovXgXIDcCVw0yrpl0fEH1fKV9qG7PGC2wJdgIckbZkWXwXsC0wjGw0zIiJerS3j2oLh3nkX38xatGI9tSYiHpfUPc/VBwK3p7d2viNpMtAvLZscEW8DSLo9rVtrMKxx0HVEfJRngcyshavHC6E6SRqfM+X7XNRTJb0kabik9VJaV1Z+nOC0lFZTeq2awaufzawxiSyQ5DMBcyKib840LI8srgY2B3oCM4BLi3wIwJo9wsvM7Aslfol8RMxckVX25P370ux0YOOcVbvxxUNlakqvkWuGZlYw5Tmt0b6ljXJmDwJeSZ9HAIMltZG0GdCD7M65cUAPSZtJak3WyTKirnxcMzSzgojiPdxV0m3AHmTXFqcB5wN7SOpJ9lzVKcCJABExUdKdZB0jS4FTImJZ2s+pwCiyoTXDI2JiXXk7GJpZwYrVSo6Iw1eTfH0t618EXLSa9JHAyPrk7WBoZgVq+s8qzIeDoZkVpLo3udw5GJpZwVwzNDOjedy762BoZgVRS3lVqJlZXdxMNjPDzWQzM6D5P9zVzKxO2dCa8o+GDoZmVjDXDM3MUNEe7tqYHAzNrCBuJpuZQXqeYWMXonAOhmZWMAdDMzNAbiabWUtXzIe7NiYHQzMrWDOIhQ6GZlY4N5MNgHnz5vLz007mjdcnIsSlf7mWPv12Yviwv3Lj9ddQWVHJXv3357wLL+buf9zGNX+5fMW2r018mQceHcO239i+EY+g+WvTqpKHrvgBrVtVUlVZwT2Pv8FvbnyCkwb24dTv7cDmXdej20FX8OEnCwDYcuP1GfaLA+m5RWcuGP4YV/zj2RX7ev3vJ/Pp54tZtjxYumw53/zRDY10VE2DgIryj4WlC4aShgMHArMiYrtS5dMUnH/2Geyx974Mu/E2Fi9ezIIFn/PUE4/y4P3/5sHHx9GmTRvmzJ4FwMGHHs7Bh2aveXjt1Vf44Q8OdSBsAIuWLGPAGbcyf+ESqioreORPR/Hgs2/xzMRpjBwzmQcvO2Kl9T/+dCFnXDma7+zaY7X7G3DGrSsCp6lZ1AxL+bTuG4ABJdx/k/DJJ/MY+/STHH7UcQC0bt2addftyM3Dr+OUn5xJmzZtAOi04Ve+tO29d93Bdw8+tEHL25LNX7gEgFZVFVRVVRABL06eyXsz531p3dlzP+e5N2awZOnyhi5m+UnjDPOZmrKSBcOIeBz4qFT7byqmvjuF9TttyM9OPYH9dt+RM087ic/nz+fttyYx9pmnOHCf3fjegfsw4fnxX9r23/f8k4EHH9YIpW6ZKirEmGuP5727fsIjz73DuNffX6P9RMC/fz+Yp64+luO/3bPIpSw/1b3J+UxNWaO/x0XSUEnjJY3/cM7sxi5OvS1dupRXXnyBo44byqjHxtKu3dpcdcUfWLZ0KXPnfsy/Rz/OeRf+lpOPP5KIWLHd8+OfZa227dh6m20bsfQty/LlwU4nDmeLw66k79Zd2KZ7pzXaz96n38wuJ/0fg86+kxMH9mbXb2xc5JKWn1K+RL6hNHowjIhhEdE3Ivpu0GnDxi5OvW3UpSsbdelK7779APj2wIN4+aUJfLVLV/Y/cCCS6NVnByoqKvjowzkrthtx9z8Y9L3vN1axW7R58xfx2IR36b/D19Zo+/fnfAZkTekRT77JDltvVMziladmEA0bPRiWu690/ipdunbjrUlvAvDkY/+lx1ZfZ8C3v8vTTzwGwNuTJ7F48WLW3yCriSxfvpx/33uXrxc2oE7rtmXdtbPrt2u1rmLvPpvxxtT6X8Vpt1Yr2rdtveLzPn03Y+KUOXVs1fwpz39NmYfWFMGvf3c5Pz7xWBYvXsym3Tfj0iuH0a7d2pzx46HsvUtvWrVuzRV//duK90SMefoJunTpxqbd16xmYvX31Q3ac90vDqSysoIKibsee437x0zmRwf15WeH7Ujn9dsz7rohPPDsW/zo0vvpvN7aPHX1sXRo14blEZz6vR3odfx1bLBuO+648GAAqioruOPhVxk97u1GPrrG18QvB+ZFudexirpj6TZgD6ATMBM4PyKur22b7Xv1iZGPPF2S8lhpbHHIZY1dBKuHReOuYvkn04oaur7+jV5x072P5rVuv807PhcRfYuZf7GUrGYYEYeXat9m1nSI5vF2PF8zNLPCFHGcoaThkmZJeiUnbX1JoyVNSv+vl9Il6c+SJkt6SVLvnG2OSetPknRMPofhYGhmBStiZ/INfPlmjbOAhyOiB/BwmgfYH+iRpqHA1ZAFT+B8YEegH3B+dQCtjYOhmRWuSNGwhps1BgI3ps83AoNy0m+KzBigo6SNgP2A0RHxUUR8DIwmj7vh3JtsZgWq17CZTpJyb8caFhHD6timc0TMSJ8/ADqnz12BqTnrTUtpNaXXysHQzApSz6fWzCmkNzkiQlJJhsC4mWxmhSvtHSgzU/OX9P+slD4dyL0XsltKqym9Vg6GZlawEt+BMgKo7hE+Brg3J/3o1Ku8EzAvNadHAf0lrZc6TvqntFq5mWxmBSvWMMPcmzUkTSPrFb4EuFPSEOBdoPqm/pHAAcBk4HPgOICI+EjSr4Fxab1fRUSd9146GJpZwYo15LqWmzVDEVn1AAAHRElEQVT2Xs26AZxSw36GA8Prk7eDoZkVpgyeSJMPB0MzK0jWm1z+0dDB0MwKVv6h0MHQzIqhGURDB0MzK1hTf3BrPhwMzaxgzeCSoYOhmRWuGcRCB0MzK0xzebirg6GZFaYMXhCfDwdDMytYM4iFDoZmVgTNIBo6GJpZgZr+O5Hz4WBoZgWp58NdmywHQzMrnIOhmZnvQDEzAzy0xswMaBatZAdDMyuQB12bmfl2PDOzFco/FDoYmlkRNIOKoYOhmRXOQ2vMzKBZtJMdDM2sYM0gFjoYmllhJL8q1MwsU/6x0MHQzArXDGKhg6GZFa4ZtJKpaOwCmFm5U97/8tqbNEXSy5ImSBqf0taXNFrSpPT/eildkv4sabKklyT1XtOjcDA0s4Jkt+PlN9XDnhHRMyL6pvmzgIcjogfwcJoH2B/okaahwNVrehwOhmZWsBIEw1UNBG5Mn28EBuWk3xSZMUBHSRutSQYOhmZWsHo0kztJGp8zDV3N7gJ4UNJzOcs7R8SM9PkDoHP63BWYmrPttJRWb+5AMbPC1K/WNyen6VuTb0bEdElfAUZLej13YUSEpFiDktbKNUMzK4jqMeUjIqan/2cB9wD9gJnVzd/0/6y0+nRg45zNu6W0enMwNLPCFSkaSlpbUofqz0B/4BVgBHBMWu0Y4N70eQRwdOpV3gmYl9Ocrhc3k82sYEW8Ha8zcE96WGwVcGtEPCBpHHCnpCHAu8D30/ojgQOAycDnwHFrmrGDoZkVrFihMCLeBrZfTfqHwN6rSQ/glGLk7WBoZoVrBnegOBiaWcGaw8NdldUymwZJs8muBzQ3nYA5jV0Iq5fmes42jYgNi7lDSQ+Q/bzyMSciBhQz/2JpUsGwuZI0Po+xVdaE+Jy1PB5aY2aGg6GZGeBg2FCGNXYBrN58zloYXzM0M8M1QzMzwMHQzAxwMDQzAxwMS0rSVpJ2ltRKUmVjl8fy43PVMrkDpUQkHQxcTPZstenAeOCGiPikUQtmNZK0ZUS8mT5XRsSyxi6TNRzXDEtAUivgMGBIROxN9uy1jYH/lbROoxbOVkvSgcAESbcCRMQy1xBbFgfD0lmH7I1dkD2t9z6gFXCE1BzeMtt8pIeIngqcDiyWdAs4ILY0DoYlEBFLgMuAgyXtFhHLgSeBCcA3G7Vw9iURMR84HrgVOBNYKzcgNmbZrOE4GJbOE8CDwFGSvhURyyLiVqALq3l4pTWuiHg/Ij6LiDnAiUDb6oAoqbekrRu3hFZqfp5hiUTEQkl/J3vt4dnpl2kR2WPN1+gdDdYwIuJDSScCf0hvZqsE9mzkYlmJORiWUER8LOk64FWy2sZC4AcRMbNxS2Z1iYg5kl4C9gf2jYhpjV0mKy0PrWkg6UJ8pOuH1sRJWg+4EzgjIl5q7PJY6TkYmtVA0loRsbCxy2ENw8HQzAz3JpuZAQ6GZmaAg6GZGeBgaGYGOBiWFUnLJE2Q9Iqkf0hqV8C+9pB0X/r8XUln1bJuR0k/WoM8LpB0Zr7pq6xzg6RD6pFXd0mv1LeMZtUcDMvLgojoGRHbAYuBk3IXKlPvcxoRIyLiklpW6QjUOxialRMHw/L1BLBFqhG9Iekm4BVgY0n9JT0j6flUg2wPIGmApNclPQ8cXL0jScdKujJ97izpHkkvpmkX4BJg81Qr/UNa7+eSxkl6SdKFOfs6V9Kbkp4EtqrrICSdkPbzoqS7Vqnt7iNpfNrfgWn9Skl/yMn7xEJ/kGbgYFiWJFWR3Sb2ckrqAfw1IrYF5gPnAftERG+yh8r+TNJawHXAd4A+wFdr2P2fgcciYnugNzAROAt4K9VKfy6pf8qzH9AT6CPpW5L6AINT2gHADnkczt0RsUPK7zVgSM6y7imPbwPXpGMYAsyLiB3S/k+QtFke+ZjVyvcml5e2kiakz08A15M9BefdiBiT0ncCtgGeSo9NbA08A2wNvBMRkwDSE1mGriaPvYCjYcXjq+alW9Ny9U/TC2m+PVlw7ADcExGfpzxG5HFM20n6DVlTvD0wKmfZnen2xUmS3k7H0B/4n5zrieumvN/MIy+zGjkYlpcFEdEzNyEFvPm5ScDoiDh8lfVW2q5AAn4bEdeuksfpa7CvG4BBEfGipGOBPXKWrXp7VKS8fxwRuUETSd3XIG+zFdxMbn7GALtK2gKypzhL2hJ4HeguafO03uE1bP8wcHLatlLSusCnZLW+aqOA43OuRXaV9BXgcWCQpLaSOpA1yevSAZiRXpVw5CrLDpVUkcr8NeCNlPfJaX0kbZmeVG1WENcMm5mImJ1qWLdJapOSz4uINyUNBf4j6XOyZnaH1eziJ8AwSUOAZcDJEfGMpKfS0JX703XDrwPPpJrpZ2SPJnte0h3Ai8AsYFweRf4lMBaYnf7PLdN7wLNkr1A4KT0j8m9k1xKfV5b5bGBQfj8ds5r5QQ1mZriZbGYGOBiamQEOhmZmgIOhmRngYGhmBjgYmpkBDoZmZgD8f27qWlX7cCWCAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot non-normalized confusion matrix\n",
    "plot_confusion_matrix(Y_test, Y_pred, classes=[0, 1],\n",
    "                      title='Confusion matrix, without normalization')\n",
    "\n",
    "# Plot normalized confusion matrix\n",
    "plot_confusion_matrix(Y_test, Y_pred, classes=[0, 1], normalize=True,\n",
    "                      title='Normalized confusion matrix')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8549471661863592"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f1_score(Y_test, Y_pred)"
   ]
  },
  {
   "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": "3.5.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}